The Historian of Islam at Work. Essays in Honor of Hugh N. Kennedy 9789004525238, 9789004525245

Citation preview

The Historian of Islam at Work

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Islamic History and Civilization studies and texts

Editorial Board Hinrich Biesterfeldt Sebastian Günther

Honorary Editor Wadad Kadi

volume 198

The titles published in this series are listed at brill.com/ihc

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

picture taken by shawkat m. toorawa

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

The Historian of Islam at Work Essays in Honor of Hugh N. Kennedy

Edited by

Maaike van Berkel Letizia Osti

leiden | boston

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

The Library of Congress Cataloging-in-Publication Data is available online at https://catalog.loc.gov lc record available at https://lccn.loc.gov/2022039036

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill‑typeface. issn 0929-2403 isbn 978-90-04-52523-8 (hardback) isbn 978-90-04-52524-5 (e-book) Copyright 2022 by Maaike van Berkel and Letizia Osti. Published by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Brill Hotei, Brill Schöningh, Brill Fink, Brill mentis, Vandenhoeck & Ruprecht, Böhlau and V&R unipress. Koninklijke Brill nv reserves the right to protect this publication against unauthorized use. Requests for re-use and/or translations must be addressed to Koninklijke Brill nv via brill.com or copyright.com. This book is printed on acid-free paper and produced in a sustainable manner.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Contents List of Figures and Tables xi Notes on Contributors xv Tabula Gratulatoria xxi A Lifelong Passion for Islamic History xxiii Maaike van Berkel and Letizia Osti Bibliography of Hugh N. Kennedy xxvii

part 1 Caliphate and Power 1

A Ḥimyarite Restorationist Prophecy Michael Cook

3

2

Kinship, Dynasty, and the Umayyads Andrew Marsham

12

3

He Reigned as Caliph; Then He Died: The Reigns of Caliphs Versified 46 Geert Jan van Gelder

4

Versifying History in Abbasid Iraq: The Universal History of ʿAlī b. al-Jahm 69 Harry Munt

5

How to found an Islamic state: The Idrisids as rivals to the Abbasid Caliphate in the Far Islamic West 91 Corisande Fenwick

6

Rethinking “the Mamlūk State” with Ibn Khaldūn: “Mamlukization,” ʿaṣabiyya, and Historiographical Imaginations of the Sultanate of Cairo (1200s–1500s) 117 Jo Van Steenbergen

7

Ibn Khaldūn and the Ḥafṣid Caliphate Allen Fromherz

140

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

viii

contents

part 2 Economy and Society 8

A Three-Centered System: Aleppo, Damascus, and Cairo in the Age of the Ayyubids 157 R. Stephen Humphreys

9

Informal and Formal Trading Associations in Egypt and Ifrīqiya, 850–1150 171 Chris Wickham

10

Good Governance in Theory and Practice: Comparing Abū Yūsuf’s Kitāb al-Kharāj with Papyri 183 Petra M. Sijpesteijn

11

A Matter of Trust: On Some Principles of Governance in the Letters of Qurra b. Sharīk 201 Arietta Papaconstantinou

12

Calculating the Population of Samarra Alastair Northedge

13

Flour for the Caliph: Watermills in the “Land behind Mosul” Cristina Tonghini

14

Bedouin, Bandits, and Caliphal Disappearance: A Reappraisal of the Qarāmiṭa and Their Success in Arabia 254 Peter Webb

15

Zinā and muḥṣanāt in the Quran Richard Kimber

210

234

283

part 3 Abbasids 16

Muslim Nostalgia: Longing for the Abbasid Past in the Mamluk Era 299 Robert Irwin

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ix

contents

17

The al-Mustanṣiriyya madrasa in Baghdad and Its Founder, al-Mustanṣir 320 Carole Hillenbrand

18

Hārūn al-Rashīd in Premodern Arabic Literary Imaginary: Ideology of Monogamy, Harem Politics, and Court Intrigues 340 Wen-chin Ouyang

19

The Representation of the Barmakids in Bodleian Manuscript Ouseley 217 and Other Monographs 356 Arezou Azad and Pejman Firoozbakhsh

20

Eutychius of Alexandria Vindicated: Muslim Sources and Christian Arabic Historiography in the Early Islamic Empire 384 Robert Hoyland

21

Bureaucrats on the Move: Messengers in Fourth/Tenth-Century Iraq 405 Maaike van Berkel, Nadia Maria El Cheikh and Letizia Osti

22

Al-Ṭabarī’s Unacknowledged Debt to Ibn Abī Ṭāhir Ṭayfūr Sarah Bowen Savant

23

Heraqleh: A New Interpretation Andrew Petersen

432

448

part 4 Frontiers and the Others 24

The Interface between Byzantium and the Ilkhanids in Fourteenth-Century Book Painting 475 Robert Hillenbrand

25

Exploring Europe through Medieval Islamic Folk Literature Niall Christie

26

The Lordship and Bishopric of Banyas in the Latin Kingdom of Jerusalem (1126–1164) 521 Alan V. Murray

503

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

x

contents

27

Fortresses and Frontiers: Castles and Northern Syria in the Sultanate of Cairo 538 Angus D. Stewart

28

The Sasanian Fort of Pānkān Balázs Major

29

Negotiating the North: Armenian Perspectives on the Conquest Era 591 Tim Greenwood

30

New Palaeoenvironmental Evidence on the Possible Impact on Agriculture of Early Arab-Islamic Raiding Activity on Crete 614 John Haldon Index

560

635

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Figures and Tables Figures 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 5.1 5.2 5.3 5.4 5.5 5.6 5.7 12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8 12.9 13.1 13.2 13.3 13.4 13.5 17.1 17.2 17.3

ʿUthmān and the Sufyanid and Marwanid Umayyad caliphs 14 The lifespans of the Umayyad caliphs 20 Abū al-ʿĀs b. Umayya’s marriages and children 22 Al-Ḥakam b. Abī al-ʿĀṣ’s marriages and children 24 Abū Sufyān b. Ḥarb’s marriages and children 26 ʿUthmān b. ʿAffān’s marriages and children 28 ʿAbd al-Malik b. Marwān’s marriages and children 36 Al-Walīd b. ʿAbd al-Malik’s marriages and children 39 Map of early medieval North Africa showing the location of the rival states 92 Map of Morocco showing the location of Idrisid mints, towns, and known silver mines 99 Silver dirham of Idrīs i minted at Walīla 101 Silver dirham of Idrīs ii minted at Walīla 102 Plan of early medieval Volubilis 105 Reconstruction of the housing in sector D 107 Reconstruction of the Idrisid complex 109 The site of Basra, showing the limits of the early Islamic city 211 The site of Kufa showing the limits of the early Islamic city 212 The site of Samarra, showing the extent of the ancient remains 216 The division of the archaeological site into zones 217 The disposition of the military cantonments 218 Al-Mutawakkiliyya, showing the classic layout of a cantonment (qaṭīʿa) at Samarra 219 Examples of blocks of cantonment houses 226 Plan of Building B at Amman, destroyed in 131/749 227 Examples of elite housing from al-Mutawakkiliyya 229 General map of the LoNAP survey area 236 LoNAP survey, settlement distribution map 239 General map of site 124, with the two groups of mills 246 Aerial view of site 124, upper group 247 Site 124, drop towers 1 and 2 248 Al-Mustanṣiriyya madrasa: Courtyard from the south 320 Al-Mustanṣiriyya madrasa: Façade of the prayer hall 324 Al-Mustanṣiriyya madrasa: Axonometric view 325

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xii 17.4 17.5 17.6 17.7 17.8 17.9 17.10 19.1 19.2 21.1 21.2 22.1 22.2 22.3 22.4

23.1 23.2 23.3

23.4 23.5 23.6 23.7 23.8

figures and tables Al-Mustansiriyya madrasa: Corridor (īwān) 327 Al-Mustansiriyya madrasa: The wall facing the Tigris from the northwest 329 Al-Mustanṣiriyya madrasa: Riparian inscription in Ayyubid-style cursive writing 330 Al-Mustanṣiriyya madrasa: Inscription on the wall facing the Tigris 331 Al-Mustansiriyya madrasa: The southwestern exterior wall with the inscription as restored in 1865 332 Al-Mustanṣiriyya madrasa: The portal inscription 334 Al-Mustanṣiriyya madrasa: Part of the portal inscription 336 Ouseley 217, Bodleian Library Persian manuscript, frontispiece on fol. 1v 361 Ḍiyāʾ al-Dīn Baranī, Akhbār-i Barmakiyān, Aga Khan Museum manuscript folio akm 126 366 Viziers appointed between 313/925 and 334/945 414 The first five amīr al-umarāʾ, 324–334/936–945 416 The first alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād and al-Ṭabarī’s Taʾrīkh 435 The seventh alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād and al-Ṭabarī’s Taʾrīkh 437 The eighth alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād and al-Ṭabarī’s Taʾrīkh 438 The passive voice as a potential indicator of text reuse in al-Ṭabarī’s Taʾrīkh. Blue dots mark occurrences of the phrases ḥaddathanā/ḥaddathanī as the first element of an isnād. Red dots indicate uses of the passive-voice phrase dhukira ʿan. Vertical lines identify section beginnings as well as the segment that corresponds to the extant volume of Ibn Abī Ṭāhir’s Kitāb Baghdād 441 Map of the Middle East showing the location of Raqqa and other sites 449 Map showing the location of Heraqleh in relation to Raqqa and other sites in the vicinity 450 corona image from 1967 showing Heraqleh in relation to the Euphrates floodplain and an adjacent partially eroded circular structure to the south-east 452 Herzfeld’s drawing of Heraqleh before modern disturbances 454 Google Earth Image of Heraqleh showing central building and concentric inner and outer enclosures 456 Plan of the central building of Heraqleh after Toueir 1983 458 View of the central building or terrace at Heraqleh from the west 460 Heraqleh, central building from SW in 2010 460

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

figures and tables 23.9 23.10 23.11 23.12 23.13 23.14 23.15 24.1 24.2 24.3 24.4 24.5 24.6 24.7 24.8 24.9 24.10 24.11 24.12 28.1 28.2 28.3 28.4 28.5 28.6 28.7 28.8 28.9 28.10

xiii

Aerial view of Heraqleh circa 1985 461 Stucco from excavations at Heraqleh with 80cm white and yellow scale divided into 10cm sections. 461 Greek cross reused within the central building at Heraqleh 463 Interior of the western gate of the outer enclosure showing the location of carved stone wall decoration 465 Detail of the wall decoration in the western gate 466 Vault inside the central building showing the use of roughly square stone for the walls and fired brick for the vaults 467 One of a number of circular well-like holes giving access to the vaults below 468 Palermo, Monreale Cathedral: Mosaic of angels visiting Lot, late twelfth century 481 Al-Bīrūnī, Chronology of ancient nations, 1307: Muḥammad appoints ʿAlī as his successor 482 Hosios Lukas, mosaic of the Anastasis, first half of the eleventh century 483 Al-Bīrūnī, Chronology of ancient nations, 1307: The Baptism of Jesus 485 Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: The birth of Muḥammad 486 Daphni, mosaic of the Nativity of Christ, ca. 1100 488 Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: Baḥīrā recognizes Muḥammad as a prophet 489 Daphni, mosaic of the Baptism of Christ, ca. 1100 490 Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: The Annunciation 492 Al-Bīrūnī, Chronology of ancient nations, 1307: The Annunciation 494 Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: Muḥammad receives the first revelation from Jabrāʾīl 495 Daphni, mosaic of the Annunciation, ca. 1100 496 Valley of the Kargh River with the rectangular enclosure of Pānkān in the foreground 561 Fort of Pānkān from the northeast 562 Plan of the fort 564 Detail of northern enclosure wall (w129) showing the remains of the formwork casting 566 Northwestern cluster of buildings in the fort 567 Excavated gate from the south 568 Construction periods of the gate 569 Sasanian-period bricks 571 Terrain model of the fort 573 Northeastern cluster of buildings in the fort 574

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xiv 28.11 28.12 28.13 28.14 28.15 28.16 28.17 28.18 30.1

figures and tables Buildings R100 and R101 from the south 575 Interior of R100 looking south 576 Excavation trench 2019/3 in the area of room R300, looking west 578 Possible courtyard S1001 in excavation trench 2019/2 579 Excavation trench 2019/5, with the openings of the two stone-lined storage pits 580 Sample of fine creamware shards 582 Map of the metal and ceramic finds of the 2016 survey 584 Bronze cosmetics mortar 586 Google Image map of the island of Crete 621

Tables 12.1 12.2 12.3a 12.3b 12.4 12.5 12.6 19.1 19.2

Areas of the Samarra mosques 222 Areas of the muṣallas at Samarra 223 Typology of buildings relative to the question of residence 224 Buildings that might or might not be residential 225 Total number of small houses by zone 228 Typology of elite housing 229 Zonal division of elite housing 230 Stemmatic Diagram of Baranī’s Akhbār-i Barmakiyān 376 Table of Premodern Persian Monographs on the Barmakids (Lost and Surviving Manuscripts and Published Editions) 377

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Notes on Contributors Arezou Azad is Senior Research Fellow and Programme Director of the Invisible East programme at the University of Oxford. She is a historian of the medieval Islamic east (Iran, Afghanistan, Central Asia), with a DPhil from Oxford, and has published multiple books and peer-reviewed articles on the social and cultural history of the region. Maaike van Berkel is Professor of Medieval History at the Radboud University Nijmegen. Her research focuses on administration, communication, and court and urban history in the medieval Middle East. Currently she is the principal investigator of a project on water management in Middle Eastern cities. Niall Christie is an instructor in history at Langara College in Vancouver, Canada, where he teaches the history of Europe and the Muslim world. He is also an adjunct professor of medieval studies at the University of Victoria. His research focuses on the Muslim response to the Crusades. Michael Cook has been teaching the history of the Muslim world in the Near Eastern Studies Department at Princeton University since 1986. Before that he taught in the History Department at the School of Oriental and African Studies in the University of London. His latest book is Ancient religions, modern politics. Nadia Maria El Cheikh is a scholar of the Abbasid Caliphate and Byzantium. Her publications include Byzantium viewed by the Arabs (Harvard Middle Eastern monographs, 2004), which was translated into Turkish and Greek. In 2013 she coauthored a book entitled Crisis and continuity at the Abbasid court: Formal and informal politics in the caliphate of al-Muqtadir (295–320/908–932) (Brill). Women, Islam and Abbasid identity was published in 2015 by Harvard University Press and was recently translated into Arabic. She served as Dean of the Faculty of Arts and Sciences at the American University in Beirut between 2016 and 2021. In 2022 she was appointed Vice Provost for Cultural and Research Engagement at nyu Abu Dhabi.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xvi

notes on contributors

Corisande Fenwick is Associate Professor in Mediterranean Archaeology at the Institute of Archaeology, ucl and Director of the Society for Libyan Studies. Her recent books include Early Islamic North Africa (Bloomsbury, 2020) and the co-edited Oxford handbook of Islamic archaeology (oup, 2020). She currently directs excavations in Morocco and Tunisia. Pejman Firoozbakhsh is a philologist of Iranian languages, focusing on the formation and development of New Persian. His research interests include the New Iranian languages and dialects, Persian codicology, historiography, and textual criticism. Pejman graduated from the University of Hamburg with a PhD in Iranian Studies in 2020. Allen Fromherz is Professor of History and Middle East Studies Center Director at Georgia State University. He authored The Almohads: Rise of an Islamic empire; Ibn Khaldun, life and times; The Near West: Medieval North Africa, Latin Europe and the Mediterranean and Qatar, a modern history; and edited The Gulf in world history and Sultan Qaboos and Modern Oman. He is a Senior Fulbright Scholar to Spain (2022). Geert Jan van Gelder (b. Amsterdam, 1947) was Lecturer in Arabic at the University of Groningen from 1975 until 1998 and Laudian Professor of Arabic at the University of Oxford from 1998 until 2012. He has published widely on classical Arabic literature. Tim Greenwood is a Reader in the School of History at the University of St Andrews. He has published widely on the political, social, and cultural history of late antique and medieval Armenia (c. 500–1100). He is preparing a monograph on law and legal culture in medieval Armenia. John Haldon studied in Birmingham, Athens, and Munich. He is a Fellow of the British Academy and current Director of the Princeton Climate Change and History Research Initiative. His research focuses on the history of the medieval eastern Roman (Byzantine) Empire, on state systems and resources in the premodern world, and on the impact of environmental stress on premodern societies.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xvii

notes on contributors

Carole Hillenbrand cbe, fba (Professor Emerita, Edinburgh; Honorary Professor, St Andrews) has published seven books plus three volumes of collected articles. She was awarded the King Faisal Prize in Islamic Studies for The Crusades: Islamic perspectives (1999), and the British Academy Nayef Al-Rodhan Prize for Global cultural understanding for Islam: A historical introduction (2016). Robert Hillenbrand fba, Professor of Islamic Art at Edinburgh and St Andrews, has published 11 books; some 200 articles; and edited, co-edited, or coauthored 14 books. He has held visiting professorships at Cambridge, Princeton, ucla, Bamberg, Dartmouth College, Leiden, New York, Cairo, and Groningen. He works on Islamic architecture, book painting, and iconography. Robert Hoyland is Professor of Middle East History at New York University’s Institute for the Study of the Ancient World, having previously taught at the universities of St Andrews and Oxford. He has published on diverse aspects of the intellectual and material culture of the late antique and early Islamic Middle East. R. Stephen Humphreys is Professor Emeritus in History and Islamic Studies at the University of California, Santa Barbara. He is the author of From Saladin to the Mongols: The Ayyubids of Damascus, 1193–1260 (1977), Islamic history: A framework for inquiry (1991), and Muʿawiya ibn Abi Sufyan: From Arabia to empire (2006). He has been a visiting member of the Institute for Advanced Study in Princeton, NJ and a visiting fellow at All Souls College, Oxford. Robert Irwin lectured in the Mediaeval Department of the University of St Andrews before leaving to become a full-time writer of fiction and nonfiction. He has published books on the Arabian nights, the Mamluks, and Orientalism. His most recent work of nonfiction is Ibn Khaldun: An intellectual biography. Richard Kimber was formerly Lecturer in Arabic Studies at the University of St Andrews. Balázs Major is an archaeologist, Arabist, and historian and holds a PhD in archaeology from Cardiff University. He is Director of the Institute of Archaeology at Pázmány Péter Catholic University. He is directing archaeological excavations in Syria, - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xviii

notes on contributors

Lebanon, and Iraqi Kurdistan, with a main interest in medieval military architecture and rural settlements. Andrew Marsham is Professor of Classical Arabic Studies at the University of Cambridge and a Fellow of Queens’ College. His publications include Rituals of Islamic monarchy (Edinburgh, 2009) and two edited volumes: Power, patronage, and memory in early Islam (Oxford, 2018), with Professor Alain George, and The Umayyad world (Routledge, 2021). Harry Munt is Senior Lecturer in Medieval History at the University of York. He is the author of several articles on early Islamic history and premodern Arabic history writing as well as The Holy City of Medina: Sacred space in early Islamic Arabia (Cambridge University Press, 2014). Alan V. Murray is Senior Lecturer in Medieval Studies at the University of Leeds. He wrote his doctoral thesis on the nobility of the kingdom of Jerusalem under the supervision of Hugh Kennedy and has published numerous works on the crusades, the principalities of Outremer, and medieval warfare, including The Crusader kingdom of Jerusalem: A dynastic history, 1099–1125 (2000), The Franks in Outremer: Studies in the Latin principalities of Syria and Palestine, 1099–1187 (2015), and Baldwin of Bourcq: Count of Edessa and King of Jerusalem (1100–1131) (2022). Alastair Northedge is Professor Emeritus of Islamic Art and Archaeology at Université de Paris 1 (Panthéon-Sorbonne). He has worked in Syria, Jordan, Saudi Arabia, Kazakhstan, and Turkmenistan, and conducted projects at Amman in Jordan, and Ana in Iraq, in addition to Samarra. He is author of Studies on Roman and Islamic Amman, joint author of Excavations at Ana, and published the Historical topography of Samarra in 2005. The second volume of the project at Samarra, the Archaeological atlas of Samarra, was published in 2015. He subsequently worked on the medieval city of Dehistan in Turkmenistan. After retirement in 2017, he is now working on the archaeological site of Old Basra at al-Zubayr in Iraq. Letizia Osti is Associate Professor of Arabic Language and Literature at the University of Milan. She has published on classical Arabic prose and narrative techniques in

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

notes on contributors

xix

biographical collections, historiography, literature, and intersections thereof. She is the author of History and memory in the Abbasid caliphate: writing the past in medieval Arabic literature (Bloomsbury, 2022). Wen-chin Ouyang is Professor of Arabic and Comparative Literature at soas, University of London. She has published extensively on classical and modern Arabic literature and critical theory, including The thousand and one nights. Arietta Papaconstantinou is a social historian of the late antique Mediterranean, focusing on the transition from the Roman to the Islamic empire. She researches rural communities and historical multilingualism and Mediterranean cultural history as a whole. She teaches late antique history at the University of Reading and is an associate member of the Faculty of Oriental Studies in Oxford and of the Institute for Byzantine Studies at the Collège de France. Andrew Petersen is Director of Research in Islamic Archaeology at the University of Wales Trinity Saint David. He studied medieval history at St Andrews, Islamic Architecture at Oxford, and wrote his PhD on medieval and Ottoman Palestine at Cardiff University. He is a Member of the Institute for Archaeologists and a Fellow of the Royal Historical Society. Sarah Bowen Savant is Professor of History at the Aga Khan University—Institute for the Study of Muslim Civilisations (aku-ismc) and the principal investigator of the European Research Council—funded kitab project (kitab‑project.org). She received her PhD from Harvard University and her MA from the University of Chicago. Her publications include The new Muslims of post-conquest Iran: Tradition, memory, and conversion (Cambridge University Press, 2013; winner of the SaidiSirjani Book Award given by the International Society for Iranian Studies on behalf of the Persian Heritage Foundation); as editor (with Helena de Felipe), Genealogy and knowledge in Muslim societies: Understanding the past (akuismc, Exploring Muslim Contexts/Edinburgh University Press, 2014); as translator (with Peter Webb), The excellence of the Arabs: A translation of Ibn Qutaybah’s Faḍl al-ʿarab wa l-tanbīh ʿalā ʿulūmihā (Library of Arabic Literature/New York University Press, 2017); and numerous articles treating ethnic identity, cultural memory, genealogy, and history writing. She is currently preparing with the kitab project team a two-volume study, entitled A cultural history of the Arabic book. She also sits on the management team of aku-ismc. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xx

notes on contributors

Petra M. Sijpesteijn is Professor of Arabic at Leiden University. She is a cultural and social historian of the medieval Middle East. Currently she is the principal investigator of a European Research Council funded project entitled Embedding conquest: Naturalising Muslim rule in the early Islamic empire (600–1000). Angus D. Stewart studied with Hugh Kennedy at the University of St Andrews. He is the author of The Armenian kingdom and the Mamluks: War and diplomacy during the reigns of Het‘um ii (1289–1307) (Leiden, 2001). Now lecturing in Mediaeval and Middle Eastern History at St Andrews, he has inherited Hugh’s former module on “The Mediaeval castle.” Cristina Tonghini is an archaeologist who specializes in the Arab world in the Islamic period. She is Full Professor at the Ca’ Foscari University of Venice. Her current research focuses on settlement, landscape, resources management, and production in northern Iraq. Recent publications include From Edessa to Urfa: The fortification of the citadel (Archaeopress 2021). Jo Van Steenbergen is Professor of Arabic and Islamic Studies at Ghent University. He has published many chapters, articles, edited volumes, and books, especially on late medieval Syro-Egyptian history, including A history of the Islamic world, 600–1800 (2021). Peter Webb is a University Lecturer in Arabic Literature and Culture at Leiden University. His research analyzes the evolution of Arab identity and Muslim interpretations of pre-Islamic history. He is author of Imagining the Arabs: Arab identity and the rise of Islam (Edinburgh, 2016) and editor/translator of several classical Arabic texts for nyp Press’s Library of Arabic Literature and Brill’s Bibliotheca Maqriziana. Chris Wickham is Chichele Professor of Medieval History (Emeritus) at the University of Oxford and taught at both Oxford and Birmingham. He has published widely on European and Eurasian history across the period 400–1200.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Tabula Gratulatoria Peter Adamson Camilla Adang Dionisius Agius Judy Ahola Samer Ali Fréderic Bauden Karen Bauer Lale Behzadi David Bennett Amira Bennison Monique Bernards Teresa Bernheimer Fanny Bessard Hinrich Biesterfeldt Kevin Blankinship Antoine Borrut Antonia Bosanquet Julia Bray Leslie Brubaker Jelle Bruning Averil Cameron Paul Cobb Lawrence Conrad Michael Cooperson Amikam Elad Tayeb El-Hibri Maribel Fierro Alison Gascoigne Antonella Ghersetti Nathan Gibson Rob Gleave Matthew Gordon Frank Griffel Beatrice Gründler Sebastian Günther Hannah-Lena Hagemann Avraham Hakim

Eric Hanne Paul Heck Stefan Heidemann Konrad Hirschler Steven Judd Alexander Key Ruqayya Yasmine Khan Hilary Kilpatrick Waardenburg István Kristó-Nagy Remke Kruk Marie Legendre Zina Maleh Eduardo Manzano Moreno Christopher Melchert Charles Melville Alex Metcalfe Benjamin Michaudel James Montgomery Suleiman Mourad Pernilla Myrne John Nawas Nassima Neggaz Bilal Orfali Walter Pohl Maurice Pomerantz Dwight Reynolds Khodadad Rezakhani Chase Robinson Everett Rowson Marina Rustow Ihab el-Sakkout Ignacio Sánchez Jens Scheiner Emily Selove Mehdy Shaddel Devin Stewart Yuko Tanaka

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxii Shawkat Toorawa Deborah Tor Isabel Toral John Turner Uwe Vagelpohl

tabula gratulatoria

Kevin van Bladel Vanessa Van Renterghem James Weaver Philip Wood Mohsen Zakeri

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

A Lifelong Passion for Islamic History Maaike van Berkel and Letizia Osti

One afternoon in July 2004, in the elegant setting of the Royal Flemish Academy of Belgium for Science and the Arts, the School of Abbasid Studies held a session on the caliphate of al-Muqtadir. It was organized by Hugh Kennedy, one of the school’s directors, who had selected a few junior scholars to contribute to the discussion. It was exciting to discover a shared interest in this topic, and two years later, at the School’s meeting in St Andrews, we could not stop talking, comparing notes, and drawing parallels. You should write a book, more than one colleague suggested; and so we did, in this very Brill series: our study of al-Muqtadir, Crisis and continuity at the Abbasid court, was published in 2013. As is often said, many friendships are killed by writing a book together. We, on the contrary, began with a book and, almost 20 years later, continue to find ways of collaborating and enjoying each other’s friendship. A picture of the Muqtadiriyya—as someone once described it—on Hugh’s desk testifies to this. Although both of us knew Hugh long before 2004, we have chosen to begin with this memory because it highlights some of the reasons for the present collection: Hugh’s ability to bring people together and communicate enthusiasm for their ideas and projects; his curiosity about the work of colleagues, no matter how junior; and his willingness to support and promote them. Yet, this book is first and foremost a tribute to Hugh’s impressive scholarship and how it has influenced the field of Islamic history at large as well as our own work. For instance, working with Hugh has taught us the importance of reading personality centered narratives, where single paradigmatic figures are responsible for major historical developments, interpreting the portrayal of such figures not only as themselves but also as representatives of a group with specific principles, practices, and aspirations. On the other hand, Hugh’s research rarely relies on chronicles alone: another important thing he has taught us is to read the rich narrative sources at our disposal in combination with material and documentary evidence. Hugh’s research also has the unusual quality of being both specific and accessible to specialists of other periods and areas: he has often collaborated on projects on the medieval Mediterranean and contributed to comparative endeavors on Europe and Eurasia, and is always an important presence at the International Medieval Congress at Leeds. Finally, Hugh has always been very active in the communication of scholarship, participating in and curating radio

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxiv

van berkel and osti

programs and giving open lectures, and contributing to public debate with accessible material. Hugh Kennedy is the quintessential historian of the Islamic world.

∵ We begin the book with a bibliography that illustrates the breadth and depth of Hugh’s research, even without listing the translations of most of his monographs into many foreign languages. This is followed by four sections, loosely organized along chronological or geographical lines, representing major themes on which Hugh has worked and published extensively. The 30 contributions together cover some, but by no means all, of the topics, areas, and periods that Hugh has worked on during over 50 years dedicated to Islamic history. The theme of the first section, “Caliphate and power,” has been a lifelong interest for Hugh. His first book, based on his dissertation, was The early Abbasid caliphate, published in 1981. A few years later, in 1987, he published The Prophet and the age of the caliphates, which became a standard and is now in its fourth edition. This volume has introduced generations of students to the study of Islamic history. More recently, the power, ideology, and history of this major Islamic institution have been the focal point of The caliphate: The history of an idea (2016). Here, Hugh demonstrates his ability to make medieval history relevant for the present. The contributions in this section of our volume cover a vast temporal and geographical span: from the time of the Prophet until the Sultanate of Cairo in the fifteenth century, and from Iraq to the Maghreb. They not only discuss the political power of a specific period but also look at how it is portrayed in different types of sources, discussing the memory of past caliphates and the ways in which the caliphate is represented by contemporary and later authors. Hugh has been one of the pioneering scholars in the field of the economic history of pre-1500 Muslim societies. This is the object of the second section: Hugh’s studies on the fiscal administration and financial problems of the Abbasid Caliphate (such as The decline and fall of the first Muslim empire, 2004) and his landmark analysis of the economic and agricultural foundations of the city of Baghdad in the Abbasid period (The feeding of the five hundred thousand, 2011) have inspired so many of us. Not only do they employ both material and textual sources, they also translate dry information into a vivid portrait of social life in Abbasid Iraq. The authors of this section of the volume make use of a diverse corpus of sources, including documents on paper and papyrus, archaeological evidence, and narrative sources, providing significant contributions to the social

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a lifelong passion for islamic history

xxv

and economic history of North Africa and the Middle East, both in rural and in urban contexts, from the second/eighth to the seventh/thirteenth century. We have arranged the chapters geographically, from west to east, from the Mediterranean to Iraq and the Arabian Peninsula. A festschrift in honor of Hugh Kennedy would not be complete without a section on the Abbasids: although Hugh has published on a wide variety of empires and rulers across the Islamic and Byzantine worlds, the Abbasids have been a constant throughout his career. As one of the directors of the School of Abbasid Studies, and as a speaker and discussant at many conferences, he has promoted interest in this period among new generations of scholars. In fact, most of the publications we have mentioned so far, including the volume we wrote together, discuss the Abbasids or have them as their focus. To these we should add his study The court of the caliphs, also published as When Baghdad ruled the Muslim world (both 2004), where he uses major historical and literary sources to portray the Abbasid caliphs as distinct characters. The chapters in this section discuss the intellectual history of the period, the Abbasids’ building activities, and their politics and administration. The importance of Abbasid history and its relevance for the present is underscored by the number of contributions devoted to the memory of the Abbasids, even up to modern times. This is why we start with contributions viewing the Abbasids from the distant future and then zoom in on Abbasid sources themselves. The fourth section of this book deals with frontiers and what is beyond them. Hugh’s interests in the Crusades, warfare, and people of the sword, as well as cultural exchange in border regions, have resulted in three major books: Crusader castles (1994), The armies of the caliphs (2001), and The great Arab conquests (2007). The latter is an illustration of Hugh’s increasing interest in bringing Islamic history into mainstream culture: like The court of the caliphs, this book is written in a manner accessible to a wide section of the public. The chapters in this final section, arranged in reverse chronological order, discuss fortresses, raiding, acculturation, and representations on the different sides of frontiers, from the ninth/fifteenth century to the time of the conquests, and from Iran to Crete. They confirm that, in Islamic history as elsewhere, contact and conflict go hand in hand.

∵ When we started this project, we were aware that we would be asking a lot of our invitees: a paper with an obvious connection to Hugh’s research, and a very strict deadline. Still, we were overwhelmed by the positive response of former students and colleagues, so much so that we kept worrying that we had

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxvi

van berkel and osti

missed contributors (and we are sure we have). And yet, this should not have surprised us. His scholarship aside, it is impossible not to like and admire Hugh. This is why, along with the 30 chapters, we include a tabula gratulatoria with the names of many colleagues who wish to convey their felicitations. A volume like this requires the help of many. We should first of all thank Peter Webb, who took the first steps in this project. Hinrich Biesterfeldt and Sebastian Günther immediately welcomed the volume in this series. Two anonymous readers reviewed the entire manuscript and made important suggestions for improvement. Peter Brown and Angus D. Stewart generously helped us with topics we felt were outside our field of expertise. We owe a particular debt to Nadia Maria El Cheikh, our Muqtadiriyya colleague-in-crime, for her constant support and friendship, and for organizing the bibliography of Hugh’s work with the help of the librarians at the American University of Beirut. The bibliography has been finalized by Rick van Brummelen, student-assistant at the Radboud University Nijmegen. Without the practical and moral support of Maurits van den Boogert and Teddi Dols from Brill this project, and especially its rapid production schedule, would not have been possible, and Rebekah Zwanzig is the best copyeditor we could have wished for. Sarah Savant gave us the perfect venue for presenting this book and secretly organized festivities. And of course, our authors were wonderful in following our orders with no complaints and providing excellent contributions.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Bibliography of Hugh N. Kennedy Books 1978 Politics and the political élite in the early Abbasid caliphate, PhD diss., University of Cambridge, ProQuest dissertations and theses global.1 1981 The early Abbasid caliphate: A political history, London. 1986 The Prophet and the age of the caliphates: The Islamic Near East from the sixth to the eleventh century, London. Second and updated edition, London 2004. Third and updated edition, London 2015. Fourth and updated edition, London 2022. 1994 Crusader castles, Cambridge. 1996 Muslim Spain and Portugal: A political history of Al-Andalus, London. 2001 The armies of the caliphs: Military and society in the early Islamic state, London. 2002 Mongols, Huns and Vikings: Nomads at war, London. 2004 The court of the caliphs: The rise and fall of Islam’s greatest dynasty, London. Also published as When Baghdad ruled the Muslim world: The rise and fall of Islam’s greatest dynasty, Cambridge, MA 2005.

1 Many of Hugh’s monographs have appeared in translation. This bibliography only lists the original English editions.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxviii

bibliography of hugh n. kennedy

2006 The Byzantine and early Islamic Near East, Aldershot. 2007 The great Arab conquests: How the spread of Islam changed the world we live in, London. 2013 with M. van Berkel, N.M. El Cheikh, and L. Osti, Crisis and continuity at the Abbasid court: Formal and informal politics in the caliphate of al-Muqtadir (295–320/908–32), Leiden. 2016 The caliphate: A Pelican introduction, London. Also published as Caliphate: The history of an idea, New York.

Edited Works 2001 The historiography of Islamic Egypt (c. 950–1800), Leiden. An historical atlas of Islam/Atlas historique de l’Islam, Leiden. 2003 with I.A. Alfonso and J.E. Monge, Building legitimacy: Political discourses and forms of legitimation in medieval societies, Leiden. 2006 Muslim military architecture in greater Syria: From the coming of Islam to the Ottoman period, Leiden. 2008 Al-Tabari: A medieval Muslim historian and his work, Princeton, NJ. 2009 with A. Papaconstantinou and M. Debie, Writing “true stories”: Historians and hagiographers in the late antique and medieval Near East, Turnhout. 2013 Warfare and poetry in the Middle East, New York.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxix

Translations 1990 Al-Manṣūr and al-Mahdī, vol. 29 of The history of al-Ṭabarī, trans., Albany.

Articles and Book Chapters 1981 Central government and provincial élites in the early ʿAbbāsid caliphate, in bsoas 44, 26–38. 1982 Succession disputes in the early Abbasid caliphate, in R. Hillenbrand (ed.), Union Européenne des Arabisants et Islamisants: 10th Congress, Edinburgh, 9–16 September 1980: Proceedings, Edinburgh, 29–33. 1985 From Polis to Madina: Urban change in late antique and Eerly Islamic Syria, in Past & Present 106, 3–27. The last century of Byzantine Syria: A reinterpretation, in Byzantinische Forschungen 10, 141–183. 1986 The desert and the sown in eastern Arabian history, in I.R. Netton (ed.), Arabia and the Gulf, Kent, 18–28. The Melkite church from the Islamic conquest to the Crusades: Continuity and adaptation in the Byzantine legacy, in 17th international Byzantine congress: The major papers, New Rochelle, 325–343. The Uqaylids of Mosul: The origins and structure of a nomad dynasty, in M. Paz Torres and M. Marin (eds.), Actas del xii Congreso de la u.e.a.i, Madrid, 391– 402. 1987 Recent French archaeological work in Syria and Jordan, in Byzantine and modern Greek studies 11, 245–252. with D. Price, Marginalia, in The Yale University library gazette 62, 56–61.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxx

bibliography of hugh n. kennedy

1989 Change and continuity in Syria and Palestine at the time of the Moslem conquest, in aram periodical 1, 258–267. 1990 The Abbasid caliphate: A historical introduction, in J. Ashtiany et al. (eds.), Abbasid belles lettres, ii, Cambridge, 1–15. The Barmakid revolution in Islamic government, in C.P. Melville (ed.), Persian and Islamic studies in honour of P.W. Avery, Cambridge, 89–98. 1992 Antioch: From Byzantium to Islam and back again, in J. Rich (ed.), The city in late antiquity, London, 181–189. The impact of Muslim rule on the pattern of rural settlement, in P. Canivet and J.P. Rey-Coquais (eds.), La Syrie de Byzance à l’islam: viie–viiie siècles, Damascus, 291–297. Nomads and settled people in Bilad al-Sham in the ninth and tenth centuries, in M.A. Bakhīt and M.Y. Abbadi (eds.), Proceedings of the fifth international conference on Bilad al-Sham, Amman, 105–113. 1995 The financing of the military in the early Islamic state, in A. Cameron (ed.), The Byzantine and early Islamic Near East, iii, Princeton, 361–378. The Muslims in Europe, in R. McKitterick (ed.), The new Cambridge medieval history, ii, Cambridge, 249–271. 1997 From oral tradition to written record in Arabic genealogy, in Arabica 44, 531– 544. 1998 Egypt as a province in the Islamic caliphate, 641–868, in C.F. Petry (ed.), The Cambridge history of Egypt, i, Cambridge, 62–85. From antiquity to Islam in the cities of al-Andalus and al-Mashriq, in P. Cressier and M. Garcia-Arenal (eds.), Genèse de la ville islamique en al-Andalus at au Maghreb occidental, Madrid, 53–64. with J.H.W.G. Liebeschuetz, Antioch and the villages of northern Syria in the fifth and sixth centuries a.d.: Trend and problem, in Nottingham medieval studies 32, 65–90.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxxi

1999 Medieval Merv: An historical overview, in G. Herrmann (ed.), Monuments of Merv, London, 27–44. Islam, in G.W. Bowersock, P. Brown, and O. Grabar (eds.), Late antiquity: A guide to the postclassical world, Harvard, 219–237. 2000 Intellectual life in the first four centuries of Islam, in F. Daftary (ed.), Intellectual traditions in Islam, New York, 17–30. Sicily and al-Andalus under Muslim rule, in T. Reuter (ed.), The new Cambridge medieval history, iii, Cambridge, 646–669. The early development of church architecture in Syria and Jordan c. 300–c. 750, in Studies in church history 36, 1–33. Gerasa and Scythopolis: Power and patronage in the Byzantine cities of Bilad al-Sham, in beo 52, 199–204. 2001 Syria, Palestine and Mesopotamia, in A. Cameron (ed.), The Cambridge ancient history, xiv, Cambridge, 588–611. 2002 Caliphs and their chroniclers in the middle Abbasid period (third/ninth century), in C.F. Robinson (ed.), Texts, documents, and artefacts: Islamic studies in honour of D.S. Richards, Leiden, 17–36. Military pay and the economy of the early Islamic state, in Historical research 75, 155–169. 2004 with J.F. Haldon, The Arab-Byzantine frontier in the eighth and ninth centuries: Military organisation and society in the borderlands, in L.I. Conrad and M. Bonner (eds.), Arab-Byzantine relations in early Islamic times, London, 141–178. The decline and fall of the first Muslim empire, in Der Islam 81, 4–30. The true caliph of the Arabian nights, in History today 54, 31–36. Byzantine-Arab diplomacy in the Near East from the Islamic conquests to the mid eleventh century, in L.I. Conrad and M. Bonner (eds.), Arab-Byzantine relations in early Islamic times, London, 81–91. Muslim Spain and Portugal: Al-Andalus and its neighbours, in D. Luscombe and J. Riley-Smith (eds.), The new Cambridge medieval history, iv, Cambridge, 599–622.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxxii

bibliography of hugh n. kennedy

2005 The caliphate, in Y.M. Choueiri (ed.), A companion to the history of the Middle East, Hoboken, 52–67. 2006 Justinianic plague in Syria and the archaeological evidence, in L.K. Little (ed.), Plague and the end of antiquity: The pandemic of 541–750, Cambridge 87–96. From Shahristan to Medina, in si 102–103, 5–34. The military revolution and the early Islamic state, in N. Christie and M. Yazigi (eds.), Noble ideals and bloody realities: Warfare in the middle ages, Leiden, 197–208. 2007 Al-Jāhiz and the construction of homosexuality at the Abbasid court, in A. Harper and C. Proctor (eds.), Medieval sexuality: A casebook, London, 175– 188. 2008 Inherited cities, in S.K. Jayyusi et al. (eds.), The city in the Islamic world, i, Leiden, 93–113. The Mediterranean frontier: Christianity face to face with Islam, 600–1050, in T.F.X. Noble and J.M.H. Smith (eds.), The Cambridge history of Christianity: Early medieval Christianities, c. 600–c. 1100, iii, Cambridge, 178–196. 2009 Survival of Iranianness, in V.S. Curtis and S. Stewart (eds.), The rise of Islam, New York, 13–29. with K. Burnside, Abbāsid caliphate, in The Oxford encyclopedia of the Islamic world, Oxford Islamic studies online. 2010 The city and the nomad, in R. Irwin (ed.), The new Cambridge history of Islam, iv, Cambridge, 274–289. How to found an Islamic city, in C. Goodson, A.E. Lester, and C. Symes (eds.), Cities, texts and social networks, 400–1500: Experiences and perceptions of medieval urban space, London, 45–63. The coming of Islam to Bukhara, in Y. Suleiman (ed.), Living Islamic history: Studies in honour of Professor Carole Hillenbrand, Edinburgh, 77–91. The late ʿAbbāsid pattern, 945–1050, in C.F. Robinson (ed.), The new Cambridge history of Islam, i, Cambridge, 360–394.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxxiii

Syrian elites from Byzantium to Islam: Survival or extinction?, in J. Haldon (ed.), Money, power and politics in early Islamic Syria: A review of current debates, London, 181–198. 2011 Great estates and elite lifestyles in the fertile crescent from Byzantium and Sasanian Iran to Islam, in A. Fuess and J-P. Hartung (eds.), Court cultures in the Muslim world: Seventh to nineteenth centuries, London, 54–79. The ribat in the early Islamic world, in H. Dey and E. Fentress (eds.), Western monasticism ante litteram: The spaces of monastic observance in late antiquity and the early middle ages, Turnhout, 161–175. The feeding of the five hundred thousand: Cities and agriculture in early Islamic Mesopotamia, in Iraq 73, 177–199. 2012 Journey to Mecca: A history, in V. Porter and M.A. Abdel Haleem (eds.), Hajj: Journey to the heart of Islam, Harvard, 68–132. Shayzar: A historical overview, in C. Tonghini (ed.), Shayzar i: The fortification of a citadel, Leiden, 2–25. with J. Haldon, Regional identities and military power: Byzantium and Islam ca. 600–750, in W. Pohl and C. Gantner (eds.), Visions of community in the post-Roman world: The west, Byzantium and the Islamic world, 300–1100, London, 317–353. Caliph, in R.S. Bagnall et al. (eds.), The encyclopedia of ancient history, Wiley online, https://doi.org/10.1002/9781444338386.wbeah12034. Elite incomes in the early Islamic state, in L.I. Conrad and F.M. Doner (eds.), The articulation of early Islamic state structures, London, 135–150. 2013 Pity and defiance in the poetry of the siege of Baghdad, in H. Kennedy (ed.), Warfare and poetry in the Middle East, New York, 149–165. Revival in the low countries, in J.M. Yeager (ed.), Early evangelicalism: A reader, Oxford, 146–152. 2014 Landholding and law in the early Islamic state, in J. Hudson and A. Rodríguez (eds.), Diverging paths? The shapes of power and institutions in medieval Christendom and Islam, Leiden, 159–181. Introduction to The eclipse of the ʿAbbasid caliphate: Classical writings of the medieval Islamic world, by Ibn Miskawayh, trans. D. Margoliouth, New York.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxxiv

bibliography of hugh n. kennedy

2015 The Middle East in Islamic late antiquity, in A. Monson and W. Scheidel (eds.), Fiscal regimes and the political economy of premodern states, Cambridge, 390–403. Introduction, in P. Sijpesteijn and A.T. Schubert (eds.), Documents and the history of the early Islamic world, Leiden, 1–7. 2016 Landed property and government finance in the early ʿAbbasid caliphate, in J. Hudson and S. Crumplin (eds.), The making of Europe: Essays in honour of Robert Bartlett, Leiden, 264–276. 2017 Baghdad as a center of learning and book production, in S. Blair and J. Bloom (eds.), By the pen and what they write: Writing in Islamic art and culture, New Haven, 91–103. The origins of the Aghlabids, in G.D. Anderson, C. Fenwick and M. Rosser-Owen (eds.), The Aghlabids and their neighbors: Art and material culture in ninthcentury North Africa, Leiden, 31–48. 2018 with A. Azad, The coming of Islam to Balkh, in A. Delattre, M. Legendre, and P. Sijpesteijn (eds.), Authority and control in the countryside: From antiquity to Islam in the Mediterranean and Near East (6th–10th century), Leiden, 284– 310. From Polis to Madina: Some reflections thirty years on, in S. Panzram and L. Callegarin (eds.), Entre civitas y Madīna: El mundo de las ciudades en la Península Ibérica y en el norte ee África (Siglos iv–ix), Madrid, 13–20. Frontiers of Islam: An essay on the varieties of frontier interactions, in K. Hebers and K. Wolf (eds.), Southern Italy as contact area and border region during the early Middle Ages: Religious-cultural heterogeneity and competing powers in local, transregional, and universal dimensions, Cologne, 51– 64. Muʾnis al-Muẓaffar: An exceptional eunuch, in A. Höfert, M. Mesley, and S. Tolino (eds.), Celibate and childless men in power, London, 79–91. 2020 The rise and fall of the early ʿAbbāsid political and military elite, in H-L. Hageman and S. Heidemann (eds.), Transregional and regional elites: Connecting the early Islamic empire, Berlin, 99–114.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxxv

2021 The emergence of new polities in the break-up of the Abbasid caliphate, in R. Kramer and W. Pohl (eds.), Empires and communities in the post-Roman and Islamic world, c. 400–1000ce, Oxford, 14–27. with W. Pohl, Comparative perspectives: Differences between the dissolution of the Abbasid caliphate and the western Roman empire, in R. Kramer and W. Pohl (eds.), Empires and communities in the post-Roman and Islamic world, c. 400–1000ce, Oxford, 64–75.

Book Reviews 1977 Review of The classical heritage in Islam, by F. Rosenthal, in The classical review 27, 320. 1981 Review of Conversion to Islam, by L. Nehemia, in ijmes 13, 249–250. Review of Conversion to Islam in the medieval period: An essay in quantitative history, by R.W. Bulliet, in ijmes 13:2, 250–252. Review of The shaping of ʿAbbasid rule, by J. Lassner, in ijmes 13, 367–369. 1982 Review of Frankincense and myrrh: A study of the Arabian incense trade, by N. Groom, in History 67, 276–277. 1983 Review of Mouvements populaires à Bagdad à l’époque ʿAbbasside, ixe–xie siècles, by S. Sabari, in bsoas 46, 347–348. Review of The Muslim discovery of Europe, by B. Lewis, in History 68, 275. Review of The political and social history of Khurasan under Abbasid rule, 747– 820, by E.L. Daniel, in ahr 88, 149–150. The multi-storey tradition, review of Sana: An Arabian Islamic city, by R.B. Serjeant and R. Lewcock (eds.), in Times literary supliment 4188, 734. 1984 Review of Atlas of the Islamic world since 1500, by C.F. Robinson, in History 69, 267–268. Review of Die Chronik des Ibn Ad-Dawādārī. 1. Teil. Kosmographie, by B. Radtke (ed.); Die Chronik des Ibn Ad-Dawādārī. 3. Teil. Der Bericht über den Proph-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxxvi

bibliography of hugh n. kennedy

eten und die Rechtgeleiteten Chalifen, by Muhammad as-Saʿīd Gamāl ad-Dīn (ed.), in bsoas 47, 412–413. Review of Islam and the west: The Moriscos, a cultural and social history, by A.G. Chejne, in History 69, 314–315. Instead of the Sassanians, review of Iraq after the Muslim conquest, by M.G. Morony, in Times literary supliment 4249, 999. 1985 Review of St James’s catapult: The life and times of Diego Gelmirez of Santiago de Compostela, by R.A. Fletcher, in History 70, 504–505. 1986 Review of Crusade and mission: European approaches toward the Muslims, by B.Z. Kedar, in History 71, 131–132. 1987 Review of Das Sendschreiben des ʿAbdalḥamīd b. Yaḥyā (Gest, 132/750) an den Kronprinzen ‘Abdallāh b. Marwān ii, by H. Schönig, in jras 2, 318. Review of God’s caliph: Religious authority in the first centuries of Islam, by P. Crone and M. Hinds, in jras 2, 320–321. Review of The history of al-Ṭabarī (Taʾrīkh al-rusul wa’l-mulūk). Vol. 27: The Abbāsid revolution, trans. G. Saliba; Vol. 38: The return of the caliphate to Baghdad, trans. F. Rosenthal, in jras 119, 321–322. Review of Humanism in the renaissance of Islam: The cultural revival during the Buyid age, by J.L. Kraemer, in jras 119, 329–330. 1988 Review of Meccan trade and the rise of Islam, by P. Crone, in Middle East Studies Association Bulletin 22, 54–55. 1989 Review of De l’iqta’ étatique à l’iqta’ militaire. Transition économique et changements sociaux à Baghdad, 247–447 de l’/hégire/861–1055 Ap. J, by H. Abdallah, in The English historical review 104, 450–451. 1990 Review of Christian martyrs in Muslim Spain, by K.B. Wolf, in The journal of ecclesiastical history 41, 335. Review of Damas au viie/xiiie siècle: Vie et structures religieuses d’une métropole islamique, by L. Pouzet, in jras 122, 388–389.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxxvii

Review of The history of al-Ṭabarī (Taʾrīkh al-rusul wa’l-mulūk). Vol. 2: Prophets and patriarchs, trans. W.M. Brinner; Vol. 4: The ancient kingdoms, trans. M. Perlmann, annotations of Iranian names and terms by S. Shaked, in jras 122, 151–152. Review of The history of al-Ṭabarī (Taʾrīkh al-rusul wa’l-mulūk). Vol. 32: The reunification of the ʿAbbāsid caliphate, trans. C.E. Bosworth; Vol. 37: The ʿAbbāsid recovery, trans. P.M. Fields, annotated by J. Lassner, in jras 122, 152–153. 1991 Review of The Arab conquest of Spain, 710–797, by R. Collins, in History 76, 483– 484. Review of Arabic historical thought in the classical period, by T. Khalidi, in bsoas 60, 131–132. Review of The early ʿAbbasid empire, by al-Tabari; The meadows of gold, by Massoudi, in History 76, 482–483. 1992 Review of Early Islam: Collected articles, by M.W. Watt, in jras 2, 435. Review of An early Islamic family from Oman: Al-Awtabi’s account of the Muhallabids, by M. Hinds, in bjmes 19, 202–203. Review of A history of the Arab peoples, by A. Hourani, in jras 2, 434. Review of The rise of humanism in classical Islam and the Christian West with special reference to scholasticism, by G. Makdsis, in jras 2, 272–273. Review of State and society in Fatimid Egypt, by Y. Lev, in Times literary supplement 4659, 24. Review of Tribes, government and history in Yemen, by P. Drescher, in bjmes 19, 218–219. 1993 Review of Abū’l-Ḥusain al-Rāzī (–347/958) Und Seine Schriften: Untersuchungen Zur Frūhen Damaszener Geschitsschreibung, by G. Conrad, in jras 3, 260– 261. Review of al-Majdal ilà taqwīm al-lisān wa-taʿlīm al-bayān, by Ibn Hišām alLajmī, ed. J.P. Lazaro, 2 vols, in jras 3, 262–263. Review of Displaying the Orient: Architecture of Islam at nineteenth-century world’s fairs, by Z. Celik, in ahr 98, 1,573–1,574. Review of The Fatimid vizierate 969–1172, by L.S. al-Imad, in jesho 36, 287–289. Review of Moorish Spain, by R. Fletcher, in History 78, 494–495.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xxxviii

bibliography of hugh n. kennedy

1994 Review of A history of Palestine, 634–1099, by M. Gil, in bsoas 57, 233. Review of The Seljuks of Anatolia: Their history and culture according to local Muslim sources, by M.F. Köprülü and G. Leiser, in History 79, 279. In Aleppo once, review of Monuments of Syria: A historical guide, by R. Burns; The arts and crafts of Syria: Collection Antoine Touma and Linden-Museum, Stuttgart, by J. Kalter, M. Pavaloi, and M. Zerrnickel, in Times literary supplement 4738, 28. Obsessed with embroidery, review of The Afghan amulet: Travels from the Hindu Kush to Razgrad, by S. Paine, in Times literary supplement 4740, 27. 1995 Review of The history of the Saffarids of Sistan and the Maliks of Nimruz (247/861 to 949/1542–3), by C.E. Bosworth, in jras 5, 427–428. Review of The legacy of Muslim Spain, by S.K. Jayyusi, in jras 5, 107–108. Review of A Mediterranean emporium: The Catalan kingdom of Majorca, by D. Abulafia; Byzantium and the Crusader state, 1096–1204, by R-J. Lilie, in History today 45, 56–57. Review of The medieval Spains, by B.F. Reilly, in History 80, 466. with I. el-Sakkout, review of Ansāb al-ashrāf. Vol. 6 B. by Aḥmad b. Yaḥyā b. Jābir al-Balādhurī, ed. K. Athamina, in jras 5, 410–413. Why the Muslims won, review of Byzantium and the early Islamic conquests, by W.E. Kaegi (ed.), in Times literary supplement 4794, 26. 1996 Review of The art and architecture of Islam, 1250–1800, by S.S. Blair and J.M. Bloom, in ahr 101, 532–533. Review of Byzantium: The decline and fall, by J.J. Norwich, in New York Times book review, 10. 1997 Review of Der Herausbildungsprozess des arabisch-islamischen Staates: Eine quellenkritische Untersuchung des Zusammenhangs zwischen den staatlichen Zentralisierungstendenzen by A. Ibrahim, in jesho 40, 305. Review of The early Arabic historical tradition: A source-critical study, by A. Noth and L.I. Conrad, in jss 42, 186. Review of Nachbarn, Bundnispartner: “Themen Und Formen” der Darstellung christlich-muslimischer Begegnungen in ausgewahlten historiographischen Quellen des islamischen Spanien, by B. Münzel, in Hispania 57, 848–849. Review of Trade and traders in Muslim Spain: The commercial realignment

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xxxix

of the Iberian peninsula, 900–1500, by O.R. Constable, in History 82, 288– 289. The descendants of Goliath, review of The Berbers, by M. Brett and E. Fentress, in tls 4913, 8. 1998 Review of The emergence of the ʿAbbasid autocracy: The ʿAbbasid army, Khurasan and Adharbayjan, by F. Amabe, in bjmes 25, 304–305. Review of Hierarchy and egalitarianism in Islamic thought, by L. Marlow, in Middle East journal 52, 627–628. Review of Religion and politics under the early ʿAbbasids: The emergence of the proto-Sunni elite, by M.Q. Zaman, in Islam and Christian-Muslim relations 9, 244. Review of The succession to Muhammad: A study of the early caliphate, by W. Madelung, in jras 8, 88–89. Review of Syria: Cradle of civilisations, by A. Cheneviere, in jras 8, 267. Rivals to the freedom of the seas, review of The blood-red Arab flag: An investigation into Qasimi piracy, 1791–1820, by C.E. Davies, in Times literary supplement 4979, 25. 1999 Review of Faḍāʾil bayt al-maqdis wa al-khalīl wa-faḍaʾil al-shām, by al-Maqdisī, ed. O. Livne-Kafri, in bsoas 62, 206. Review of The late Christian communities of Palestine, by R. Schick, in Journal of Roman archaeology 12, 813–814. Review of State and provincial society in the Ottoman empire: Mosul, 1540–1834, by D. Khoury, The English historical review 114, 447–448. Review of Syrian Ismailism, by N.A. Mirza, in History 84, 137. 2000 Review of The Arabic literary heritage: The development of its genres and criticism, by R. Allen, in jras 10, 227–228. Review of The theory and the practice of market law in medieval Islam: A study of Kitāb al-niṣāb al-iḥṭisāb of ʿUmar b. Muḥammad al-Sunāmī, by M.I. Dien, in jras 10, 94–95. 2002 Review of The breaking of a thousand swords: A history of the Turkish military of Samarra (ah200–275/815–889ce), by M.S. Gordon, in jras 12, 189–190. Review of The Crusades: Islamic perspectives, by C. Hillenbrand, in Asian affairs 33, 259. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

xl

bibliography of hugh n. kennedy

Review of Empire and elites after the Muslim conquest, by C.F. Robinson, History 97, 392. Review of Le château, by M. Bur, in Speculum 77, 886–887. My neighbour’s sins, review of Commanding right and forbidding wrong in Islamic thought, by M. Cook, in Times literary supplement 5167, 5. 2003 Review of The Crusades from the perspective of Byzantium and the Muslim world, by A. Laiou and R. Mottahedeh (eds.), in Byzantine and modern Greek studies 27, 315–316. Review of The formation of Turkey: The Seljukid sultanate of Rum, eleventh to fourteenth centuries, by C. Cahen, in The English historical review 118, 189–190. Review of Exploring an Islamic empire: Fatimid history and its sources, by P.E. Walker, in History 88, 479–480. Review of Unknown Crusader castles, by K. Molin, in History 88, 481–482. Review of White banners: Contention in ‘Abbasid Syria, 750–880, by P.M. Cobb, in jss 48, 397–398. 2004 Review of The templar of Tyre: Part iii of the deeds of the Cypriots, by P. Crawford, in The English historical review 119, 1,390–1,391. 2008 Review of Founding the Fatimid state: The rise of an early Islamic empire. An annotated English translation of al-Qāḍī al-Nuʿmān’s Ifititāḥ al-daʿwa by H. Haji, in Early medieval Europe 16, 107–108. 2011 Review of The hadīth, by M. Shah, in jqs 13, 110–112. Review of Samarcande et Samarra: Élites d’Asie Centrale dans l’Empire Abbaside, by E. de la Vaissiere, in bsoas 74, 152–153. 2016 Review of Medieval Muslim historians and the Franks in the Levant, by A. Mallett, in The English historical review 131, 641–642. 2018 Review of Longing for the lost caliphate: A transregional history, by M. Hassan, in The English historical review 133, 1,266–1,268.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bibliography of hugh n. kennedy

xli

2019 Review of The merchant of Syria: A history of survival, by D. Darke, in Maghreb review 44, 278. Review of Violence in Islamic thought from the Mongols to European imperialism, by R. Gleave and I.T. Kristó-Nagy, in Maghreb review 44, 281. Review of The works of Ibn Wāḍiḥ al-Yaʿqūbī: An English translation. Volume 1–3, by M.S. Gordon et al. (eds.), in bsoas 82, 356–358. 2020 Review of Non-Muslim provinces under early Islam: Islamic rule and Iranian legitimacy in Armenia and Caucasian Albania, by A. Vacca, in jis 31, 122–123.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

part 1 Caliphate and Power

∵

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 1

A Ḥimyarite Restorationist Prophecy Michael Cook

The city of Ḥimṣ, as Hugh Kennedy has written, was probably the first in Syria to have a substantial Muslim population.*,1 Regrettably, no work by an early historian of the city is extant,2 but it is our good fortune that a rich and early collection of apocalyptic traditions, the Fitan of Nuʿaym b. Ḥammād al-Marwazī (d. 228/843), survives to this day and contains a mass of material that sheds a vivid light on the hopes and fears of the inhabitants of the city. This material has been studied to good effect by Wilferd Madelung.3 Among the traditions included in the work is one ascribed to a prominent early authority, Kaʿb alAḥbār (d. 34/654–655). In one printed edition of the work it runs as follows (I footnote points where the editor diverges from the unique British Library manuscript):4 qāla:5 li-mani l-mulku Ẓafāri? qāla: li-Ḥimyari l-akhbāri,6 qīla: li-mani lmulku Ẓafāri? qāla lil-Ḥabashi l-shirāri, qīla: li-mani l-mulku Ẓafāri? qāla: li-Fārisa l-aḥrāri, qīla: li-mani l-mulku Ẓafāri? qāla: li-Qurayshi ttijārin,7 qīla: li-mani l-mulku Ẓafāri? qāla: ilā Ḥimyari biḥārin, wa-qāla l-Ḥakamu: li-Ḥimyari l-biḥāri.8

* My thanks to Yaara Perlman for a sharp-eyed reading of this chapter that saved me from a number of errors and persuaded me that the meter of the poem is best taken as catalectic. 1 Kennedy, Great Arab conquests 86. 2 For the works that once existed, see Rosenthal, History of Muslim historiography 333, 391. 3 Madelung, Apocalyptic prophecies in Ḥimṣ, with a survey of the population of early Muslim Ḥimṣ at 141–143. 4 Nuʿaym b. Ḥammād, Fitan, ed. M. b. M. b. S. al-Shūrā 76 no. 285 (my references to this work are to this edition unless otherwise stated). I owe my knowledge of this tradition and my sense of its interest to Michael Payne, who included it in a paper he presented at the mehat conference in Chicago in 2015. 5 The qāla of the printed text reads qāla qīla in the manuscript (Nuʿaym b. Ḥammād, Fitan f. 27b). 6 The manuscript reads al-akhyāri. 7 I take the liberty of disregarding the hamza placed on the initial alif by the editor; it is not in the manuscript, and reading atjār does not help. 8 There is a brief parallel sharing much of the isnād in Nuʿaym b. Ḥammād, Fitan f. 112b, almost unrecognizable in the printed text cited above, 288–289 no. 1145.

© Michael Cook, 2022 | doi:10.1163/9789004525245_002

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

4

cook

Here, al-Ḥakam is one of the traditionists from whom Nuʿaym has the tradition; we will come to the complete chain of transmission in a moment. At this point a rather painfully literal translation of what we have in front of us might look like this: He [Kaʿb] said: To whom is the kingdom, Ẓafār? He [Kaʿb] said: To Ḥimyar of the reports. It was said: To whom is the kingdom, Ẓafār? He [Kaʿb] said: To the evil Abyssinians. It was said: To whom is the kingdom, Ẓafār? He [Kaʿb] said: To the noble Persians. It was said: To whom is the kingdom, Ẓafār? He [Kaʿb] said: To Quraysh of trading. It was said: To whom is the kingdom, Ẓafār? He [Kaʿb] said: To Ḥimyar of seas; al-Ḥakam said: To Ḥimyar of the seas. The site of Ẓafār is located near Yarīm,9 between Taʿizz and Ṣanʿāʾ; as the ancient Ḥimyarite capital, it fits well in a tradition about the Ḥimyarites, the South Arabian people whose state dominated Yemen for several centuries before the rise of Islam. It is, however, awkward to have to take Ẓafāri as in apposition to almulk,10 and in general the text does not make much sense. Indeed some of it is plain nonsense. How can we improve things? One nonsensical item is easily rectified: “Ḥimyar of the reports.” Ḥimyar al-akhbāri should, of course, be Ḥimyar al-akhyāri, so that the good Ḥimyarites match the bad Abyssinians. And this indeed is the reading found in the manuscript, and also in Suhayl Zakkār’s edition.11 Adopting this reading further provides us with a clue to the point of the tradition: it is pro-Ḥimyarite. We can also ease things a little if we take account of the other point where the printed text reproduced above diverges from the manuscript: Kaʿb does not ask the question (qāla: li-mani l-mulku Ẓafāri?), he merely reports it (qāla: qīla: limani l-mulku Ẓafāri?). Another improvement in our understanding of the tradition follows from this with a little thought. Kaʿb apparently gives no less than five different answers to the question, naming four distinct ethnic groups. While there is no indication in the text that these answers refer to successive historical periods, the first four would fit that assumption well: first, the indigenous Ḥimyarites rule, then the Abyssinian invaders, then the Persian invaders, then the Qurashī caliphs (be they patriarchal, Umayyad, or Abbasid). All this is historically accurate. The fifth answer would then relate to a future as yet unrealized: a Ḥimyarite restoration. 9 10 11

Müller, Ẓafār. The masculine verbs (qāla) preclude our understanding it as a vocative. Nuʿaym b. Ḥammād, Fitan, ed. Suhayl Zakkār 67.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

5

a ḥimyarite restorationist prophecy

This understanding would fit well with most of the chain of transmission by which Nuʿaym has the tradition. It runs as follows: ʿUthmān [b. Saʿīd] b. Kathīr (Ḥimṣī, Qurashī, Umayyad mawlā, d. 209/ 824–825)12 and al-Ḥakam b. Nāfiʿ (Ḥimṣī, Bahrānī, mawlā, d. 221/835–836)13 from Saʿīd b. Sinān (Ḥimṣī, Ḥanafī or Kindī, d. 163/779–780)14 from al-Walīd b. ʿĀmir al-Yazanī (Ḥimyarite)15 from Yazīd b. Khumayr (Ḥimṣī, Yazanī, Ḥimyarite)16 from Kaʿb [al-Aḥbār] (Ḥimṣī, Ḥimyarite, d. 34/654–653)17 The transmitters are thus typically Ḥimṣīs of South Arabian, or more specifically Ḥimyarite, descent. It should thus be no surprise that they transmit a tradition that describes the Ḥimyarites as good and rewards them with an eventual restoration of their rule. Our tradition can indeed be compared with another that likewise foretells a Ḥimyarite restoration, though in a quite different wording:18 kāna hādhā l-amru fī Ḥimyara, fa-nazaʿahu llāhu taʿālā minhum wa-ṣayyarahu fī Qurayshin wa-sa-yaʿūdu ilayhim. This matter [rulership] was in Ḥimyar, but God took it away from them and transferred it to Quraysh; it will return to them [Ḥimyar]. This is subversive, as can be seen from the way Ibn Ḥanbal (d. 241/855) is said by his son to have handled the last part of the tradition in his Musnad:19 in his book (kitāb) he took the precaution of spelling out wa-sa-yaʿūdu ilayhim with 12 13

14 15

16 17 18

19

Al-Mizzī, Tahdhīb xix, 377, 379. Al-Mizzī, Tahdhīb vii, 146, 154. Bahrāʾ is part of Quḍāʿa, and thus on one view southern (see Caskel, Ğamharat an-nasab i, tables 279 and 328; for the disagreement on the question, see Kister, Ḳuḍāʿa 315). Al-Mizzī, Tahdhīb x, 495, 498. Al-Mizzī, Tahdhīb, has no entry on him, but see x, 495, and xxxii, 119; there is, for example, an entry on him in al-Bukhārī, al-Taʾrīkh al-kabīr, iv, ii, 149, but it does not help. Dhū Yazan is a clan of Ḥimyar, see Caskel, Ğamharat an-nasab i, tables 274, 277, and 278. Al-Mizzī, Tahdhīb xxxii, 119; see also xxiv, 190. Al-Mizzī, Tahdhīb xxiv, 189, 193; he is a Ḥimṣī, see 190. Nuʿaym b. Ḥammād, Fitan, 275–276 no. 1091; the isnād is of the same Ḥimṣī and southern (sometimes specifically Ḥimyarite) character as that of our tradition. This tradition is translated in Madelung, Apocalyptic prophecies in Ḥimṣ 151. Ibn Ḥanbal, Musnad iv, 91.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

6

cook

each letter written separately (muqaṭṭaʿ), though when he transmitted the tradition orally he pronounced the words in the normal fashion (takallama ʿalā l-istiwāʾ). Sometimes the tendentious part of this tradition is simply omitted.20 Against all this we can set, for example, an antithetical tradition ascribed to ʿAlī b. Abī Ṭālib (d. 40/661), and obviously aiming to counter such subversive speculation:21 laysa baʿda Qurayshin illā l-Jāhiliyya. After Quraysh [sc. the Qurashī caliphs] there is only the [future] Jāhiliyya. We now understand the point of the tradition, but we have yet to resolve a number of difficulties. Can we really assume that the tradition spells out a succession of ethnic dominations? There is nothing in its wording to indicate a temporal sequence. Is it really Kaʿb who answers the question? Since he has reported the question, we would expect him to report the answers, not supply them himself. Is Ẓafār really in apposition to “the kingdom”? Here, as already mentioned, the wording of the question is awkward: we might expect Kaʿb to ask “To whom is the kingdom of Ẓafār?” (li-man mulku Ẓafāri?), rather than “To whom is the kingdom, Ẓafār?” (li-mani l-mulku, Ẓafāri?).22 In fact, the first is exactly what we find in a version of the tradition quoted by Yāqūt (d. 626/1229) in his entry on Ẓafār in his geographical dictionary, and also found in his likely source, Ibn Khurradādhbih (third/ninth century).23 Also, can ittijār really be the right reading? The reference to trade in the context of Quraysh is certainly in place, but we would expect “Quraysh the merchants,” not “Quraysh of trade.” And finally, what sense could we hope to make of “Ḥimyar of (the) seas”? We can quickly solve several of these problems if we turn to a version of our prophecy quoted by al-Masʿūdī (d. 345/956). He reports an inscription allegedly written on the gate of Ẓafār, in the ancient script, on a black stone; the text, in Pellat’s edition, runs as follows:24 20 21 22 23

24

See, for example, al-Ṭabarānī, Musnad al-Shāmiyyīn ii, 135. Nuʿaym b. Ḥammād, Fitan 275 no. 1086; compare also 274–275 no. 1085, according to which rulership will remain in Quraysh as long as two men are left alive. For the invariant vocalization of Ẓafāri, see Wright, Grammar of the Arabic language i, 243–244. Yāqūt, Muʿjam al-buldān iv, 60b: li-man mulku Ẓafāri; Ibn Khurradādhbih, al-Masālik wal-mamālik 145. That Ibn Khurradādhbih is the source, direct or indirect, of Yāqūt’s version is suggested by the ending of the tradition, to be discussed below. Al-Masʿūdī, Murūj al-dhahab ii, 211; I omit the last two lines, which have no bearing on the text of our tradition.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

7

a ḥimyarite restorationist prophecy

yawma shīdat Ẓafāri qīla li-man ʾan- / ti fa-qālat li-Ḥimyara l-ʾakhyār thumma sīlat25 mā baʿda dhāka fa-qālat / ʾinna mulkī lil-ʾAḥbushi lʾashrār thumma sīlat mā baʿda dhāka fa-qālat / ʾinna mulkī li-Fārisa l-ʾaḥrār thumma sīlat mā baʿda dhāka fa-qālat / ʾinna mulkī ʾilā Qurayshi l-tijār26 thumma sīlat mā baʿda dhāka fa-qālat / ʾinna mulkī li-Ḥimyarin27 saḥḥār. At the time Ẓafār was built, she was asked “To whom do you belong?”; she replied “To the good Ḥimyarites.” Then she was asked “What then?” and replied “My kingdom goes to the evil Abyssinians.” Then she was asked “What then?” and replied “My kingdom goes to the noble Persians.” Then she was asked “What then?” and replied “My kingdom goes to Quraysh the merchants.” Then she was asked “What then?” and replied “My kingdom goes to the Ḥimyarites, a sorcerer.” With this we get some illumination. First, the temporal succession of ethnic dominations is now explicit.28 Second, it is not Kaʿb who answers the question; it is put to the city of Ẓafār, and each time it is asked the city itself replies to it. Third, this means that Ẓafāri is not in apposition to “the kingdom” but rather is in the vocative. Fourth, “Quraysh the merchants” fits better than “Quraysh of trade.” In addition to all this, we can now see that the tradition is in verse—and it scans (the meter is khafīf ).29 The major problem we are left with is the nonsense of the final reference to “the Ḥimyarites, a sorcerer (saḥḥār).” It will be obvious that this is a variant of the equally nonsensical “Ḥimyar of (the) seas.”30 Clearly there was something at this point in the original tradition that was easily misconstrued. We could 25 26 27 28

29 30

For sīla as a form of suʾila, see Lane, Arabic-English lexicon 1283a. A plural of tājir, see Lane, Arabic-English lexicon 297c. For the nunation of diptotes through poetic license, see Wright, Grammar of the Arabic language ii, 387–388. A succession of thummas is likewise found in a brief parallel passage in a report of a Ḥimyarite inscription on a tomb, see Nuʿaym b. Ḥammād, Fitan f. 105b, corresponding to 276 no. 1092 (on this tradition, see Madelung, Apocalyptic prophecies in Ḥimṣ 151 n. 47; much of its text is either Ḥimyarite or gibberish). Wright, Grammar of the Arabic language ii, 367–368 (trimeter, catalectic). Unless we were to link the latter to the statement of Procopius that the Ḥimyarites

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

8

cook

improve things a little by discarding Pellat’s vocalization saḥḥār in favor of suḥḥār, plural of sāḥir: “the Ḥimyarites, sorcerers.”31 But why would a tradition promising a Ḥimyarite restoration want to describe the Ḥimyarites in this opprobrious fashion? What we would expect Ẓafār to say in her final answer is that her kingdom would return to the Ḥimyarites, just as in the parallel tradition quoted above (wa-sa-yaʿūdu ilayhim). The simplest way to resolve this issue is to go back to the entry on Ẓafār in Yāqūt’s geographical dictionary.32 When we looked at it before, we saw a corruption introduced by a copyist who did not realize that the occurrences of Ẓafāri in the tradition were in the vocative.33 But at the point that concerns us here, Yāqūt’s text is superior, giving what must be the original reading: liḤimyara sa-yuḥārū, “it will be returned to Ḥimyar.”34 The verb ḥāra means to return, and it has a transitive fourth form, aḥāra,35 of which here we have the passive. Clearly the verb was selected for the sake of the rhyme.36 It is not a common verb, which is why it is immediately followed by a gloss for the benefit of less erudite readers: “that is to say, it will return to Yaman” (ay yarjiʿu ilā l-Yaman).37 The rarity of this verb also explains the tendency of copyists and editors to corrupt it. So the last line and its translation should be as follows:

31 32

33 34

35 36

37

had a port from which they sailed to Ethiopia (Power, Red Sea 49, citing Procopius, History of the wars i, 182, 183). The old Paris edition has a shadda on the ḥāʾ but no vocalization (al-Masʿūdī, Murūj aldhahab, iii, 178). Yāqūt, Muʿjam al-buldān iv, 60b. In the report of the tomb inscription mentioned above, we find ḥāra maḥār at the corresponding point, see Nuʿaym b. Ḥammād, Fitan f. 105b, corresponding to 276 no. 1092; the phrase also occurs in the variant version of the tradition found in Ibn Durayd, Jamharat al-lugha ii, 311b (where in place of Ẓafār we find Dhamār), cited in Madelung, Apocalyptic prophecies in Ḥimṣ 151 n. 47. This is not the only corruption: we also find here li-Fārisa l-aḥbār, depriving the Persians of their nobility and reducing them to the rank of scholars. So in the Beirut printing of 1957. Wüstenfeld’s text at this point preserves the correct consonantal skeleton, but points it as li-Ḥimyara sa-tujār (Yāqūt, Muʿjam al-buldān iii, 577). The text of Ibn Khurradādhbih has li-Ḥimyarin yuḥār (al-Masālik wa-l-mamālik 145), whereas that of Yāqūt, doubtless dependent on a more correct text of Ibn Khurradādhbih than ours, preserves the sa-. Lane, Arabic-English lexicon 665. If we took the meter to be acatalectic, the -rū would rhyme imperfectly with the -rī of the previous lines, though for examples of this fault, see Wright, Grammar of the Arabic language ii, 357. This gloss is not the work of Yāqūt, since it already appears in Ibn Khurradādhbih, alMasālik wa-l-mamālik 145 (li-Ḥimyarin yuḥār, ay yarjiʿu ilā Ḥimyar).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a ḥimyarite restorationist prophecy

9

thumma sīlat mā baʿda dhāka fa-qālat / ʾinna mulkī li-Ḥimyarin sa-yuḥār. Then she was asked “What then?” and replied “My kingdom will return to the Ḥimyarites.” This leaves a couple of more general observations to be made. First, I have assumed that al-Masʿūdī’s version, with its last line corrected in light of Yāqūt’s, is broadly representative of the original text of the prophecy, subsequently corrupted in versions such as Nuʿaym’s. Chronologically this may seem a little strange: Nuʿaym composed his book long before al-Masʿūdī or Yāqūt. Particularly striking in this connection is the indication at the end of the tradition that gross corruption may already have occurred even before the time of Nuʿaym: al-Ḥakam’s al-biḥār, with the insertion of the definite article, is the work of a transmitter who was rightly puzzled by biḥār but was unable to restore the original text. In other words, it would seem that sa-yuḥār had already fallen casualty to corruption in second/eighth-century Ḥimṣ. In principle it could, of course, be the case that al-Masʿūdī and Yāqūt, or their sources, were seeking to impose a secondary appearance of order on primary chaos. But on the whole it seems more likely that they were preserving rather than inventing a text that made sense. This derives some support from an account of an inscription—or the same inscription—from Ẓafār that is preserved in a very early source, the biography of Muḥammad by Ibn Isḥāq (d. 150/767–768) as transmitted by the Kufan Yūnus b. Bukayr (d. 199/814–815). Here, too, the inscription is said to have been written in the ancient script (al-kitāb al-awwal), and it runs as follows:38 li-mani l-mulku, Ẓafāri? li-Ḥimyara l-akhyāri, li-mani l-mulku Ẓafāri? liFārisi l-akhyāri, li-mani l-mulku Ẓafāri? li-Qurayshin il-tijāri To whom is the kingdom, Ẓafār? To the good Ḥimyarites. To whom is the kingdom, Ẓafār? To the good Persians. To whom is the kingdom, Ẓafār? To Quraysh the merchants. Here, there is no sign of the Abyssinians, or of the return of the kingdom to Ḥimyar, and the duplication of al-akhyāri is inelegant. But this passage shows at least that the jingle goes back to the second/eighth century. The other general observation concerns the text of Nuʿaym’s Fitan. If the small sample probed in this chapter is anything to go by, we are dealing with a

38

Ibn Isḥāq, Kitāb al-Siyar 55.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

10

cook

manuscript that abounds in textual corruptions. In other words, we are much in need of a critical edition of the book, as opposed to one that merely prints the manuscript more or less accurately—though such editions, and translations based on them, are nevertheless a major service to scholarship.39 Preparing a critical edition would undoubtedly involve a great deal of editorial labor. Fortunately, we are much better placed to do it today than our predecessors were, thanks to our access to databases of Arabic texts.40 But a scholarly edition of the Fitan would still be a daunting task.

Bibliography Sources al-Bukhārī, al-Taʾrīkh al-kabīr, 4 vols. each in two parts, Hyderabad 1360–1378. Caskel, W., Ğamharat an-nasab: Das genealogische Werk des Hišām ibn Muḥammad alKalbī, 2 vols., Leiden 1966. Cook, D., “The book of tribulations”: The Syrian Muslim apocalyptic tradition, Edinburgh 2017. Ibn Durayd, Jamharat al-lugha, 4 vols., Hyderabad 1345. Ibn Ḥanbal, Musnad, 6 vols., Būlāq 1313. Ibn Isḥāq, Kitāb al-Siyar wa-l-maghāzī, ed. Suhayl Zakkār, Beirut 1978. Ibn Khurradādhbih, al-Masālik wa-l-mamālik (bga 6), ed. M.J. de Goeje, Leiden 1889. al-Masʿūdī, Murūj al-dhahab, eds. C. Barbier de Meynard and Pavet de Courteille, 9 vols., Paris 1861–1877. al-Masʿūdī, Murūj al-dhahab, ed. C. Pellat, 5 vols., Beirut 1965–1974. al-Mizzī, Tahdhīb al-Kamāl, ed. B. ʿA. Maʿrūf, 35 vols., Beirut 1985–1992. Nuʿaym b. Ḥammād, Fitan, ms. British Library, Or. 9449. Nuʿaym b. Ḥammād, Fitan, ed. S. Zakkār, Mecca n.d. Nuʿaym b. Ḥammād, Fitan, ed. M. b. M. b. S. al-Shūrā, Beirut 1997. Procopius, History of the wars, trans. H.B. Dewing, 7 vols., Cambridge, MA 1914–1940. al-Ṭabarānī, Musnad al-Shāmiyyīn, ed. Ḥ. ʿA. al-Salafī, 4 vols., Beirut 1989–1996. Yāqūt, Muʿjam al-buldān, ed. F. Wüstenfeld, 6 vols., Leipzig 1866–1873. Yāqūt, Muʿjam al-buldān, 5 vols., Beirut 1957.

39

40

Since this chapter was drafted a useful translation of the entire work has been published by David Cook (Cook, “Book of tribulations”; for our tradition, see 51 no. 248; and for the parallel with the repetition of thumma, see 220 no. 1117). In this way many more versions of our tradition can be found than I have cited in this chapter. I have confined myself to what I hope is a strategic selection.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a ḥimyarite restorationist prophecy

11

Studies Kennedy, H., The great Arab conquests: How the spread of Islam changed the world we live in, London 2007. Kister, M.J., Ḳuḍāʿa, in ei2, v, 315–318. Lane, E.W., An Arabic-English lexicon, London 1863–1893. Madelung, W., Apocalyptic prophecies in Ḥimṣ in the Umayyad age, in jss 31 (1986), 141–185. Müller, W.W., Ẓafār, in ei2, xi, 379–380. Power, T., The Red Sea from Byzantium to the caliphate: ad500–1000, Cairo 2012. Rosenthal F., A history of Muslim historiography, Leiden 1952. Wright, W., A grammar of the Arabic language, 2 vols., Cambridge 1981.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 2

Kinship, Dynasty, and the Umayyads Andrew Marsham 1

Introduction

Like many other undergraduate students, I first encountered the Umayyad dynasty while reading Hugh Kennedy’s The Prophet and the age of the caliphates. Since then, I have learned much more about Umayyad history from Hugh Kennedy and from his published work—especially foundational aspects of the Umayyads’ power, including the army, taxation, military pay, and land tenure, but also historiography and genealogy.1 This chapter, which is dedicated with much gratitude to Hugh, presents some preliminary conclusions about the biology and the social practices that shaped the Umayyad dynasty.2 “The Umayyad dynasty” is of course shorthand for a more complicated historical reality. Descendants of Umayya b. ʿAbd Shams claimed authority over the new empire formed by the Arabian conquests for most of the 106 solar years between 23/644 and 132/750—fairly successfully for about 90 of them. However, the Banū Umayya’s claims were never universally, or perhaps even widely, acknowledged, nor were they ever a group that was unified or cohesive in its political action. The chapter takes as a starting point the importance of succession in a patrimonial context, where claims to legitimate leadership could be made by many leading members of a kin group, rendering its members both collaborators and competitors, often at the same time. It examines the role of

1 On genealogy and historiography, see Kennedy, From oral tradition. I would like to thank the editors for the kind invitation to contribute the chapter, to acknowledge an Arts and Humanities Research Council grant that supported some of the initial research (grant no. 1026731/1), and to thank colleagues at the 2019 Society for the Promotion of Byzantine Studies Spring Symposium, “Blood in Byzantium,” for comments and discussion. 2 On dynasties in world history, but with a focus on the later medieval and early modern periods, see Duindam, Dynasties; and Duindam, Dynasty. For a discussion of dynastic politics in the Islamic Middle East down to the early twenty-first century, see McMillan, Fathers and sons. © Andrew Marsham, 2022 | doi:10.1163/9789004525245_003 © Andrew Marsham, 2022distributed | doi:10.1163/9789004525245_003 This is an open access chapter under the terms of the CC BY-NC 4.0 license.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

13

women and marriage in alliances in the political economy and highlights some of the ways that biology and human frailty impacted the way Umayyad dynastic politics played out. After a short summary of the history of the dynasty, some of the main conclusions are presented before a more detailed survey of the marriage and reproductive patterns of the Umayyad rulers from ʿUthmān (r. 23–35/644–656) to Hishām (r. 105–125/724–743), as well as three of their ancestors, Abū l-ʿĀṣ b. Umayya (fl. ca. 600), his son al-Ḥakam b. Abī l-ʿĀṣ (d. ca. 31/651–652), and his nephew Abū Sufyān b. Ḥarb b. Umayya (d. 32–34/653–655).3 These three ancestors are included because the genealogical material shows that their marriage and reproductive choices were an important foundation for the later success of their progeny.

2

The Umayyad Dynasty

Two Umayyad patriarchs are famous as dynasts, Muʿāwiya b. Abī Sufyān (r. ca. 41–60/661–680) and Marwān b. al-Ḥakam (r. 64–65/684–685), whence the “Sufyanid” and “Marwanid” Umayyads. Before them another Umayyad, ʿUthmān b. ʿAffān (r. 23–35/644–656), had also led the West Arabian conquerors (see fig. 2.1).4 ʿUthmān’s Umayyad successors’ claims were often grounded in their shared inheritance of legitimacy from him.5 There are hints in the later tradition of ʿUthmān himself having had plans for the succession prior to his death in 35/656. As discussed below, his marriages and reproductive choices do indeed suggest that he had dynastic ambitions. Muʿāwiya’s success as a dynast was limited; his son Yazīd (r. 60–64/680– 683) succeeded him but immediately faced opposition and then died young, a little under 40, as he fought widespread opposition to his rule.6 Yazīd’s son Muʿāwiya b. Yazīd (r. 64/683–684) also died young, at about 20 years of age, and his brother Khālid b. Yazīd then lost out to the claims of a second cousin twice removed, Marwān b. al-Ḥakam.7

3 Keshk, Abū Sufyān. 4 Summaries of this history include Kennedy, Prophet 43–106; Hawting, First dynasty. 5 See for example panegyrics by al-Akhṭal that glorify the descendants of Abū l-ʿĀṣ, who include both ʿUthmān and the Marwānids and refer to them as “the protection of ʿAbd Shams”; alAkhṭal, Sharḥ dīwān al-Akhṭal 189, 287. 6 Hawting, Yazīd (i). 7 Bosworth, Muʿāwiya (ii); Ullmann, Khālid b. Yazīd.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

marsham

figure 2.1 ʿUthmān and the Sufyanid and Marwanid Umayyad caliphs a. marsham

14

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

15

Marwān may have been about 60 years old at his accession in 684. He died within about 18 months and was succeeded by one of his sons, ʿAbd al-Malik (r. 65–86/685–705), whose long reign allowed him to assert and then tighten his grip on the caliphate and install a series of sons as his successors.8 ʿAbd al-Malik was aided in his dynastic success by the timely death of the paternal half-brother, ʿAbd al-ʿAzīz (d. ca. 85/704), whom Marwān had nominated to succeed ʿAbd al-Malik. ʿAbd al-Malik was then succeeded by—in this order—two sons, a fraternal nephew and son-in-law, two more sons, three grandsons, and another fraternal nephew, before the Abbasid Revolution overtook his dynasty in 132/750. In hindsight this familiar succession of rulers from three branches of the Umayyad descent line appears almost natural, but as with all monarchic successions, it was contingent on the vagaries of politics and biology. The “Great Arab Conquests” were an era of permanent competition over power and authority.9 Important variables in this competition were social and biological: marriage, reproduction, and longevity were some of the main foundations of power in the tribal societies of West Arabia.

3

Summary Conclusions

Because of the character of the prosopographic material upon which this chapter is based, the usual sequence is inverted, and the main conclusions are presented here. The fourth section, “Umayyad marriages and children” follows with the detailed evidence upon which these summary conclusions are founded. The chapter ends with a short fifth section, “Final remarks,” which points to some avenues for further research. The four main conclusions discussed in this third section build on the advances made by Asad Ahmed and Majied Robinson in their prosopographical work on early Islam.10 They concern the social and political roles of women and marriage, the changing patterns in Umayyad marriage, and the political importance of fertility and mortality. First, the role of women in the formation of political networks is foregrounded by this kind of analysis in a way that it is not always in the narrative sources

8 9 10

For Marwān’s age and regnal span, see Bosworth, Marwān (i). On his reign and successors, Robinson, ʿAbd al-Malik, passim; Kennedy, Prophet 78–97; Hawting, First dynasty 46–103. Glubb, Great Arab conquests; Kennedy, Great Arab conquests. Robinson, Marriage; Ahmed, Religious elite.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

16

marsham

and, until recently, still less so in the modern accounts based on them.11 In contrast, many of the genealogical primary sources do pay attention to the place of women in family networks. This interest is retained in a vestigial form in some of the narrative works—that is, the importance of the women was self-evident to the medieval Arabic scholars.12 Because of the patrimonial distribution of power and a culture of women retaining their connections with their original kin group after marriage, paternal uncles (ʿumūm, sing. ʿamm) were often a man’s political rivals, whereas maternal uncles (akhwāl, sing. khāl) could sometimes be crucial allies.13 The importance of maternal connections are also reflected in naming patterns, with a mother’s father often lending his name to her first-born son.14 The extent and character of any agency for the women themselves in these arrangements is less evident because the sources tend to present the women in relation to male forms of authority and power.15 A second conclusion, which likewise builds on the conclusions of Ahmed and Robinson, is that there are discernible patterns in the Umayyads’ marriage arrangements.16 These patterns relate to the social and political context, especially the importance of women in concepts of honor and shame and in forging political alliances. With the changing social and political context of the conquests and then of the political success of the Umayyads, their marriage choices, which were guided by these considerations, changed.

11

12

13 14 15

16

An early exception is Abbott, Women and the state. For recent discussions, see, for example, Urban, Conquered populations; El-Azhari, Queens; Robinson, Marriage; Tayyara, Matrilineal lineages; Ahmed, Religious elite; Ali, Marriage. There is forthcoming work by Leone Pecorini Goodall, Abdulla Haider, and Marie Legendre. Clarke, Heirs and spares, addresses questions of gender and masculinity in elite political culture. Al-Zubayrī’s Kitāb Nasab Quraysh includes much information about women. Al-Balādhurī’s Ansāb often notes maternal relationships. Al-Yaʿqūbī’s Taʾrīkh, which is organized by caliphal reign, notes the maternal ancestry of each caliph, as does al-Ṭabarī’s annalistic Taʾrīkh. Ibn Ḥabīb’s Kitāb al-Muḥabbar is unusual for its focus on matrilineal connections, see Tayyara, Matrilineal lineages. For the possibility that matrilineal connections may have been especially important before Islam and in very early Islam, see Webb, Imagining 195– 205. Tapper and Tapper, Marriage. Marsham, Rituals 120 and n. 43. However, there is scope to do much more with the sources on this question. See for example the female perspective on kinship relations attributed to ʿĀʾisha, “There was never anything in the past between me and ‘Alī other than what usually happens between a woman and her male in-laws […]”; al-Ṭabarī, Taʾrīkh i, 3,231; translation from al-̣Tabarī, History xvi, 170. Hereafter al-Ṭabarī’s Taʾrīkh is abbreviated to Ṭab. Ahmed, Religious elite; Robinson, Marriage.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

17

The status and protection of men’s female blood relatives were crucial to preserving a descent group’s honor in West Arabian culture.17 Ideally, a kin group’s women were to marry high status men who were closely allied with the woman’s own family—this is indeed the pattern of most of the marriages of Umayyad women. Few if any Islamic-era Umayyad women married outside the Quraysh.18 In contrast, men could marry or form sexual unions for a wider variety of reasons, and lower-status brides and slave women were not unsuitable partners. Women who might be perceived as lower status could be important wives for an alliance with their families. For the caliphs and their sons, some lower-status women could also be useful marriage partners because their families could not easily claim the caliphate for themselves. This applied especially to reproductive sexual partnerships with female slaves, who by virtue of their slave status were usually no longer in contact with their kin group. Abū l-ʿĀṣ and his descendants, both before and after Islam, showed a marked preference for wives from allies within ʿAbd Shams at Mecca, and sometimes from the Makhzūm branch of the Quraysh. Beyond Mecca, some of the Thaqīf of neighboring al-Ṭāʾif supplied many of their wives. Likewise, many of Abū Sufyān b. Ḥarb’s marriages were with ʿAbshāmīs, and his branch of the Umayyad clan also had close connections with the Thaqīf. In both these Umayyad descent lines there were also links with the wider group of Kināna (within which the Quraysh was one descent line) and with Daws, remembered as a subgroup of the “southern” Azd.19 It is notable, too, that later, at the end of the seventh century and into the eighth, the Marwanid Umayyads consistently took wives descended from ʿUthmān via two of his sons, ʿAmr and Abān, and from Muʿāwiya via his son Yazīd and granddaughter ʿĀtika bt. Yazīd. These marriages with the descent lines of other Umayyad caliphal lines were intended to confer prestige on the Marwanids’ offspring and secure political support from their maternal relatives. With the conquests, and concomitant power struggles on a much larger stage, there was a shift to greater exogamy in order to cement alliances.20 The 17 18 19

20

See Tapper and Tapper, Marriage; and Schneider, Of vigilance and virgins, which suggest useful models. See also Robinson, Marriage 163–166. On the longstanding relationship between the Quraysh and Thaqīf, see Hosein, Tribal alliance, passim; and on the possible prevalence of links between the ʿAbd Shams and Banū Umayya and Thaqīf, 111–112, 118–119. On the prominence of Makhzūm at Mecca ca. 600 ce, see Hinds, Makhzūm. On Kināna, see Watt, Kināna. On al-Azd, see Webb, al-Azd. It is also notable that Azd and Kināna were among the tribes that did not oppose the Meccans in the so-called “Wars of Apostasy”; Donner, Early Islamic conquests 201. Robinson, Marriage 148–159.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

18

marsham

Banū Kalb of the Syrian frontier became the foundation of Umayyad military power in the seventh century, and this is reflected in a series of marriages with their women. Different branches of the Umayyad clan made marriages with three main branches of Kalb, respectively, in a pattern that reflects competition both within Kalb and within the Umayyad kin-group. Muʿāwiya, ʿUthmān, ʿUthmān’s son-in-law Saʿīd b. al-ʿĀṣ, ʿUthmān’s first cousin Marwān, Marwān’s son ʿAbd al-ʿAzīz and grandson ʿUmar b. ʿAbd al-ʿAzīz, and Marwān’s paternal cousin ʿAmr b. Saʿīd al-Ashdaq all married women from the Banū Kalb.21 Meanwhile, ʿAbd al-ʿAzīz’s half-brother, ʿAbd al-Malik b. Marwān, made extensive marriage alliances with other tribal groups, including Ghaṭafān, Fazāra, and Makhzūm, all of whom had also supplied wives to his ancestors. With the death of ʿAbd al-ʿAzīz in 85/704, before ʿAbd al-Malik, Kalb’s close kinship connection with the caliphate was severed. After al-Walīd, who married into the wider Quraysh, ʿAbd al-Malik’s three other successors among his sons became much more endogamous in their marriage choices, favoring Umayyad women. Only ʿUmar b. ʿAbd al-ʿAzīz’s brief caliphate temporarily restored an Umayyad with Kalbī relatives (notably a grandmother and a wife) to power. Where an alliance was being forged or reinforced, a marriage exchange could take place, with a daughter marrying into the allied family and a woman from that family marrying a corresponding male relative.22 This kind of arrangement tied together the fortunes of the two groups more closely by creating shared interests in the honor of the women and the success of the sons. It could manage the risk of conflict between potential rivals or reinforce the support of an ally. Some examples within the Umayyad descent group include the marriages contracted between the families of ʿUthmān b. ʿAffān and his paternal uncle alḤakam b. Abī l-ʿĀṣ, those between the two ʿAbshāmīs, Muʿāwiya b. Abī Sufyān and ʿAbdallāh b. ʿĀmir b. Kurayz, and those between the half-brothers ʿAbd al-Malik and ʿAbd al-ʿAzīz. There are also prominent women who married a sequence of senior figures, such as ʿAbd al-ʿAziz’s daughter Umm al-Ḥakam, who married al-Walīd, then Sulaymān, and then Hishām. Once political power was secured for one family there was a tendency to narrow the range of marriages to close relatives while at the same time having most children with less well connected or even deracinated women. These two changes freed the ruling descent line from political entanglement with other powerful groups. Majied Robinson has shown that from the 690s on, and after

21 22

For ʿAmr’s marriage, see Ibn Saʿd, Ṭabaqāt v, 237; Ṭab. ii, 787. For Saʿīd’s, see Ibn ʿAsākir, Taʾrīkh lxx, 137. On marriage and alliance, see Landau-Tasseron, Alliances 164.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

19

ʿAbd al-Malik’s sons moved away from marrying outside the most elite Qurashī circles and toward marrying other Umayyads, they chose to have most of their children with slave women (“concubines”), thus ensuring that these children had no maternal relatives.23 Similar marriage and reproduction decisions appear already to have been made in the mid-seventh century by (or for) Yazīd b. Muʿāwiya, who, as the nominated inheritor of power from his father, would have benefitted from the removal of his offspring from other political connections. Third, not just marriage but also fertility was politically important. The founding fathers of dynasties are literally that. The significance of sons is acknowledged in the narrative sources, which have the Prophet prophesy Marwanid success because of the number of al-Ḥakam’s sons.24 Four men among those discussed below stand out for their exceptional fertility, having had more than 20 children each—al-Ḥakam b. Abī l-ʿĀṣ, his paternal nephew ʿUthmān b. ʿAffān, and his son and his grandson, ʿAbd al-Malik and al-Walīd. While their exceptional fertility may in part be an illusion created by sources that record more details about famous people, long-lived and politically successful men probably did tend to have more children; in an environment of political competition between descent groups, having numerous children improved the odds of those children’s success. This was in part simply through creating more chances, but also by providing sons with brothers who could be allies as well as rivals. In the competition between the progeny of al-Ḥakam, ʿUthmān, and Abū Sufyān, al-Ḥakam’s had the advantage of numbers, compounded by ʿUthmān’s childrens’ youth at his death, Muʿāwiya’s infertility, and Muʿāwiya’s son Yazīd’s early death. Indeed, viewed from the perspective of marriage and reproduction, the dynastic ambitions of the third caliph, ʿUthmān b. ʿAffān, can likely be glimpsed. Although ʿUthmān was killed before he could make any arrangements for the succession, his specific marriage and reproductive choices look like those of a man who had political ambition for his offspring, as some of the narrative sources also suggest.25 Unlike his father, ʿAffān b. Abī l-ʿĀṣ, who seems to have had only three children, ʿUthmān fathered more than 20. However, ʿUthmān’s children lost out, first to their older and better-connected second cousin once removed, Muʿāwiya, and then to their first cousins in the line of their father’s paternal half-brother, al-Ḥakam. 23 24 25

Robinson, Marriage 179–186. On this, see also the interesting letter attributed to ʿAbd alḤamīd: ʿAbbās, ʿAbd al-Ḥamīd 193–194. Humphreys, Muʿāwiya 35; Madelung, Succession 342. On these sources, see Madelung, Succession 88–90.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

20

marsham

figure 2.2 The lifespans of the Umayyad caliphs (hatching indicates life before accession; black indicates life as caliph) a. marsham; s. fagg; f. barraud

Fourth, a dynastic weakness for the Marwanid branch was their own misfortune with early mortality. After ʿUthmān, no ruling Umayyad of Islamic times seems to have lived beyond his 60s, and many died in their 40s. This problem was exacerbated by their patrimonial succession arrangements, where power often passed “horizontally” within one generation rather than “vertically” from one generation to another—that is, between paternal uncles from the perspective of the next generation (see fig. 2.2). Of ʿAbd al-Malik’s nine successors, most were short lived, often dying in their 40s. As a consequence, besides the six years of civil war under Marwān ii (r. 744–750), at the end of the Umayyad era, only two of ʿAbd al-Malik’s eight other successors ruled for more than four years. These were al-Walīd b. ʿAbd al-Malik (r. 86–96/705–715) and Hishām b. ʿAbd al-Malik (r. 105/724–743). The critical decade (96–105/715–724) between these two longer caliphates saw the deaths of four caliphs. At this time, resistance from rivals within the conquest elite had been substantially suppressed and victory over the Roman Empire was briefly within reach, but the instability caused by a succession of very short caliphates contributed to the stalling of expansion and the revival of internal conflict. The long reign of Hishām then saw something of a reassertion of Marwanid imperial power, but momentum was not fully regained, and the final unravelling of the Marwanid caliphate had already begun before Hishām’s death in his early 50s.26 26

Ṭab ii, 1,728–1,729 records three ages: 55, 52, and 54. For the political narrative, see Marsham, Umayyad Empire; Hawting, First dynasty 72–119; Kennedy, Prophet 90–106.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

21

In the next section of this chapter, the marriages of the ruling Umayyads down to the time of Hishām are surveyed, together with those of three of their ancestors. The main sources used are the Nasab Quraysh by Abū ʿAbdallāh Muṣʿab b. ʿAbdallāh b. al-Zubayrī (d. ca. 236/851) and the Ansāb al-ashrāf by Aḥmad b. Yaḥyā al-Balādhurī (d. ca. 279/892).27 These have occasionally been supplemented by other sources, such as Ibn Saʿd’s (d. 230/845) al-Ṭabaqāt alkubrā, al-Ṭabarī’s (d. 310/923) Taʾrīkh, and Ibn ʿAsākir’s (d. 571/1176) Taʾrīkh madīnat Dimashq.28 As Watt and Crone observed, and Ahmed and Robinson have shown in more detail, these genealogical materials are consistent and apparently roughly accurate for the most part—only very occasionally is there clear evidence of fabrication.29 At the same time, the available material is highly uneven, with more complete attention paid to key historical figures and a bias toward recording men rather than women.30

4

Umayyad Marriages and Children

4.1 Abū l-ʿĀṣ b. Umayya’s Marriages The paternal grandfather of the future caliphs ʿUthmān and Marwān, Abū l-ʿĀṣ b. Umayya, married three Qurashī women. They were Āmina bt. ʿAbd al-ʿUzza b. Ḥurthān of the Banū ʿAdī b. Kaʿb, Ruqayya bt. al-Ḥārith b. ʿUbayd b. ʿUmar of Makhzūm, and Ṣafiyya bt. Rabīʿa b. ʿAbd Shams.31 According to al-Zubayrī, he also married one woman from Thaqīf, Arwā bt. Asīd b. ʿIlāj b. Abī Salama32 (see fig. 2.3).

27

28

29

30 31 32

For al-Zubayrī and his Nasab Quraysh, see Robinson, Marriage 65–66 et passim. On alBalādhurī’s death date, see Su, Representations 10; Lynch, Arab conquests 38. Hereafter al-Zubayrī’s Nasab is abbreviated to Zub. The editions of al-Balādhurī’s Ansāb are cited as follows: Bal. (D) for the Damascus edition; Bal. (B). for the Beirut edition. This latter edition was accessed via al-Maktaba al-shāmila at https://al‑maktaba.org/book/9773. Now at https://shamela.ws/book/9773, last accessed March 3, 2022. Ibn ʿAsākir’s Taʾrīkh was accessed via al-Maktaba al-shāmila at https://al‑maktaba.org/​ book/71. Now at https://shamela.ws/book/71, last accessed March 3, 2022. Ibn Saʿd’s Ṭabaqāt was accessed via https://shamela.ws/book/9351, last accessed March 7, 2022. Watt, Muhammad at Mecca xiv–xv; Crone, Slaves on horses 3–17; Ahmed, Religious elite 1–12; Robinson, Marriage 15–81. See also, Heidemann and Hagemann, Transregional and regional elites. Al-Zubayrī records “2328 sons to 610 daughters”; Robinson, Marriage 74. Zub. 100–101; Bal. (B) v, 479, § 1207. This Arwā is thus from the aḥlāf Ghiyāra branch of the Thaqīf; Hosein, Tribal alliance, 115,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

22

marsham

figure 2.3 Abū al-ʿĀs b. Umayya’s marriages and children (Z = mentioned in al-Zubayrī’s Nasab Quraysh; B = mentioned in al-Balādhurī’s Ansāb al-ashrāf ) a. marsham

Abū l-ʿĀṣ’s 13 or more children by these unions included ʿAffān, the father of Caliph ʿUthmān, Ṣafiyya, wife of Abū Sufyān (and the mother by him of Ramla, one of the Prophet’s wives), and al-Ḥakam, the father of Caliph Marwān i. As children of Āmina, ʿAffān and Ṣafiyya were full siblings, while al-Ḥakam was their half-brother by Ruqayya.33 ʿAffān would go on to marry his second paternal cousin, Arwā bt. Kurayz b. Rabīʿa b. Ḥabīb b. ʿAbd Shams, with whom he would have two daughters and ʿUthmān. Arwā’s mother, Umm Ḥakīm bt. ʿAbd al-Muṭṭalib b. Hāshim, was a paternal aunt of the Prophet Muḥammad, making ʿUthmān b. ʿAffān a first

33

228. Al-Balādhurī has the same Arwā as the mother of Asīd b. Abī l-ʿĪṣ b. Umayya; Bal. (B) v, 456, § 1163. Zub. 100–101; Bal. (B) v, 479, § 1207.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

23

cousin once removed of the Prophet on his mother’s side (see fig. 2.1 above).34 As Asad Ahmed has noted, this maternal connection with the Prophet probably helps to explain ʿUthmān’s early conversion when members of his paternal descent group, the ʿAbd Shams, are generally said to have resisted Muḥammad’s new movement.35 Beyond the Quraysh, Abū l-ʿAṣ had close connections with the Thaqīf at al-Ṭāʾif. For example, his son al-Ḥakam had at least six children by Thaqafī women.36 Both al-Zubayrī and al-Balādhurī also note the marriages of Abū lʿĀṣ’s daughters Rayḥāna and Khālida to Thaqafīs.37 When al-Ḥakam was exiled after the fall of Mecca, he was sent to Waḍī Wajj, where al-Ṭāʾif was situated.38 However, al-Ḥakam also married women from Kināna and Murra b. ʿAwf.39 4.2 Al-Ḥakam b. Abī l-ʿĀṣ’s Marriages and Children Abū l-ʿĀṣ’s son al-Ḥakam (d. ca. 31/651–652)40 was the father of Marwān and grandfather of ʿAbd al-Malik. Al-Ḥakam’s large number of children is specifically remarked upon in the genealogical literature.41 Al-Ḥakam had four or five wives and some slave women, by whom he had certainly more than 20, and likely rather more than 30, offspring42 (see fig. 2.4). Some details differ in the sources, although some of the differences may be attributable simply to copyists’ errors. Al-Ḥakam married more exogamously than his father. His first wife was probably Āmina bt. ʿAlqama b. Ṣafwān b. Umayya b. Muḥarrith al-Kinānī, certainly she was the mother of his eldest son, ʿUthmān b. al-Ḥakam. She was also the mother of Marwān, who seems to have been born around 2 or 4/623 or 625.43 The genealogists list Mulayka bt. Awfā of the Murra b. ʿAwf next among

34 35 36 37 38 39 40

41

42 43

Zub. 101; Bal (D) i, 100; Bal. (B) v, 480, § 1212. Ahmed, Religious elite 107–108. See below, section 4.2. Zub. 101; Bal. (B) v, 479, § 1207. Bal. (D) v, 80. See below, section 4.2, for al-Ḥakam’s marriages to women from Kināna and Murra b. ʿAwf. Al-Dhahabī, Taʾrīkh iii, 363–365. Accessed via al-Maktaba al-shāmila al-ḥadīthīya, https://​ al‑maktaba.org, last accessed September 5, 2021. Now at https://shamela.ws/book/12397, last accessed March 3, 2022. Zub. 159: “… he fathered twenty-one men and women”; al-Dhahabī, Taʾrīkh iii, 365: “twenty men and eight women.” For him being a statistical outlier, see Robinson, Qurashi marriage 528. Zub. 159–160; Bal. (B) vi, 255, 301–305. Bosworth, Marwān (i). Her brother was later appointed as governor to Mecca by ʿAbd alMalik; Bal. (B) xi, 144, where his genealogy is given as ʿAlqama b. Ṣafwān b. Muḥarrith.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

24

marsham

figure 2.4 Al-Ḥakam b. Abī al-ʿĀṣ’s marriages and children (Z = mentioned in al-Zubayrī’s Nasab Quraysh; B = mentioned in al-Balādhurī’s Ansāb al-ashrāf ) a. marsham

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

25

al-Ḥakam’s wives.44 Al-Ḥakam’s marriages also reflect his father’s and then his own association with the Thaqīf. Al-Ḥakam had at least six (al-Zubayrī) or seven (al-Balādhurī) children with a Thaqafī woman, Umm al-Nuʿmān bt. al-Ḥārith b. Abī ʿAmr.45 A second union with a woman from the Thaqīf was— according to al-Zubayrī—with an unnamed daughter of Munabbih b. Shabīl b. al-ʿAjlān b. ʿAttāb b. Mālik b. Kaʿb, or with an unnamed “concubine” according to al-Balādhurī. This relationship produced five or three children according to the two respective genealogists. Al-Ḥakam had just one Qurashī wife, Umm Yūsuf al-Baʿītha bt. Hāshim b. ʿUtba b. Rabīʿa b. ʿAbd Shams. 4.3 Abū Sufyān Ṣakhr b. Ḥarb b. Umayya’s Marriages and Children Al-Ḥakam shares some biographical similarities with his first cousin Abū Sufyān. Both were powerful ʿAbshāmī leaders in Mecca who opposed Muḥammad but fathered future caliphs. Both fathered many children, albeit al-Ḥakam likely more than double Abū Sufyān; the sources record about 15 children for Abū Sufyān, of whom about seven were boys, by seven women.46 An additional son, Ziyād—the future governor of Iraq for ʿAlī (r. 35–41/656–661) and then for Muʿāwiya—was attributed as a son of Abū Sufyān in a politically motivated fiction from the time of Ziyād’s surrender to Muʿāwiya in 44/66547 (see fig. 2.5). Among Abū Sufyān’s daughters, Ramla, whose matronymic was Umm Ḥabība, married the Prophet.48 Her other sisters married into the Quraysh and Thaqīf.49 Among the seven boys Abū Sufyān fathered, two predeceased him, killed in battle (Ḥanẓala) or killed by the plague on campaign (Yazīd), and two were without issue (the same Yazīd and ʿAmr). This left his son Muʿāwiya with few surviving brothers and nephews and so contributed to the weakness of Abū Sufyān’s progeny in comparison to those of al-Ḥakam, his contemporary.

44

45

46 47 48 49

Zub. 159; Bal. (B) vi, 301; Ibn ʿAsākir, Taʾrīkh xxxviii, 331. Her genealogy is given by al-Zubayrī as Mulayka bt. Awfā b. Khārija b. Sinān b. Abī Ḥāritha b. Murra b. Nushba b. Ghayẓ b. Murra b. ʿAwf. Al-Balādhurī gives Mulayka bt. Awfā b. al-Ḥārith. On the Banū Murra b. ʿAwf, see Landau-Tasseron, Murra. Her full genealogy is given by al-Zubayrī as Umm al-Nuʿmān bt. al-Ḥārith b. Abī ʿAmr b. ʿAmr b. Wahb b. ʿAmr b. ʿĀmir b. Sayyār b. Mālik b. Ḥuṭayṭ b. Muʿṣab b. Quṣayy alThaqafiyya. This would appear to place her in the Mālik branch of the Thaqīf, see Hosein, Tribal alliance 113. For the marriages and children of Abū Sufyān, see Zub. 123–127; and Bal. (D) iv, 9–12. Bal. (D) iv, 212–220; Hasson, Ziyād b. Abīhi, with further references. Bal. (D) i, 526, where it is noted that she is sometimes called Hind, but Ramla is more reliable. Zub. 124; Bal. (D) iv, 10. On her brother-in-law at the siege of al-Ṭāʾif, see Hosein, Tribal alliance 155.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

26

marsham

figure 2.5 Abū Sufyān b. Ḥarb’s marriages and children (Z = mentioned in al-Zubayrī’s Nasab Quraysh; B = mentioned in al-Balādhurī’s Ansāb al-ashrāf ) a. marsham

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

27

4.4 ʿUthmān b. ʿAffān b. Abī l-ʿĀṣ’s Marriages Like his paternal uncle al-Ḥakam, ʿUthmān fathered more than 20 children. However, ʿUthmān is also said to have contracted an exceptional eight or nine marriages. They are listed according to a tentative chronology in figure 2.6 below, derived in part from al-Ṭabarī’s list of ʿUthmān’s wives, from Madelung’s Excursus in his Succession to Muḥammad, and from Ahmed’s Religious elite.50 ʿUthmān’s age at his death is confused in the sources. He is said to have lived either to 63 years old, which would give him a birth year of ca. 593 and would have him enter his mid-teens in 608, or to 75 years old, which would place his birth in ca. 581 and his mid-teens in ca. 596, or into his 80s, which would place his birth year in the 570s and would have him aged 15 in the mid- to late 580s.51 Of these ages, 75 perhaps seems most plausible as Wilferd Madelung has argued.52 Famously, ʿUthmān is said to have had one son by Ruqayya, a daughter of the Prophet. However, this son, ʿAbdallāh, died as a child.53 After Ruqayya’s death in ca. 624, he married another one of the Prophet’s daughters, Umm Kulthūm—a union that was without issue. Seven children are said to have been born to three women from Quraysh— Ramla bt. Shayba of ʿAbd Shams, Fāṭima bt. al-Walīd of the Banū Makhzūm, and Asmāʾ bt. Abī Jahl of the Banū Hāshim. He is said to have had three daughters with Ramla, two sons and a daughter with Fāṭima, and a son with Asmāʾ. Asmāʾ was also Fāṭima bt. al-Walīd’s mother, and the marriage is not widely attested, so there is doubt about this marriage’s historicity.54 If it was an authentic marriage, then it was probably his first, before he followed Muḥammad.55 Three more sons were born from two more marriages with daughters of leading tribal leaders—one, who died very young, to Fākhita bt. Ghazwān of the tribe of Māzin, a sister of ʿUtba, an early convert and Companion,56 and two to Umm al-Banīn bt. ʿUyayna al-Fazarī, whom ʿUthmān is said to

50 51 52

53 54 55 56

Ahmed, Religious elite 106–111; Zub. 104–105; Bal. (B) v, 600–601, §1548; Ṭab. i, 3,056. For eight or nine marriages as exceptional, see Robinson, Qurashi marriage 526–30. Ṭab. i, 3,052–3,054. Madelung notes that 63 may be a hagiographic figure, matching the age of the Prophet at his death: Madelung, Succession 370. This would also allow for his marriage to Asmāʾ before Muḥammad's mission. Bal. (D) v, 252; Watt, Ruḳayya. Ahmed, Religious elite 109. Madelung, Succession 363–364. Ibn Saʿd, Ṭabaqāt iii, 98–100; Bal. (D) v, 252; Ṭab. i, 3,056; Bosworth, ʿUtba b. Ghazwān. On ʿAbdallāh, her son, see Ṭab. i, 3,056.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

28

marsham

figure 2.6 ʿUthmān b. ʿAffān’s marriages and children, with possible date spans for some of the marriages a. marsham

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

29

have married after Abū Bakr’s (r. 11–13/632–634) defeat of her father, the leading shaykh of Ghaṭafān.57 At least eleven other children were born to two women from tribes involved in the conquest of Syria—Umm ʿAmr al-Dawsī, a daughter of Jundab b. ʿAmr of the Azd, whose father died in battle in Syria in ca. 636, and Nāʾila bt. alFarāfiṣa of the Syrian tribe of Kalb, whom he married in 28/648–649.58 ʿUthmān fathered five known children with Umm ʿAmr—four sons and a daughter, Maryam. The eldest son, ʿAmr, is said by al-Zubayrī to have been the oldest of ʿUthmān’s sons, who himself fathered children but was still young at the time of ʿUthmān’s death.59 ʿAmr’s sister Maryam is said to have married three Umayyads, Saʿīd b. al-ʿĀṣ, the widow of her half-sister Umm ʿAmr, then ʿAbd al-Raḥmān b. al-Ḥārith al-Makhzūmī, and finally ʿAbd al-Malik b. Marwān.60 ʿUthmān fathered five daughters and one son with Nāʾila. That eleven of ʿUthmān’s 22 known children are said to have been born to daughters of leading figures in Syria, probably in the 630s, 640s, and early 650s, suggests strongly that ʿUthmān was seeking to produce heirs whose maternal relatives were politically and militarily powerful men in Bilād al-Shām, perhaps as alternatives to the progeny of his second cousin and governor in Syria, Muʿāwiya b. Abī Sufyān. However, most of these children were likely under 20 years old when ʿUthmān was killed. Furthermore, the children of the bestconnected mother, Nāʾila, were five girls and only one (likely very young) son. This pattern suggests possible dynastic ambitions on ʿUthmān’s part, thwarted by his murder and by his children’s sex and their youth at the time of his death. It is notable that among the eleven is the one man who kept his distance from the Umayyad clan and sought political connections elsewhere: ʿAmr b. ʿUthmān.61 4.5 Muʿāwiya b. Abī Sufyān’s Marriages One of ʿUthmān’s most powerful allies and potential competitors was his second cousin and governor in Syria, Muʿāwiya b. Abī Sufyān. As a long-standing commander in the frontier province of Syria, who was well-connected with the Banū Kalb, Muʿāwiya was ideally placed to take power after ʿUthmān’s death, but his own limited fertility—said to have been ended by a wound

57 58 59 60 61

Fück, Ghaṭafān. For the date of the marriage to Nāʾila, see Ahmed, Religious elite 131. Zub. 105–106. Cf. Bal. (D) v, 254. Bal. (D) v, 253. Ahmed, Religious elite 108. For ʿAmr’s prestige among the sons of ʿUthmān, see Zub. 110; Bal. (D) v, 254. See also Madelung, Succession 89.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

30

marsham

from a failed assassination attempt in ca. 40/660–661—left him with only one heir, who would be supported by the dominant Kalbī element in the Syrian army.62 Muʿāwiya contracted two marriages with women from the Qurashī line of Nawfil bt. ʿAbd Manāf, by whom he is said to have had at least one boy, ʿAbdallāh (described as “weak”), and two girls, Hind and Ramla.63 A marriage to an ʿAbshāmiyya first cousin, Kuthna bt. Umayya b. Abī Sufyān b. Umayya, seems to have been without issue. He had another boy, ʿAbd al-Raḥmān, and another girl, ʿĀʾisha, by concubines. He also married two women from the Banū Kalb— Maysūn bt. Bahdal and Nāʾila bt. ʿUmāra. Maysūn was famously the sister of the leading shaykhs of the Banū Kalb; Nāʾila was her paternal first cousin. Of these two Kalbī women, only Maysūn had children—a girl who died very young and one son, Yazīd, who was born around 24/645.64 4.6 Muʿāwiya’s Children’s Marriages Yazīd b. Muʿāwiya’s status as his father’s nominated heir is reflected in his marriages and reproductive choices, which anticipate those of his Marwanid second cousins a couple of decades later; he separated his line from other groups by producing most of his children with concubines and with close cousins. Yazīd had at least 18 children by three wives and an unspecified number of concubines.65 Most of these children—ten of them—were by concubines. A related development was a move toward close cousin marriage: two of Yazīd’s three child-producing marriages were with women from leading proximate branches of the ʿAbd Shams—a daughter of Abū Hāshim b. ʿUtba b. Rabīʿa (her name and matronymic vary) and Umm Kulthūm bt. ʿAbdallāh b. ʿĀmir b. Kurayz,66 who was the daughter of a maternal cousin of ʿUthmān, ʿAbdallāh b. ʿĀmir, who served as a leading commander in Iran and Iraq under ʿUthmān and then Muʿāwiya.67 These two marriages yielded a further seven

62 63 64 65 66 67

For the political implications of Muʿāwiya’s few children in the Arabic sources, see Humphreys, Muʿāwiya 35. On the injury that made him infertile, see Ṭab i, 3,464–3,465. For one Fākhitah as ʿĀtika, and accompanying Muʿāwiya on campaign, see Ṭab i, 288–289. Zub. 126, 127–128; Bal. (D) iv, 315–316; Ṭab ii, 204–205. On Yazīd’s age, see Hawting, Yazīd (i). Zub. 128–130; Bal. (B) v, 290, § 779, 355–356, § 923. Zub. 128; Bal. (B) v, 355, § 923. Al-Zubayrī gives the former as Umm Hāshim and alBalādhurī calls her Umm Khālid and gives her ism as Fākhita. Morony, ʿAbd Allāh b. ʿĀmir; Gibb, ʿAbd Allāh b. ʿĀmir; Bal. (D) vii, 687, 689; Ṭab. i, 2,802, 2,828.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

31

children between them, including Khālid and Muʿāwiya b. Yazīd, who briefly succeeded him in 64/683–684. Yazīd also had a daughter, Ramla (by an unnamed “Ghassānid woman” according to al-Balādhurī).68 This Ramla is said to have married ʿAbbād, a son of the Iraqi governor Ziyād (g. 45–53/665–673), as did her sister, Umm ʿAbd al-Raḥmān, after her.69 A fourth marriage, with a great-granddaughter of the second caliph, ʿUmar b. al-Khaṭṭāb, was childless and ended in divorce.70 The marriages of Muʿāwiya’s other five children who lived to adulthood reflect his alliances with the family of ʿUthmān and with two governors in the East—the same ʿAbdallāh b. ʿĀmir, whose daughter married Yazīd, and the same Ziyād b. Abī Sufyān, whose son married Yazīd’s daughter. In 41/661 Muʿāwiya reappointed ʿAbdallāh b. ʿĀmir to the governorship of Basra that he had lost after ʿUthmān’s murder. Then, in 44/665, when ʿAlī’s former governor of Basra, Ziyād b. Abī Sufyān, surrendered, Muʿāwiya deposed ʿAbdallāh and installed Ziyād, adopting him as a brother.71 The relevant marriages probably occurred in this sequence, reflecting Muʿāwiya’s shift from being part of a network of alliances with ʿUthmān’s relatives to establishing his own connections with men who owed more to him personally. Hind, Muʿāwiya’s daughter by Fākhita bt. Qaraẓa, married ʿAbdallāh b. ʿĀmir b. Kurayz, who was also the father-in-law of Hind’s full brother, ʿAbdallāh (and, as noted above, of her halfbrother Yazīd). Ramla, who was Muʿāwiya’s daughter by Fākhita’s sister, Kanūd (or Katwa), married ʿAmr b. ʿUthmān. Muʿāwiya’s daughter by a concubine, ʿĀʾisha (or Ṣafiyya), married a son of Ziyād.72 4.7 Al-Ḥakam b. Abī l-ʿĀṣ’s Childrens’ Marriages While al-Ḥakam’s marriages largely reflect the politics of Mecca in the transitional decades from the era just before Muḥammad’s mission through to the first waves of conquests in Syria, Iraq, and Egypt (ca. 600–ca. 650), the marriages of his many children contributed to the consolidation and development

68 69

70 71 72

Bal. (B) v, 290, § 779. Al-Zubayrī lists only Ramla’s second cousin, ʿUtba b. ʿUtba b. Abī Sufyān, as her husband and gives Umm ʿAbd al-Raḥmān as the wife of ʿAbbād b. Ziyād; Zub. 130. Al-Balādhurī says Ramla and Umm ʿAbd al-Raḥmān married ʿAbbād “one after the other”; Bal. (B) v, 356, § 923. Both scholars relate the account that the marriage was arranged by Khālid b. Yazīd and was criticized by ʿAbd al-Malik. On Ziyād, see Hasson, Ziyād b. Abīhi. Zub. 360–361; Bal. (B) v, 290–291, § 779. For these events, see above, n. 44. For all three of these marriages, see Zub. 128; Bal. (B) v, 285, §763. ʿAbdallāh b. Kurayz was the father of Amāt al-Ḥamīd, wife of Hind’s full brother, ʿAbdallāh b. Muʿāwiya, Zub. 169.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

32

marsham

of alliances in the context of the new “conquest society” (ca. 620–ca. 670).73 They were a crucial foundation for the future political success of the Marwanids. Al-Ḥakam’s closeness to ʿUthmān b. ʿAffān is reflected in his childrens’ marriages. Two of al-Ḥakam’s children married daughters of ʿUthmān: Marwān b. al-Ḥakam had about five male children and at least one daughter by Umm Abān, ʿUthmān’s daughter by al-Ramla bt. Shayba al-ʿAbshāmī;74 Marwān’s full brother, al-Ḥārith b. al-Ḥakam, married Umm Abān’s full sister, ʿĀʾisha, by whom he had at least two children.75 A third child, Abān, who was al-Ḥakam’s son by Mulayka of the Banū ʿAwf, married Umm ʿUthmān bt. Khālid b. ʿUqba b. Abī Muʿayṭ, a niece of ʿUthmān b. ʿAffān’s maternal half-brother, al-Walīd b. ʿUqba.76 Al-Ḥakam’s other sons married women from other branches of the Umayyad clan. ʿAbd al-Raḥmān, who was a full brother of Marwān and al-Ḥārith, married Umm al-Qāsim bt. ʿAbdallāh b. Khālid b. Asīd. Yaḥyā, who was a son by Mulayka, married two women from the Quraysh—Zaynab bt. ʿAbd al-Raḥmān of the Banū Makhzūm, by whom he had at least two children, and Umm Kulthūm bt. Muḥammad b. Rabīʿa of the Banū Hāshim, by whom he had at least another two.77 Al-Ḥakam’s sons also married women from beyond the Quraysh. Besides his children by Umm Abān bt. Shayba, al-Ḥārith also had at least four children by Mufāda, a daughter of the Companion, Tamīmī shaykh, and poet, al-Zibriqān b. Badr.78 Ḥabīb, who was a son by Mulayka, married Maryam bt. ʿAbdallāh of the ʿAws of Medina.79 Yaḥyā, also a son by Mulayka, married Umm Sulaymān, a daughter of ʿĀmir b. al-Ḥarash of the Banū Kaʿb b. Qays, in addition to his two Qurashī wives.80 Many of al-Ḥakam’s daughters married Qurashīs and Thaqafīs. Umm alBanīn, a full sister of Marwān and his brothers, married Saʿīd b. al-ʿĀṣ al-Umawī (d. ca. 59/679), another close ally of Caliph ʿUthmān, by whom she had the future candidate for the caliphate and rival of ʿAbd al-Malik, ʿAmr b. Saʿīd al73 74 75 76 77 78 79 80

For “conquest society,” see Crone, Slaves on horses 27 and 29. Zub. 161. Cf. Bal. (B) vi, 307. Zub. 169–170. Bal. (B) v, 601, § 1550. Zub. 171. Bal. (B) vi, 302. For Arwā as the mother of both ʿUthmān and al-Walīd, see Bal. (B) iv, 311. See also Bosworth, Al-Walīd b. ʿUqba. Zub. 171. Zub. 170. On al-Zibriqān, see Lecker, al-Zibriḳān b. Badr. Lecker does not note the marriage with al-Ḥārith, but notes two others with Thaqafī commanders. Zub. 172–173. Zub. 171.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

33

Ashdaq.81 According to al-Balādhurī, Umm Abān and then her half-sister, Umm al-Ḥakam, both married ʿAbdallāh b. al-Muṭṭalib b. Ḥanṭab al-Makhzūmī. (In contrast, al-Zubayrī records only the marriage of Umm Abān to ʿAbdallāh’s brother ʿAbd al-Malik, and elsewhere al-Balādhurī has their sister Umm Salama as a wife of ʿAbdallāh b. Ḥanṭab.)82 Umm Abān’s full sister, Umāma (or Thumāma), married ʿAbd al-Raḥmān b. Abī Dhiʿb from the Qurashī clan of ʿĀmir b. Luʾayy. Umm al-Banīn’s full sister, Zaynab, married Usayd b. al-Akhnas alThaqafī.83 4.8 Marwān b. al-Ḥakam’s Marriages and Children Marwān b. al-Ḥakam is said to have had about 16 children by five wives and one concubine and to have married a sixth wife without issue. As with his father, the number of his children is remarked upon in some of the sources; it does appear to be statistically exceptional, albeit less so than his father’s 30 or more children.84 Marwān is said to have been born in ca. 2 or 4/623 or 626 and to have been over 60 years old at his death in 65/685.85 At his cousin ʿUthmān’s accession in 23/644, Marwān would have been a young man of about 20, and so his marriages probably began at or just before this time. They reflect his close associations with the networks of power in the era of ʿUthmān’s caliphate. The two sons of Marwān who would go on to achieve the most prominence were born early in ʿUthmān’s caliphate: the future caliph ʿAbd al-Malik (d. ca. 86/705), who is said to have been born in 26/646–647, and the future governor of Egypt and crown prince, ʿAbd al-ʿAzīz (d. ca. 85/704).86 Their mothers were, respectively, ʿĀʾisha bt. Muʿāwiya b. al-Mughīra b. Abī l-ʿĀṣ, who was Marwān’s paternal first cousin once removed, and Laylā bt. Zabbān b. al-Aṣbagh b. ʿAmr al-Kalbī, who was the daughter of a leading figure from the powerful Syrian tribe, the Banū Kalb, and a second cousin of ʿUthmān and Saʿīd b. al-ʿĀṣ’s wives from Kalb.87 A third son, Bishr, is almost as well-known as ʿAbd al-Malik and ʿAbd al-ʿAzīz; Bishr was in Egypt with his paternal brother ʿAbd al-ʿAzīz and then was gov81 82 83 84 85 86 87

Zub. 159. On these connections, see Bosworth, Saʿīd b. al-Āṣ. Zub. 159; Bal. (B) vi, 302; ix, 126. Zub. 159. Zub. 160 says he had eleven children, “men and women”; Ibn Saʿd v, 36 says he had 13, “men and women.” On the statistics, see Robinson, Qurashi marriage 527 and 530. Bosworth, Marwān (i). For their dates, see Gibb, ʿAbd al-Malik b. Marwān; Zetterstéen, ʿAbd al-ʿAzīz b. Marwān. Zub. 160. On Zabbān and his father, Abū Tumādir, see Ibn al-Kalbī, Nasab Maʿadd waYaman 566, 568; al-Ṭabarī, History viii, 95, n. 408.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

34

marsham

ernor of Iraq under his paternal half-brother ʿAbd al-Malik; Bishr’s mother was Quṭiyya bt. Bishr b. Mālik, who was from one of the leading branches of the Banū Kilāb branch of ʿĀmir b. Ṣaʿṣaʿa.88 As noted above, Marwān also had about five male children and at least one daughter by Umm Abān bt. ʿUthmān b. ʿAffān and at least two children by Zaynab bt. ʿAmr from the Makhzūm branch of the Quraysh. He also had at least one son by a concubine, the famous frontier commander Muḥammad (d. ca. 100/719–720), whose mother is often said to have been Kurdish.89 Marwān’s sons’ maternal connections help to explain whom among them were eventually promoted as candidates for the caliphate. ʿAbd al-Malik was a son by his father’s paternal cousin once removed, and so was descended from his great-grandfather Abū l-ʿĀṣ on both sides. He had one full brother, Muʿāwiya, but Muʿāwiya is remembered as “stupid” in the tradition and is said to have married to a daughter of ʿAlī b. Abī Ṭālib, both things would have made him unsuitable for the caliphate in the eyes of most of his other relatives and their allies.90 ʿAbd al-ʿAzīz’s mother, Laylā, was from the Syrian Banū Kalb, who were the mainstay of the military support of the Umayyads in the seventh century. ʿAbd al-ʿAzīz also had a full brother, but he died in childhood.91 Bishr also appears to have had caliphal ambitions.92 His noble but non-Qurashī mother and later his governorship in Iraq gave him the connections that allowed for this. However, his death in 73–75/692–695, at about 40 years of age, curtailed them.93 The marriages of Marwān’s two nominated heirs, ʿAbd al-Malik and ʿAbd alʿAzīz, are discussed in detail in the following section. Among Marwān’s other children, the marriage patterns often reveal the political reach and status of their father, with marriages into the families of ʿUthmān b. ʿAffān, ʿAlī b. Abī Ṭālib, Saʿīd b. al-ʿĀṣ, the Banū Makhzūm, and the Banū Kalb.94 4.9 ʿAbd al-Malik and ʿAbd al-ʿAzīz’s Marriages and Children Like his father and grandfather, ʿAbd al-Malik married and fathered children prodigiously.95 However, his marriages took place in a very different political

88 89 90 91 92 93 94 95

Zub. 160–161. Zub. 161. Zub. 45; Bal. (D) v, 341–342. Bal. (D) v, 340. See, e.g., Treadwell, Orans drachm. Bosworth, Bishr b. Marwān. Zub. 45, 121, 160, 161, 170; Bal. (D) v, 340, 342. Cf. Bal.(D) v, 340; Bal. (B) ix, 177. For his statistical outlier status, see Robinson, Qurashi marriage 528.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

35

context, reflecting his position as one of the oldest and best connected of Marwān’s sons, and then as his heir and as caliph during and after the second civil war. ʿAbd al-Malik reached adulthood (ca. 15 years old) and so the age of potential fatherhood in the early 40s/660s, when his father was achieving prominence as a leading figure in Muʿāwiya’s caliphate. The matches he made reflect his father’s position and then his own status, first as candidate for the caliphate and then as caliph. ʿAbd al-Malik is said to have contracted six marriages that produced ten male children and three girls and up to a further five that did not produce progeny or where the progeny are unknown. The majority of ʿAbd al-Malik’s sons were by Wallāda, ʿĀtika, and his concubines. At least a further eight children by the latter are recorded in the sources96 (see fig. 2.7). The children who became caliphs were two sons by Wallāda bt. al-ʿAbbās al-ʿAbsī—al-Walīd (r. 86–96/705–715) and Sulaymān (r. 96–-99/715–717)—and two by ʿĀtika bt. Yazīd b. Muʿāwiya b. Abī Sufyān—Yazīd ii (r. 101–105/720– 724)—and ʿĀʾisha bt. Hishām b. Ismāʿīl al-Makhzūmī—Hishām (r. 105-–125/ 724–743). These four sons succeeded in order of seniority: al-Walīd was probably born in ca. 54/674, and Sulaymān shortly after; Yazīd and Hishām were born in ca. 71/690 and ca. 72/691, respectively.97 At ʿAbd al-Malik’s death in 86/705, al-Walīd and Sulaymān would both have been about 30 years old, whereas Yazīd and Hishām would only have been about 15. Al-Walīd and Sulaymān had two more brothers, Marwān and Dāʾūd, neither of whom are said to have had any children, and one sister, ʿĀʾisha, who married Khālid b. Yazīd.98 Yazīd’s mother was ʿĀtika bt. Yazīd b. Muʿāwiya, who is celebrated in the tradition as the daughter of one caliph, the granddaughter of another, the wife of a third, the mother of a fourth, and the grandmother of a fifth. Thus, Yazīd b. ʿAbd al-Malik’s maternal uncles and cousins were grandsons and greatgrandsons of Muʿāwiya b. Abī Sufyān; he is sometimes known in the sources as Ibn ʿĀtika.99 Yazīd had two full brothers, Marwān and Muʿāwiya, and one sister, Umm Kulthūm; Muʿāwiya died young, without children.100 Hishām’s mother, ʿĀʾisha, was a daughter of the Makhzūmī shaykh, Hishām b. Ismāʿīl, whom ʿAbd al-Malik appointed governor of Medina after retaking

96 97

98 99 100

Zub. 83, 111, 161–165; Bal. (d) vi, 302–6; Ṭab. ii, 1,174. For al-Walīd, see Kennedy, Al-Walīd 1. Al-Walīd (i); Ibn ʿAsākir, Taʾrīkh lxiii, 185, 187. For Sulaymān, see Eisener, Sulaymān b. ʿAbd al-Malik. For Yazīd, see Lammens and Blankinship, Yazīd (ii). For Hishām, see Gabrieli, Hishām b. ʿAbd al-Malik. Zub. 161–162; Bal. (D) vi, 302; Ṭab. ii, 1,174. Marsham, Rituals 120 and n. 43. Zub. 163; Bal. (d) vi, 302; Ṭab. ii, 1,174.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

36

marsham

figure 2.7 ʿAbd al-Malik b. Marwān’s marriages and children (Z = mentioned in al-Zubayrī’s Nasab Quraysh; B = mentioned in al-Balādhurī’s Ansāb al-ashrāf ; Ṭ = mentioned in al-Ṭabarī’s Taʾrīkh) a. marsham

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

37

the town from Ibn al-Zubayr. ʿAbd al-Malik’s marriage with Hishām b. Ismāʿīl’s daughter was contracted at the same time and brought a son immediately, in 72/691.101 Thus, the marriage was one of alliance, bringing a rival family into association with ʿAbd al-Malik’s. It is notable that it produced only the one son; this may have been deliberate as well as a consequence of a fairly late union, when ʿAbd al-Malik was about 45. Notably, ʿĀʾisha bt. Mūsā b. Ṭalḥa and Umm Ayyūb bt. ʿAmr b. ʿUthmān were also both from non-Marwanid Qurashī branches whose ancestors had claimed the caliphate, and each bore ʿAbd alMalik only one son.102 ʿAbd al-ʿAzīz’s marriages likewise reflect his leading status among the sons of Marwān, albeit in the shadow of his half-brother ʿAbd al-Malik. Unlike ʿAbd al-Malik, ʿAbd al-ʿAzīz did not make many marriages of alliance or have many children by concubines, differences that are probably attributable to him never becoming caliph. He is said to have had four wives, fathering seven sons and four daughters by them, as well as at least one son, al-Aṣbagh, by a concubine.103 One of ʿAbd al-ʿAzīz’s sons did later become a caliph: ʿUmar ii (r. 99–101/717– 720). ʿUmar was born in 61/681, when ʿAbd al-ʿAzīz was perhaps about 35 years old. His mother was Umm ʿĀṣim Ḥafṣa bt. ʿĀṣim b. ʿUmar b. al-Khaṭṭāb, who was a granddaughter of her son’s namesake, the second caliph, ʿUmar i (r. 13–23/634–644). ʿUmar b. ʿAbd al-ʿAzīz’s ancestry contributed to him being nominated as Sulaymān b. ʿAbd al-Malik’s heir after the crisis caused by Sulaymān’s death in 99/717. ʿUmar had three full brothers, ʿĀṣim, Abū Bakr, and Muḥammad.104 ʿAbd al-ʿAzīz’s three other marriages reflect his grand-maternal connections and the building of alliances with the Sufyanids and their allies, most likely during Muʿāwiya’s caliphate. Like his father, he married a woman from the Banū Kilāb, Laylā bt. Suhayl, by whom he fathered two boys and a girl. Like his brother ʿAbd al-Malik, he married a granddaughter of Muʿāwiya (but not a daughter of Yazīd i), ʿĀʾisha bt. ʿAbdallāh b. Muʿāwiya, by whom he had only daughters. ʿAbd al-ʿAzīz also had a son and a daughter by Umm ʿAbdallāh bt. ʿAbdallāh b. ʿAmr b. al-ʿĀṣ, which was a marriage that was probably contracted to secure his position

101 102 103 104

Zub. 163–164; Bal. (d) vi, 302; Hinds, Makhzūm; Gabrieli, Hishām b. ʿAbd al-Malik. Zub. 164; Bal. (D) vi, 302. Zub. 168–169; Bal. (D) v, 369; vii, 5–6. See further on these brothers, Ibn ʿAsākir, Taʾrīkh xxv, 271–274, §3016, §3017, where ʿĀṣim is said to have been killed by the Kharijites and Muḥammad is said to have been childless; and lxvi, 38–40, § 8388, on Abū Bakr. For ʿĀṣim as the prayer leader at Kūfa for his paternal half-brother, ʿAbdallāh b. ʿUmar, see Bal. (B) viii, 222.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

38

marsham

as the governor of Egypt; Umm ʿAbdallāh was the granddaughter of the conqueror of Egypt, ʿAmr b. al-ʿĀṣ (g. 19–ca. 25/640–644, ca. 42 or 43/ca. 660–662 or 664), and the daughter of another governor and renowned scholar, ʿAbdallāh (g. 43/664; d. 65/684).105 4.10 Al-Walīd, Sulaymān, Yazīd, and Hishām’s Marriages and Children Among ʿAbd al-Malik’s 19 named sons, his first nominated heir, al-Walīd, made an exceptional number of marriages during his 42 or so years, for which he is remembered in the sources. At least nine wives are named between al-Zubayrī, al-Balādhurī, and al-Ṭabarī106 (see fig. 2.8.) Al-Walīd’s marriages reflect both his seniority in age among ʿAbd al-Malik’s sons and his prestige as a likely successor to his father. Most matches were political alliances, uniting ʿAbd al-Malik’s family with potential rivals but yielding only two known children—perhaps deliberately so. Indeed, al-Walīd fathered 15 of his 22 known children with his concubines, including his eldest son, the military commander al-ʿAbbās b. al-Walīd. Al-Walīd’s marriages outside the line of Abū l-ʿĀṣ included: Nafīsa and Zaynab, who were two great-granddaughters of ʿAlī b. Abī Ṭālib; a daughter of Saʿīd b. al-ʿĀṣ, Āmina, whose brother, ʿAmr, rebelled against ʿAbd al-Malik and was killed in 69/689;107 Umm ʿAbdallāh, who was a niece of ʿAbd al-Malik’s wife from the same ʿUthmānid line and, during his caliphate, one of her nieces, ʿIzza, whom he later divorced; a daughter of Ibn al-Zubayr’s Qurashī Kufan governor, ʿĀtika, whom he divorced; and a “Fazārī woman.”108 However, these seven marriages produced only two known children, ʿAbd al-Raḥmān by Umm ʿAbdallāh and Abū ʿUbayda by the Fazārī woman. Al-Walīd had more children with his first cousin Umm al-Banīn bt. ʿAbd al-ʿAzīz, in a union that tied the fortunes of ʿAbd al-ʿAzīz and ʿAbd al-Malik. Al-Walīd is said to have attempted to promote ʿAbd al-ʿAzīz as his heir, without success.109 (Two of al-Walīd’s sons did eventually succeed to the caliphate, long

105 106

107 108 109

On ʿAmr, see Keshk, ʿAmr b. al-ʿĀṣ. On ʿAbdallāh’s death date, see Ibn ʿAsākir, Taʾrīkh xxxi, 290. For al-Walīd’s marriages and children, see Zub. 52, 116, 165, 168; Bal. (D) vii, 5–6; Ṭab. ii, 1,270. A report from al-Madāʾinī claims he had 63 wives (!), without naming them; Bal. (D) vi, 6. For his exceptional status, see Robinson, Qurashi marriage 528. For 42 as his age at death, see Kennedy, al-Walīd. 1. Al-Walīd (i); Bal. (B) viii, 93 says he was 49 when he died, which would place his birth at around 47/667–668 as opposed to Jacobi’s estimate of 54/674. Bal. (B) viii, 91, for a story about her remarks on ʿAbd al-Malik’s death. Bal. (B) x, 481, for ʿAbdallāh b. Muṭīʿ and Ibn al-Zubayr. On this, see Marsham, Rituals 114, 121–122. On marriages between al-Walīd and Sulaymān’s children, see Bal. (B) viii, 68. Cf. Bal. (B) viii, 86–87.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

39

kinship, dynasty, and the umayyads

figure 2.8 Al-Walīd b. ʿAbd al-Malik’s marriages and children (Z = mentioned in al-Zubayrī’s Nasab Quraysh; B = mentioned in alBalādhurī’s Ansāb al-ashrāf ; Ṭ = mentioned in al-Ṭabarī’s Taʾrīkh) a. marsham

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

40

marsham

after their father’s death—Yazīd iii in 126/744 and Ibrāhim for a month or two in the same year.110) Al-Walīd also married one of Umm al-Banīn’s sisters, Umm al-Ḥakam, but had no children by her.111 In contrast to al-Walīd, his three brothers married more endogamously. Sulaymān’s six marriages were all to members of the Umayyad branch of the Quraysh, by whom he had eight known children. In addition, he had at least six children by concubines, making at least 14 children in all.112 Yazīd and Hishām were about 15 or more years younger than al-Walīd and Sulaymān and so would have reached their mid-teens in about 86/705, around the time of their father’s death. Yazīd contracted two marriages, one with a Qurashī and one with a Thaqafī, by whom he had six known children, and had at least a further eight children by slave women.113 Hishām made five marriages—all with Umayyad women—by whom he had eight known children, and at least a further seven children by slave women.114 The Umayyad lines with which these three brothers married were quite specific and consistent. Four of their marriages were in the lines of their paternal great-great-grandfather, al-Ḥakam b. Abī l-ʿĀṣ,115 five were in the line of ʿAmr b. ʿUthmān b. ʿAffān (four of ʿAmr’s granddaughters, one of whom, Suʿda, first married Yazīd and then Hishām after Yazīd’s death), two were nieces of ʿĀtika bt. Yazīd b. Muʿāwiya, mother of Yazīd b. ʿAbd al-Malik,116 and one (Umm ʿAmr, a wife of Sulaymān) was a daughter of the Umayyad governor of Fars and Kufa, ʿAbdallāh b. Khālid b. Abī l-ʿĪṣ.117 The only non-Qurashī was Yazīd’s Thaqafī wife, Umm al-Ḥajjāj bt. Muḥammad b. Yūsuf, the niece of the governor of Iraq, alḤajjāj b. Yūsuf (r. 75–95/694–714). His son, al-Walīd b. Yazīd, who became caliph in 125/743 and was killed in 126/744, was a son by this Umm al-Ḥajjāj and was supported by her male relatives.118

110 111 112 113 114 115 116 117 118

Hawting, Yazīd (iii); Cremonesi, Ibrāhīm b. al-Walīd. Zub. 168. On Sulaymān’s marriages, see Zub. 165–166; and Bal. (D) vii, 40, with further references in what follows. Zub. 166–167; Ṭab. ii, 1,465. Zub. 167–168, 171; Bal. (D) v, 253; Ṭab. ii, 1,465. Zub. 165, 168, 171. Zub. 165–167; Ṭab. ii, 1,465. Zub. 166; Bal. (D) vii, 40. Zub. 166–167; Ṭab. ii, 1,465. For the dates, see Dietrich, al-Ḥadjdjādj b. Yūsuf. Kennedy, AlWalīd 2. Al-Walīd (ii). On the succession of al-Walīd ii, see Marsham, Rituals 122; Crone, Slaves on horses 45; Crone, Qays and Yemen 55.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

41

4.11 ʿUmar b. ʿAbd al-ʿAzīz’s Marriages and Children ʿUmar was probably less than a decade younger than his paternal cousins alWalīd and Sulaymān, and about a decade older than Yazīd and Hishām. He would have come of age in about 75/695, when his father was both established as governor in Egypt and the widely recognized heir apparent to his uncle ʿAbd al-Malik. ʿUmar’s three marriages were with his paternal cousin Fāṭima bt. ʿAbd al-Malik—making ʿUmar the brother-in-law of ʿAbd al-Malik’s sons—and with two daughters of tribal leaders, Lamīs bt. ʿAlī of the Banū al-Ḥārith b. Kaʿb and Umm Shuʿayb, or Umm ʿUthmān, bt. (Saʿīd, or Shuʿayb, b.) Zabbān. The latter was his maternal cousin from the Banū Kalb. He had seven known children by these wives, and seven by concubines.119

5

Final Remarks

This brief survey has examined the marriages of the Umayyad rulers and some of their most important ancestors and children down to the mid-Marwanid era. There is much more to say. For example, the marriages of the wider Umayyad clan might be explored, and matrilineal lines of kinship examined more extensively. In terms of sources, both the narrative historiography and poetry remain under-explored.120 Here, the survey ends with some very brief comments about the last Umayyad caliphs, who were three of ʿAbd al-Malik’s grandsons (alWalīd ii, Yazīd iii, and Ibrāhīm) and a nephew (Marwān ii). The last three of these rulers had slave mothers, in part reflecting the unusual political circumstances of the third civil war.121 This war between the progeny of al-Walīd i and Yazīd ii might also be seen as a conflict between the marginalized Syrian Kalb, who had been closely connected to the Umayyad clan in the mid-seventh century through to the early eighth but had lost power and influence to predominantly Qaysī groups on the northern Roman frontier. It is also connected to conflicts in Iraq, with al-Walīd ii’s mother being a Thaqafī relative of the Iraqi governor Yūsuf b. ʿUmar.122 Although the ruling Marwanid branch of the 119

120

121 122

On ʿUmar’s marriages and children, see Ibn Saʿd, Ṭabaqāt v, 330; Bal. (D) vii, 161; Ibn ʿAsākir, Taʾrīkh vii, 85; xxxi, 218. On the Banū al-Ḥārith b. Kaʿb in very early Islam, see Schleifer, Banū al-Ḥārith b. Kaʿb. On Fāṭima, see Marsham, Rituals 120 and n. 41. Some suggestions about the poetry are made in Marsham, Rituals 86–112. Crone and Hinds, God’s caliph, makes good use of the poetry; and Alain George and Nadia Jamil have recently shown the great potential of the poetry for Umayyad history, see George, Umayyad mosque. For their mothers, see Bal.(D), vii, 170, 474, 540, 548. Ṭab., ii, 1740–1879 gives a detailed narrative of the civil war. Note that some allegiances

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

42

marsham

Umayyad dynasty had sought to disentangle themselves from close kinship with their tribal allies, they nonetheless fell victim to the politics of their Syrian armies and the challenges of ruling such a vast empire.

Bibliography Sources ʿAbbās, I. (ed.), ʿAbd al-Ḥamīd b. Yaḥyā al-Kātib wa-mā tabqā min rasāʾilihi wa-rasāʾil Sālim Abī al-ʿAlāʾ, Amman 1988. al-Akhṭal, Sharḥ dīwān al-Akhṭal al-Taghlibī, Beirut 1968. al-Balādhurī, Ansāb al-Ashrāf, ed. M. al-Fardūs al-ʿAẓm, 25 vols., Damascus 1996–2004 = Bal. (D). al-Balādhurī, Jumal min Ansāb al-Ashrāf, ed. Suhayl Zakkār and Riyāḍ al-Ziriklī, 13 vols., Beirut 1996 = Bal. (B). al-Dhahabī, Taʾrīkh al-Islām, ed. ʿU.ʿA. al-Salām al-Tadmurī, 52 vols., Beirut 1413/1993. Ibn ʿAsākir, Taʾrīkh Madīnat Dimashq, ed. ʿA. b. G. al-ʿUmrawī, 80 vols., Beirut 1995. Ibn al-Kalbī, Nasab Maʿadd wa-Yaman, ed. R. Ḥasan, Beirut 2004. Ibn Saʿd, al-Ṭabaqāt al-kubrā li-Ibn Saʿd, 10 vols., Beirut 1968. al-Maktaba al-shāmila, https://shamela.ws/, last accessed March 3, 2022. al-Maktaba al-shāmila al-ḥadīthīya, https://al‑maktaba.org, last accessed September 8, 2021. (See now https://shamela.ws/.) al-Ṭabarī, The history of al-Ṭabarī, ed. E. Yar-Shater, trans. A. Brockett, 40 vols., Albany, 1985–2007. al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, ed. M.J. de Goeje et al., 3 vols., Leiden 1979–1901 = Ṭab. al-Yaʿqūbī, Taʾrīkh, ed. M. Houtsma, 2 vols., Leiden 1883. al-Zubayrī, Kitāb Nasab Quraysh, ed. É. Levi-Provençal, Cairo 1953 = Zub.

Studies Abbott, N., Women and the state in early Islam, in Journal of Near Eastern Studies 1 (1942), 341–368. Ahmed, A.Q., Religious elite of the early Islamic Ḥijāz: Five prosopographical case studies, Oxford 2011. Ali, K., Marriage and slavery in early Islam, Cambridge, MA and London 2010.

cut across the Quḍāʿa/Qays divide—some ʿAbsīs were allies of al-Walīd i’s family, as one might expect. Somewhat contrasting discussions of the power dynamics involved include Crone, Slaves on horses, 42–8; Kennedy, Prophet and the age, 97–101; Blankinship, End of the jihad, 223–36.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

43

Blankinship, K.Y., The end of the Jihād state: The reign of Hisham Ibn ʿAbd al-Malik and the collapse of the Umayyads, New York 1994. Bosworth, C.E., Bishr b. Marwān, in ei2 i, 1,242–1,243. Bosworth, C.E., Marwān (i) b. al-Ḥakam, in ei2 vi, 621–623. Bosworth, C.E., Muʿāwiya (ii) b. Yazīd, in ei2. vii, 268–269. Bosworth, C.E., Saʿīd b. al-Āṣ, in ei2. viii, 853. Bosworth, C.E., ʿUtba b. Ghazwān, in ei2. x, 944. Bosworth, C.E., Al-Walīd b. ʿUqba, in ei2. xi, 130. Clarke, N., Heirs and spares: Elite fathers and their sons in the literary sources of Umayyad Iberia, in Al-Masāq 28 (2016), 67–83. Cremonesi, V., Ibrāhīm b. al-Walīd, in ei2. iii, 990–991. Crone, P., Slaves on horses: The evolution of the Islamic polity, Cambridge 1980. Crone, P., Were the Qays and Yemen of the Umayyad period political parties? in Der Islam 71 (1994), 1–57. Crone, P., and M. Hinds, God’s caliph: Religious authority in the first centuries of Islam, Cambridge 1986. Dietrich, A., al-Ḥadjdjādj b. Yūsuf, in ei2 iii, 39–43. Donner, F.M., The early Islamic conquests, Princeton 1981. Duindam, J., Dynasties: A global history of power, 1300–1800, Cambridge 2016. Duindam, J., Dynasty: A very short introduction, Oxford 2019. Eisener, R., Sulaymān b. ʿAbd al-Malik, in ei2. ix, 821–822. El-Azhari, T.K., Queens, eunuchs and concubines in Islamic history, 661–1257, Edinburgh 2019. Fück, J.W., Ghaṭafān, in ei2 iii, 1,023–1,024. Gabrieli, F., Hishām b. ʿAbd al-Malik, in ei2 iii, 493–495. George, A., The Umayyad mosque of Damascus: Art, faith and empire in early Islam, London 2021. Gibb, H.A.R., ʿAbd Allāh b. ʿĀmir, in ei2 i, 43. Gibb, H.A.R., ʿAbd al-Malik b. Marwān, in ei2 i, 76–77. Glubb, J.B., The great Arab conquests, London 1963. Hasson, I., Ziyād b. Abīhi, in ei2. xi, 519–522. Hawting, G.R., Yazīd (i) b. Muʿāwiya, in ei2. xi, 309–311. Hawting, G.R., Yazīd (iii) b. al-Walīd, in ei2. xi, 311–312. Hawting, G.R., The first dynasty of Islam: The Umayyad Caliphate ad661–750, London 22000. Heidemann, S., and H.–L. Hagemann, Transregional and regional elites: Connecting the early Islamic empire, Berlin 2020. Hosein, R.I.C., Tribal alliance formations and power structures in the Jahiliyah and early Islamic periods: Quraysh and Thaqif (530–750ce), PhD diss., University of Chicago 2010.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

44

marsham

Humphreys, R.S., Muʿawiya Ibn Abi Sufyan: The saviour of the caliphate, Oxford 2006. Kennedy, H., Al-Walīd 1. Al-Walīd (i) b. ʿAbd al-Malik b. Marwān (r. 86–96/705–15), in ei2. xi, 127–129. Kennedy, H., Al-Walīd 2. Al-Walīd (ii) b. Yazīd b. ʿAbd al-Malik (r. 125–6/743–4), in ei2 xi, 128. Kennedy, H., The Prophet and the age of the caliphates: The Islamic Near East from the sixth to the eleventh century, London 11986, Abingdon and New York 32016. Kennedy, H., From oral tradition to written record in Arabic genealogy, in Arabica 44 (1997), 531–544. Kennedy, H., The great Arab conquests: How the spread of Islam changed the world we live in, London 2007. Keshk, K.M.G., Abū Sufyān, in ei3, 41–42. Keshk, K.M.G., ʿAmr b. al-ʿĀṣ, in ei3, 68–69. Lammens, H., and K.H. Blankinship, Yazīd (ii) b. ʿAbd al-Malik, in ei2 xi, 311. Landau-Tasseron, E., Murra, in ei2. vii, 628–631. Landau-Tasseron, E., Alliances among the Arabs, in al-Qantara 26 (2005), 141–173. Lecker, M., al-Zibriḳān b. Badr, in ei2. xi, 496. Lynch, R.J., Arab conquests and early Islamic historiography: The Futuh al-Buldan of alBaladhuri, London and New York 2020. Madelung, W., The succession to Muḥammad: A study of the early caliphate, Cambridge 1997. Marsham, A., Rituals of Islamic monarchy: Accession and succession in the first Muslim empire, Edinburgh 2009. Marsham, A., The Umayyad Empire, Edinburgh (forthcoming). McMillan, M.E., Fathers and sons: The rise and fall of political dynasty in the Middle East, New York and Basingstoke 2013. Morony, M.G., ʿAbd Allāh b. ʿĀmir, in ei3, 15–16. Robinson, C.F., ʿAbd al-Malik, Oxford 2005. Robinson, M., Marriage in the tribe of Muhammad: A statistical study of early Arabic genealogical literature, Berlin and Boston 2020. Robinson, M., Qurashi marriage and the roots of revolt: The rebellion of ʿAbd Allāh b. Muʿāwiya, 744–747, in A. Marsham (ed.), The Umayyad world, Abingdon and New York 2021, 518–538. Schleifer, J., Banū al-Ḥārith b. Kaʿb, in ei2 iii, 223. Schneider, J., Of vigilance and virgins: Honor, shame and access to resources in Mediterranean Societies, in Ethnology 10 (1971), 1–24. Su, I-W., Representations of the Marwanids in the Ansāb al-Ashrāf and the reception of its audience in the ninth-century cultural milieu, master’s thesis, University of Edinburgh 2013. Tapper, R., and N. Tapper, Marriage, honour and responsibility: Islamic and local mod-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

kinship, dynasty, and the umayyads

45

els in the Mediterranean and the Middle East, in Cambridge Anthropology 16 (1992), 3–21. Tayyara, A., Matrilineal lineages as a signifier of social links in the context of Badr and Uḥud, in J.S. Lowry and S.M. Toorawa (eds.), Arabic belles lettres, Atlanta 2019, 19–38. Treadwell, L., The “Orans” drachms of Bishr ibn Marwān and the figural coinage of the early Marwānid period, in J. Johns (ed.), Bayt al-Maqdis: Jerusalem and early Islam (Oxford Studies in Islamic Art 9.2), Oxford 1999, 223–269. Ullmann, M., Khālid b. Yazīd, in ei2 iv, 929–930. Urban, E., Conquered populations in early Islam: Non-Arabs, slaves and the sons of slave mothers, Edinburgh 2020. Watt, W.M., Muhammad at Mecca, Oxford 1953. Watt, W.M., Muhammad: Prophet and statesman, London 1961. Watt, W.M., Kināna, in ei2 v, 116. Watt, W.M., Ruḳayya, in ei2. viii, 594–595. Webb, P., Imagining the Arabs: Arab identity and the rise of Islam, Edinburgh 2016. Webb, P., al-Azd, in ei3, 1–5.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 3

He Reigned as Caliph; Then He Died: The Reigns of Caliphs Versified Geert Jan van Gelder

In Sellar and Yeatman’s immortal classic, 1066 and all that: A memorable history of England, all English monarchs are given in a rationalized genealogical tree, in which all seven Edwards are found in a straightforward line of fathers and sons, and the same with all Henrys, Georges, and Williams.1 Accuracy, alas, has fallen victim to rationalization, and more useful is an anonymous mnemonic poem that lists all English monarchs since William the Conqueror in eight couplets, beginning “Willie Willie Harry Stee / Harry Dick John Harry three” and ending with “After which Elizabeth / And that’s all folks until her death.”2 A Muslim or Arabic equivalent of Sellar and Yeatman, 622 and all that, has to my knowledge never been published,3 but there is no lack of Arabic mnemonic poems that list the caliphs. In view of my own specialism and that of the recipient of this volume, whose numerous publications that contain the word “caliphs” or “caliphate” would qualify him as caliphologist par excellence, it seems good to honor him by devoting the following essay to such poems.4 These poems are considerably longer than the doggerel verse on English kings and queens. The poem on the subject by the polymath al-Suyūṭī (d. 911/ 1505), found nearly at the end of his History of the caliphs (Tārīkh al-khulafāʾ) has 116 lines in monorhyme and basīṭ meter.5 It lists the caliphs in order, with dates and ultra-short information on their reigns, including the “post-Abbasid” Abbasid caliphs in Cairo as far as al-Mustamsik, just short of the last of the line before the Ottoman conquest of Egypt in 923/1517. Al-Suyūṭī introduces the 1 Sellars and Yeatman, 1006 and all that 39. 2 Found online at https://en.wikipedia.org/wiki/Mnemonic_verses_of_monarchs_in_England (last accessed July 6, 2021). Those who know their English history will identify “Stee” as King Stephen. 3 Long ago I wrote a short, irreverent essay with that title, in the vein of Sellar and Yeatman, which cannot be published, at least not in my lifetime. 4 On the genre, see, e.g., Rosenthal, History 336; Ullmann, Untersuchungen 55–56. 5 Al-Suyūṭī, Tārīkh al-khulafāʾ (Beirut 1408/1988) 587–591 (this edition lacks vs. 86; I have been unable to consult its first edition, Cairo 1371/1952); ibid., (Qatar 1424/2013) 782–786. Hereafter, page references are to the first-mentioned edition unless indicated otherwise.

© Geert Jan van Gelder, 2022 | doi:10.1163/9789004525245_004

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

he reigned as caliph; then he died

47

poem as follows: “A certain earlier poet (baʿḍ al-aqdamīn) composed an urjūza (poem in rajaz meter) on the names of caliphs and their death dates, ending with the days of al-Muʿtamid. I have composed a qaṣīda that is better, thinking fit to seal the book with it.”6 It seems likely that, with the reference to his unnamed predecessor, al-Suyūṭī meant the poet ʿAlī b. al-Jahm (d. 249/863),7 for he, patronized by Caliph al-Mutawakkil, is known not only for his regular panegyric poems but also for a lengthy poem in paired rhyme and rajaz meter on the caliphs, called al-Muḥabbara fī l-tārīkh.8 The poem, which begins with the beginning of human history, ends with the Caliph al-Mustaʿīn (r. 248– 252/862–866), not with al-Muʿtamid, as al-Suyūṭī says, and since the poet died a few years before al-Muʿtamid’s reign (256–279/870–892), one must conclude that al-Suyūṭī refers to a version of the poem with a continuation, possibly that by Abū l-Ḥasan Aḥmad b. Muḥammad al-Asadī (d. 320/932), mentioned and partly quoted by Abū Bakr al-Ṣūlī.9 These two poems are not the only ones concerned with caliphal history. In al-Andalus, Abū Ṭālib ʿAbd al-Jabbār (d. 516/1122–1123 or 510/1116–1117)10 composed a long historical urjūza (452 couplets) on universal history, including the reigns of the eastern caliphs, reaching al-Mustarshid, followed by the Spanish Umayyads and subsequent western dynasties.11 Not to be outdone, the polymath and vizier Lisān al-Dīn Ibn al-Khaṭīb (d. 776/1375), who had a few more centuries to cover and who probably knew ʿAbd al-Jabbār’s poem, composed a similar but much longer urjūza (1,157 couplets) entitled Raqm al-ḥulal fī naẓm al-duwal.12 Like ʿAbd al-Jabbār’s poem, it is not restricted to caliphs but also lists

6 7 8

9 10 11 12

Al-Suyūṭī, Tārīkh al-khulafāʾ 587. On him, see Gibb, ʿAlī b. al-D̲ j̲ahm; gas ii, 580–581; Kennedy, ʿAlī b. al-Jahm. ʿAlī b. al-Jahm, Dīwān 228–250 (al-Muḥabbara fī l-tārīkh; caliphal history from p. 243); also published as Urjūza fī tārīkh al-khulafāʾ, caliphs starting p. 60 at couplet 231. See also al-Ṣūlī, Awrāq (nb retrograde pagination) 429–428, 460, 480–479, with a sequel by Abū l-Ḥasan Aḥmad b. Muḥammad al-Asadī, ibid. 388, 360; al-Maqdisī, al-Badʾ wa-l-taʾrīkh ii, 85–86. On the poem, see also Ullmann, Untersuchungen 55. The title, al-Muḥabbara (the adorned [poem]) seems strange in view of its distinctly unornate style, but one may compare the several works by Muḥammad b. Ḥabīb (d. 240/860) written in prosaic prose, entitled al-Muḥabbar, al-Munammaq, al-Muwashshaḥ, etc.—I was pleasantly surprised on learning that Harry Munt chose to write, as his contribution to this book (see the next chapter), a study of precisely this poem—in far more detail than I can give here. gas i, 310, ii, 581; see, e.g., Yāqūt, Muʿjam al-udabāʾ iv, 197; al-Ṣafadī, Wāfī viii, 32. He died in 516/1123 according to Ibn Bashkuwāl, Ṣila 554; and in 510/1116–1117 according to al-Ṣafadī, Wāfī xviii, 35. Ibn Bassām, Dhakhīra (Beirut) i, 920–944, (Cairo) ii, 405–431; for the passage on caliphs, see 929–940 (Beirut), 414–427 (Cairo). Ibn al-Khaṭīb, Raqm al-ḥulal (Tunis; the edition by ʿAdnān Darwīsh was not accessible

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

48

van gelder

rulers with other titles, as does Amīn al-Dawla Abū l-Ghanāʾim Muslim b. Maḥmūd al-Shayzarī (d. after 622/1225), who wrote an Urjūza fī l-tārīkh, which he included in his anthology Jamharat al-Islām. It ends with the Ayyubid rulers; the anthology was dedicated to the last Ayyubid ruler in Yemen.13 In 641/1243, Taqī al-Dīn Abū l-ʿAbbās Aḥmad b. al-Mubārak al-Naṣībī (d. 663/1265) composed a historical urjūza reaching the last caliph of Baghdad, al-Mustaʿṣim, whose reign from 640/1242 came to a violent end in 656/1258.14 Al-Sakhāwī (d. 902/1497), in al-Iʿlān bi-l-tawbīkh li-man dhamma ahl al-tawrīkh,15 mentions an urjūza on the caliphs by Abū Muḥammad Jaʿfar b. Aḥmad b. al-Ḥusayn alSarrāj (d. 500/1106–1107); another by al-Shams Muḥammad b. Aḥmad al-Bāʿūnī al-Dimashqī (d. 861/1456–1457), entitled Tuḥfat al-ẓurafāʾ fī tawārīkh al-mulūk wa-l-khulafāʾ,16 reaching al-Ashraf Barsbāy, with a sequel (dhayl) by his nephew al-Bahāʾ Muḥammad b. al-Qāḍī l-Jamāl Yūsuf (d. 910/1505);17 and an urjūza on the caliphs by Ibn Abī l-Baqāʾ.18 What Ullmann says about ʿAlī b. Jahm’s urjūza, “eine trockenen Aufzahlung von Fakten … eine blecherne, stupide Reimerei” (a dry list of facts and crude, silly doggerel verse),19 could be applied, though in varying degrees, to all these poems. This is not necessarily because the poets were incompetent; ʿAlī b. alJahm, for one, despite his doggerel versification of caliphal reigns, is an excellent poet in his other compositions. What to us may look silly was surely not considered silly by the standards of the poets and their intended readership. Instead of intending to compose beautiful poetry, with eloquent diction, original metaphors and comparisons, and striking motifs, they aimed at versifying generally well-known facts for the sake of memorization, in some cases combined with praise of a ruler to whom the poem was dedicated. The aims of the several poets mentioned above differed. ʿAlī b. al-Jahm and al-Suyūṭī provide

13 14 15

16 17 18

19

to me). After each major historical period, a commentary in prose (rather patchy for the earlier periods) is provided. The poem is not included in his Dīwān. Al-Shayzarī, Jamharat al-Islām 299–309. gals i, 590. Al-Sakhāwī, Iʿlān 305–306; cf. Rosenthal, History 336. Instead of the unusual al-tawrīkh (on which see below, note 20) one also finds al-taʾrīkh (as in ei2 viii, 882) or al-tawārīkh (as in gals ii, 35, 32). In Ḥājjī Khalīfa, Kashf al-ẓunūn i, col. 1243, the poem is entitled Farāʾid al-sulūk fī tārīkh al-khulafāʾ wa-l-mulūk and its sequel al-Ishāra al-wafiyya ilā l-khaṣāʾiṣ al-Ashrafiyya. According to Brockelmann (gals ii, 54) this poem in rajaz was composed in 901/1495. He may be the Andalusian poet Muḥammad b. Muḥammad b. Sulaymān b. Abī l-Baqāʾ (d. 616/1219–1220; see al-Ṣafadī, Wāfī i, 215). No doubt there is more. An urjūza fī l-khulafāʾ by Badr al-Dīn Muḥammad b. Jamāʿa (d. 733/1333) is mentioned by Brockelmann (gals ii, 81). Ullmann, Untersuchungen 55.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

he reigned as caliph; then he died

49

the durations of the successive reigns, in years and months, and sometimes mention the year of an event. Expressing numbers in metrical verse has forced the poets to use some awkward constructions, and even when the numbers fit in naturally they do not enhance the poetic quality. The Andalusians ʿAbd alJabbār and Ibn al-Khaṭīb, however, dispense with dates and give potted political and cultural history, which makes their poems more readable as literature. One could call their efforts versified tārīkh (history), while Ibn al-Jahm and al-Suyūṭī produced versified taʾrīkh (chronography or dating).20 Almost all poets in this genre chose the meter commonly used for didactic versification of facts, rajaz, with paired rhyme (aabbccdd …), which allows for length. Rajaz is an easy and flexible meter, with a basic foot of four syllables of which the third is always short, the fourth always long, and the first and second syllables either short or long at the poet’s discretion (xxsl). Al-Suyūṭī, however, chose not to follow this custom. In his prose introduction translated above, he calls his own poem a qaṣīda, saying that it is better than his predecessor’s urjūza. He composed his poem in the basīṭ meter, which has a higher status that the lowly rajaz. Moreover, instead of paired rhyme, he opted for the monorhyme used in the “standard” qaṣīda form.21 Every line ends in -rā, preceded by two short syllables. Although didactic poems that do not use rajaz are not rare, they are a minority, and al-Suyūṭī obviously wanted to distinguish his poem and himself by employing a more prestigious form. It is not clear whether he believed his poem to be superior merely by his use of this form, or whether the poem was also better, in his view, in terms of stylistic quality and historical content. We can attempt to decide for ourselves by looking at the two poems, or rather parts of them, in some detail and contrasting them with corresponding passages from the poems by ʿAbd al-Jabbār and Ibn al-Khaṭīb. We can only guess what prompted al-Suyūṭī to choose the rhyme in -rā, but in a poem that mentions dozens of death dates the rhyme word qubirā (he was buried) comes in handy. He uses it six times (ll. 6, 21, 57, 66, 91, 111). The rhyme word ʿumurā ([so-and-so many years] of age) occurs three times (ll. 3, 27, 54). Expressing dates in verse often requires breaking up the number. Thus, the 20

21

Tārīkh is obviously derived from taʾrīkh but has become a different word with a different plural (tawārīkh in contrast to taʾārīkh). The distinction is by no means strict; a historian or historiographer is always muʾarrikh, never muwarrikh, and the verbal noun tawrīkh is known to me only from the title of al-Sakhāwī’s Iʿlān (see above, note 15). With this phrase I refer to prosody, and not to the thematic sequence that is often found in the qaṣīda but (contrary to what is often stated) is not a necessary element of it. The term qaṣīda is ambiguous, for it is often used for any longer poem, and ʿAlī b. al-Jahm’s urjūza is sometimes called a qaṣīda muzdawija (poem in paired rhyme), e.g., in al-Ṣūlī, Awrāq, 460; Ibn al-ʿImrānī, Inbāʾ 95; Yāqūt, Muʿjam al-udabāʾ iv, 196; al-Ṣafadī, Wāfī viii, 32.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

50

van gelder

Prophet Muḥammad died in the year “three followed by ten” (l. 4), the caliph alMustaẓhir died in the year “two followed by ten of the sixth century ah” (l. 69), and the “post-Abbasid” caliph al-Mustaʿīn died in “five followed by ten” (of the ninth century; l. 95). All three lines rhyme in talī ʿasharā. Repetition of rhyme words, provided they are not too close together (a distance of seven lines is stipulated by theoreticians of traditional prosody), is allowed, but in a proper ode or elegy such repetitions would no doubt count as blemishes. Al-Suyūṭī’s rhyme words only rarely give essential or useful information; often, their sole function is to close the line with a tolerable rhyme. ʿAlī b. al-Jahm’s method of paired rhyme leaves much more freedom and provides more variation, but his rhymes, too, often seem feeble, as in Wa-qāma ʿAbdu l-Maliki bnu Marwān / mustanhiḍan li-l-ḥarbi ghayra wasnān, “Then reigned ʿAbd al-Malik ibn Marwān, / inciting warfare, not slumbering” (l. 262), with more examples in the quoted passage below. Unlike al-Suyūṭī’s monorhyme, the changing rhymes of the other poems also enable the caliphs’ names to appear from time to time in rhyme position, which, one would think, may facilitate memorization. Finding Arabic rhymes for al-Manṣūr, al-Rashīd, al-Amīn, for instance, is much easier than finding English rhymes for Henry, William, George, or Charles. Not only rhyme words are repeated in al-Suyūṭī’s qaṣīda: the recurrence of formulas can hardly be avoided. There is an abundance of expressions such as wa-qāma (here meaning “then came to reign”) at the beginning of lines (ll. 6, 8, 13, 29, 37–38, 44, 53, 55, 57, 61, 69–70, 73, 75–78, 81, 83, 86–87, 96–98), or (wa-)thumma, “(and) then” (ll. 16, 21, 27, 32–34, 40–43, 47, 58, 64–67, 71). It is the same with the other poems. Historians of Arab and Islamic history will not be much concerned with the poetic and stylistic qualities of these poems. They often neglect poetry, either because it is thought to offer little valuable historical information or because it is often difficult. Poetry may, however, fruitfully be used as an important historical source at times. But this cannot be said of the versifications listed above. It may be worthwhile, nevertheless, to investigate what is said about the individual rulers, and which facts were selected for inclusion in what I called the “ultra-short” information about each of them. One usually associates one or two keywords with individual caliphs of the first few centuries: ridda with Abū Bakr, conquests with ʿUmar b. al-Khaṭṭāb, civil war with ʿAlī, piety with ʿUmar b. ʿAbd al-ʿAzīz, the “good Umayyad caliph,” rakishness and poetry with al-Walīd b. Yazīd, the foundation of Baghdad with al-Manṣūr, civil war again and miḥna with al-Maʾmūn, the conquest of the Byzantine city of Amorium with al-Muʿtaṣim, and so on. One expects these events and facts to be mentioned in the poems, but although many of them are, there are some surprising omissions. Many of the later caliphs were nonentities, listed only with their dates. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

‫‪51‬‬

‫‪he reigned as caliph; then he died‬‬

‫‪Four Poems Compared‬‬

‫‪1‬‬

‫‪1.1‬‬ ‫‪ʿAlī b. al-Jahm‬‬ ‫‪In order to illustrate the differences between the four selected poems, I will‬‬ ‫‪focus on the successive reigns of six early Abbasids, al-Manṣūr, al-Mahdī, al‬‬‫‪Hādī, al-Rashīd, al-Amīn, and al-Maʾmūn, covering an important and colorful‬‬ ‫‪period of some 80 years (136–218/754–833). Since the historical events and‬‬ ‫‪names in the following excerpts should be well-known to almost all readers of‬‬ ‫‪this chapter, extensive annotation would seem unnecessary. Readers keen on‬‬ ‫‪poetry should be warned, again, that what follows is versification, not poetry.‬‬ ‫‪To begin, here is ʿAlī b. al-Jahm:22‬‬ ‫وقـــام بالخلافـــة المنصـــور ُ‬

‫ت بحزَ ْم ـه‪23‬الأم ـور ُ‬ ‫فاستوسق ـ ْ‬

‫سنـَـه ْ‬ ‫فعاش إثن َينِ‪24‬وعشر ي ـن َ‬

‫ي َحمي ِحم َى الملُك و يفُ ْ ني الخ َو َن َـه ْ‬

‫كـــه ْ‬ ‫مح ْرِمـــا ً بم َ ّ‬ ‫ثـــّم توُ فُ ـّــي ُ‬

‫ك ـه ْ‬ ‫ي عن ـه م ُلـ ْ َ‬ ‫ث المْه ـد ُ ّ‬ ‫ف ـو َرِ َ‬

‫شــر َ ِح ـَج ـٍح وشْهــرا‬ ‫فعــاش ع ْ‬

‫ف شه ـٍر ث ـّم زار القب ْـرا‬ ‫ونص ـ َ‬

‫ي موسى بعد َه ُ‬ ‫واستخلف الهادِ َ‬

‫ل ع َْه ـد َهْ‬ ‫وك ـان ق ـد ولّاه ُ قب ـ ُ‬

‫فعاش موس ـى سنـة ً وشهر يَ ـْـْن‬

‫ينق ُص يومـا ًواحـدا ً أوِ اثن َي ْـْن‬

‫وقـــام بالخلافـــة الرشيـــد ُ‬

‫ك المم َن ّـــع السعيـــد ُ‬ ‫الملَـِ ــ ُ‬

‫فعاش عشر ي ـن ووفـّـى عَّدهـا‬

‫ن وعامـا بً عد َهـا‬ ‫وعـاش عام َيـ ِ‬

‫ل‬ ‫جـ ْ‬ ‫ف شهٍر ثم ّ وافاه الَأ َ‬ ‫ونص َ‬

‫ل‬ ‫ت فاْنهَّد الجبَ َ ْ‬ ‫س يوم َ السب ِ‬ ‫بط ُو َ‬

‫و بايعـــوا محمـّ ــد َ الأمينـــا‬

‫ونكثـــوا الب َي ْعـــة َ أجمعينـــا‬

‫ل أحمـــد ُ‬ ‫إلّا قليلا ً والقليـــ ُ‬

‫ت للنـاس جميعـا ًم َو ْعِـد ُ‬ ‫والمـو ُ‬

‫فأ َمّنــــوه ثــــّم قتلــــوه ُ‬

‫مــا هكــذا عاه َدهــْم أبــوه ُ‬

‫م ـا ع ـاش إلّا أر بع ـا ً وأشه ُـرا‬

‫ســه معَّفــرا‬ ‫حتـّـى تهــاد َْوا رأ َ‬

‫و بايعَ ــوا المأمــونَ عبــد َ اللـّـه ِ‬

‫فبايعــوا يْقظــانَ غيــر َ ساهــي‬

‫وف ّاهـُ ــم ُ خلافـــة َ المنصـــورِ‬

‫ن والشهــورِ‬ ‫فــي عــددِ السنيــ َ‬

‫ث ـّم أت ـى ال ـروم َ فم ـات غاز ي ـا‬

‫ل القاصي ـا‬ ‫كان الب َد َن ْدونُ المحَ َـ َ ّ‬

‫‪Urjūza 68–69; Dīwān 248 (vss. 292–307).‬‬ ‫‪Dīwān: bi-ʿazmihī.‬‬ ‫‪Dīwān: thintayni.‬‬

‫‪- 978-90-04-52524-5‬‬ ‫‪Downloaded from Brill.com 01/29/2024 12:42:59PM‬‬ ‫‪via Università Statale degli Studi di Milano‬‬

‫‪22‬‬ ‫‪23‬‬ ‫‪24‬‬

52

van gelder

Then al-Manṣūr took on the caliphate and through his prudence matters turned into good order. He lived [i.e., reigned] for twenty-two years, defending the sanctuary of his realm and annihilating traitors. Then he died while on pilgrimage in Mecca. Al-Mahdī inherited from him his rule, Who then lived [i.e., reigned] for ten years and one month and a half; then he visited the grave, Leaving behind as caliph Mūsā al-Hādī, having appointed him before. Then Mūsā lived one year and two months bar one or two days. Then reigned as caliph al-Rashīd, the felicitous and mighty king.25 He lived [i.e., reigned] for full twenty years and two years and one after that, And half a month. Then his time was up and he died in Ṭūs on a Saturday. The mountain fell. They pledged allegiance to Muḥammad al-Amīn, but they broke their pledge, all of them, Except for a few—and those few were more praiseworthy. Death is appointed to all people. They gave him a safeguard and then they killed him. It was not thus that his father had made a covenant with them. He only reigned for four years and some months, until they presented his dust-covered head. They pledged allegiance to ʿAbdallāh al-Maʾmūn and thus they pledged allegiance to a vigilant man, not negligent. He completed, in number of years and months, the time of al-Manṣūr’s caliphate. Then he went to the Byzantines and died while on campaign, in the remote location of al-Badandūn. This passage is singularly lacking in characterization or interesting detail. One notes the bias toward al-Amīn in the conflict with his brother al-Maʾmūn, and 25

It is sometimes said that the title of malik (king) was disliked by pious Muslims, as denoting mere temporal, worldly rule in contrast to khalīfa or imām (cf. Kennedy, Caliphate 48). But one does find malik referring, in a positive context, to Abbasid caliphs; the same below in Abū Ṭālib ʿAbd al-Jabbār’s poem for Hārūn’s predecessor al-Hādī.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

53

he reigned as caliph; then he died

the lukewarm praise of the latter (“vigilant, not neglectful”). Only the lengths of the reigns are given, not the dates. 1.2 Al-Suyūṭī Al-Suyūṭī is even more terse and uninformative, giving the period a mere seven lines, but he does provide the dates:26 ‫مح ْرِمـا ًق ُبـِـرا‬ ٍ ‫خمسيـن بعـد ثمـا‬ ُ ‫ن‬

‫ت في‬ َ ّ ‫وقام م ِن بعدِه ِ المنصور ُ ث َم‬

‫ب حت ّى أْمرُه ُْم د َثرَ ا‬ َ ْ ‫ل الع ُر‬ َ ‫و َأهم‬

ُ َ‫ص أعمال ًا م َوال ِيه‬ َّ ‫خ‬ َ ‫وه ْو الذي‬

‫تْسٍع وست ّين مسموما ً كما ذ ُك ِرا‬

‫ي مات لدى‬ ّ ُ ‫ثم ّ ابن ُه ُ وه ُو َ المهَ ْد‬

‫في عاِم سبعين لم ّا هـَـّم أن غ َـد َرا‬

ُ ‫ثم ّ ابنـُـه ُ وهـو الهـادي ومو ْت َتـُـه‬

‫ثلاثة ً مات في الغ َز ْو الرفيـع ذرا‬

ً ‫ثم ّ الرشيـد وفـي تسعيـن تاليـة‬

‫ل كم ـا ق ُـدِرا‬ ٌ ‫ثماني ـا ً جاء ـه ق َت ْـ‬

ً ‫ثم ّ الأمين وف ـي تسعي ـن تالي ـة‬

‫ت فاعت َبـِـرا‬ ُ ‫ن عش ْرة َكان المـو‬ ِ ‫ثما‬

‫ت في‬ َ ّ ‫وقام من بعده المأمون ثم‬

After him there reigned al-Manṣūr. In eight-and-fifty, while on pilgrimage, he was buried. He is the one who entrusted governorships to his mawālī and neglected the Arabs, whose influence dwindled. Then his son, al-Mahdī, who died in sixty-nine, poisoned, it is said. Then his son, al-Hādī, whose death occurred in the year seventy, when he was about to betray.27 Then al-Rashīd, who died in three-and-ninety, while on an ambitious campaign.28 Then al-Amīn, and in ninetyand-eight a killing occurred, as was fated. After him reigned al-Maʾmūn, and then in [two hundred and] eighteen death came. Contemplate!

26 27 28

Al-Suyūṭī, Tārīkh al-khulafāʾ (ʿAbd al-Ḥamīd) 588–589, (al-Ḥusaynī) 783. Al-Hādī intended to annul the future succession of his brother al-Rashīd that had been arranged by their father. The interpretation of fī l-ghazwi l-rafīʿi dhurā (vowelled dharā in some editions) is not wholly clear. Hārūn died while on campaign in Khorasan.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

‫‪54‬‬

‫‪van gelder‬‬

‫‪1.3‬‬ ‫‪Abū Ṭālib ʿAbd al-Jabbār‬‬ ‫‪By contrast, the Andalusian Abū Ṭālib ʿAbd al-Jabbār, who is not interested in‬‬ ‫‪dates or the lengths of reigns, uses 44 couplets to relate the events in those‬‬ ‫‪years:29‬‬ ‫صي ِّــر الأمــر ُ إلــى المنصــورِ‬ ‫ف ُ‬

‫فأحكـــم َ التدبيـــر َ للأمـــورِ‬

‫حــز ِْم‬ ‫إذ كــان ذا سياســة ٍ و َ‬

‫ي الع َــز ِْم‬ ‫ي ق َــوِ َ ّ‬ ‫مســَّدد َ الــر ْأ ِ‬

‫ب‬ ‫كـــة ٍ و ي َث ْـــرِ ِ‬ ‫ت بم ّ‬ ‫فخرجـــ ْ‬

‫ب‪30‬‬ ‫طالبـــة ٌ آل أبـــي طالـــ ِ‬

‫ج‬ ‫فآل ـت الح ـر ُ‬ ‫ب إل ـى اهتي ـا ِ‬

‫ج‬ ‫سلـِ ـٍم ال َ ّ‬ ‫مــع أبــي م ُ ْ‬ ‫ســراّ ِ‬

‫ل حت ّى اغتال ـه المنص ـور ُ‬ ‫فاحتا َ‬

‫لمـّــا أتـــاه الق َـــد َر ُ المقـــدور ُ‬

‫ص الأم ـر ُ لأب ـي جعف ـرِ‪31‬‬ ‫فخل َ َ‬

‫مه َن ّــأ ً مــن غيــرِ مــا تكــُّدرِ‬

‫ك ـه ْ‬ ‫قم َّ‬ ‫حت ّى توف ّي‪32‬ف ـي طر ي ـ ِ‬

‫كــه ْ‬ ‫ت الأيـّـام ُ عنــه م ُلـ ْ َ‬ ‫و بــ َز ّ ِ‬

‫ي‬ ‫ي الأمــر َ ابنـُـه ُ المهَ ْــد ّ ُ‬ ‫فو َل ِــ َ‬

‫ي‬ ‫سن َى الر ِ ّضا ال َس ّر ُ ّ‬ ‫سيرة الح ُ ْ‬ ‫ذو ال ِّ‬

‫ح أبــي الع َتاه ِيـَـه ْ‬ ‫وهـُـو َ ممــدو ُ‬

‫ف ـي غي ـرِ م ـا قصي ـدة ٍ وقافيِ ـَـه ْ‬

‫مشب ِّبـــا ً بعتُ ْبـــة ٍ محبو ب َت ِـــه ْ‬

‫صتـِـه ْ‬ ‫ك ـر ُ ق ِ ّ‬ ‫ب التار يـخ ذِ ْ‬ ‫كت ُ ِ‬ ‫في ُ‬

‫شعْــر ٌ ف َشــا‬ ‫لابنْ تـِ ـه ِ ع ُل َيـّ ـة ٍ ِ‬

‫لٍ ور َشا‬ ‫ن َط ّ‬ ‫صــة ٌ فــي شــأ ِ‬ ‫وق ّ‬

‫وكــان يشتــُّد علــى الزنادِق َــه ْ‬

‫وم َن غ َلا يرُ ْضـي بـذاك خالق َـه ْ‬

‫سطا‬ ‫إذ كان في العْدل إماما ًم ُْق ِ‬

‫حي ْن ُــه فاعت ُب ِطــا‬ ‫حتـّ ـى أتــاه َ‬

‫فولَ ِي َ اله ـادي ابنـُـه م ـن بع ـدِه ِ‬

‫سيرتــه وقصــدِه ِ‬ ‫فســار فــي ِ‬

‫ت أي ّام ُـه ُ‬ ‫ع َـْدلا ًإلـى أن ذهبـ ْ‬

‫ق عــن مأمولــه ِحمام ُــه ُ‬ ‫فعــا َ‬

‫فص ـار ه ـارونُ الرشي ـد ُ تالي ـا‬

‫للم َلـِـك الهــادي إمامــا ً واليــا‬

‫كعْبـَـه ُ‬ ‫ك وأعلــى َ‬ ‫فش َي ّــد المل ُـْـ َ‬

‫ل صعْبـَـه ُ‬ ‫ح َْزمــا ً وع َْزمــا وأذ ّ َ‬

‫ك الأمجــادا‬ ‫واستــو ْزر البرامــ َ‬

‫فاستو ْســق الأمــر ُ بهــْم وزادا‬

‫ث الأيـّـاِم‬ ‫حت ّى د َهاهـْم حـاد ُ‬

‫ش فإلــى انصــراِم‬ ‫وكــ ُ ّ‬ ‫ل عي ْــ ٍ‬

‫ثم ّ دهى الحـيَ ْنُ الرشيد َ فاخت ُـرِْم‬

‫حتـِـْم‬ ‫حت ْم ٌ في العبِ اد قد ُ‬ ‫ت َ‬ ‫والمو ُ‬

‫‪Ibn Bassām, Dhakhīra (Beirut) i, 934–936; (Cairo) ii, 420–423. There are some minor dif‬‬‫‪ferences between the editions.‬‬ ‫‪The hemistich is corrupt, rhyme and meter being faulty.‬‬ ‫‪The meter is corrupt.‬‬ ‫‪The meter requires reading this as either tuwuffī, as a licence for tuwuffiya, or as the less‬‬ ‫‪correct tawaffā.‬‬ ‫‪- 978-90-04-52524-5‬‬ ‫‪Downloaded from Brill.com 01/29/2024 12:42:59PM‬‬ ‫‪via Università Statale degli Studi di Milano‬‬

‫‪29‬‬ ‫‪30‬‬ ‫‪31‬‬ ‫‪32‬‬

‫‪55‬‬

‫‪he reigned as caliph; then he died‬‬

‫ن‬ ‫ثـــّم و َل ِـــي محمـّ ــد ُ الأميـــ ُ‬

‫ل به الت ِّن ِّينُ‬ ‫فـــي طـــالعٍ ح َ ّ‬

‫فلــم يــزْل مشتغلِ ا ً بال َل ّْهــوِ‬

‫فـــي غِـــر ّة ٍ ومه ْلـــة ٍ وز َه ْـــوِ‬

‫ن‬ ‫شــده أبــو ن ُــواس الح َسَــ ُ‬ ‫ي ُن ْ ِ‬

‫ن‬ ‫جــ ُ‬ ‫وكــان ممـّ ـْن شأن ُــه ُ التم ُ ّ‬

‫ن‬ ‫أشع ـار ُه ف ـي الخمـْـر والغلِ مْ ـا ِ‬

‫ن هان ـي‬ ‫فيحت ـذي م ـا قال ـه ُ اب ـ ُ‬

‫ن‬ ‫ف بالمأم ـو ِ‬ ‫حتـّـى أت ـاه الحت ْـ ُ‬

‫ن‬ ‫فصــار ر َه ْنــا ً فــي يــدِ الم َنــو ِ‬

‫أنحــى عليــه طاهــر ٌ فاغتال َــه ُ‬

‫سلطانــه أزال َــه ُ‬ ‫قت ْلا ً وعــن ُ‬

‫ب فـي بغـدادِ‬ ‫ت الحـرو ُ‬ ‫ودار ِ‬

‫ل أمرُهـــا إلـــى الف َســـادِ‬ ‫وآ َ‬

‫فجاءهــا المأمــونُ عبــد ُ اللـّـه ِ‬

‫ل أمــٍر داه ِ‬ ‫فانــزاح عنهــا كــ ُ ّ‬

‫س‬ ‫ت في زِ ينة الع َرو ِ‬ ‫حت ّى اغتد ْ‬

‫س‬ ‫ب النحـو ِ‬ ‫وغاب عنها كوكـ ُ‬

‫س لــه فسَلمِ ــوا‬ ‫إذ بايــَع النــا ُ‬

‫وأشــرق الدهــر ُ وكــاد ي ُْظلـِـم ُ‬

‫سيرتــه المأمــونُ‬ ‫وكــان فــي ِ‬

‫ن‬ ‫ع َـْدلا ً رض ـا ً ل ـه تقُ ً ـى ودِي ـ ُ‬

‫صـــٍر بالعلـــم والكلاِم‬ ‫ذا ب َ َ‬

‫مف َو ّهـــا ً بال َن ّث ْـــر والن ِّظـــاِم‬

‫ن أكثـْم‬ ‫وكـان فـي أي ّامـه ابـ ُ‬

‫قاضيه ِ يحي ـى ال َل ّو ْذ َع ـُّي المفُ ْ ه ِـم‬

‫ف‬ ‫ث مع ـه مستط ـر َ ُ‬ ‫ل ـه حدي ـ ٌ‬

‫ف‬ ‫وكــان ذا ف ِْقــه ٍ لــه تصــ ُر ّ ُ‬

‫وثـار َ إ براهيـم ُ إ بـن‪33‬المهَ ْ ـدي‬

‫سعْــدِ‬ ‫عليــه والطالــُع غيــر ُ َ‬

‫فعاق َـــه عمـّ ــا أراد الق َـــد َر ُ‬

‫فجاءــــه منهزمــــا ً يعتــــذِر ُ‬

‫ل‬ ‫نج ْـ َ‬ ‫ن َ‬ ‫واستـوزر الح َسَـ َ‬ ‫ل س َْهـ ِ‬

‫ل‬ ‫سـ ّ َ‬ ‫ن ِ‬ ‫إذ ناهـَـز َ الحس ـ ُ‬ ‫ن الـك َْه ـ ِ‬

‫م ُصاه ِــرا ً لــه ببـُـورانَ ابن َتـِـه ْ‬

‫منوِ ّهــا ً مــن جاه ِــه ِ وح ُْرم َتـِـه ْ‬

‫فصــَّد عمـّ ـا ينتحيــه الح َسَنــا‬

‫ك ِحمــاٍم بدفــاعٍ قد د َنا‬ ‫و َْشــ ُ‬

‫ن‬ ‫فأصبح المأم ـونُ بع ـد َ الح َسَـ ِ‬

‫ن‬ ‫ب الح َــز َ ِ‬ ‫مرزّءــا ً يلَبْ ـَـس ثــو ْ َ‬

‫م ُو َرِّ يــا ً إذ كــان قــد سقــاه ُ‬

‫حشــاه ُ‬ ‫سُمـّـا ً وخيمــا ً‪34‬قاطعــا ً َ‬

‫و بايَع المأمونُ موسى ألر ِ ّض ـا‪35‬‬

‫ثم ّ قضى اللـّـه ُ لموسـى مـا قضـى‬

‫فد ُفـِ ـن الرضــا مــع الرشيــدِ‬

‫طو بى لموسـى مـن فتـًـى شهيـدِ‬ ‫ُ‬

‫ثم ّ ثـوى المأمـونُ فـي جهـادِه ِ‬

‫ره ْنــا ً بمــا قّدمــه مــن زادِه ِ‬

‫‪The irregular hamza is required by the meter (unless one reads it as Ibrāhīmuni bnu).‬‬ ‫‪ in the editions.‬وحياّ ً‪My emendation of‬‬ ‫‪The irregular hamza of the article is required by the meter.‬‬

‫‪- 978-90-04-52524-5‬‬ ‫‪Downloaded from Brill.com 01/29/2024 12:42:59PM‬‬ ‫‪via Università Statale degli Studi di Milano‬‬

‫‪33‬‬ ‫‪34‬‬ ‫‪35‬‬

56

van gelder

Then al-Manṣūr was made to rule and he consolidated the arrangement of affairs, For he was prudent in governance, having sound views and resolve. In Mecca and Yathrib [i.e., Medina] some claimants of the family of Abū Ṭālib rebelled.36 The war led to commotion with Abū Muslim the saddler.37 Al-Manṣūr plotted until he had him murdered when the divine decree came to him. Then Abū Jaʿfar [al-Manṣūr] ruled uncontested, happily, without further troubles, Until he died on his way to Mecca and Time stripped him of his rule. Then his son al-Mahdī came to power, a man of excellent conduct and lofty good will. He was eulogized by Abū l-ʿAtāhiya in many an ode and poem,38 While the story of his [i.e., Abū l-ʿAtāhiya’s] love poems on his beloved, ʿUtba, is mentioned in books of history. Al-Mahdī’s daughter ʿUlayya also composed poetry that was widely known, as was her story with Ṭall (“Dew”) and Rashaʾ (“Fawn”).39 He dealt severely with the Manichaean heretics and the Shiʿite extremists, thereby pleasing his Creator, For in matters of justice he was a just Imam, until his death came to him in the prime of life. Then his son al-Hādī succeeded him and came to power and followed him in his conduct and purpose,

36

37 38

39

This probably refers to the ʿAlid rebellion of Muḥammad b. ʿAbdallāh b. Ḥasan b. Ḥasan b. ʿAlī, called al-Nafs al-Zakiyya; he was killed in 145/762–763. The line is clearly corrupt and the translation tentative. Abū Muslim al-Khurāsānī is sometimes called a saddler (sarrāj), see, e.g., Agha, Revolution 58. Abū l-ʿAtāhiya is more famous for his renunciant or ascetic poems. For the story of how al-Mahdī became his patron, involving a caliphal slave girl called ʿUtba, see Rowson, Abu l-ʿAtahiyah 13–14. On ʿUlayya (d. 210/825), see Gordon, Place of competition; and al-Samarai, Die Macht der Darstellung 47–61, 95–120.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

57

he reigned as caliph; then he died

With justice, until his days came to end and his death prevented his ambition. Then al-Rashīd became the next, following King al-Hādī, to lead and to govern. He fortified his realm and raised its glory, with prudence and resolve, and overcame its difficulties. As viziers he appointed the members of the glorious Barmak family and with them affairs turned to good order, increasingly, Until the events of Time struck them down. Each life is due to be cut off eventually. Then death struck al-Rashīd and he was carried off; death is decreed for all people. Then succeeded Muḥammad al-Amīn, with his ascendant in Draco.40 He never stopped being preoccupied with amusing himself, heedless, at his leisure, arrogantly, While Abū Nuwās al-Ḥasan would recite poems to him, a poet specializing in shameless wantonness, Whose poems are about wine and youths, and the caliph followed what Ibn Hāniʾ composed,41 Until Death came to him (in his conflict) with al-Maʾmūn and he became a hostage in Death’s hand. Ṭāhir42 went for him and had him murdered and removed him from power. Fighting went on in Baghdad, which turned to ruins. Then ʿAbdallāh al-Maʾmūn arrived in it and thus every calamity was swept away from it, So that it appeared in the finery of a bride and its inauspicious star disappeared. When the people pledged their allegiance to him they thrived and Time, which had almost darkened, shone again.

40

41 42

This is apparently a bad astrological sign. Al-Yaʿqūbī, who often gives astrological data in connection with reigns, mentions that on the day of al-Amīn being sworn in as caliph the Ascending Node (al-raʾs) was in Cancer (al-Yaʿqūbī, Tārīkh ii, 525; idem, Works 1190). But ʿAbd al-Jabbār may be referring to al-Amīn’s ascendant at birth. Abū Nuwās is the name whereby al-Ḥasan b. Hāniʾ is usually known. Ṭāhir b. al-Ḥusayn, the most important general of al-Maʾmūn’s army in the war on his brother al-Amīn.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

58

van gelder

In his conduct al-Maʾmūn was just of good will, god-fearing and religious, Well-versed in knowledge and theology, eloquent in prose and verse. In his days lived Yaḥyā b. Aktham, his qadi, that man of quick-witted intellect. There is a curious anecdote about him and the caliph;43 he was well-versed in jurisprudence. Ibrāhīm b. al-Mahdī rebelled against him [viz., the caliph] but his star was inauspicious, And the divine decree thwarted his ambition, so he came to him, apologizing. He appointed al-Ḥasan, son of Sahl as his vizier when al-Ḥasan was of the age of maturity,44 And married the latter’s daughter Būrān, as a tribute to his standing and respect. But al-Ḥasan was turned away from his course by a speedy death […]45 And al-Maʾmūn, after al-Ḥasan’s death, was afflicted, donning the cloth of grief, Feigning, since it was he who had given him to drink a lethal46 poison that cut his bowels.47

43

44 45 46 47

Yaḥyā b. Aktham (d. 242/857), chief qadi of Baghdad, boon companion of al-Maʾmūn, described in anecdotes especially for his pederasty. The “curious anecdote” could be the story about his condemnation of mutʿa marriage (temporary marriage) in the presence of al-Maʾmūn, which much impressed the caliph, as told in al-Khaṭīb al-Baghdādī, Tārīkh Madīnat al-Salām xvi, 291–293; and Ibn Khallikān, Wafayāt vi, 149–151; but there are more stories. This somewhat gratuitous information is obviously prompted by the rhyme (Sahl—kahl). I do not know what is intended by bi-difāʿin qad (or idh) danā. Reading wakhīman instead of what looks like wa-ḥayyan in the editions, which does not make sense. There is considerable confusion in this and the following couplet, which are about alMaʾmūn’s appointment of ʿAlī b. Mūsā al-Riḍa, the eighth imam of the Twelver Shiʿi, as his successor—not Mūsā as in the poem. Al-Maʾmūn’s motives for this are not clear. It caused consternation among the leading Abbasids and ʿAlī’s death, shortly afterwards, possibly by poison, has been attributed to the caliph in some sources. Al-Faḍl b. Sahl (the brother of al-Ḥasan b. Sahl), who apparently was behind the appointment, was murdered and al-Maʾmūn’s involvement in this is suspected, although he displayed grief and executed the actual murderers. Our poet is wrong: al-Ḥasan was not murdered and died of natural causes.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

‫‪59‬‬

‫‪he reigned as caliph; then he died‬‬

‫‪Al-Maʾmūn acknowledged Mūsā al-Riḍā as his successor,‬‬ ‫‪but then God decreed for Mūsā what was decreed‬‬ ‫‪And he was buried alongside al-Rashīd:‬‬ ‫!‪Blessed be he, Mūsā, as a martyr‬‬ ‫‪Then al-Maʾmūn, waging holy war, found a resting place,‬‬ ‫‪a hostage to the provisions he had prepared.48‬‬ ‫‪1.4‬‬ ‫‪Lisān al-Dīn b. al-Khaṭīb‬‬ ‫‪Ibn al-Khaṭīb, too, dispenses with dates and lengths of reigns (some of these‬‬ ‫‪are given in the prose commentary) and summarizes some historical events.‬‬ ‫‪He deals with the same period in 37 couplets:49‬‬ ‫ثــّم تولـّـى بعــده المنصــور ُ‬

‫الأســـد ُ المس َل ّـــط اله َصـــور ُ‬

‫ت الأبه ـاء ُ والقص ـور ُ‬ ‫فضاء ـ ِ‬

‫ت بعص ـره العص ـور ُ‬ ‫وافتخ ـر ْ‬

‫شجاع َــه ْ‬ ‫العلــم ُ والتدبيــر ُ وال َ ّ‬

‫ش بغَ ْتـا ًكقيِ ـام الساع َـه ْ‬ ‫ي َبط ِـ ُ‬

‫ض وأرسى الدول َه ْ‬ ‫فدَّوخ الأر َ‬

‫صو ْل ـه ْ‬ ‫ض ـّده م ـن َ‬ ‫فلم يك ـن ل ِ‬

‫ج غيـر َ مـر ّة ٍ مـن عمُ ـرِه ِ‬ ‫وح ـ َ ّ‬ ‫وخلـّـص الأمــر َ بلا م ُنــازِِع‬

‫صـر ْ فـي جميـِع أمـرِه ِ‬ ‫ولـم يق ِّ‬ ‫ف النـازِع‬ ‫كـ ّ‬ ‫سب َ‬ ‫لـكنّها القـو ْ ُ‬

‫ك ـه ْ‬ ‫ت الأيـّـام ُ عنـه م ُلـ َ‬ ‫وابتـ َز ّ ِ‬

‫ك ـه ْ‬ ‫ق مَ َ ّ‬ ‫وم ـات و َه ْ ـو بطر ي ـ ِ‬

‫ي‬ ‫و ُ‬ ‫صيـّـر الأم ـر ُ إل ـى المْه ـد ّ ِ‬

‫ي‬ ‫َ‬ ‫بح ْــرِ ال َن ّــدى وقمــرِ ال َن ّــد ّ ِ‬

‫وكان مرهو ب ـا ً م ُط ـاعَ الأم ـرِ‬

‫حظ م َـن خاَطبـه عـن جَم ْـرِ‬ ‫يل َ ْ َ‬

‫ش َْهمــا ً جُ‬ ‫ش اعــا ً ب َطَلا ً أديبــا‬

‫ج ـودِه ِ الجديب ـا‬ ‫ث ُ‬ ‫يرَ ْوي بغِ َي ْـ ِ‬

‫ح أب ـي الع َتاه ِيـَـه ْ‬ ‫وه ـو مم ـدو ُ‬

‫أي ّام ُـــه ُ م ُش ْر ِقـــة ٌ وزاه ِيـَــه ْ‬

‫حتـّـى إذا أقفـر منـه النـادي‬

‫قام ابن ُه ُ بالملُ ْك موسى الهـادي‬

‫ج ـه ِ‬ ‫ل الأم ـر َ عل ـى م ِنها ِ‬ ‫فحم َـ َ‬

‫جه ِ‬ ‫ك عن إنها ِ‬ ‫وصان رسم َ الملُ ِ‬

‫ت الف ـؤادِ‬ ‫وك ـان ش َْهم ـا ً ثاب ـ َ‬

‫ي‬ ‫صــ ُ‬ ‫حديثـُـه ُ مت ّ ِ‬ ‫ل ال َت ّــرداد ِ‬

‫ل م ُّدت ُـه أن د َلـ َك ـا‬ ‫ول ـم ت َط ُـ ْ‬

‫ثـّم الرشيـد ُ بعـده قـد م َلـ َكـا‬

‫ظـــم الخلافـــة َ الرشيـــد ُ‬ ‫فع َ ّ‬

‫ق والتسديــد ُ‬ ‫فظهــر التوفيــ ُ‬

‫ج ـودِه ِ‬ ‫وكان بحرا ً زاخرا ً ف ـي ُ‬

‫جــودِه ِ‬ ‫وغ ُــر ّة ً غ َــراّ ء َ فــي و ُ ُ‬

‫‪i.e., his fate in the hereafter depending on the good and bad deeds performed in his life‬‬‫‪time.‬‬ ‫‪Ibn al-Khaṭīb, Raqm al-ḥulal 20–22; for the corresponding prose commentary, see 24–26.‬‬ ‫‪- 978-90-04-52524-5‬‬ ‫‪Downloaded from Brill.com 01/29/2024 12:42:59PM‬‬ ‫‪via Università Statale degli Studi di Milano‬‬

‫‪48‬‬ ‫‪49‬‬

‫‪60‬‬

‫‪van gelder‬‬

‫خبـَـر ْ‬ ‫س بِشعْ ـٍر و َ‬ ‫وأعل ـم َ الن ـا ِ‬

‫حض َرْ‬ ‫ب منه الأصمع ُيّ إْن َ‬ ‫ج ُ‬ ‫يعْ َ‬

‫ك الـك ِرامــا‬ ‫واستــوزر البرام ِــ َ‬

‫ن شيئـا ًدامـا‬ ‫خت َص ّهـْم لـو أ ّ‬ ‫وا ْ‬

‫ت لهْم علـى يديـْـه ال َن ّْكبـَـه ْ‬ ‫جر َ ْ‬

‫ث سر ي ـِع الوثَ ْبـَـه ْ‬ ‫للـّـه ِ م ِـن ل َي ْـ ٍ‬

‫ن‬ ‫وع َق َـــد َ البي ْعـــة َ للأميـــ ِ‬

‫ن‬ ‫وإذ دعا الحـيَ ْنُ أتى في الحيِ ـ ِ‬

‫ن‬ ‫ي الأمــر َ ابنـُـه ُ الأميــ ُ‬ ‫فو َل ِــ َ‬

‫ن‬ ‫جــود ُه ُ م َعيــ ُ‬ ‫وكــان ن َْدبــا ً ُ‬

‫ن‬ ‫ف م َكي ـ ُ‬ ‫ش ـر َ ٍ‬ ‫وق َـْدر ُه ُ ف ـي َ‬

‫ن‬ ‫ح م ُبيـــ ُ‬ ‫ضـــ ٌ‬ ‫ضل ُـــه ُ م َت ّ ِ‬ ‫وف ْ‬

‫لـكنـّ ــه أخْل َـــد َ للب َطال َـــه ْ‬

‫س ـوء َ الحال َـه ْ‬ ‫ج ـ َر ّ علي ـه ذاك ُ‬

‫س‬ ‫ن‪50‬وك ـا ِ‬ ‫ب ـاع العلُ ا بش ـاد ٍ‬

‫صحبــة ِ الشيــ ِ‬ ‫س‬ ‫خ أبــي نــوا ِ‬ ‫و ُ‬

‫وغ َي ّــر َ الع ُهــدة َ فــي مأم َن ِهــا‬

‫وأخ ـرج الحيَ ّ َـة َ م ـن م َ ْ‬ ‫كمنَ ِه ـا‬

‫ل م ُّدت ُـه ُ أن خ ُلع ـا‬ ‫ول ـم تط ُ ـ ْ‬

‫ل إذ عثـر الدهـر لُ عَ ـا‬ ‫ولـم يق ُـ ْ‬

‫ن‬ ‫فغ َم َرتـْ ــه فتِ نـــة ُ المأمـــو ِ‬

‫ن‬ ‫وخاطب َت ْـه بالخط ـوب الج ُـو ِ‬

‫ل إلى اصطلام ِـه ِ‬ ‫ت الحا ُ‬ ‫ومال ِ‬

‫وق ـام عب ـد ُ اللـّـه ف ـي م َقام ِـه ِ‬

‫ك العال ـم الحلي ـم ُ‬ ‫و َه ْ ـو الملي ـ ُ‬

‫ساع َــده السعــد ُ بمــا يــروم ُ‬

‫من بعـدِ مـا كاب َـد َ أمـر َ عم ِ ّـه ِ‬

‫وفــر ّج اللـّـه ُ لــه مــن غمَ ِ ّــه ِ‬

‫ك الُأ َمّ ـه ْ‬ ‫فق ـ َر ّ بالمأم ـون م ُلـْـ ُ‬

‫ب دائ ـٍم وغمَُ ّـه ْ‬ ‫بع ـد َ اضط ـرا ٍ‬

‫ق السعد ُ عل ـى الخلاف َـه ْ‬ ‫وأشر َ‬

‫ن بلا مخاف َــه ْ‬ ‫ل الأْمــ ُ‬ ‫وانســد َ َ‬

‫وك ـان حب ْـرا ً عالم ـا ً حكيم ـا‬

‫عــْدلا ً تقَ يـّـا ً حازمــا ً حليمــا‬

‫ل المْهـدي‬ ‫نجـ ُ‬ ‫وثـار َ إ براهيـم ُ َ‬

‫ســرا ً بغيــر عْهــدِ‬ ‫ونال َــه ق ْ‬

‫فآثرَ َ العْفو َ وأْغضى عـن د َم ِـه ْ‬

‫م َن ْق َبـــة ً شاهـــدة ً بكرَ َم ِـــه ْ‬

‫ومـات فـي غْزوتـه المعلوم َـه ْ‬

‫كانــت بهــا أعمال ُــه مختوم َــه ْ‬

‫‪Then, after him [viz., al-Saffāḥ] reigned al-Manṣūr,‬‬ ‫‪the mighty, bone-crushing lion,‬‬ ‫‪And courtyards and castles shone‬‬ ‫‪and epochs boasted of his epoch:‬‬ ‫;‪Learning, organization, courage‬‬ ‫‪he would strike suddenly, like the Last Hour.‬‬ ‫‪He subjugated the earth, firmly anchoring the dynasty.‬‬ ‫‪Anyone who opposed him was powerless to assail him.‬‬ ‫‪ in the edition.‬بشاذن ‪Wrongly‬‬

‫‪- 978-90-04-52524-5‬‬ ‫‪Downloaded from Brill.com 01/29/2024 12:42:59PM‬‬ ‫‪via Università Statale degli Studi di Milano‬‬

‫‪50‬‬

61

he reigned as caliph; then he died

More than once in his lifetime he went on pilgrimage and he never fell short in any matter. He ruled alone, not contested by anyone; but the bow was in the hand of the one who drew it.51 But Time stripped him of his rule and he died on his way to Mecca. Then al-Mahdī was made to rule, sea of generosity and moon of any assembly. He was feared, his command obeyed, staring at those he addressed as if from burning embers, Astute, brave, heroic, cultured, watering barren land with the rain of his generosity. He was eulogized by Abū l-ʿAtāhiya; his days were shining, radiant, Until he left the assembly [i.e., he died] and his son Mūsā al-Hādī reigned. He carried on, on the same path, and preserved the realm from being obliterated. He was astute, with a firm heart; his story is always repeated. But it did not last long before he declined. Then al-Rashīd came to reign. Al-Rashīd enhanced the greatness of the caliphate and gave evidence of success and sound opinion. He was an overflowing sea of generosity, a shining highlight while he lived, The most knowledgeable of people about poetry and history, admired by al-Aṣmaʿī when he was present. He appointed the noble Barmak family as viziers and conferred special distinction on them—if only it lasted! Through him a disaster befell them: Ah, what a lion he was, quick to pounce! He let people swear allegiance to al-Amīn and Death called him when his time had come. Then his son al-Amīn came into power; he was an energetic man, his generosity a well-spring, 51

Apparently a variation on the proverb “Give the bow to the one who made it” (i.e., let the right man do the job); Abū ʿUbayd al-Bakrī, Faṣl 298–299; al-ʿAskarī, Jamhara i, 66–67. The prose commentary refers to a couplet mentioning “the saddler” (al-sarrāj), viz. Abū Muslim, but this is not found in the poem.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

62

van gelder

His standing noble and firm, his excellence manifest and evident, But he was inclined toward futile pastimes, and this drew him to ruin. He sold a lofty status for a fawn-like lad, a cup of wine, and the company of shaykh Abū Nuwās.52 He altered the covenant that had been secure and brought out a snake from its hiding-place. It was not long before he was deposed, and he could not say, when Fate stumbled, “On your feet again!” The civil war with al-Maʾmūn overwhelmed him and addressed him with black distress.53 The situation moved toward his being overthrown and ʿAbdallāh [al-Maʾmūn] took his place. He was a learned, wise54 ruler, aided by good fortune in what he desired, After having suffered the matter of his uncle55 and God had relieved him of his grief. Then, through al-Maʾmūn, the realm of the community was firmly settled after continual disturbance and distress. A propitious star shone on the caliphate and security descended, fear disappeared. He was a learned scholar, a philosopher, just, pious, prudent, wise. Ibrāhīm, son of al-Mahdī, rebelled and he got hold of him by force, unconditionally, But he preferred to forgive him, sparing his life: a virtue testifying to his noble nature. He died on his conquering campaign, as is well-known, with which his works were sealed. 52

53 54 55

The “fawn” was al-Amīn’s favorite servant who was called Kawthar. It is unusual to see Abū Nuwās called a shaykh, but it is true that he was in his fifties during al-Amīn’s reign. This couplet is one of a handful quoted by al-Maqqarī in his Nafḥ al-ṭīb (vii, 101), who praises Raqm al-ḥulal (“extremely charming, sweet, and eloquent”) and says that he had memorized most of it but had forgotten it. The rather strange metaphor is prompted by paronomasia (khāṭabatʾhu—khuṭūb). Al-Maʾmūn is called ḥalīm twice; in addition to “wise” it also means “gentle-mannered, lenient, forbearing.” Ibrāhīm b. al-Mahdī, as mentioned after four couplets.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

he reigned as caliph; then he died

2

63

Conclusion

Summarizing, we find that al-Manṣūr is praised by ʿAlī b. al-Jahm, ʿAbd alJabbār, and especially by Ibn al-Khaṭīb, while al-Suyūṭī refrains from praise and seems to criticize him for favoring non-Arabs at the expense of the Arabs. That he died on his way to Mecca, mentioned by all four, is apparently more important than, for instance, al-Manṣūr’s foundation of Baghdad, which is ignored by all four poets even though it is arguably his greatest contribution to Abbasid history. The dubious plotting of Abū Muslim’s murder is mentioned by ʿAbd alJabbār. Al-Mahdī is a nonentity for ʿAlī b. al-Jahm, and al-Suyūṭī only says he was said to have been poisoned, although most sources ascribe his death to a hunting accident.56 ʿAbd al-Jabbār and Ibn al-Khaṭīb, by contrast, are both very positive about al-Mahdī. ʿAbd al-Jabbār is the only one to refer to the caliph’s persecution of the zanādiqa, the Manichaean “heretics,” and Shiʿite ghulāh, obviously a good thing. He also includes a bit of literary history in mentioning Abū lʿAtāhiya’s odes and the interesting love poems by the caliph’s gifted daughter, ʿUlayya, about two eunuchs, “Dew” and “Fawn.” Ibn al-Khaṭīb, who may have been familiar with ʿAbd al-Jabbār’s poem, also mentions Abū l-ʿAtāhiya. Of the six caliphs in the chosen range, al-Hādī is the least important, due to his short reign; ʿAlī b. al-Jahm only mentions its duration. Al-Suyūṭī refers to al-Hādī’s attempt to secure the succession of his son contrary to what alMahdī had arranged. This “betrayal,” though normal enough in dynastic histories, was thwarted by his early death, possibly by poison. ʿAbd al-Jabbār and, following him, Ibn al-Khaṭīb are surprisingly positive about him. “His story will always be repeated”: but one wonders what story the poet had in mind. In Hugh Kennedy’s The caliphate al-Hādī is not even named but merely mentioned as a son of al-Mahdī and Hārūn’s older brother.57 In popular memory Hārūn al-Rashīd’s rule became the epitome of glory. He “could be described as the greatest caliph of them all,”58 in the Arab world and, through the popularity of the Thousand and one nights, also in Europe. ʿAlī b. al-Jahm has nothing specific to offer but calls him a “felicitous and mighty king” and compares his death with the falling of a mountain. Al-Suyūṭī merely notes that he died while on campaign. The two Andalusian poets are more fulsome in their eulogies. Both briefly mention the rise and fall of the Barmakid family, 56 57 58

In the main text of his history, al-Suyūṭī mentions the riding accident, adding “it is also said that he died having been poisoned” (Tārīkh al-khulafāʾ 330). Kennedy, Caliphate 103–104 (al-Hādī is named only in the list of caliphs at the end, 384). Kennedy, Caliphate 104.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

64

van gelder

their fall being the single most dramatic event of his reign, while Ibn al-Khaṭīb stresses the caliph’s knowledge of poetry and history (khabar), admired by the great philologist al-Aṣmaʿī. The killing of al-Amīn is mentioned by all. ʿAlī b. al-Jahm seems to be shocked by the fact that the oath of allegiance to him was broken and that he was killed in spite of a safeguard promised to him; he does not mention anything about the caliph’s behavior that may have caused one thing and another. The Andalusians, by contrast, dwell on al-Amīn’s dissolute conduct and clearly see in his murder a moral lesson. Ibn al-Khaṭīb blames him, not others, for breaking a covenant, but he is the only one who also sees positive qualities, describing alAmīn as energetic, generous, evidently excellent though drawn to ruin by sex and wine, with Abū Nuwās as good poet but evil genius. Al-Maʾmūn is praised in rather vacuous terms by ʿAlī b. al-Jahm, and he is said to have died while campaigning against the Byzantines. Al-Suyūṭī does not praise him at all, even though he had described him in glowing terms in his book as “the most excellent of the Abbasids in prudence, resolve, wisdom, knowledge, astuteness, esteem, courage, lordly qualities, and generosity.”59 ʿAbd al-Jabbār and Ibn al-Khaṭīb laud him for restoring Baghdad and the realm to their former glory. The city once more appears as a “bride,” while the caliph’s wedding to a real bride, the daughter of his vizier, was a memorable event mentioned by both. The failed rebellion by his uncle Ibrāhīm gives them an opportunity to praise al-Maʾmūn’s clemency. Both, however, hint at less felicitous matters, such as the ill-advised recognition of an ʿAlid as his successor and the subsequent suspicious deaths of the candidate and the vizier who had been behind the plan. It is striking that none of the poems mention what is surely one of the most-often discussed events of al-Maʾmūn’s reign: the institution of the miḥna or “inquisition” and his controversial (and ultimately failed) attempt to assert caliphal control in matters of dogma. One could perhaps have expected a certain anti-Abbasid bias from the two Andalusian authors, in view of the glorious Andalusian Umayyad past after the overthrow of the Umayyads in the east by the Abbasids, but no such sentiment can be detected. ʿAbd al-Jabbār opens his account of the Abbasids with the words “Then the Abbasids came to power, and there was nothing wrong with their rule” ([…] fa-lam yakun fī ḥukmihim min bāsī),60 while Ibn al-Khaṭīb introduces the Abbasid caliphs as “People of lofty qualities, generosity, courage, pure nobility, without ambiguity.”61 59 60 61

Al-Suyūṭī, Tārīkh al-khulafāʾ 364. Ibn Bassām, Dhakhīra (Beirut) i, 933; (Cairo) i, ii, 419. Ibn al-Khaṭīb, Raqm al-ḥulal 20.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

65

he reigned as caliph; then he died

In this chapter only relatively small samples of a limited number of poems could be discussed. Some poets round off their composition with a proper peroration, such as ʿAlī b. al-Jahm who, having reached the caliphate of al-Mustaʿīn, concludes the poem with a eulogy and a blessing:62 ْ‫ت عن الأضداد والمشار َك َه‬ ْ َ ‫خ َل‬

ْ‫فنحن في خلافة ٍ مبار َك َه‬

ِ ‫جميَع هذا الأمـرِ مـن أحكام ِـه‬

ِ ‫فالحمـد للـّـه ِ عل ـى إنعام ِـه‬

‫علــى النبــّي باطنــا ً وظاهــرا‬

‫خ ـرا‬ ِ ‫ث ـّم السلام ُ أّول ًا وآ‬

Thus we are living in a Caliphate, free from opposites63 and partnership. So, praise be to God for His kind bestowal of all this, among His decrees. Then, peace, first and last, on the Prophet, inwardly and outwardly. Likewise, Ibn al-Khaṭīb, who ends his poem in the Jumādā ii of the year 763 (March 1362), when the Naṣrid ruler Muḥammad b. Ismāʿīl (r. 755–760/1354– 1359 and 763–793/1362–1391) began his long second reign, concludes as if with a happy ending, with a touch of real poetry in the final flowery verse:64 ِ ‫جهــادِه‬ ِ ‫مهنئّ ـا فــي م ُن ْت ـدى‬

ِ ‫واقتعـد السلطـانُ فـي م ِهـادِه‬

ِ ‫ف في عِبادِه‬ ِ ‫واستأنف الألطا‬

ِ ‫ل اللـّـه ِ ف ـي ب ِلادِه‬ ُ ‫وع ـّم فض ـ‬

‫ب‬ ِ ‫كفيلــة ً بالغ َــر َض المطلــو‬ َ

‫ب‬ ِ ‫ت على وف ْق الهوى المرغو‬ ْ ّ ‫تم‬

ِ‫علــى النبــّي أحمــد َ المحمــود‬

ِ‫ثــّم صلاة ُ اللـّـه ِ فــي ترديــد‬

ِ‫ق الع َمـود‬ ُ ِ ‫ش ـر‬ ْ ُ‫ح م‬ ٌ ‫صب ـ‬ ُ ‫ح‬ َ ‫ولا‬

ِ‫ت و َْرقـاء ُ فـي ُأملـود‬ ْ ‫ما غّرد‬

Authority has settled in its proper resting place, to be congratulated in the gathering of its Holy War, And God’s favour is spread all over His lands and He has renewed his benevolences among His servants. Completed (is the poem), according to what was desired,

62 63 64

ʿAlī b. al-Jahm, Dīwān 250; Urjūza 72 (vss. 331–333). According to the editor of the urjūza, all manuscripts have al-aḍdād, whereas the Dīwān edition has al-aḍrār, “harmful things.” I take aḍdād to mean “opposing rivals” here. Ibn al-Khaṭīb, Raqm al-ḥulal 115.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

66

van gelder

guaranteed to provide the purpose that was sought. Next, may God’s blessing be bestowed repeatedly on the Prophet most praiseworthy, As long as a grey dove coos on a twig and a dawn breaks with a shining shaft of light. Other poems, such as ʿAbd al-Jabbār’s urjūza and al-Suyūṭī’s rāʾiyyah, end abruptly, having reached the ruler contemporaneous with the poet; the history having caught up with the present, as if the poem were to be continued. It is perhaps a truer way, for history never stops, not even that of caliphs.

Bibliography Sources Abū ʿUbayd al-Bakrī, Faṣl al-maqāl fī sharḥ al-amthāl, wa-huwa sharḥ li-kitāb al-Amthāl li-Abī ʿUbayd al-Qāsim b. Sallām, eds. Iḥsān ʿAbbās and ʿAbd al-Majīd ʿĀbidīn, Beirut 1983. ʿAlī b. al-Jahm, Dīwān, ed. Khalīl Mardam, Beirut 1949, repr. Beirut 1400/1980. ʿAlī b. al-Jahm, Urjūza fī tārīkh al-khulafāʾ, ed. Yūsuf al-Sinnārī, Cairo 2018. al-ʿAskarī, Abū Hilāl al-Ḥasan b. ʿAbdallāh, Jamharat al-amthāl, eds. Aḥmad ʿAbd alSalām and Abū Hājar Muḥammad Saʿīd b. Basyūnī Zaghlūl, 2 vols., Beirut 1408/1988. Ḥājjī Khalīfa, Kashf al-ẓunūn ʿan asāmī l-kutub wa-l-funūn, eds. Şerefettin Yaltkaya and Rifat Bilge, 2 vols., Istanbul 1941–1943. Ibn Bashkuwāl, al-Ṣila, ed. Ibrāhīm al-Abyārī, Cairo 1989. Ibn Bassām al-Shantarīnī, al-Dhakhīra fī maḥāsin ahl al-jazīra, 4 vols., Cairo 1939– 1945. Ibn Bassām al-Shantarīnī, al-Dhakhīra fī maḥāsin ahl al-jazīra, ed. Iḥsan ʿAbbās, 8 vols., Beirut 1978–1979. Ibn al-ʿImrānī, Muḥammad b. ʿAlī b. Muḥammad, al-Inbāʾ fī tārīkh al-khulafāʾ, ed. Qāsim al-Sāmarrāʾī, Cairo 1999. Ibn Khallikān, Wafayāt al-aʿyān wa-anbāʾ abnāʾ al-zamān, ed. Iḥsān ʿAbbās, 8 vols., Beirut 1968–1972. Ibn al-Khaṭīb, Lisān al-Dīn, Dīwān, ed. Muḥammad Miftāḥ, Casablanca 1989. Ibn al-Khaṭīb, Lisān al-Dīn, Raqm al-ḥulal fī naẓm al-duwal, Tunis 1316[/1898–99]. Ibn al-Khaṭīb, Lisān al-Dīn, Sharḥ Raqm al-ḥulal fī naẓm al-duwal, ed. ʿAdnān Darwīsh, Damascus 1990 [not seen]. al-Khaṭīb al-Baghdādī, Tārīkh Madīnat al-Salām, ed. Bashshār ʿAwwād Maʿrūf, 17 vols., Beirut 2001–2006. al-Maqdisī, al-Badʾ wa-l-taʾrīkh, ed. Clément Huart, 6 vols., Paris 1899–1919.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

he reigned as caliph; then he died

67

al-Maqqarī, Nafḥ al-ṭīb min ghuṣn al-Andalus al-raṭīb, 8 vols., ed. Iḥsān ʿAbbās, Beirut 1968. al-Ṣafadī, Ṣalāḥ al-Dīn Khalīl b. Aybak, al-Wāfī bi-l-Wafayāt / Das biographische Lexikon des Ḫalīl ibn Aibak aṣ-Ṣafadī, 30 vols. (bi 6), Beirut, Wiesbaden, and Berlin 1931–2005. al-Sakhāwī, al-Iʿlān bi-l-tawbīkh li-man dhamma ahl al-tawrīkh, ed. Sālim b. Gh.t.r (?) b. Sālim al-Ẓafīrī, Riyadh 1438/2017. al-Shayzarī, Amīn al-Dawla Abū l-Ghanāʾim Muslim b. Maḥmūd, Jamharat al-Islām dhāt al-nathr wa-l-niẓām / Assembly of Islam in prose and poetry, facs. ed., Frankfurt am Main 1986. al-Ṣūlī, Abū Bakr Muḥammad b. Yaḥyā, al-Awrāq, eds. V.I. Belyaev and A.B. Khalidov, St. Petersburg 1998. al-Suyūṭī, Jalāl al-Dīn ʿAbd al-Raḥmān, Tārīkh al-khulafāʾ, ed. Muḥammad Muḥyī l-Dīn ʿAbd al-Ḥamīd, 1Cairo 1371/1952, Beirut 1408/1988. al-Suyūṭī, Jalāl al-Dīn ʿAbd al-Raḥmān, Tārīkh al-khulafāʾ, ed. Muḥammad Ghassān Naṣūḥ ʿAzqūl al-Ḥusaynī, Qatar 1424/2013. al-Yaʿqūbī, Aḥmad b. Abī Yaʿqūb, al-Tārīkh, ed. M.Th. Houtsma, 2 vols., Leiden 1883. al-Yaʿqūbī, Ibn Wāḍiḥ, The works of Ibn Wāḍiḥ al-Yaʿqūbī: An English translation, eds. Matthew Gordon, Chase F. Robinson, Everett Rowson, and Michael Fishbein, 3 vols. (ihc 152), Leiden 2018. Yāqūt, Muʿjam al-udabāʾ, ed. Aḥmad Farīd Rifāʿī, 20 vols., Cairo 1936–1938.

Studies Agha, Saleh Said, The revolution which toppled the Umayyads: Neither Arab nor ʿAbbāsid, Leiden 2003. gals = Brockelmann, Carl, Geschichte der arabischen Litteratur, 2 vols. + 3 suppl. vols., Leiden 1937–1943. gas = Sezgin, Fuat, Geschichte des arabischen Schrifttums, 17 vols. Leiden and Frankfurt /M 1967–2015. Gibb, H.A.R., ʿAlī b. al-D̲ j̲ahm, in ei2 i, 386. Gordon, Matthew, The place of competition: The careers of ʿArīb al-Maʾmūniyyah and ʿUlayyah bint al-Mahdī, sisters in song, in James E. Montgomery (ed.), ʿAbbasid studies: Occasional papers of the School of ʿAbbasid Studies, Cambridge 6–10 July 2002, Leuven 2004, 61–81. Kennedy, Hugh, The caliphate: A Pelican introduction, London 2016. Kennedy, P.F., ʿAlī b. al-Jahm, in Julie Scott Meisami and Paul Starkey (eds.), Encyclopedia of Arabic literature, 2 vols., i, London 1998, 79. Rosenthal, Franz, A history of Muslim historiography, Leiden 1968. Rowson, Everett K., Abu l-ʿAtahiyah, in Michael Cooperson and Shawkat M. Toorawa (eds.), Arabic Literary culture, 500–925 (Dictionary of Literary Biography 311), Detroit 2005, 12–20.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

68

van gelder

al-Samarai, Nicola Lauré, Die Macht der Darstellung. Gender, sozialer Status, historische Re-Präsentation: zwei Frauenbiographien aus der frühen Abbasidenzeit, Wiesbaden 2001. Sellars, W.C., and R.J. Yeatman, 1006 and all that: A memorable history of England, 11930, Harmondsworth 1973. Ullmann, Manfred, Untersuchungen zur Raǧazpoesie: Ein Beitrag zur arabischen Sprach- und Literaturwissenschaft, Wiesbaden 1966.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 4

Versifying History in Abbasid Iraq: The Universal History of ʿAlī b. al-Jahm Harry Munt

Early Islamic history-writing has, of course, generated a huge amount of modern scholarship and much of this has focused on works we tend to label “universal histories.”1 There have been debates over whether, strictly speaking, anything like “universal” history existed in the premodern Islamic world, at least before the work of the Ilkhanid vizier Rashīd al-Dīn (d. 718/1318).2 As the phrase is commonly understood in studies of Islamic history-writing, however— covering histories that begin with Creation and end at some point in the Islamic era3—extant examples are usually identified from the works of Ibn Qutayba (d. 276/889), Abū Ḥanīfa al-Dīnawarī (d. 282/895), and/or al-Yaʿqūbī (d. after 295/908) onward.4 Earlier, now lost, projects are known to have existed;5 but there are also at least a couple earlier extant works as well. The best known of these is the Taʾrīkh of the Andalusi traditionist and jurist ʿAbd al-Malik b. Ḥabīb (d. 238/852–853), a work that in its extant version—it survives in the recension of his student Yūsuf b. Yaḥyā al-Maghāmī (d. 288/900–901)—begins with Creation and moves through a survey of some pre-Islamic prophets and the history of the Quraysh before an overview of Muḥammad’s career and subsequent caliphal history down to the reigns of al-Walīd b. ʿAbd al-Malik (r. 86–96/705– 1 For some of Hugh Kennedy’s own contributions to this scholarship, see Kennedy, Caliphs and their chroniclers; idem, Sources; idem (ed.), al-Ṭabarī. For broad discussions of universal history-writing in the premodern Islamic world, see, for example, Rosenthal, History 133–150; Radtke, Weltgeschichte; idem, Towards a typology; Robinson, Islamic historiography 134–138; Goodman, Islamic humanism 161–211; Marsham, Universal histories; idem, Past. 2 Chabbi, Représentation du passé 28, for example, questions the existence of universal historiography in the early Islamic centuries. For the identification of Rashīd al-Dīn as the first truly universal historian, see for example the discussion in Goodman, Islamic humanism 202. 3 See, for example, the definitions in Radtke, Towards a typology 2; Robinson, Islamic historiography 135; Marsham, Universal histories 431. 4 For example, in Rosenthal, History 133; Radtke, Weltgeschichte 9; Robinson, Islamic historiography 136; Marsham, Universal histories 446. 5 Robinson, Islamic historiography 135, for example, notes the three parts of Ibn Isḥāq’s (d. ca. 150/767–768) oeuvre, a Kitāb al-Mubtadaʾ on pre-Islamic history, a Kitāb al-Mabʿath wa-lmaghāzī on the Prophet’s biography, and a Taʾrīkh al-khulafāʾ on subsequent Islamic history.

© Harry Munt, 2022 | doi:10.1163/9789004525245_005

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

70

munt

715) and Hishām b. ʿAbd al-Malik (r. 105–125/724–743), and then it shifts focus to the governors and rulers of al-Andalus, the prosopography of some scholars, and some other brief legal material.6 The other extant example of universal history from the mid-third/ninth century, however, is less well-known and only rarely discussed in the context of early Islamic universal history-writing: this is the 330- (or 333-)verse poem, in rajaz meter with paired rhyming couplets (urjūza muzdawija), composed by the sometime Abbasid courtier ʿAlī b. alJahm (d. 249/863).7 Given Hugh Kennedy’s longstanding interest in and considerable contribution to the history of the Abbasid caliphs and the histories written about and patronized by them, I hope that this brief study of ʿAlī b. alJahm’s versified universal history is of some interest in this volume.

1

ʿAlī b. al-Jahm

Our poet, or his family, were originally from Khurasan and were from the Banū Sāma b. Luʾayy, a tribe whose status—they claimed to be Qurashīs— generated a certain amount of controversy.8 His father, al-Jahm b. Badr, seems to have been a midranking Abbasid official, serving as overseer of the Yemeni barīd and that province’s ṭirāz, as well as of the frontier (al-thaghr) for alMaʾmūn (r. 198–218/813–833) and then of the shurṭa of a district of Baghdad for al-Wāthiq (r. 227–232/842–847).9 Some have assumed that ʿAlī b. alJahm is the brother of one Muḥammad b. al-Jahm, who was a close associ-

6 Ibn Ḥabīb, Taʾrīkh. For an overview of the work’s contents and al-Maghāmī’s role in its transmission, see Aguadé’s introduction to this edition at 100–103 and 75, 87–88, respectively. 7 It is briefly discussed, among other universal histories, in Radtke, Weltgeschichte 66; references to further brief discussions appear in notes throughout this chapter. The poem is also discussed in Geert Jan van Gelder’s chapter in this volume. I am very grateful to him for sharing his chapter with me and for offering comments on mine. In this chapter, the term “verse” is used to describe a full rhyming couplet. 8 An important study of ʿAlī’s life and poetry remains Khalīl Mardam’s introduction to ʿAlī b. alJahm, Dīwān (henceforth referred to as Mardam, Muqaddima). The principal sources I have consulted for his biography, alongside Mardam’s study, are Ibn al-Muʿtazz, Ṭabaqāt al-shuʿarāʾ 319–322; Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,667–3,698; al-Marzubānī, Muʿjam al-shuʿarāʾ 140– 141; al-Khaṭīb al-Baghdādī, Taʾrīkh Baghdād xiii, 290–292 (no. 6170); Ibn Khallikān, Wafayāt al-aʿyān iii, 355–358 (no. 462); al-Ṣafadī, al-Wāfī bi-l-wafayāt xx, 481–484. There are also plenty of anecdotes about ʿAlī b. al-Jahm throughout al-Ṣūlī, Awrāq. Anything not otherwise referenced in what follows is common across most or all of these accounts. On the debate about whether the Banū Sāma b. Luʾayy were to be considered Qurashīs or not, see Abū l-Faraj alIṣfahānī, Aghānī x, 3,667–3,669; Goldziher, Muslim studies i, 173–175. 9 Al-Khaṭīb al-Baghdādī, Taʾrīkh Baghdād viii, 166.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

71

ate of al-Maʾmūn, whom he served for a time as governor of Fārs and Jibāl, and also of the Barmakids and al-Jāḥiẓ (d. 255/868–869), and to whom was attributed a Kitāb Siyar mulūk al-Furs that may have been a translation of the Khwadāynāmag.10 This assumption may be a mistake, however; I have not seen a source that links this particular Muḥammad, who is almost always known as Muḥammad b. al-Jahm al-Barmakī, to ʿAlī. There is a Muḥammad b. al-Jahm al-Sāmī, who according to Abū l-Ḥusayn al-Rāzī (d. 347/958) was appointed governor of Damascus in 225/839–840 during the caliphate of al-Muʿtaṣim (r. 218–227/833–842).11 Given the nisba, which refers to membership in the Banū Sāma b. Luʾayy, this individual was presumably ʿAlī’s brother, but he may well have been a different person to the better-known Muḥammad b. al-Jahm alBarmakī. ʿAlī was himself well-known for his poetry, among which were panegyrics for several Abbasid caliphs, including al-Muʿtaṣim and al-Wāthiq as well as alMutawakkil (r. 232–247/847–861), the caliph who had a lasting impact on ʿAlī’s life. He was initially a prominent poet at the latter caliph’s court, although he does not seem to have been particularly politically astute and formed some serious rivalries with other courtiers. These eventually led to his downfall, and he was imprisoned on several occasions before eventually being exiled to Khurasan in 239/853–854.12 In Khurasan, following al-Mutawakkil’s orders, the governor Ṭāhir b. ʿAbdallāh b. Ṭāhir had him crucified (or perhaps better, exposed in public) naked in al-Shādhiyākh for one day before he was taken down and put into prison.13 Upon his release from prison, he returned to Iraq, but does not seem to have been as close to al-Mutawakkil after that. At some point, he

10

11

12 13

For a brief overview of this Muḥammad b. al-Jahm, see Zakeri, Al-Ṭabarī 30–31; HämeenAnttila, Khwadāynāmag 67–68. That he was ʿAlī’s brother, see Mardam, Muqaddima 4, 5–6, 18; Ali, Arabic literary salons 76. Ibn ʿAsākir, Taʾrīkh madīnat Dimashq lii, 241 (no. 6182; reading al-Sāmī instead of the edition’s al-Shāmī); see also al-Ṣafadī, Umarāʾ Dimashq 77. On al-Rāzī’s lost work on the governors of Damascus, see Conrad, Abūʾl-Ḥusain al-Rāzī 23–51. Scepticism over the identification of this Muḥammad b. al-Jahm al-Barmakī as ʿAlī’s brother is also found in Larsen, Crucified speaks 69, n. 7. The date of his exile is given by al-Ṭabarī, Taʾrīkh iii, 1419. For an account of his exile and punishment in Khurasan, see Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,672. For his verses discussing his crucifixion/exposure in particular, see Ullmann, Motiv der Kreuzigung 119–125; Larsen, Crucified speaks. Al-Shādhiyākh, just outside Nishapur, was established by the Tahirids as their residence and administrative center. The Saffarid Yaʿqūb b. al-Layth (r. 247–265/861–879) then destroyed the Tahirid complex there and turned it into an area of gardens before it became an urban development over subsequent years; see Yāqūt, Muʿjam al-buldān iii, 228–231; al-Khalīfa al-Naysābūrī, Tārīkh-i Nīshābūr 201 (§ 2722).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

72

munt

seems to have been in charge of the maẓālim court in Ḥulwān in Iraq; Mardam assumes this was during the reign of al-Muʿtaṣim, but in any case it was presumably before his exile to Khurasan.14 ʿAlī was firmly associated with Iraqi traditionalists (aṣḥāb al-ḥadīth).15 We are told that Aḥmad b. Abī Duʾād (d. 240/854), the chief qadi and considered one of the architects of the miḥna, shunned him because he was one of the ḥashwiyya, a term used, often disparagingly, to refer to the aṣḥāb al-ḥadīth.16 He is described as “one of the Sunnis” (min ahl al-sunna) opposed by a group of the Shiʿa (rawāfiḍ).17 He also posed questions to the foremost Iraqi traditionalist of the day, Aḥmad b. Ḥanbal (d. 241/855), on theological matters, specifically predeterminism/free will (al-qadar).18 This traditionalist stance is reflected in much of ʿAlī’s poetry.19 For example, in a poem he apparently composed during his first stint in prison and sent to his (unnamed) brother, ʿAlī noted, “The rawāfiḍ, the Christians and the Muʿtazilīs all worked together versus my invective.”20 Another poem was composed to lampoon Ibn Abī Duʾād and included the anti-Muʿtazilite line, “What are these dangerous innovations which, in your ignorance, you have labelled justice and monotheism?”21 Al-Mutawakkil, of course, is famous for having begun to bring the anti-traditionalist miḥna to an end, and several of ʿAlī’s panegyrics for that caliph praise him for this act that restored correct doctrine.22

14 15

16

17 18 19 20 21

22

Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,674; Mardam, Muqaddima 8–9. See also the discussion in Mardam, Muqaddima 6, 18–19, 24–26. The term “traditionalists” is used in this chapter, as is common now, to refer to holders of a particular theological/epistemological position, while “traditionists” is used to refer to those who collect and transmit ḥadīths; see Makdisi, Ashʿarī and the Ashʿarites 49. Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,681. On the term ḥashwiyya, see briefly Hoover, Ḥashwiyya. For a contemporary discussion of the ḥashwiyya, to which we will return later, by Jaʿfar b. Ḥarb (d. 236/850–851), see van Ess, Frühe muʿtazilitische Häresiographie 65–67 (Ar., §§ 110–114), with corresponding commentary at 51–53 (Ger.). Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,694. Ibn Abī Yaʿlā, Ṭabaqāt al-ḥanābila i, 223 (no. 309). Some examples are given in Mardam, Muqaddima 25–26. Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,670–3,671; ʿAlī b. al-Jahm, Dīwān 84 (no. 25, l. 21); see also Larsen, Crucified speaks 71. Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,682; ʿAlī b. al-Jahm, Dīwān 125–126 (takmila no. 30, l. 2). The Muʿtazilites, staunchly opposed to the traditionalists, were known to refer to themselves as the people of justice and monotheism (ahl al-ʿadl wa-l-tawḥīd); see Watt, Islamic philosophy 50. See, for example, ʿAlī b. al-Jahm, Dīwān 71–76 (no. 23); also his qaṣīda ruṣāfiyya, which is discussed and translated in Ali, Arabic literary salons 75–116. For the argument that the winding down of anti-traditionalism was more gradual than one act during the reign of al-Mutawakkil, see Melchert, Religious policies.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

73

Some of ʿAlī’s poetry was controversial beyond satirizing various Abbasid courtiers and officials. He was, for example, accused of stealing verses from other poets, especially from Ibrāhīm b. al-ʿAbbās al-Ṣūlī (d. 243/857), who was a close associate of Ibn Abī Duʾād.23 His poetry was also well-known for its irreverence towards ʿAlī b. Abī Ṭālib and his family.24 There is a fairly lengthy anecdote explaining that al-Mutawakkil disagreed with his predecessors’—alMaʾmūn, al-Muʿtaṣim, and al-Wāthiq are specifically named—affection for ʿAlī b. Abī Ṭālib and how this caliph was surrounded by courtiers who encouraged his dislike of the ʿAlids, courtiers who included ʿAlī b. al-Jahm.25 Despite the controversies he encouraged through his poetry and through other means at the court, and also in spite of his stints in prison and exile, ʿAlī’s eventual end did not come at the command of a vengeful caliph or through the machinations of a rival but rather at the hands of the Banū Kalb, who raided the caravan in which he was traveling to the Syrian frontier while it was in the region of Aleppo in 249/863.26

2

Universal History in Verse

According to Ibn al-Nadīm (wr. late fourth/tenth century), a collection of ʿAlī b. al-Jahm’s poetry was put together by Abū Bakr al-Ṣūlī (d. 335/947).27 Dīwāns were known to al-Khaṭīb al-Baghdādī (d. 463/1071), who noted it was “famous,” and to Ibn Khallikān (d. 681/1282), who remarked that it was “short.”28 Contemporaneous with Ibn Khallikān, ʿAlī b. al-Jahm’s Dīwān was included as part of a multiwork volume (majmūʿ) catalogued in the Ashrafiyya library in Damascus.29 We have no way of knowing if any of these disparate references are

23 24 25 26

27 28 29

Abū l-Faraj al-Iṣfahānī, Aghānī x, 3,675, 3,684–3,685. For Ibrāhīm’s friendship with Ibn Abī Duʾād, see Yāqūt, Irshād al-arīb i, 274. See, for example, al-Marzubānī, Muʿjam al-shuʿarāʾ 140; and the discussion in Mardam, Muqaddima 27–28. Ibn al-Athīr, Kāmil vi, 130–131. The anecdote is appended to a discussion of al-Mutawakkil’s destruction of the tomb of al-Ḥusayn b. ʿAlī at Karbala. As well as being discussed in the sources mentioned above, see n. 8, his death is also noted by al-Ṭabarī, Taʾrīkh iii, 1,514. At one point al-Ṣūlī gives ʿAlī’s death date as 239/853–854, but this is clearly a mistake since he elsewhere has him involved in events that took place after this date and explicitly dates his death to 249/863; see al-Ṣūlī, Awrāq 432, 434–5, 446, 481, 499 (§§ 179, 224, 309, 332, 335). Ibn al-Nadīm, Fihrist i/ii, 466. Al-Khaṭīb al-Baghdādī, Taʾrīkh Baghdād xiii, 290; Ibn Khallikān, Wafayāt al-aʿyān iii, 356. Hirschler, Medieval Damascus 313 (no. 1,175a).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

74

munt

to similar recensions or copies of the Dīwān, although Ibn Khallikān’s notice that it was “short” fits with its inclusion in a majmūʿ as in the Ashrafiyya catalogue. The description also fits a surviving collection of ʿAlī b. al-Jahm’s poetry, the manuscript used by Khalīl Mardam for his edition, copied in 1002/1594 and held in the Escorial, although it seems clear that this is again a different recension from that known to either al-Ṣūlī or Ibn Khallikān.30 The universal history, however, like much of ʿAlī’s known output, does not survive in this extant short Dīwān. Very brief extracts, dealing with Adam, Eve, Cain, and Abel, from the opening lines were cited in two fourth/tenthcentury universal histories, al-Masʿūdī’s (d. 345/956) Murūj al-dhabab and alMaqdisī’s (wr. ca. 355/965–966) Kitāb al-Badʾ wa-l-taʾrīkh;31 and some further lines dealing with the caliphates of al-Wāthiq and al-Mutawakkil were cited by Abū Bakr al-Ṣūlī in his Kitāb al-Awrāq.32 Al-Ṣūlī was also aware that one Aḥmad b. Muḥammad al-Asadī al-Anbārī had composed a continuation of ʿAlī b. al-Jahm’s historical poem and quoted some verses from this continuation, which ran to the caliphate of al-Muʿtamid (r. 256–279/870–892).33 On the basis of those brief extracts that were known to him at the time, von Grunebaum assumed that ʿAlī b. al-Jahm had composed two separate urjūzas, one on Creation and the Fall together with one on the Abbasids.34 We do, however, have the full text of ʿAlī b. al-Jahm’s urjūza on universal history, which demonstrates that it was in fact one work that, over 330 (or 333) verses, began with Creation and ended with the caliphate of al-Mustaʿīn (r. 248–252/862–866). Khalīl Mardam did not know of this full text when he initially edited ʿAlī b. al-Jahm’s Dīwān, although he tells us in an appendix to his edition of the Dīwān that shortly after the initial publication the Iraqi scholar Muḥammad al-Samāwī, based in Najaf at the time, sent him a copy of the full poem based on two older copies.35 Mardam then provided an edition of this (with 330 verses), which was added as an appendix to a reprinted edition of the Dīwān.36 Yūsuf al-Sinnārī

30 31

32 33 34 35 36

This manuscript is also a majmūʿ; see Derenbourg, Manuscrits arabes 242 (no. 369); Mardam, Muqaddima 45–47. Al-Masʿūdī, Murūj al-dhahab i, 38 (§ 49; = ll. 22, 24); al-Maqdisī, al-Badʾ wa-l-taʾrīkh ii, 85– 86 (= ll. 3–14, 17–20, 22). These lines are also provided in ʿAlī b. al-Jahm, Dīwān 157–159 (takmila no. 68). Al-Ṣūlī, Awrāq 554 (§ 73 = ll. 310–312), 551 (§ 78 = ll. 313–315). Al-Ṣūlī, Awrāq 360, 388, 428–429, 478–479 (§§ 225, 340, 406, 479); see also Yāqūt, Irshād al-arīb ii, 62. Von Grunebaum, On the origin 11. For the career of al-Samāwī, see the brief overview in Kaḥḥāla, Muʿjam al-muʾallifīn x, 97. This appendix was initially published in Majallat al-majmaʿ al-ʿilmī al-ʿarabī 26 (1951) and

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

75

has also now published an edition (with 333 verses) based on manuscripts unknown to Mardam, the earliest of which was copied in 623/1226 and is held in the Prophet’s Mosque in Medina.37 Mardam, for what it is worth, gave the poem the title al-Muḥabbara fī l-taʾrīkh (The polished [poem] on history), but this title is not given in other manuscripts.38 Between the various surviving versions of the text—the various known manuscripts as well as the extracts cited by al-Ṣūlī, al-Masʿūdī, and al-Maqdisī—there are numerous discrepancies in individual words and phrases (sometimes entire verses differ), but these rarely alter the basic meaning, and almost never the gist, of the narrative.39 The version(s) of the poem we can read today was composed right at the end of ʿAlī b. al-Jahm’s lifetime, since it covers the accession of al-Mustaʿīn in 248/862, and ʿAlī died the following year. As mentioned, the poem is in rajaz meter, and each verse is a paired rhyming couplet (so it is often called an urjūza muzdawija). This style is often used for poetry considered didactic, or sometimes also epic.40 The origins of such poetry is debated, with von Grunebaum’s argument that it was a feature of Sasanian-era Persian culture that was later incorporated into Arabic literary

37

38

39

40

was then reprinted in an updated edition of ʿAlī b. al-Jahm, Dīwān, 227–250; references in this chapter are to the latter. ʿAlī b. al-Jahm, Urjūza. For discussion of the manuscripts, see al-Sinnārī’s introduction to this edition (henceforth al-Sinnārī, Muqaddima) at 20–22. All references to the poem in this chapter (including the numbers given to verses) are to Mardam’s edition and to the specific readings on offer there unless otherwise stated. For discussion of the title, see al-Sinnārī, Muqaddima 14, 18. Muḥabbara presumably means something like “adorned” or “embellished,” but following Lane (Arabic-English lexicon ii, 498) we could also interpret it as something “made plain.” I have used “polished” to attempt to reflect this. Mardam did acknowledge (ʿAlī b. al-Jahm, Dīwān 228, n. 1) that rather than al-Muḥabbara we might read al-Mukhbira. The most troubling discrepancy comes from al-Ṣūlī, Awrāq 460 (§276), where four verses are cited from ʿAlī b. al-Jahm’s “poem in paired rhyme” ( fī qaṣīdatihi al-muzdawija) on the death of al-Muntaṣir in 248/862, a topic which is covered in the extant full poem on universal history, but these particular verses cannot be found there. They may have been cited from a different composition, or they may have actually come from Aḥmad b. Muḥammad’s aforementioned continuation of ʿAlī’s poem, since we know from al-Ṣūlī’s citation of verses from this continuation concerning al-Mutawakkil’s reign (see Awrāq 478–479 [§ 225]) that it also went over some of the ground already covered in ʿAlī’s poem in a different way. The possibility remains, however, that the version of ʿAlī b. al-Jahm’s poem known to al-Ṣūlī had some lines very different to those in the extant full version. For didactic poetry, see van Gelder, Arabic didactic verse; for discussion of poetry in such a style as potentially epic, see Reinert, Anfänge (esp. 239 for what he understands as epic in this instance).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

76

munt

culture finding opposition and support to varying degrees.41 We need not worry about the question of the origins of the form here, but we will return to the issue of the potentially didactic aim of ʿAlī’s poem later. Khalīl Mardam highlighted the significance of ʿAlī’s interest in historical poetry, although he went a bit too far in describing his work as, “The composition of an Arabic Iliad, encompassing the lives of the caliphs and their conquests.”42 To be fair, Mardam made this comment in the introduction to the initial publication of the Dīwān, so before he had actually seen the full version of the poem. Franz Rosenthal’s judgement was scathing, describing the poem as a “mediocre product” that is “quite unworthy of so gifted a lyrical poet as ʿAlî b. al-Jahm.”43 It is certainly not a classic of either poetical or historical composition. It is, however, one of the very earliest surviving works of universal history in Arabic, and as such it is of much greater interest for this than for its literary qualities alone. The poem opens with a traditional passage of praise for God and the Prophet and then outlines the poet’s sources before launching into the story of Adam’s creation: 1 2 3 4 5 6 7 8

41 42 43 44 45

All praises be to God, the Creator and the Reviver, for He is deserving of praise; And may blessings, now and ever more, be upon the Prophet, privately and openly. O questioner about the origins of creation, a question of one seeking the truth, I have been told by a number of trustworthy people, knowledgeable in many sciences and with other good qualities,44 At the forefront in the search for traditions (āthār) and knowing the veracity of reports (akhbār), Acquainted with the Torah and the Gospels, well-versed in the Revelation and its interpretation, That He Who does what He wills, Who is Mighty and Eternal Began with the creation of Adam and cut from him his wife Eve.45 Von Grunebaum, On the origin 12–13; cf. Ullmann, Untersuchungen 47–50. There is some further discussion in Rosenthal, History 180, n. 4; Reinert, Anfänge 237–238, 243–245. Mardam, Muqaddima 39. Rosenthal, History 183; see also the similarly disparaging comments in Ullmann, Untersuchungen 4, 55. In ʿAlī b. al-Jahm, Urjūza 32: “knowledgeable in many sciences and of perspicacious reasoning.” ʿAlī b. al-Jahm, Dīwān 228.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

77

versifying history in abbasid iraq

We are then taken through history down to the time of Lamech, son of Methuselah, before Noah arrives on the scene in line 58. The story continues with the Flood and then Noah’s children and their descendants, which includes an expected foray into the Arabian tribes ʿĀd and Thamūd and their prophets Hūd and Ṣāliḥ (at lines 75–92). In line 93 we get to Abraham, which of course leads into verses on Ishmael, Hagar, and the origins of Mecca, followed by a brief overview of the groups that controlled Mecca down to Quṣayy and his descendants (at lines 117–121). At this point, the poem returns (line 122) to Ishmael’s brother Isaac, followed by the prophets who succeeded him, reaching Moses by line 138 and David by line 167. Solomon’s reign is covered surprisingly briefly, in lines 177–179, followed by one line on the entire period of the kings of Israel and Judah after Solomon down to Nebuchadnezzar’s conquest of Jerusalem. At line 184 we have our first of five verses that discuss pre-Islamic Iran’s history, with the story of one Dārā killing the last of Nebuchadnezzar’s heirs and taking over his realm. The poem then progresses through a number of remaining prophets, down to Alexander and then Jesus (the latter at lines 193–194), before a brief notice at lines 195–196 on the Sasanians’ overthrow of the Arsacids, the latter referred to here as al-Ashghānūn.46 There is then a survey of political history across Arabia, Syria, and Iran before the pre-Islamic section is brought to an end at lines 203– 207: 203 204 205 206 207

This is a summary of the histories of the nations, as transmitted by Arabs and non-Arabs. Every people has deep-thinkers, though rarely are their affairs summarized. Histories become obscure during periods between prophets, apart from those accompanied by poems. The Persians and the Romans had periods of rule/great wars (ayyām), which Islam prevents us glamourizing;47 Intelligent people are content with God’s Books and the sayings of the messengers.48

So, 207, or 63%, of the poem’s 330 verses deal with pre-Islamic history.

46 47 48

ʿAlī b. al-Jahm, Urjūza 57 has al-Ashghānīnā. Cf. al-Ṭabarī, Taʾrīkh i, 823, where they are al-Ashkāniyya. I follow Mardam’s editorial suggestion (and the text in ʿAlī b. al-Jahm, Urjūza 58) and read tafkhīmihā instead of tafḥīmihā. ʿAlī b. al-Jahm, Dīwān 241–242.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

78

munt

We then get 20 verses on the career of Muḥammad, 19 on the Rāshidūn caliphs, 35 on the Umayyads, and, finally, 49 on the Abbasids. The main theme of these 123 verses on the Islamic era, at least those following Muḥammad’s death, is to outline the succession of caliphs, one after the other, and give an idea of the length of their reigns. For some caliphs, we get no further information. The two verses on Muʿāwiya b. Abī Sufyān (r. 41–60/661–680), for example, read simply: Then their affairs were overseen by Muʿāwiya, who lived ten [years] after ten had passed, 248 Until, when he had fulfilled 20 for them, he died when the date was 60 [ah].49 247

These two verses—describing Muʿāwiya’s 20-year reign, which ended with his death in 60/680—also give a good idea of how many lines are filled with slightly contorted ways of counting the number of years (and in other verses it is months as well) to fill the required meter and rhyme. Such ostensibly uninteresting information, however, is offered alongside some features of the poem that are more significant. Partly, this is simply because of its existence, and it is again worth emphasizing that it provides us with one of the earliest extant Arabic universal histories, alongside Ibn Ḥabīb’s Taʾrīkh. We can go further, however, for with ʿAlī b. al-Jahm we have an author of a universal history who lived, at least for lengthy parts of his life, in both Abbasid court and traditionalist scholarly circles. The importance of both caliphal imperialism and traditionalism (or at least traditionism) as drivers of historical production over the third/ninth century is often highlighted.50 Not every individual historian, however, was as invested directly in both of these driving forces as ʿAlī b. al-Jahm. His work, therefore, gives us valuable insight into the vision of history promoted by someone whom we might label an Abbasid traditionalist, active after al-Mutawakkil had started to integrate traditionalists more fully into caliphal circles and administration after his predecessors’ miḥna.51 The main message of ʿAlī’s history seems to be that caliphs are the successors of prophets and that while prophecy’s absence between Jesus and Muḥammad

49 50 51

ʿAlī b. al-Jahm, Dīwān 244–245. See, for example, Robinson, Islamic historiography 24–38, 83–102; Marsham, Universal histories 445–450; idem, Past 325–326. The impact of al-Mutawakkil’s return to traditionalism on the writing of history is emphasized in Borrut, Entre mémoire et pouvoir 94.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

79

led to political disharmony and rivalry between competing empires, Muḥammad’s prophetic mission was succeeded by a period of unifying, if not always perfect, caliphal rule. He is not interested in presenting caliphal or Islamic history in succession to Roman or Iranian history, as can be seen in, for example, line 206 quoted above. Jaakko Hämeen-Anttila, in his recent study of the Sasanian-era Khwadāynāmag and its transmission among Muslim historians, suggests that, “The Arab tradition of world histories was primarily based on Persian historiography and sacred history.”52 This may be a plausible conclusion to draw from many extant Arabic histories, but it is scarcely one you could draw from ʿAlī b. al-Jahm’s (or for that matter Ibn Ḥabīb’s) work. There are only five verses that deal with Iranian history in the poem;53 and even these do not seem to be doing the work of emphasizing the Iranian origins of good rulership and social structure, an approach that we can see in many other works from the third/ninth century.54 The poem perhaps gives us the most abundant information to consider ʿAlī b. al-Jahm’s attitude toward the caliphate and in particular toward individual caliphs. On the whole, ʿAlī is staunchly pro-Abbasid. This comes across from his panegyrics, not only in their praise of individual caliphs but also in their occasional comment on the legitimacy of the Abbasid family in general. One such example comes from a poem in praise of al-Muʿtaṣim, in which he exclaims, “It is you, sons of al-ʿAbbās, who are most deserving among all people of the inheritance of the Prophet.”55 In his historical urjūza, the onset of Abbasid rule is presented in a salvific light after the era of the Umayyads: 282 283

52 53

54

55

Until God, the patron of grace, brought the truth, which included mercy and compassion, And chose for the people Abū l-ʿAbbās, among the most supportive and the best of people,

Hämeen-Anttila, Khwadāynāmag 101. The poem does, therefore, technically meet Hoyland’s observation (in History 6) that, “All Muslim chronicles that take Creation as their starting point recount the histories of three particular pre-Islamic peoples—they may well treat others, but these three always feature. They are the Israelites … the pre-Islamic Arabs … and the Persians.” It is worth adding, however, that Ibn Ḥabīb’s Taʾrīkh barely mentions pre-Islamic Iranian history. For just one contemporaneous example, see Marlow, Hierarchy and egalitarianism 66– 67, discussing the Kitāb al-Tāj, wrongly attributed to al-Jāḥiẓ and probably composed by one Muḥammad b. al-Ḥārith al-Taghlibi/al-Thaʿlabī between 232/847 and 247/861; for this argument about the authorship, see Schoeler, Verfasser und Titel. ʿAlī b. al-Jahm, Dīwān 11 (no. 1, l. 50).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

80

munt

Of the family of the Prophet from the Banū l-ʿAbbās, astute and distinguished imams. 285 The blade of rulership returned to its scabbard and legitimacy (ḥaqq) returned to its holders (aṣḥāb).56

284

In the verses that follow, most of the Abbasid caliphs come across as decent rulers and good people, or at least remain relatively free from direct criticism. The principal exception comes near the end of the poem in line 324, where ʿAlī seems to praise God for the death of al-Muntaṣir in 248/862, soon after he had helped rebellious soldiers infamously murder his father and predecessor al-Mutawakkil in 247/861: “All of a sudden he met with his fate; all praise to He who hastens His revenge.”57 Umayyad caliphs, however, come in for more regular criticism. The section on Yazīd b. Muʿāwiya (r. 60–64/680–683), for example, begins, “Then his son Yazīd came to rule; he was neither prudent of opinion nor well-guided.”58 ʿAbd al-Malik b. Marwān—who incidentally is here the caliph who succeeds ʿAbdallāh b. al-Zubayr after his death in 73/69259—is presented as a warmonger who terrified humankind: Then came ʿAbd al-Malik b. Marwān, inciting war without rest Until the horizons yielded to him and Iraq became free of Muṣʿab [b. al-Zubayr] 261 As the sacred town [i.e., Mecca] did his brother; all humankind lay in fear of his assault. 262 He died having lived 13 [years] and four months during his reign.60 259 260

The poem’s material on the Rāshidūn era and the first fitna is also interesting in light of third/ninth-century discussions surrounding the relative legitimacy of Abū Bakr (r. 11–13/632–634), ʿUmar (r. 13–23/634–644), ʿUthmān (r. 23–35/644– 656) and ʿAlī b. Abī Ṭālib (r. 35–40/656–661) as caliphs.61 ʿAlī b. al-Jahm’s fam-

56 57 58 59

60 61

ʿAlī b. al-Jahm, Dīwān 247. ʿAlī b. al-Jahm, Dīwān 249. ʿAlī b. al-Jahm, Dīwān 245 (l. 249). That Ibn al-Zubayr was often considered a reigning caliph, rather than simply as a rebel, by premodern scholars, who also often only dated ʿAbd al-Malik’s caliphate from Ibn alZubayr’s death, see Robinson, ʿAbd al-Malik 31–35. ʿAlī b. al-Jahm, Dīwān 245–246. Line 262 confirms that ʿAlī counted ʿAbd al-Malik’s caliphate as beginning only with the death of Ibn al-Zubayr, so he reigned 73–86/692–705. See among very recent studies, Melchert, Rightly guided caliphs; and Su, Early Shiʿi Kufan traditionists. This question also dominates much of the discussion in Crone, Medieval Islamic political thought 51–141.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

81

ously dismissive attitude toward the ʿAlids, discussed above, is not reflected in this historical urjūza.62 Here, ʿAlī b. Abī Ṭālib takes his rightful place as the fourth caliph before dying as a martyr, just like his predecessor ʿUthmān; apart from these references to martyrdom, the first fitna receives no mention. Furthermore, Abū Bakr was the rightful first caliph, “Whom he put in charge of the people’s prayer and with whom he was content.”63 After ʿAlī b. Abī Ṭālib’s death, there is no mention of the caliphal claims of his son al-Ḥasan. These are the lines covering ʿUthmān’s and ʿAlī’s caliphates directly: 239 240 241 242 243 244 245 246

ʿUthmān b. ʿAffān, the chosen one, oversaw affairs for 12 [years] and then passed, Martyred on the path of truth, which all the little alleyways could not prevent.64 Authority was entrusted to ʿAlī the Hāshimī, the virtuous, the pure. He oversaw affairs legitimately (sharʿan) for four years and nine months, Then passed as a blessed martyr; his life was a blessing and his passing a great loss. This was the year 40 had passed since the reckoning of the years. Authority was transferred from Medina; such is true as was related by Safīna From the Prophet concerning the rulers of the community among kings and imams.65

Safīna was a mawlā of the Prophet, having previously been a slave belonging to Muḥammad’s wife Umm Salama.66 He was particularly well-known as the transmitter from Muḥammad of the famous ḥadīth referred to here, often used by traditionalists advancing a four “rightly-guided” caliph understanding of succession to the Prophet. This ḥadīth comes in several versions, with the essential common point summarized nicely in one version in Abū Dāwūd’s (d. 275/889) Sunan: “The messenger of God (ṣ) said, ‘There will be succession to 62 63

64 65 66

As also noted in al-Sinnārī, Muqaddima 10–12. ʿAlī b. al-Jahm, Dīwān 243 (l. 228). The identity of the “he” in this line is ambiguous—it could be God or Muḥammad—although the specific reference to Abū Bakr being put in charge of prayer favors Muḥammad. For discussion of Abū Bakr’s possible appointment by Muḥammad to lead prayer while he was in his final illness, see for example Ibn Hishām, al-Sīra al-nabawiyya iv, 301–304. I follow the reading of this verse suggested by Mardam in an editorial comment and in ʿAlī b. al-Jahm, Urjūza 61. ʿAlī b. al-Jahm, Dīwān 244. Ibn Saʿd, Ṭabaqāt v, 100–101 (no. 741); al-Mizzī, Tahdhīb al-Kamāl xi, 204–206 (no. 2420).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

82

munt

prophecy (khilāfat al-nubuwwa) for thirty years. Then God will bring dominion (mulk) to whomever He wishes, or His dominion to whomever He wishes.’ ”67 All of this puts ʿAlī b. al-Jahm’s views on the early caliphate in line with a group of traditionalists.68 We saw earlier that he was considered by some to belong among the so-called ḥashwiyya; his contemporary Jaʿfar b. Ḥarb described four different groups of the ḥashwiyya.69 The expression of early caliphal legitimacy in this poem fits well with one of these groups, identified by Jaʿfar as the Basrans and Wāsiṭīs, who we are told argued that the first four caliphs did indeed rule in order of precedence and who would support this argument with reference to the ḥadīth of Safīna.70 It also demonstrates ʿAlī’s commitment to establishing the importance and reality of caliphal and Muslim communal unity. When discussing subsequent history, ʿAlī does admit conflict—his account of the second fitna admits plenty, as can be seen in the verses cited above about ʿAbd al-Malik—but he also highlights the importance of maintaining orderly succession and of honoring oaths of allegiance. This can be seen nicely in his discussion of the caliphate of al-Amīn (r. 193–198/809–813), whose reign was consumed by a civil war with his brother al-Maʾmūn, which he eventually lost. Their father, Hārūn al-Rashīd (r. 170– 193/786–809), famously orchestrated a succession arrangement in which alAmīn was to succeed him, followed by al-Maʾmūn, but after Hārūn’s death al-Maʾmūn and his supporters were not content to wait.71 ʿAlī b. al-Jahm was strongly opposed to this disregard of the oath of allegiance sworn to al-Amīn:

67

68

69 70

71

Abū Dāwūd, Sunan iv, 293 (no. 4647: Kitāb al-Sunna, bāb fī l-khulafāʾ). For discussion, see for example Zaman, Religion and politics 170–173; Melchert, Rightly guided caliphs 69; Su, Early Shiʿi Kufan traditionists 29–33. The point is, if you add the lengths of the reigns of Abū Bakr, ʿUmar, ʿUthmān, and ʿAlī you get to 30 years, though this sometimes means finding a few extra months here or there. It overlaps nicely in many ways with the approach in the chronicle of the Islamic era by the contemporary traditionalist Khalīfa b. Khayyāṭ (d. ca. 240/854); see Andersson and Marsham, First Islamic chronicle 32–33; Andersson, Early Sunnī historiography 226–249. See above, n. 16. Van Ess, Frühe muʿtazilitische Häresiographie 65–66 (Ar., §111). Zaman interestingly identified (in Religion and politics 171–173) the isnāds of the surviving versions of the Safīna ḥadīth as reflecting Basran and Wāsiṭī origins, which fits nicely with Jaʿfar b. Ḥarb’s account, although Su has recently (in Early Shiʿi Kufan traditionists 29–33) noted that the ḥadīth seems to have been circulating in Kufa as well by the late second/eighth century. On the succession arrangement, see Marsham, Rituals 219–227; for the subsequent civil war between the brothers, see Kennedy, Early Abbasid caliphate 135–150.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

83

They gave allegiance to Muḥammad al-Amīn, but all betrayed the oath Except for a few, the few who are most blessed; death is promised for all people. 300 They promised him security, then they killed him; this is not that to which his father had sworn them.72

298 299

We can also see his concern for appropriate succession in his efforts to establish that nobody ended up benefitting from the murder of al-Mutawakkil: After him they pledged allegiance to al-Muntaṣir, but the winners among them became losers; 323 He lived while ruling for six months and expelled them from his realm and the army. 324 All of a sudden he met with his fate; all praise to He who hastens His revenge.73 322

It is worth remembering here that ʿAlī b. al-Jahm himself died the year following the death of al-Muntaṣir and the accession of al-Mustaʿīn, and so he was oblivious to the years of chaos and disaster that were to follow. The poem ends on an upbeat note, reflecting a belief—or at least a desire to suggest to the new caliph—that the problems of the recent past were now resolved and caliphal succession was back in order: God chose for them an imam, may God support Islam through him, And they pledged allegiance in light of this selection to Aḥmad who has recourse (al-mustaʿīn) to the one God.74 327 He is among 20 rulers and protectors from the family of al-ʿAbbās.75 328 We are with a blessed caliphate, free from opposition and partnership.76 325 326

72

73 74

75 76

ʿAlī b. al-Jahm, Dīwān 248. This is somewhat different to the contemporary traditionalist Khalīfa’s approach to this period, for which see Andersson and Marsham, First Islamic chronicle 27; Andersson, Early Sunnī historiography 276–278. For discussion of the different attitudes among early Muslim historians to this civil war and the killing of al-Amīn, see El-Hibri, Reinterpreting Islamic historiography 59–94. ʿAlī b. al-Jahm, Dīwān 249. This is a rather different view of al-Mustaʿīn than that offered by another roughly contemporary historian who did see the turmoil of the subsequent years and judged that al-Mustaʿīn was “unworthy of the caliphate”; al-Yaʿqūbī, Taʾrīkh ii, 604. There have only been 12 ʿAbbasid caliphs mentioned in the poem. In ʿAlī b. al-Jahm, Urjūza 72, the verse reads “he is the twelfth” instead. Mardam’s edition reads al-iḍrār, “(doing) harm,” instead of al-aḍdād, “opposites [opposi-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

84

munt

All praises to God for His grace; all of this rule is through His jurisdiction. 330 And may greetings, now and ever more, be with the Prophet, privately and openly.77

329

3

A Didactic poem?

Emphasizing the status of caliphs as successors to prophets and the importance of the unity they provide through legitimate and orderly succession for the Muslim community was a key message of ʿAlī’s poetry; but who was the audience(s) for that message? Versified histories were not particularly common among the different communities living in the Islamic world at the time of ʿAlī b. al-Jahm,78 and those that are known before and around his time do not seem to have been as universal in scope. The Creation and the Fall were established topics for poetry already in late antiquity, as seen for example in the early seventh-century ce Hexaemeron of George of Pisidia;79 there was also pre-Islamic Christian Arabic poetry on these themes, a well-known example being ʿAdī b. Zayd’s (active late sixth century ce).80 Other prophetic narratives from the era before Muḥammad were the subject of pre-Islamic Arabic poetry.81 Some material on pre-Islamic Iran was also versified in Arabic paired rhyme by Abān b. ʿAbd al-Ḥamīd al-Lāḥiqī (d. ca. 200/815–816) and, later in the third/ninth century, by al-Balādhurī (d. before 279/892).82 Contemporaneous with ʿAlī b. al-Jahm, an urjūza was composed on the conquest of al-Andalus by Yaḥyā b. al-Ḥakam al-Ghazāl (d. 250/864), as was another by Tamīm b. ʿAlqama on Islamic Andalusi history down at least to the end of the reign of ʿAbd alRaḥmān (ii) b. al-Ḥakam in 238/852.83 In the decades following ʿAlī b. al-Jahm,

77 78

79 80 81 82

83

tion],” found in ʿAlī b. al-Jahm, Urjūza 72; al-aḍdād is the reading of all manuscripts known to the latter’s editor. ʿAlī b. al-Jahm, Dīwān 250. For a brief discussion of Arabic versified histories, see Ullmann, Untersuchungen 55–56; Rosenthal, History 179–185; and van Gelder’s chapter in this volume. For discussion of premodern versified histories in Syriac (and some in Armenian), see Debié, Écriture 419–426. Whitby, Devil in disguise. Dmitriev, Early Christian Arabic account; Toral-Niehoff, Eine arabische poetische Gestaltung. See, for example, Sinai, Religious poetry. Ibn al-Nadīm, Fihrist i/ii, 348, 369–370; some discussion in Ullmann, Untersuchungen 50– 51, 54; Reinert, Anfänge 239–241, 242–245; Hämeen-Anttila, Khwadāynāmag 36, 42, 44; Lynch, Arab conquests 42. Rosenthal, History 181–182; Reinert, Anfänge 245–258.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

85

a well-known historical urjūza was composed by Ibn al-Muʿtazz (d. 296/908) on the caliphate of his cousin al-Muʿtaḍid (r. 279–289/892–902), as was an annalistic urjūza on events during the early years of the reign of the Andalusi Umayyad ʿAbd al-Raḥmān (iii) b. Muḥammad (r. 300–350/912–961) by Ibn ʿAbd Rabbih (d. 328/940).84 The latter also composed a lost broader historical urjūza.85 And around the same time Masʿūdī-yi Marwazī composed a Persian muzdawija—also seemingly titled, like ʿAlī b. al-Jahm’s, al-Muḥabbara—on Iranian history from Gayūmard down to the end of the Sasanian Empire, which of course preceded the far more famous Shāhnāmas of Daqīqī (d. ca. 367/977– 978) and Firdawsī (completed 400/1010).86 Some of these poems have been understood as “epic,” but this is not really a label that works for all of them, including ʿAlī b. al-Jahm’s.87 Instead, ʿAlī’s poem has been understood as being somehow didactic, designed to help listeners learn key facts of Islamic salvation history easily.88 Such didactic poetry became very popular over subsequent centuries in a number of different fields, with urjūza muzdawija being “the didactic form par excellence.”89 Yet many features of ʿAlī b. al-Jahm’s universal history do not lend themselves easily to teaching the uninitiated. What, for example, would you make of the following verses if you were not already aware of more information about the Battle of the Ḥarra in 63/683 than this lets on? (We are in the discussion of ʿAbdallāh b. al-Zubayr’s caliphate.) Until al-Ḥajjāj took care of killing him after all the roads through the mountains had become too narrow for him; 258 And there was the destruction of the inviolable Kaʿba and the Battle of the Ḥarra in Medina. 259 Then came ʿAbd al-Malik b. Marwān […]90 257

Slightly earlier, in line 253, an otherwise unexplained al-Ḍaḥḥāk is killed by Marwān b. al-Ḥakam; listeners are presumably expected to know that this is a reference to the defeat and killing of al-Ḍaḥḥāk b. Qays at the Battle of Marj Rāhiṭ in 64/684. The treatment of many pre-Islamic figures displays the

84 85 86 87 88 89 90

Lang, Muʿtaḍid als Prinz; Monroe, Historical arjūza; Reinert, Anfänge 258–264. Toral-Niehoff, History in adab context 68. For Masʿūdī-yi Marwazī, see Hämeen-Anttila, Khwadāynāmag 133–134. Reinert, Anfänge. Reinert, Anfänge 239; Radtke, Weltgeschichte 66. Both label ʿAlī’s poem a “Lehrgedicht.” Van Gelder, Arabic didactic verse (quote from 107). ʿAlī b. al-Jahm, Dīwān 245.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

86

munt

same lack of concern with explaining much about them, other than providing them with a rough place in the chronology. The poem seems, therefore, to have been more of an aide-mémoire for those already well-acquainted with the subject matter than a tool for teaching or learning something new. Geert Jan van Gelder has noted that this is actually true of many so-called “didactic” poems.91 Perhaps more, however, than a simple aide-mémoire, ʿAlī b. al-Jahm’s poem was designed to help listeners configure (or emplot) the facts they already knew into a framework that emphasized his key themes of caliphs as successors to the prophets, the importance of orderly caliphal succession for ensuring the maintenance of the Muslim community’s unity, and the particular legitimacy of the first four caliphs and of almost all the Abbasid caliphs.

Bibliography Sources Abū Dāwūd, Sunan, ed. M.M-D. ʿAbd al-Ḥamīd, 4 vols., Cairo 21369–1370/1950–1951. Abū l-Faraj al-Iṣfahānī, Kitāb al-Aghānī, ed. I. al-Abyārī, 31 vols., with continuous pagination, Cairo 1389–1399/1969–1979. ʿAlī b. al-Jahm, Dīwān ʿAlī b. al-Jahm, ed. Kh. Mardam Bik, Damascus 1369/1949, 2n.d. (rev. ed.). ʿAlī b. al-Jahm, Urjūza fī taʾrīkh al-khulafāʾ, ed. Y. al-Sinnārī, Cairo 1439/2018. Ibn Abī Yaʿlā, Ṭabaqāt al-ḥanābila, ed. M.Ḥ. al-Fiqī, 2 vols., Cairo 1371/1952. Ibn ʿAsākir, Taʾrīkh madīnat Dimashq, eds. ʿU. al-ʿAmrawī and ʿA. Shīrī, 80 vols., Beirut 1415–1421/1995–2000. Ibn al-Athīr, al-Kāmil fī l-taʾrīkh, ed. ʿU.ʿA.-S. Tadmurī, 11 vols., Beirut 2012. Ibn Ḥabīb, Kitāb al-Taʾrīkh, ed. J. Aguadé, Madrid 1991. Ibn Hishām, al-Sīra al-nabawiyya, eds. M. al-Saqqā, I. al-Abyārī, and ʿA.-Ḥ. Shalabī, 4 vols., Cairo 1355/1936. Ibn Khallikān, Wafayāt al-aʿyān wa-anbāʾ abnāʾ al-zamān, ed. I. ʿAbbās, 8 vols., Beirut 1397–1398/1977–1978. Ibn al-Muʿtazz, Ṭabaqāt al-shuʿarāʾ, ed. ʿA.-S.A. Farrāj, Cairo n.d. (after 1375/1956). Ibn al-Nadīm, al-Fihrist, ed. A.F. Sayyid, 2 parts in 4 vols., London 1430/2009. Ibn Saʿd, Kitāb al-Ṭabaqāt al-kabīr, ed. ʿA.M. ʿUmar, 11 vols., Cairo 1421/2001. al-Khalīfa al-Naysābūrī, Tārīkh-i Nīshābūr, ed. M.R.Sh. Kadkanī, Tehran 1374/1996. al-Khaṭīb al-Baghdādī, [Taʾrīkh Baghdād] Taʾrīkh Madīnat al-Salām, ed. B. ʿA. Maʿrūf, 17 vols., Beirut 1422/2001. 91

Van Gelder, Arabic didactic verse 108.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

87

al-Maqdisī, Kitāb al-Badʾ wa-l-taʾrīkh, ed. and trans. C. Huart, 6 vols., Paris 1899–1919. al-Marzubānī, Muʿjam al-shuʿarāʾ, ed. ʿA.-S.A. Farrāj, Damascus 1960. al-Masʿūdī, Murūj al-dhahab wa-maʿādin al-jawhar, eds. C. Barbier de Meynard and P. de Courteille, revised and corrected by C. Pellat, 7 vols., Beirut 1965–1979. al-Mizzī, Tahdhīb al-Kamāl fī asmāʾ al-rijāl, ed. B. ʿA. Maʿrūf, 35 vols., Beirut 1402– 1413/1982–1992. al-Ṣafadī, Kitāb al-Wāfī bi-l-wafayāt, eds. S. Dedering et al., 30 vols., Leipzig, Istanbul, and Beirut, 1931–2010. al-Ṣafadī, Umarāʾ Dimashq fī l-islām, ed. Ṣ.-D. al-Munajjid, Damascus 1955. al-Ṣūlī, Kitāb al-Awrāq, eds. and trans. V.I. Beliaeva and A.B. Khalidova, St. Petersburg 1998. al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, eds. M.J. de Goeje et al., 3 parts in 13 vols., Leiden 1879–1901. al-Yaʿqūbī, Taʾrīkh, ed. M.T. Houtsma, 2 vols., Leiden 1883. Yāqūt, Irshād al-arīb ilā maʿrifat al-adīb, ed. D.S. Margoliouth, 7 vols., London 1907– 1927. Yāqūt, Muʿjam al-buldān, ed. F. Wüstenfeld, 6 vols., Leipzig 1866–1873.

Studies Ali, S.M., Arabic literary salons in the Islamic middle ages: Poetry, public performance, and the presentation of the past, Notre Dame 2010. Andersson, T., Early Sunnī historiography: A study of the Tārīkh of Khalīfa b. Khayyāṭ, Leiden 2019. Andersson, T., and A. Marsham, The first Islamic chronicle: The Chronicle of Khalīfa b. Khayyāṭ (d. ad 854), in M. Campopiano and H. Bainton (eds.), Universal chronicles in the high middle ages, Woodbridge 2017, 19–41. Borrut, A., Entre mémoire et pouvoir: L’espace syrien sous les derniers Omeyyades et les premiers Abbassides (v. 72–193/692–809), Leiden 2011. Chabbi, J., La représentation du passé aux premiers âges de l’historiographie califale: Problèmes de lecture et de méthode, in R. Curiel and R. Gyselen (eds.), Itinéraires d’orient: Hommages à Claude Cahen, Bures-sur-Yvette 1994, 21–47. Conrad, G., Abūʾl-Ḥusain al-Rāzī (-347/958) und seine Schriften: Untersuchungen zur frühen Damaszener Geschichtsschreibung, Stuttgart 1991. Crone, P., Medieval Islamic political thought, Edinburgh 2004. Debié, M., L’écriture de l’histoire en syriaque: Transmissions interculturelles et constructions identitaires entre hellénisme et islam, Leuven 2015. Derenbourg, H., Les manuscrits arabes de l’Escurial, vol. 1, Paris 1884. Dmitriev, K., An early Christian Arabic account of the creation of the world, in A. Neuwirth, N. Sinai, and M. Marx (eds.), The Qurʾān in context: Historical and literary investigations into the Qurʾānic milieu, Leiden 2010, 349–387.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

88

munt

El-Hibri, T., Reinterpreting Islamic historiography: Hārūn al-Rashīd and the narrative of the ʿAbbāsid caliphate, Cambridge 1999. Goldziher, I., Muslim studies, eds. and trans. C.R. Barber and S.M. Stern, 2 vols., London, 1967–1971. Goodman, L.E., Islamic humanism, Oxford 2003. Hämeen-Anttila, J., Khwadāynāmag: The Middle Persian Book of kings, Leiden 2018. Hirschler, K., Medieval Damascus: Plurality and diversity in an Arabic library, Edinburgh 2016. Hoover, J., Ḥashwiyya, in ei3 Online, http://dx.doi.org/10.1163/1573‑3912_ei3_COM_3037 (last accessed April 19, 2022). Hoyland, R., The “History of the kings of the Persians” in three Arabic chronicles: The transmission of the Iranian past from late antiquity to early Islam, Liverpool 2018. Kaḥḥāla, ʿU.R., Muʿjam al-muʾallifīn: Tarājim muṣannifī al-kutub al-ʿarabiyya, 15 vols., Damascus 1376–1381/1957–1961. Kennedy, H., Caliphs and their chroniclers in the middle Abbasid period (third/ninth century), in C.F. Robinson (ed.), Texts, documents and artefacts: Islamic studies in honour of D.S. Richards, Leiden 2003, 17–35. Kennedy, H., The early Abbasid caliphate: A political history, London 1981. Kennedy, H., The sources of al-Ṭabarī’s history of the ʿAbbāsid caliphate, in H. Kennedy (ed.), al-Ṭabarī: A medieval Muslim historian and his work, Princeton 2008, 175–185. Kennedy, H. (ed.), al-Ṭabarī: A medieval Muslim historian and his work, Princeton 2008. Lane, E.W., Arabic-English lexicon, 8 vols., London 1863–1893. Lang, C., Muʿtaḍid als Prinz und Regent: Ein historisches Heldengedicht von Ibn el Muʿtazz, in zdmg 40 (1886), 563–611; 41 (1887), 232–729. Larsen, D., The crucified speaks: ʿAlī ibn al-Jahm on his day-long exposure at Nishapur, in J.E. Lowry and S.M. Toorawa (eds.), Arabic humanities, Islamic thought: Essays in honor of Everett K. Rowson, Leiden, 2017, 67–93. Lynch, R.J., Arab conquests and early Islamic historiography: The Futuh al-buldan of alBaladhuri, London 2020. Makdisi, G., Ashʿarī and the Ashʿarites in Islamic religious history, in si 17 (1962), 37–80; 18 (1963), 19–39. Marlow, L., Hierarchy and egalitarianism in Islamic thought, Cambridge 1997. Marsham. A., The past in the early and medieval Islamic Middle East (circa 750– circa 1250), in K.A. Raaflaub (ed.), Thinking, recording, and writing history in the ancient world, Malden, MA 2014, 314–339. Marsham, A., Rituals of Islamic monarchy: Accession and succession in the first Muslim empire, Edinburgh 2009. Marsham, A., Universal histories in Christendom and the Islamic world, c. 700–c. 1400, in S. Foot and C.F. Robinson (eds.), The Oxford history of historical writing, Volume 2: 400–1400, Oxford 2012, 431–456.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

versifying history in abbasid iraq

89

Melchert, C., Religious policies of the caliphs from al-Mutawakkil to al-Muqtadir, a.h. 232–295/a.d.847–908, in ils 3 (1996), 316–342. Melchert, C., The rightly guided caliphs: The range of views preserved in ḥadīth, in S. alSarhan (ed.), Political quietism in Islam: Sunnī and Shīʿī practice and thought, London 2020, 63–79. Monroe, J.T., The historical arjūza of Ibn ʿAbd Rabbihi, a tenth-century Hispano-Arabic epic poem, in jaos 91 (1971), 67–95. Radtke, B., Towards a typology of Abbasid universal chronicles, in Occasional papers of the School of Abbasid Studies 3 (1990), 1–18. Radtke, B., Weltgeschichte und Weltbeschreibung im mittelalterlichen Islam, Beirut 1992. Reinert, B., Die Anfänge des spanisch-arabischen Epos, in Saeculum 29 (1978), 231–268. Robinson, C.F., ʿAbd al-Malik, Oxford 2005. Robinson, C.F., Islamic historiography, Cambridge 2003. Rosenthal, F., A history of Muslim historiography, Leiden 21968 (rev. ed.). Schoeler, G., Verfasser und Titel des dem Ǧāḥiẓ zugeschriebenen sog. Kitāb al-Tāǧ, in zdmg 130 (1980), 217–225. Sinai, N., Religious poetry from the Quranic milieu: Umayya b. Abī al-Ṣalt on the fate of Thamūd, in bsoas 74 (2011), 397–416. Su, I-W., The early Shiʿi Kufan traditionists’ perspective on the rightly guided caliphs, in jaos 141 (2021), 27–47. Toral-Niehoff, I., Eine arabische poetische Gestaltung des Sündenfalls: Das vorislamische Schöpfungsgedicht von ʿAdī b. Zayd, in D. Hartwig et al. (eds.), “Im vollen Licht der Geschichte”: Die Wissenschaft des Judentums und die Anfänge der kritischen Koranforschung, Würzburg 2008, 235–256. Toral-Niehoff, I., History in adab context: “The book on caliphal histories” by Ibn ʿAbd Rabbih (246/860–328/940), in jas 2 (2015), 61–85. Ullmann, M., Das Motiv der Kreuzigung in der arabischen Poesie des Mittelalters, Wiesbaden 1995. Ullmann, M., Untersuchungen zur Raǧazpoesie: Ein Beitrag zur arabischen Sprach- und Literaturwissenschaft, Wiesbaden 1966. van Ess, J., Frühe muʿtazilitische Häresiographie: Zwei Werke des Nāšiʾ al-Akbar (gest. 293 h.), Beirut 1971. van Gelder, G.J., Arabic didactic verse, in J.W. Drijvers and A.A. MacDonald (eds.), Centres of learning: Learning and location in pre-modern Europe and the Near East, Leiden 1995, 103–117. von Grunebaum, G.E., On the origin and early development of Arabic muzdawij poetry, in jnes 3 (1944), 9–13. Watt, W.M., Islamic philosophy and theology: An extended survey, Edinburgh 21985. Whitby, M., The devil in disguise: The end of George of Pisidia’s Hexaemeron reconsidered, in jhs 115 (1995), 115–129.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

90

munt

Zakeri, M., Al-Ṭabarī on Sasanian history: A study in sources, in H. Kennedy (ed.), AlṬabarī: A medieval Muslim historian and his work, Princeton 2008, 27–40. Zaman, M.Q., Religion and politics under the early ʿAbbāsids: The emergence of the protoSunnī elite, Leiden 1997.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 5

How to Found an Islamic State: The Idrisids as Rivals to the Abbasid Caliphate in the Far Islamic West Corisande Fenwick

1

Introduction

By the late second/early ninth century, North Africa was lost to the Abbasid caliphate.* Rival states had emerged in the Maghrib al-Aqṣā (roughly modern Morocco and western Algeria) a few short decades after its conquest (fig. 5.1). The earliest were the Ṣāliḥids at Nakūr (91/709–710), which followed in the wake of the Kharijite revolt by the Barghawāṭa (127/744–745), the Midrārids of Sijilmāsa (140/757–758), the Rustamids at Tāhart (160/776–777), and finally, the Idrisids, who ruled much of northern Morocco (172/788–789).1 Their emergence coincided with a phase of intense urbanization that transformed the social and economic landscape of the Maghrib al-Aqṣā: new cities were built, old settlements re-occupied or expanded, and new trade links were established.2 By 184/800, caliphal rule had failed entirely in North Africa, and the Abbasids abandoned even Ifrīqiya to their vassals, the Aghlabids, independent rulers in all but name.3 From this point on, North Africa was ruled by Islamic dynasties who were independent from, and rivals to, Baghdad. Yet these shadowy—often short-lived—North African states are all too often overlooked * This chapter was written as part of the project everydayislam—Becoming Muslim: Cultural Change, Everyday Life and State Formation in early Islamic North Africa (600–1000). This project received funding from the European Research Council (erc) under the European Union’s Horizon 2020 research and innovation program (Grant agreement no. 949367). I thank Andy Merrills and Elizabeth Fentress for their helpful comments on an earlier draft, Hassan Limane for his support and encouragement, as well as audiences at the University of Chicago and the University of Oxford for useful discussions. 1 Talbi, Hérésie, acculturation et nationalisme; Aillet, Tāhart et l’imamat rustumide; Capel and Fili, Sijilmāsa; Cressier, Nakūr; García-Arenal and Manzano Moreno, Idrīssisme. For an account of early Islamic North Africa, see Fenwick, Early Islamic North Africa. 2 Brett, Islamisation of Morocco; Cressier, Urbanisation; Rosenberger, Les premières villes islamiques. 3 See Talbi, L’Émirat Aghlabide; Anderson, Fenwick, and Rosser-Owen (eds.), Aghlabids and their neighbors. © corisande fenwick, 2022 | doi:10.1163/9789004525245_006 This is an open access chapter distributed under the terms of the cc by-nc-nd 4.0 license.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

92

fenwick

figure 5.1 Map of early medieval North Africa showing the location of the rival states corisande fenwick

in discussions of Islamic state formation, despite their important positions in the power spectrum and the developmental trajectory of Muslim statehood. That new Islamic political forms emerged as the paradigm for state authority in the Maghrib al-Aqṣā is surprising, and by no means inevitable. In stark contrast to al-Andalus and Ifrīqiya, which had urban-based societies, tax economies, and well-developed state apparatuses inherited from Romano-Byzantine rule, the western Maghrib was sparsely populated, with only a handful of larger settlements, and may not even have had a monetary economy on the eve of the Arab conquest. The region was never fully conquered or subdued by the caliphate; indeed, though the Kharijite revolt started there in 122/739–740, the presence of an Umayyad garrison was restricted to the districts of Ṭanja, Tilimsān, and Ṭarqala (thus far unlocated).4 What, if any, Abbasid activity took place in the region is unclear, but dirham were minted south of the Atlas Mountains at Tudgha in the name of Yazīd b. Ḥātim, the Abbasid governor of Ifrīqiya, and circulated widely in the caliphate and beyond. Why was it in the Maghrib alAqṣā that the earliest rivals to the caliphate emerged? How did their rulers justify their rule? And what was the relationship between these new states and the Abbasid caliphate? These questions invite discussion of the relations between Islam, rulers, and local populations on the one hand and the dynamics between these newly established rival states and the Abbasid caliphate on the other, a topic I have frequently debated with Hugh Kennedy over the years.

4 Djaït, Wilāya 97–98. On the nature of Umayyad rule in North Africa, see also Fenwick, Umayyads.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

93

This brief piece will explore the process of establishing an Islamic state on the western borders of the Abbasid caliphate in the late second/eighth and early third/ninth centuries using the example of the better-known Idrisid state established by Idrīs I and his son, Idrīs ii, so as to highlight some of the issues at stake. Though the Idrisid state of northern Morocco was the latest of the polities to be founded in the Far West, it was arguably the most significant rival— and the greatest perceived threat—to Abbasid claims. Its ʿAlid rulers claimed their right to rule from their descent from ʿAlī and the Prophet—a sufficiently powerful claim that Idrīs I was reportedly poisoned by an Abbasid secret agent on the orders of the caliph, Hārūn al-Rashīd. His descendants continued to be a thorn in the Abbasid flank, increasingly mediated by the Aghlabids in Ifrīqiya, as well as their immediate neighbors—the Midrārids of Siljilmasa, the Rustamids of Tāhart, and, across the straits, the Umayyad emirate in al-Andalus—who occasionally came together against the Idrisid threat. Several modern studies have already made significant headway in exploring the relationship between state formation and urbanization in Morocco,5 the Idrisid mint and economy,6 and the archaeology of Idrisid towns.7 However, a full history of Idrisid state formation remains to be written, and the Idrisid state continues to be almost entirely ignored by scholars working on the Abbasids and the ʿAlid successor movements.8 This chapter draws together literary, numismatic, and archaeological evidence to explore different elements of the state formation process. I first present a brief introduction to the formative period of the Idrisid state and the ʿAlidAbbasid rivalry. I then explore the often-neglected claims that the Idrisids themselves made on their coins about their sovereignty and right to rule; I argue that these provide an important counterpoint to the narratives refracted through later texts and explain why the Idrisids were perceived as such a threat to Abbasid rule. I finally turn to explore relations between the Idrisids and their Berber subjects at the first Idrisid capital, Walīla (Roman Volubilis), where extensive excavations provide a unique glimpse into daily life in a town during the second/eighth and third/ninth centuries. Taken together, I suggest

5 See especially, Cressier, Urbanisation; García-Arenal and Manzano Moreno, Idrīssisme; as well as García-Arenal, Messianism 44–51, on the relationship with Mahdism. 6 Eustache, Corpus; Manzano Moreno, El desarrollo económico. 7 Excavations and field surveys have taken place at Walīla, al-Baṣra, Fez, Āghmāt, and Ḥajar al-Nasr. 8 Though, see Talbi L’Émirat Aghlabide 362–378 for an excellent analysis of the relationship between the Idrīsids and the Aghlabids.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

94

fenwick

that this evidence sheds new light on the claims made by the Idrisids, the workings of their new state on the ground, and its relations with the caliphate.

2

Challengers to Empire: Idrīs, the ahl al-bayt, and the Foundation of the Idrisid State

As an ʿAlid and member of the Ḥasanid branch of the family of the Prophet (ahl al-bayt), Idrīs b. ʿAbdallāh b. Ḥasan (d. 175/791) was a leading figure within the larger Zaydī community in the Ḥijāz. Members of the Prophet’s family provided a rallying point for those Muslim elites who did not accept Abbasid rule.9 Indeed, the first serious challenge to the Abbasids was a simultaneous, but unsuccessful, uprising in Mecca (the Ḥijāz) and Basra (Iraq) in 145/762– 763, led by two of Idrīs’s elder brothers, Muḥammad al-Nafs al-Zakiyya and Ibrāhīm.10 Idrīs played a pivotal role in a second ʿAlid revolt in 169/786 led by his cousin or nephew al-Ḥusayn b. ʿAlī that erupted in Medina and culminated in the disastrous Battle of Fakhkh and the exodus of Idrīs and other surviving family members from the Ḥijāz to the furthest margins of Abbasid rule—North Africa and the Caspian Sea.11 There are well-established challenges with the literary sources that describe Idrīs’s arrival in Morocco in 172/788–789 and the subsequent history of the Idrisid state. The earliest surviving accounts were written through the lens of ʿAlid-Abbasid rivalry; indeed, the flight and subsequent assassination of Idrīs I on caliphal orders became a panegyric theme in Abbasid historiography.12 Pro-Abbasid sources in particular often cast the Idrisids as credulous and naïve and their Berber followers as unfaithful and misguided. As Najam Haider has shown, there are important differences between these early narratives and an alternative Zaydī tradition that appeared in the mid-fourth/ tenth century and was framed within contemporary theological polemics about the broader Zaydī community.13 A newly published edition of Ahmad b. Sahl alRāzī’s (fourth/ tenth century) Akhbār Fakhkh wa-khabar Yaḥyā b. ʿAbdallāh wa-akhīhi Idrīs b. ʿAbdallāh (The stories of the battle of Fakhkh and Yaḥyā 9

10 11 12 13

On the insecurity of early ʿAbbasid rule, see Lassner, Shaping of ʿAbassid rule; for the rivalry between Abbasids, Ḥusaynids, and Ḥasanids, see Elad, Struggle for the legitimacy of authority; on the ʿAlids, see Bernheimer, ʿAlids. For the revolt, see Elad, Rebellion. For the revolt, see van Arendonk, Les débuts de l’imāmat zaidite 62–63. See, for example, the poem reproduced by al-Ṭabarī, History 30. Haider, Community divided, for an illuminating study of different versions of Idrīs’s journey and subsequent murder.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

95

b. ʿAbdallāh and his brother Idrīs b. ʿAbdallāh) offers significant potential to shed new light on early Idrisid history and politics, but it has yet to be integrated into Idrisid historiography.14 The most detailed information about the Idrisid rulers comes from the much later but local ninth/fourteenth-century Merinid sources, particularly Ibn Abī Zarʿ’s Rawḍ al-Qirṭās, which as Herman Beck, Mercedes García-Arenal, Eduardo Manzano Moreno, Simon O’Meara, and others have shown, sought to simultaneously reclassify the Idrisids as leaders of Mālikī Sunnism, to assert their relationship to the Prophet as members of the ahl al-bayt, and to glorify the status of the city of Fez (Fās) and Idrīs ii.15 Even so, the sources do provide a broadly reliable narrative of the establishment of the early Idrisid state from Idrīs’s arrival in the Maghrib to the division of northern Morocco among the sons of Idrīs ii, though there are challenges in resolving the different chronologies provided by authors. Idrīs, accompanied by his client Rāshid (in some accounts a Berber), arrived at the town of Walīla (Roman Volubilis) in 172/788–789 after a long and torturous journey from the Ḥijāz.16 Here he was welcomed by the Awraba tribe, whose leader, Abū Layla Isḥāq b. Muḥammad b. ʿAbd al-Ḥamīd (who according to al-Bakrī had already converted to Islam and was a Muʿtazilite), declared him imam shortly thereafter. By incorporating a member of the ahl al-bayt into their community, the Awraba acquired prestige and Islamic legitimacy. Idrīs subsequently undertook a series of military expeditions with his new Berber allies, which united a large swath of Morocco for the first time, as well as perhaps capturing silver mines south of the Atlas in the Draʾa and at Tudgha, where Abbasid dirham had been minted between 163–166/ 779–783 by Yazīd b. Ḥātim, the governor of Ifrīqiya. Idrīs minted silver dirham (at Tudgha and then Walīla) immediately to pay his followers,17 and as we will see below, these reveal his millennial claims and threat to Abbasid authority. Idrīs’s rule was short-lived, however, and he died in 175/791, apparently poisoned by a secret agent, probably in his tooth powder, sent by the Abbasid governor of Ifrīqiya (in some traditions it is Ibrāhīm b. al-Aghlab, the governor of the Zāb who became amīr of Ifrīqiya in 184/800) on the orders of the Abbasid

14 15 16 17

Al-Rāzī, Akhbār. First published in 1995 by Maher Jarrar. Beck, L’image d’Idrīs ii; O’Meara, Foundation legend of Fez; García-Arenal and Manzano Moreno, Idrīssisme. See Haider, Community divided. The Kitab al-Istibsar suggests that Rāshid was a Berber; see Beck, L’image d’Idrīs ii 25. Rosenberger, Les premières villes islamiques 238.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

96

fenwick

caliph, Hārūn al-Rashīd.18 That the murder of Idrīs was facilitated by Ibrāhīm is both a reflection of the limits of caliphal authority in North Africa as well as of the early rise of the Aghlabids as regional power brokers. Fortunately, Idrīs’s Berber wife or concubine, who was called Kanza or Kathīra, was pregnant and gave birth to a posthumous son, Idrīs ii, who in turn was proclaimed imam in a mosque at Walīla in 187/803 and received an oath of loyalty (bayʾa) from the tribesmen.19 Rāshid served as regent during Idrīs ii’s minority and continued to expand Idrisid control over northern Morocco; supposedly, he was so successful that he prepared a military campaign against Ifrīqiya. He was either assassinated or killed in a fight in either 186/802 or 188/804 through the machinations of Ibrāhīm b. al-Aghlab, and his head was sent to Baghdād.20 Ibrāhīm continued to be on the offensive against the Idrisids, bribing Berber chiefs such as Bahlūl al-Matgharī (who may have been Rāshid’s successor as regent) to come over to the Abbasid cause. Bahlūl was replaced by Abu Hālid Yazīd b. Ilyās al-ʿAbdī, presumedly of Arab rather than Berber origin.21 Idrīs ii apparently then wrote to Ibrāhīm a supplication for peace, making reference to his relationship with the Prophet and exhorting Ibrāhīm to submit to his authority and to renounce his pretensions to rule.22 Though the written sources are obscure and contradictory on these developments, coins provide firm evidence of Aghlabid campaigning—or at least political maneuverings— in Morocco: silver dirham were minted at Tudgha (south of the Atlas) in the name of Ibrāhīm b. al-Aghlab (r. 184–196/ 800–811) in 192/ 807–808, 193/808– 809, and perhaps also 186/802–803.23 The minting of Aghlabid dirham at Tudgha coincides with a significant shift in Moroccan power dynamics. In 192/ 808–809, Idrīs ii executed the chief of

18

19

20 21

22 23

On different traditions, see Haider, Community divided; and on the pivotal role of Ibrāhīm b. al-Aghlab, who was appointed governor in 179/795, see Talbi, L’Émirat Aghlabide 369– 371. As he points out, this could support a later date of 179/795 for the death of Idrīs, which is further supported by the coins that continued to be minted in Idrīs’s name until this date. Al-Bakrī, Description 239. While the Arabic term may be anachronistic, an oath of allegiance of a contractual nature has precedents in late antique North Africa, as well as pre-Islamic Arabia. See Talbi, L’Émirat Aghlabide 373–374; and Benchekroun, Rašid et les Idrissides; Ibn alAbbār, Kitāb 99–100. He may have been one of the 500 horsemen from al-Andalus and Ifrīqiya who reportedly came to Walīla in 189/805 and offered their services to Idrīs ii, though this is only mentioned by Ibn Abī Zarʿ, Histoire 14. Ibn al-Abbār, Kitāb 55. Al-Ush, Monnaies aglabide 99, nos. 188 and 183; 97, no. 175 (the date of 186/802–803 is deduced on stylistic grounds—no date is legible).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

97

the Awraba, Isḥāq, because of his treacherous dealings with the Aghlabids and dispatched his head eastward in a power play demonstrating the fate of disloyal chiefs.24 In the same year, he established a new quarter at Fez, which become his new center, displacing the dominance of Walīla and the Awraba further still. In so doing, he was following the well-established tradition of Muslim rulers as city-founders that Hugh Kennedy and others have explored elsewhere in detail.25 Our knowledge of Fez is almost entirely based on later written descriptions that draw heavily on established Arabic literary tropes for city-foundation.26 A first medina, Madīnat Fās, was built—most probably— by Idrīs i in 172/789 as a military camp for some of his Berber followers. This does not seem to have been a substantial settlement, and no coins were minted there.27 A second walled medina, called al-ʿAliyya (the higher), was founded by his son Idrīs ii in 192/808–809, and the earliest known coin was minted there in 197/812–813.28 Its foundation coincides with challenging political circumstances and tense relations between Idrīs ii and the Awraba at Walīla following his execution of Isḥāq.29 According to al-Bakrī, Idrīs ii gained the land for alʿAliyya through an agreement with the Berber chief of the Zawāgha who resided there in tents; the agreement was presumably similar to that made by his father with the Awraba at Walīla several years earlier.30 The ninth/fourteenth-century Rawḍ al-Qirṭās provides a crucial detail about this agreement: Idrīs purchased this land for 6,000 dirham and then started building.31 In the Quarter of the Andalous, Idrīs ii built a mosque called the “Mosque of the Shaykhs,” where he gave a Friday sermon, and in the al-Qarawiyyīn quarter, he built a mosque called the Mosque of the Shurafāʾ (today the Zāwiya of Moulay Idrīs ii), where the sermon was also given; apparently, next to it he built a qaysāriyya where coins were minted. Finally, he told the people to build their houses and cultivate the ground.32 Our limited archaeological knowledge of early medieval Fez supports a rapid investment in the infrastructure of the city in the second

24 25 26

27 28 29 30 31 32

Al-Bakrī, Description 376. Kennedy, How to found an Islamic city; Bacharach, Administrative complexes. On the history and organization of Fez, see Lévi-Provençal, La fondation de Fès; GarcíaArenal and Manzano Moreno, Idrīssisme. On literary tropes and city foundation legends, see O’Meara, Foundation legend of Fez; Valérian, Récits. No Idrisid coins have been found with the name Fās, only al-ʿAliyya and al-ʿAliyya Madīnat Idrīs. O’Meara, Foundation legend of Fez. See Eustache, Corpus 146–160. Al-Bakrī, Description 376. Al-Bakrī, Description 240. Ibn Abī Zarʿ, Histoire 30–34. Ibn Abī Zarʿ, Histoire 42–44; Beck, L’image d’Idrīs ii 67.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

98

fenwick

half of the third/ninth century. Rescue excavations have discovered stucco fragments with Quranic inscriptions as well as the foundations of the Idrisid phases of the al-Qarawiyyīn mosque, which according to tradition was built by a Kairouanese woman, Fāṭima al-Fihriyya, in 242/857.33 A painted wooden beam dedicated by Dāwūd b. Idrīs in 263/877 indicates that the decor of Idrisid mosques in Morocco was similar to that in Ifrīqiya and al-Andalus, unsurprising given the level of mobility in this period. The construction and monumentalizing of Fez did not occur in a vacuum. The rise of the Idrisids coincides with a phase of intense urbanization in the Maghrib al-Aqṣā: old settlements were re-occupied and still more towns were newly built or significantly expanded by the addition of fortification walls, mosques, hammams, markets, hydraulic systems, and so on.34 The building of Fez also needs to be situated within the political dynamics of third/ninthcentury Africa: in a short span of a few decades, the Idrisids’ rivals had already built the dynastic capitals of Sijilmāsa and Tāhart, and in Ifrīqiya, the Aghlabids built al-ʿAbbāsiyya in honor of the Abbasids.35 Now in his majority, Idrīs ii continued to expand his control over the northwest Maghrib and in 197/812 made a campaign against the Maṣmūda of the High Atlas, followed by an expedition to capture Tilimsān in 199/814–815, where he apparently established a mosque. This substantial territorial expansion is reflected in a proliferation of new mints in 197/812–813 (al-ʿAliyya, Tahlīṭ, Wargha, Wazeqqūr) and 198/813–814 (Āghmāt, Ṭanja—dubious), which are assumed to reflect the foundation of new towns (fig. 5.2).36 Less is known about the other so-called Idrisid foundations, but it is becoming increasingly clear that many were not new towns but rather were established on, or near, existing Berber settlements.37 The best known archaeologically is the town of al-Baṣra. Dirham were first minted there in 80/796–797, that is, during the regency of

33

34 35

36

37

Ettahiri, À la aube. Further rescue excavations in the al-Qarawiyyīn quarter have also found traces of early occupation, probably from the Idrisid period, see El Omari, Laoukili, and Akasbi, Fouilles archéologiques 222. Rosenberger, Les premières villes islamiques; Cressier, Urbanisation; Cressier, Quelques remarques. The model of the Idrisids as city-founders and benefactors is largely due to the testimony of al-Bakrī, our fullest pre-Merinid source for the region, and to whom we also owe our earliest detailed information on the foundation of the Midrārids at Sijilmāsa, the Rustamids at Tāhart, and the Salīḥids at Nakūr. For the mints, see Eustache, Corpus; and for their relations with urbanism, see GarcíaArenal and Manzano Moreno, Idrīssisme 23–30; Rosenberger, Les premières villes islamiques 236–241. Cressier, Quelques remarques.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

99

how to found an islamic state

figure 5.2 Map of Morocco showing the location of Idrisid mints, towns, and known silver mines corisande fenwick

Rāshid. The choice of name is certainly deliberate—the city of Basra in Iraq was where Idrīs’s elder brother Ibrāhīm had raised an abortive rebellion against the Abbasids in 145/ 762–763, and it continued to be a major center of both Shiʿi and Kharijite agitation in subsequent centuries. Though al-Baṣra was traditionally thought to be an Idrisid foundation of the late second/eighth century,38 it was not a wholly new creation. A substantial settlement of stone and mudbrick buildings with tiled roofs and flagstone pavement existed there as early as the

38

Eustache, Corpus 119–120; supplemented by Ettahiri and Meftah, Quelques émissions.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

100

fenwick

fifth or sixth century ce.39 This settlement seems to have collapsed, followed by a phase of ephemeral occupation, perhaps dating to the late first/seventh century. The town was subsequently refounded and covered an area of about 30 hectares protected by an enceinte in an ovoid form flanked by semicircular towers. These fortifications are described by al-Bakrī, who also mentions a seven-naved congregational mosque, two hammams, and two cemeteries.40 Territorial expansion and urbanization were not matched by increased Aghlabid intervention into the west, and on the guidance of his advisors, Ibrāhīm left to Idrīs ii those areas of the western Maghrib that were in his control and kept the status quo.41 Idrīs ii died at Walīla in 213/828 at the age of 33 after choking on a raisin. He left behind many sons, and when the eldest of these, Muḥammad b. Idrīs, succeeded, on the advice of his Berber grandmother he divided the territory among his brothers.

3

Striking Sovereignty

The Idrisids seem to be the first and only successor state in the Maghrib alAqṣā to mint silver coins in the second/eighth and third/ninth centuries, a claim for sovereignty that placed them in direct opposition to the Abbasid Caliphate in Baghdad and their Aghlabid vassals in Ifrīqiya and—across the straits—the Umayyad emirate in al-Andalus. The production of silver and gold coinage (sikka) was held to be a caliphal prerogative alongside the khuṭba (the ruler’s right to have his name proclaimed during the Friday sermon); in contrast, petty coinage in copper was not regulated and could be struck in the name of governors and other officials.42 Within the early Islamic world, coins were also a well-established medium for propaganda and revolutionary claims, particularly those about succession and the relationship between Islam and rulership.43 The Idrisid coins therefore offer what the later written accounts cannot: contemporary evidence for the claims that the Idrisids made themselves about their sovereignty and right to rule, rather than narratives recast through the lens of later sources. Coins have other advantages, too—when 39 40

41 42 43

Benco, Anatomy. Al-Bakrī, Description 216–217. Excavations have revealed a residential zone consisting of courtyard housing built in brick and pisé with roof tiles in the center and an industrial zone on the west; there may also have been gardens and orchards housed within the walls. Al-Raqīq (d. 417/1026) cited in Beck, L’image d’Idrīs ii, 20. Darley-Doran. et al., Sikka. Bacharach, Signs of sovereignty; Heidemann, Evolving representation; Gaiser, What do we learn.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

101

figure 5.3 Silver dirham of Idrīs i minted at Walīla (M113, insap-ucl Volubilis 2000–2005 excavations) photograph corisande fenwick

legible, they provide evidence on the date, the ruler, and the place of minting.44 However, the full potential of the numismatic evidence can only be extracted when stratigraphic evidence about their find spots is recorded, which is only rarely available for Morocco. Idrīs minted silver dirham as well as copper fulūs as soon as he was proclaimed imam in 172/788–789, and his successors continued this practice. The early Idrisid dirham mimic Abbasid dirham in most respects, but with some significant modifications that provide contemporary evidence of their claims for sovereignty (fig 5.3). First, they were made distinctive by the highly visible addition of ʿAlī below the name of Idrīs in reference to his lineage to the Prophet.45 Second, they replace the standard “Prophetic Mission” verse (Q 9:33) with the Quranic verse on “Truth” (Q 17:81). The reverse margin of the earliest dirham reads “Among the things ordered by Idrīs b. ʿAbdallāh: the truth has come ( jāʾa al-ḥaqqu), and falsehood has perished; surely falsehood is bound to perish.” As Luke Treadwell has shown, the phrase jāʾa al-ḥaqq had multiple connotations, referring to the arrival of the true religion of Islam and the salvation offered by the ʿAlids.46 The verse was a rallying cry of the ahl al-bayt: Idrīs’s brother alNafs al-Zakiyya referred to his partisans as aṣḥāb al-ḥaqq (companions of the truth) in his correspondence with the Abbasid caliph al-Manṣūr.47 Precisely the same verse appears on a dirham minted some 20 years earlier by Idrīs’s brother Ibrāhīm b. ʿAbdallāh in Basra (Iraq) during his abortive rebellion in 145/762– 763 against the Abbasid caliphs as well as on the “al-Mahdī al-Ḥaqq” dirham minted by an ʿAlid who may have been a third brother, Yahyā b. ʿAbdallāh, who raised a revolt in Tabaristan in 176/792.48 A recently published letter attributed 44 45

46 47 48

Eustache, Corpus, provides the definitive catalogue of Idrisid coins; supplemented by El Harrif, Les monnaies. There is even an example of an Idrisid dirham found in an Azerbaijan hoard minted in the name of ʿAlī b. Abī Ṭālib at Tudgha in 173/790–791; Noonan, When and how dirhams first reached Russia 453. Treadwell, Qurʾanic inscriptions 58. Traini, La corrispondenza. This was first observed by Lowick, Une monnaie ʿalide; and more recently revisited in a

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

102

fenwick

figure 5.4 Silver dirham of Idrīs ii minted at Walīla (M115, insap-ucl Volubilis 2000–2005 excavations) photograph corisande fenwick

to Idrīs and preserved in tenth-century and later Zaydī sources supports Treadwell’s reading of the verse and its use on coins. This daʿwa to the Berber Zanāta, Zawāgha, Ṣanmā, Ṣanhāja, and Lawāta tribes frames Idrīs’s claim to legitimacy in religious terms—as a restorer of religion and justice—and urges them to join the cause of the defenders of the Quran and the Sunna and to support justice, equality, and the rights of the descendants of the Prophet.49 We should therefore read the addition of ʿAlī and the selection of the “Truth verse” on the dirham as a purposeful rallying call that supported Idrīs’s claim to rule as a member of the ahl al-bayt by right of descent from ʿAlī.50 The use of the “Truth verse” was short-lived. By 197/812–813, Idrīs ii had reformed the text used on the dirham as well as the weight standard (fig. 5.4).51 The coins reverted to the standard but abbreviated “Prophetic Mission” verse used in the Islamic west, bringing them in line with those minted by the Aghlabids in Ifrīqiya and the Umayyads in al-Andalus, as well as by the Abbasids in the Middle East. Though the millennial message is removed, references to Idrīs ii’s descent from ʿAlī take on a more prominent place and are repeated in multiple places: on the obverse, ʿAlī now appears below the profession of faith,

49

50 51

thought-provoking article by Treadwell, Qurʾanic inscriptions 59 (Treadwell does not reference Lowick’s article). See also Miles, al-Mahdi al-Haqq 340. Eustache suggested that the slogan referred to the Idrisid war against the Kharijites and the unbelievers (Jews, Christians, and animists), but this is less convincing now; Corpus 63. Jarrer, Arbaʿa rasāʾil. These letters are excerpted from his 1995 edition of Ahmad b. Sahl al-Rāzī’s Akhbār Fakhkh wa-khabar Yaḥyā b. ʿAbdallāh wa-akhīhi Idrīs b. ʿAbdallāh [The stories of the battle of Fakhkh and Yaḥyā b. ʿAbdallāh and his brother Idrīs b. ʿAbdallāh]. On the importance of descent from the Prophet and Mahdism in Morocco, see also GarcíaArenal, Messianism. There seems to be a gap between in the mint output between (193/808–9) and (197/812– 13).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

103

as well as in the circular legend, which reads “There is no god but Allāh alone, who has no equal, ʿAlī.” On the reverse the name Idrīs (for Idrīs b. Idrīs) appears alongside the message “Muḥammad is God’s Prophet” and again with the name of ʿAlī. The dirham weight also drops from 2.6–7 grams to 2.3 grams (this fell further to about 2.1 grams by the end of the ninth century). It is tempting to link this coin reform and debasement—for a reform is what it was—to the expansion of Idrisid control across much of northern Morocco and the need for more silver to pay troops as well as to fund the foundation of Fez (the first coin from al-ʿAliyya dates to 197/812–813). Occasionally, the coins also reveal the honorific titles used by Idrīs or his successors, who are usually referred to as imams in the written sources.52 Ibn Khurdādhbih, writing in ca. 870, remarks that one did not address an Idrisid sovereign as a caliph in the sermon (khuṭba) but in his function as Ibn Rasūl Allāh, that is, as a descendant of the messenger of God.53 An intriguing undated fals found at Volubilis has a circular legend on the margin that states “on the order of the amīr Idrīs”—this must be either Idrīs or, perhaps more likely, his son Idrīs b. Idrīs.54 The title of amīr was a pre-Islamic Arabic term commonly used for governors and military commanders, and it also appears on two dirham minted at al-ʿAliyya by Muḥammad b. Idrīs in 216/831–832.55 Idrīs ii may have used the title in other contexts. A 198/813 letter from Pope Leo iii to Charlemagne reports that the patrician Gregory of Sicily had received a Muslim ambassador sent by the “amiralmumin” in Africa, a slightly scrambled Latin transliteration of the Arabic for amīr al-muʾminīn (commander of the believers).56 Talbi has convincingly identified this individual as Idrīs ii (rather than the Aghlabid amīr Abdallāh i), though he suggests that it is unlikely Idrīs ii used the full title.57 A contemporary, but as yet unparalleled, coin minted at Tahlīṭ in 197/812–813 bears the inscription “Muḥammad is the messenger of God and the mahdī is Idrīs b. Idrīs.”58 This use of the messianic title of al-mahdī (the guided one) was used by other members of the ahl al-bayt, especially Idrīs I’s brother Muḥammad, who was known both as al-mahdī and al-nafs al-zakiyya (“The Pure Soul”).

52 53 54 55 56 57 58

Ibn Khaldūn explains that this is because they were Shiʿite; Histoire des Berberes i, 453. Ibn Khurradādhbih, Kitāb 88–89. ‫امر به مما الاميرادر يس‬. El Harrif, Les monnaies 316. No picture or find spot provided. The title amīr is pre-Islamic, and can refer to any leader, see Athamina, Pre-Islamic roots. Migne, pl 98, 544–548; Senac, Le maghreb al-aqsâ 40–45. Talbi, L’Émirat Aghlabide 396. Eustache, Corpus 199, pl. v and vi (no. 140).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

104

fenwick

The claims on the coins were an explicit challenge to the authority of the Abbasid caliphs, a claim that would have been obvious far beyond northwest Africa given the astonishingly wide circulation of Idrisid dirham in the early second/late eighth and late second/early ninth century. Idrisid dirham have been found in Abbasid hoards in the Levant, Iraq, and Iran, as well as beyond its borders in the Caucasus and Russia, and in lesser quantities in the Baltic, Scandinavia, Central Europe, and Sicily.59 It is perhaps in this light that we should read al-Bakrī’s (writing ca. 460/1068) satirical verses about Idrīs ii that accuse him of wanting to be a caliph and Ibn al-Abbār’s (d. 658/1260) claim that Idrīs ii minted gold dīnār. He explains that the Aghlabid amīr, Ziyādat Allāh i (r. 201– 223/817–838) sent to Caliph al-Maʾmūn a sum of 1,000 dīnār minted in the name of Idrīs al-Ḥasanī so that he would understand the threat to the caliphate posed by the new state.60 This is probably historiographic fiction (certainly there is no numismatic evidence for the Idrisids minting gold, and it was their rivals, the Midrārids and Rustamids, who controlled Saharan trade with gold-rich West Africa), but the anecdote reflects the threat that the Idrisids as members of the Prophet’s family continued to pose to Abbasid claims to rule, as well as Aghlabid authority in Ifrīqiya.

4

Walīla: From Berber Center to the Center of the Idrisid State

When Idrīs arrived at Walīla in 172/788–789, he would have found a bustling small medieval town that looked very different from the towns of the Ḥijāz or the Mediterranean cities with their Romano-Byzantine heritage he had passed through on his way west (fig. 5.5).61 Though the town had served as the provincial capital of Mauretania Tingitana until the retreat of the Roman authorities in the third century ce, it was abandoned at least partially by the mid-fifth century ce following an earthquake. In the sixth century ce, Berber groups occupied the site and concentrated the population in the western third of the

59 60 61

Manzano Moreno, El desarrollo económico, especially table 1, 356–367. See also Noonan, Ninth-century dirham hoards. Al-Bakrī, Description 122. Ibn al-Abbār, Kitāb 104–105. See Talbi, L’Émirat Aghlabide 343– 345. This section draws on the results of the ongoing insap-ucl project at Volubilis directed by myself, Elizabeth Fentress, and Hassan Limane, as well as the work of earlier scholars on medieval Volubilis. See especially, Akerraz, Recherches; Fentress and Limane (eds.), Volubilis après Rome; Bennison, Walīla-Volubilis; Fentress, Fenwick, and Limane, Early medieval Volubilis.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

105

figure 5.5 Plan of early medieval Volubilis corisande fenwick and insap-ucl volubilis archaeological project

city, near the wadi; their elite continued to memorialize their dead in Latin and use Roman titles and the Roman provincial calendar until at least 34–35/655.62 After the Muslim conquest of Morocco, the town took on new importance as the residence of the Berber Awraba tribe who had converted to Islam (pre-

62

Akerraz, Note sur l’enceinte tardive.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

106

fenwick

sumably, but not certainly, the same group who had settled there in the sixth and seventh century ce). Over a century of excavations at Idrīs’s first base at Walīla (Roman Volubilis) offer exceptional—and untapped—potential to explore how the new Idrisid rulers interacted with their Berber subjects.63 The town of the Awraba was a substantial agro-settlement that minted its own coins, produced its own ceramics, and was linked with larger networks of trade and exchange. The residents of the town seem to have lived in rather simple, scattered housing within the late walls: small two-room houses—divided into domestic spaces and stabling/workrooms with storage silos outside— dating from the late first/seventh to third/ninth centuries have been excavated in two different areas (fig. 5.6).64 Probably in the early second/eighth century, the so-called “quartier arabe” was established outside the west gate and by the wadi. Built partly over the city midden and the ruins of an early Roman pagan cemetery, this new quarter contained densely built medieval housing, industrial activity, a canal, and possibly a mosque. We believe that a Muslim garrison may have been housed here, at least initially, and a gold dīnār, several dirham hoards, and hundreds of copper fulūs, as well as an intaglio inscribed in Arabic, have been found here.65 Copper coins with Arabic legends were minted at the town, most probably in the late second/eighth century, some inscribed simply with its name “Walīla” and others with the names of otherwise unknown individuals (Rashīd b. Kadīm, Muḥammad b. Khalifa, Ibrāhīm al-Ḥasanī).66 Ceramics were produced in kilns in the ruins of the Roman city, in the “Maison au Compas,” and in the center of the Berber settlement.67 The early conversion of Walīla’s inhabitants to Islam is supported by an absence of faunal evidence for pork consumption in second/eighth-century contexts, as well as burials laid out in accordance with Muslim funerary rites in the latest layers of the late antique cemeteries.68

63 64 65 66

67 68

For the archaeology of this period, see Fenwick, Early Islamic North Africa. Fentress and Limane, Volubilis après Rome 74–81. Eustache, Monnaies musulmanes; Fentress, Fenwick, and Limane, Early medieval Volubilis. These individuals have variously been identified as Abbasid generals, see Eustache, Monnaies musulmanes; and leaders of an independent Berber city-state similar to those established at Nakūr and Tilimsān in this period, see El Harrif, Monnaies islamiques trouvées à Volubilis. Most recently, it has been suggested that Rashīd b. Kadīm was the Rāshid who acted as regent for Idrīs ii during his minority; Benchekroun, Rašid et les Idrissides. Akerraz, Recherches 299–302. For the cemeteries, see Akerraz, Recherches 297, 303; for the animal bones, see King, Faunal remains.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

107

how to found an islamic state

figure 5.6 Reconstruction of the housing in sector D fernanda palmieri, insap-ucl volubilis archaeological project

Under Idrisid rule, in the late second/eighth and third/ninth centuries, the town grew rapidly and traces of occupation discovered outside the Roman walls to the northwest, as well as in the extramural zone, suggest that its extent far exceeded that of the late antique Berber settlement. There was at least one mosque in the Idrisid town. As we saw above, Idrīs ii was apparently proclaimed imam in a mosque at Walīla where he received the bayʿa.69 In the much later, but local, ninth/fourteenth-century account, Ibn Abī Zarʿ indicates also that a mosque existed at Walīla, and it was there that Rāshid presented Idrīs ii to the people in order to have him recognized as sovereign.70 This mosque was probably a relatively simple structure: a possible candidate in the “quartier arabe” is a large building with nine column bases just off the main street that was excavated but unrecorded by French excavations in the 1950s. Its identification and chronology remain to be confirmed by our ongoing excavations. The new rulers seem to have lived separately from the inhabitants of the town outside the city walls. A complex of three courtyard buildings and a hammam was constructed in the late second/eighth to early third/ninth cen-

69 70

Al-Bakrī, Description 239. Ibn Abī Zarʿ, Histoire 24, 29.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

108

fenwick

tury by the wadi, which Elizabeth Fentress and Hassan Limane suggest served as the residence of Idrīs I and his son (fig. 5.7).71 Each courtyard seems to have been reserved for different functions: from south to north a domestic space with a kitchen, a reception space, with rooms carefully floored with plaster and a raised platform decorated in red ochre at one end, and finally a storage space, with very large grain silos in the center that may have been a collective granary. These buildings are wholly different in plan and organization from the simple rectangular houses excavated in the Berber town, and Fentress has demonstrated that the courtyard plan has its origins in the Middle East and only appears in North Africa after the Muslim conquests.72 The small hammam was also organized on a Middle Eastern rather than typically North African plan and consisted of a vestibule, a cold room lined with benches and a pool at one end, a warm room, and a hot room. Analysis of consumption patterns suggests that those living in the extramural complex were living in a more privileged way than those living in the town: they were eating choice cuts of meat butchered elsewhere, dining off simple decorated tableware, discarding more coins, using imported glazed lamps, and processing cotton that must have been cultivated south of the Atlas Mountains.73 If this complex was indeed the residence of Idrīs and Idrīs ii, it is worth underscoring its small size and the simplicity of its surviving décor, even in comparison to contemporary housing elsewhere in Morocco. The only suggestion of monumentality is in the vestibule of the hammam, where a large spoliated block with a shield in relief from the still-standing Arch of Caracalla was built into the wall, perhaps as a “deliberate, symbolic, and ideologically charged transfer” from the old town.74 In contrast, excavations at the medieval Saharan trading entrepôt of Sijilmāsa, the Midrārid capital, uncovered a roughly contemporary elite residential complex consisting of rooms off a courtyard, a garden and a kitchen tentatively identified as the early dār al-imāra of the Midrārids.75 This house was lavishly decorated with painted and carved plaster, some inscribed with Quranic verses, and wooden ceilings painted with geometric patterns in an eastern style. Similar signs of wealth are seen in another third/ninth- to fourth/tenth-century house at Sijilmāsa that contained stone columns and similar painted plaster décor, as well as a filigree gold ring and

71 72 73 74 75

Fentress and Limane, Volubilis après Rome 82–102. Fentress, Reconsidering Islamic housing. Fentress and Limane, Volubilis après Rome 97–100. Fentress, Idris i 523–524. Messier and Fili, Earliest ceramics. The building is dated by by radiocarbon to 785–875 Cal ad. A detailed account of the excavations remains to be published.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

109

how to found an islamic state

figure 5.7 Reconstruction of the Idrisid complex fernanda palmieri, insap-ucl volubilis archaeological project

carved ivory acquired through Saharan trade. Excavations at Walīla have not yet revealed such signs of imported wealth. As the archaeological evidence reveals, Walīla seems to have remained very much the town of the Muslim Awraba. Life in the Berber town continued much as it had before Idrīs’s arrival, albeit perhaps with a more monetized economy: houses, ceramics, eating practices, and funerary habits follow the same pattern. The new rulers seem to have been very much outsiders: they resided outside the city walls in a separate compound consisting of houses and hammam built on a Middle Eastern rather than North African architectural plan. Even though they seem to have had access to a broader range of goods and foodstuffs than those living in the Berber town, there is no sign of drastically discrepant wealth. Indeed, Ibn Khaldūn states firmly that “his [Idrīs’s] rule over them cannot be considered an Arab rule, because the Berbers were in charge of it, and there were not many Arabs in it.”76 It is perhaps not surprising, then, that the Berber town of Walīla was not enough for Idrīs ii, and he sought to establish his own new capital at Fez.

76

Ibn Khaldūn, Muqaddimah ii, 288.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

110 5

fenwick

Conclusion

By the late second/early ninth century, North Africa was ruled by Muslim states but not by the caliphate. The exodus of Idrīs to the Maghrib recast the board for the powerplay between the Abbasids and the ʿAlids, as encounters between the two groups became increasingly dependent on intermediaries like the Aghlabids.77 Even in the Far West, however, the Idrisids, as members of the ahl al-bayt, remained a substantial threat to the Abbasids in the late second/eighth and early third/ninth century. This brief exploration of the early Idrisid state identifies some key points for future exploration. The first point is the fundamental driving role that local tribes played in state formation and urbanization. Idrīs was an outsider from the east who grouped around him different Berber tribes who selected him to lead for reasons that escape us but might include his connection to the Prophet, personal charisma, a saintly role, arbitrating role, as well as his military success. As Patrice Cressier and others have highlighted, this pattern is also replicated in later written accounts of the genesis of other North African states: new states and cities only formed as a result of a symbiosis between old “Berber” and new “Arab” populations.78 This fits well with the influential “internal frontier” model developed by Igor Kopytoff to describe the rise of political entities in sub-Saharan Africa.79 Kopytoff argues that the majority of African “states” did not evolve from simpler political forms but rather were established by immigrant groups who migrated into the “internal frontiers” between fully formed regional political systems. These frontier zones were occupied by small, decentralized groups like the Awraba. The new immigrant groups, often of chiefly or royal origins, brought with them pre-existing social and political models from their former polity. To legitimate their authority and claim to the land, newcomers had to tap into the principle of “first-comer primacy”;80 hence the role of the charismatic, saintly, but neutral, founder and the importance of foundation myths. Initially, newcomers sought to attract followers and new members as kinsmen or pseudo-kinsmen using a corporate kin-group model, but as the new polity became established, new followers were integrated under a contractual model between ruler and subjects.

77 78 79 80

For an excellent, and the fullest, analysis of the politics between the Idrisids and the Aghlabids to date, see Talbi, L’Émirat Aghlabide 362–378. For a full exploration of these issues, see especially Cressier, Nakūr; and for a detailed examination of the formation of the Rustamid state, see Aillet, Tāhart. Kopytoff, Internal African frontier. Kopytoff, Internal African frontier 51.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

111

how to found an islamic state

Islam, as a religion, played a fundamental role in legitimizing and justifying the rule of individual rulers, but Islamic structures of power were also sophisticated and flexible enough to act as a model and framework for ambitious rulers on the edges of the caliphate. In the case of Idrīs and his descendants, their claims for legitimacy seem to have been based in part on descent (consanguinity with the Prophet), an imam-centric conception of the religio-political order, and a model of reciprocal loyalty between ruler and elite that was periodically reaffirmed through ceremonies such as the bayʿa.81 The performance of the bayʿa in the mosque at Walīla (and perhaps the minting and distribution of silver dirham to pay troops) provides an illustration of the contractual model in practice using the language of pledges and allegiances provided by Islam. The messages inscribed on the Idrisid dirham provide contemporary evidence into the legitimizing discourse used to justify their rule and how it shifted from an original emphasis on ʿAlid millennial claims for justice and the truth under Idrīs to descent from the Prophet and ahl al-bayt membership under his son. The act of urban foundation—of establishing a new city for a new dynasty—proved one of the most visible and enduring acts of legitimization in the Maghrib alAqṣā, just as it did in the imperial center. These acts of striking coins and urban foundation in Idrisid Morocco were not purely a local phenomenon but part of a far wider conversation in the medieval world about how to be a Muslim ruler—a dialogue that the rival states in the Far West brought to a head for the Muslim world.

Bibliography Sources al-Bakrī, Description de l’Afrique septentrionale par El-Bekri, trans. W.M. de Slane, Algiers 1913 (rev. ed.). Ibn al-Abbār, Kitāb al-ḥulla al-siyarāʾ, ed. H. Mu’nis, Cairo 1964. Ibn Abī Zarʿ, Histoire des souverains de la Berbérie occidentale, (171–726H/788–1326J.-C.): Le Kitâb rawḍ al-Qirtâs, trans. A. Beaumier, Paris 1860. Ibn Khaldūn, Histoire des Berberes et des dynasties musulmanes de l’Afrique Septentrionale, trans. W.M. de Slane, 4 vols., Algiers, 1852–1856. Ibn Khaldūn, The Muqaddimah. An Introduction to History, trans. F. Rosenthal, 3 vols., Princeton, NJ 1958. Ibn Khurradādhbih, Kitāb al-masālik [wa-] al-mamālik, ed. Goeje, 2nd ed., Leiden 1967.

81

On ceremonial, see Marsham, Rituals of Islamic monarchy.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

112

fenwick

al-Rāzī, Akhbār Fakhkh wa-Khabar Yaḥyā ibn ʿAbdallāh wa-Akhīhi Idrīs ibn ʿAbdallāh [The stories of the battle of Fakhkh and Yaḥyā b. ʿAbdallāh and his brother Idrīs b. ʿAbdallāh], ed. M. Jarrar, Beirut 1995. al-Ṭabarī, The history of al-Tabari, vol. 30: The Abbasid Caliphate in equilibrium: The caliphates of Musa al-Hadi and Harun al-Rashid a.d.785–809/a.h. 169–193, trans. C.E. Bosworth, New York 2015.

Studies Aillet, C., Tāhart et l’imamat rustumide (c. 160/777–296/909), in ai 45 (2011), 47–78. Akerraz, A., Note sur l’enceinte tardive de Volubilis, in Bulletin archéologique du Comité des travaux historique 19B (1983), 429–436. Akerraz, A., Recherches sur les niveaux islamiques de Volubilis, in P. Cressier and M. García-Arenal (eds.). Genèse de la ville islamique en al-Andalus et au Maghreb occidentale, Madrid 1998, 295–304. Al-Ush, M. A-l. F., Monnaies aglabides étudiées en relation avec l’histoire des Aglabides, Damascus 1982. Anderson, G., C. Fenwick, and M. Rosser-Owen (eds.), The Aghlabids and their neighbors: Art and material culture in ninth-century North Africa (Oriental handbook), Leiden 2017. Athamina, K., The pre-Islamic roots of the early Muslim caliphate, in Der Islam 76 (1999), 10–11. Bacharach, J.L., Administrative complexes, palaces and citadels: Changes in the loci of medieval Muslim rule, in I. Bierman, R. Abou-El-Haj, and D. Preziosi (eds.), The Ottoman city and its parts: Urban structure and social order, New Rochelle, NY 1991, 111–128. Bacharach, J.L., Signs of sovereignty: The Shahāda, Qurʾanic verses, and the coinage of ʿAbd al-Malik, in Muqarnas 27 (2010), 1–30. Beck, H.L., L’image d’Idrīs ii, ses descendants de Fās et la politique s̱ẖarīfienne des sultans marīnides: 656–869/1258–1465, Leiden 1989. Benchekroun, C.T., Rašid et les Idrissides: L’histoire “originelle” du Maroc entre marginalisation et glorification, in Al-Qantara 35 (2014), 7–27. Benco, N.L., Anatomy of a medieval Islamic town: Al-Basra, Morocco, Oxford 2004. Bennison, A.K., Walīlā-Volubilis dans la tradition textuelle arabe, in E. Fentress and H. Limane (eds.), Volubilis àpres Rome: Les fouilles ucl/insap, 2000–2005, Leiden 2018, 61–73. Bernheimer, T., The ʿAlids: The first family of Islam, 750–1200, Edinburgh 2013. Brett, M., The Islamisation of Morocco: From the Arabs to the Almoravids, in Morocco 1 (1992), 57–71. Capel, C., At the dawn of Sijilmasa: New historical focus on the process of emergence of a Saharan state and a caravan city, in M. Sterry and D.J. Mattingly (eds.), State

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

113

formation and urbanisation in the ancient Sahara and beyond, Cambridge 2020, 594– 620. Capel, C., and A. Fili, Sijilmāsa au temps des Midrārides: Nouvelles approches historiques et premier bilan archéologique, in C. Aillet (ed.), L’ibadisme dans les sociétés de l’Islam medieval, Berlin 2018, 137–168. Cressier, P., Urbanisation, arabisation, islamisation au Maroc du Nord: Quelques remarques depuis l’archéologie, in P. Cressier, J. Aguadé, and Á. Vicentes (eds.), Peuplement et arabisation au Magreb occidental, Madrid 1998, 27–39. Cressier, P., Nakur: Un émirat rifian pro-omeyyade contemporain des Aghlabides, in G. Anderson, C. Fenwick, and M. Rosser-Owen (eds.), The Aghlabids and their neighbors: Art and material culture in ninth-century North Africa, Leiden 2017, 491–513. Cressier, P. Quelques remarques sur la genèse des villes islamiques au Maghreb occidental, in S. Panzram and L. Callegarin (eds.), Entre civitas y madīna. El mundo de las ciudades en la Península Ibérica y en el norte de África (siglos iv–ix), Madrid 2018, 317–330. Darley-Doran, R.E., et al., Sikka, in ei2 online. Djaït, H., La Wilāya d’Ifrīqiya au iie/viiie siècle: Étude institutionnelle (suite et fin), in si 28 (1968), 79–107. El Harrif, F-Z., Monnaies islamiques trouvées à Volubilis: Liberté locale et pouvoir abbaside, in Actes de Ières journées nationales d’Archéologie et du patrimoine, Rabat, 1–4 juillet 1998, vol. 3, Rabat 2001, 142–159. El Harrif, F.-Z., Les monnaies, in E. Fentress and H. Limane (eds.), Volubilis àpres Rome: Les fouilles ucl/insap, 2000–2005, Leiden 2018, 308–338. Elad, A., The struggle for the legitimacy of authority as reflected in the ḥadīth of alMahdī, in J. Nawas (ed.), Abbasid Studies ii: Occasional papers of the School of Abbasid Studies, Leuven 2010, 39–96. Elad, A., The rebellion of Muḥammad al-Nafs al-Zakiyya in 145/672, Leiden 2016. El Omari, M.E.I., M. Laoukili, and M. Akasbi, Fouilles archéologiques à Funduq Ṣāgha à Fès: Eléments pour une lecture de l’urbanisme de la ville, in Hespéris-Tamuda 52 (2017), 209–227. Ettahiri, A.S., À l’aube de la ville de Fès. Découvertes sous la mosquée al Qarawiyyin, in Dossiers de l’archéologie 365 (2014), 42–49. Ettahiri, A.S., and N. Meftah, Quelques émissions monétaires de l’atelier d’al-Basra (alMaghrib al-Aqsa), in P. Sénac and S. Gasc (eds.), Monnaies du haut Moyen Âge: Histoire et archéologie, Toulouse 2015, 265–280. Eustache, D., Monnaies musulmanes trouvées à Volubilis, in Hespéris 43 (1956), 133– 195. Eustache, D., Corpus des dirhams idrīsites et contemporains: Collection de la Banque du Maroc et autres collections mondiales, publiques et privées, Rabat 1970. Fentress, E., Reconsidering Islamic housing in the Maghreb, in I. Grau Mira and S. Guti-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

114

fenwick

érrez Lloret (eds.), De la estructura domestica al espacio social. Lecturas arqueologicas del uso social del espacio, Alicante 2013, 237–244. Fentress, E., Idris i and the Berbers, in G. Anderson, C. Fenwick, and M. Rosser-Owen (eds.), The Aghlabids and their neighbors: Art and material culture in ninth-century North Africa, Leiden 2017, 514–530. Fentress, E., and H. Limane (eds.), Volubilis après Rome. Les Fouilles ucl/ insap 2000– 2005, Leiden 2018. Fentress, E., C. Fenwick, and H. Limane, Early medieval Volubilis: The archaeology of a Berber town, in Hésperis-Tamuuda 56 (2020), 209–235. Fenwick, C., Early Islamic North Africa: A new perspective, London 2020. Fenwick, C., The Umayyads and North Africa: Imperial rule and frontier society, in A. Marsham (ed.), The Umayyad world, London 2021, 293–313. Gaiser, A.R., What do we learn about the early Khārijites and Ibāḍiyya from their coins? in jaos 130 (2010), 167–187. García-Arenal, M., Messianism and puritanical reform: Mahdis of the Muslim west, Leiden 2006. García-Arenal, M., and E. Manzano Moreno, Idrīssisme et villes idrīssides, in si 82 (1995), 5–33. Haider, N., The community divided: A textual analysis of the murders of Idrīs b. ʿAbdallāh (d. 175/791), in jaos 128 (2008), 459–475. Heidemann, S., The evolving representation of the early Islamic empire and its religion on coin imagery, in A. Neuwirth, N. Sinai, and M. Marx (eds.), The Qurʾān in context, Leiden 2009, 149–196. Jarrer, M., Arbaʿa rasāʾil Zaydiyya mubakkira [Four early Zaydī epistles], in I. al-Saʿāfīn (ed.), Fī Miḥrāb al-Maʿrifa. Festschrift for Iḥsān ʿAbbās, Beirut 1997, 267–298. Kennedy, H., How to found an Islamic city, in C. Goodson, A.E. Lester, and C. Symes (eds.), Cities, texts and social networks 400–1500, Aldershot 2010, 45–63. King, A., The faunal remains, in E. Fentress and H. Limane (eds.), Volubilis après Rome. Les Fouilles ucl/ insap 2000–2005, Leiden 2018, 369–386. Kopytoff, I., The internal African frontier: The making of African political culture, in I. Kopytoff (ed.), The African frontier: The reproduction of traditional African societies, Bloomington 1987, 3–84. Lassner, J., The shaping of Abbasid rule, Princeton, NJ 1980. Lévi-Provençal, É., La fondation de Fès, in É. Lévi-Provençal (ed.), Islam d’Occident: Études d’histoire médiévale, Paris 1948, 14–42. Lowick, N., Une monnaie ʿalide d’al-Basrah datée de 145 H (762–3 après J.-C.), in Revue numismatique 6 (1979), 218–224. Marsham, A., Rituals of Islamic monarchy: Accession and succession in the first Muslim empire, Edinburgh 2009. McDougall, J., Histories of heresy and salvation: Arabs, Berbers, community and the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

how to found an islamic state

115

state, in K.E. Hoffman and S.G. Miller (eds), Berbers and others: Beyond tribe and nation in the Maghrib, Bloomington and Indianapolis 2010, 15–38. Manzano Moreno, E., El desarrollo económico de las ciudades idrisíes: La evidencia numismática, in P. Cressier and M. García-Arenal (eds.), Genèse de la ville islamique en al-Andalus et au Maghreb occidentale, Madrid 1998, 353–377. Messier, R., and A. Fili, The earliest ceramics of Sijilmasa, in P. Cressier and E. Fentress (eds.), La céramique maghrébine du haut moyen âge, viiie–xe siècle: État des recherches, problèmes et perspectives, Rome 2011, 129–146. Migne, J-P., pl [Patrologiae cursus completus, seria Latina], Leiden 1841–1855. Miles, G.C., Al-Mahdi al-Haqq, amīr al-muʾminīn, in Revue numismatique 6 (1965), 329– 341. Noonan, T., When and how dirhams first reached Russia. A numismatic critique of the Pirenne theory, in Cahiers du Monde Russe et Soviétique 21 (1980), 401–469. Noonan, T.S., Ninth-century dirham hoards from northwestern Russia and the southeastern Baltic, in Journal of Baltic Studies 13 (1982), 220–244. O’Meara, S., The foundation legend of Fez and other Islamic cities in light of the life of the Prophet, in A.K. Bennison and A.L. Gascoigne (eds.), Cities in the pre-modern Islamic world: The urban impact of religion, state and society, London 2007, 27–41. Rosenberger, B., Autour d’une grande mine d’argent du Moyen Age marocain: Le Jebel Aouama, in Hespéris-Tamuda 5 (1964), 15–77. Rosenberger, B., Les premières villes islamiques du Maroc: Géographie et fonctions, in P. Cressier and M. García-Arenal (eds.), Genèse de la ville islamique en al-Andalus et au Maghreb occidentale, Madrid 1998, 229–256. Senac, P., Le maghreb al-aqsâ et l’occident chrétien (viiie–ixe siècles), in si 98/99 (2004), 29–48. Savage, E., and A.A. Gordus, Dirhams for the empire, in P. Cressier and M. García-Arenal (eds.), Genèse de la ville islamique en al-Andalus et au Maghreb occidentale, Madrid 1998, 377–402. Talbi, M., L’Émirat Aghlabide, Tunis 1966. Talbi, M., Hérésie, acculturation et nationalisme des berbères Bargawata, in Premier congrès des cultures Méditerranéennes d’influence arabo-berbère, Alger 1973, 217– 223. Traini, R., La corrispondenza tra al-Manṣūr e Muḥammad an-Nafs az-zakiyya, in aiuon 14 (1964), 773–798. Treadwell, L., Qurʾanic inscriptions on the coins of the ahl al-bayt from the second to fourth century ah, in jqs 14 (2012), 47–71. van Arendonk, C., Les débuts de l’imāmat zaidite au Yemen, Leiden 1960. Valérian, D., Récits de fondations et islamisation de la mémoire urbaine au Maghreb, in C. Aillet and B. Tuil Leonetti (eds.), Dynamiques religieuses et relation au sacré dans le Maghreb médiéval. Éléments d’enquête, Madrid 2015, 151–168.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

116

fenwick

Wansbrough, J., On recomposing the Islamic history of North Africa: A review article, in jras 101 (1969), 161–170. Wurtzl, C., The coinage of the revolutionaries in the late Umayyad period, Museum Notes (American Numismatic Society) 23 (1978), 176–178.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 6

Rethinking “the Mamlūk State” with Ibn Khaldūn: “Mamlukization,” ʿaṣabiyya, and Historiographical Imaginations of the Sultanate of Cairo (1200s–1500s) Jo Van Steenbergen

A state exists chiefly in the hearts and minds of people; if they do not believe it is there, no logical exercise will bring it to life.1

∵ For over four decades now, Professor Hugh Kennedy has been one of the foremost scholars of the political, social, military, and economic history of the early Islamic period. Time and again he has confirmed the validity of this reputation by arguing and demonstrating—in countless lectures and papers as well as in many high-impact publications—that the organization of the Abbasid Caliphate’s leadership between the second/eighth and the fourth/tenth centuries concerned a far more complex, more dynamic, and more negotiated set of arrangements than is generally taken for granted. Interestingly, in Professor Kennedy’s opening lecture in the prestigious series “Vorlesungen über islamische Geschichte und Kultur in memoriam C.H. Becker” at the Asien-Afrika Institut of Hamburg University (Germany)—published as a journal article in Der Islam in 2004—an additional interpretive layer was added to his more traditional social, economic, and military forays into the study of this Abbasid complexity. In this most ambitiously generalizing (and tantalizing) specimen of his seminal scholarship, Professor Kennedy concluded that “[t]he breakup of the Islamic empire was essentially a product of the success of the dominant culture”: whereas this breakup in the fourth/tenth century “was no simple matter, but the result of a complex interaction of political forces,” a major

1 Strayer, On the medieval origins of the modern state, 5.

© Jo Van Steenbergen, 2022 | doi:10.1163/9789004525245_007

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

118

van steenbergen

factor obviously was the Abbasid imperial center’s growing irrelevance within the worldview and the frameworks of reference of the new Islamic commonwealth’s diverse subject populations and their local Muslim leaderships.2 This chapter consists of an equally ambitious reflexive essay on the historical relationship between political organization and a dominant culture, especially between the complexity of leadership arrangements on the one hand and, on the other, their relevance (and irrelevance) within the worldviews and the frameworks of reference of those who partake in those arrangements. Rather than Professor Kennedy’s early Islamic Abbasid imperial complexity, however, this chapter will foreground the case study of leadership organization in another one of premodern Islamic history’s more successful transregional political formations: the late medieval sultanate that dominated the eastern Mediterranean, Egypt, Greater Syria (Bilād al-Shām), southeastern Anatolia, and the Ḥijāz from its capital of Cairo between the mid-seventh/thirteenth and the early tenth/sixteenth centuries. Following Professor Kennedy’s identification of a political center’s need to also retain cultural relevance, this essay’s central question revolves around the historical agency, in the maintenance of this relevance, of the cultural imaginations and representations of this sultanate’s leadership organization. In fact, it will be argued that today’s standard appreciations of these imaginations and representations can, and should, move in new directions. Just as has been happening for many years now in the Abbasid case, these new directions should build on the acknowledgment that the organization of the sultanate’s leadership between the seventh/thirteenth and the tenth/sixteenth centuries concerned a far more complex, more dynamic, and more negotiated set of arrangements than is generally taken for granted. In modern scholarship, this sultanate’s long and complex history has always been imagined, and defined, as that of “the Mamluk Empire” or, especially, “the Mamluk State.” The first part of this chapter summarizes this paradigmatic imagination and relates it to the historical thinking of the eighth/fourteenthcentury scholar, historian, and courtier Ibn Khaldūn (d. 808/1406). In the second part, I question the validity of this paradigmatic imagination by sketching the far more complex patterns of leadership organization that are increasingly acknowledged to have been at play in the long history of the sultanate’s power elites. The third part returns to Ibn Khaldūn’s thinking. In pursuit of a different perspective to think about the sultanate’s complexity and continued relevance, this part connects Ibn Khaldūn’s conceptualization of

2 Kennedy, Decline and fall 27–29 (quotes from 27, 29).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

119

social power with the turn that modern state studies have been making to cultural readings of the state phenomenon, and to taking account of meaningmaking practices in their analyses of state-formation processes. In the fourth part, I adopt this different perspective to examine the relationship between a diverse handful of late medieval cultural representations of the sultanate’s complexity and relevance, especially by Ibn Khaldūn and his colleague historians. Finally, in this chapter’s concluding observations I tie these different parts together in an alternative interpretation of the continuity of the sultanate’s internal relevance and coherence, which arguably outlasted that of any other major Islamic polity—including the Abbasid one—before the early modern period.

1

From Francis Fukuyama to Ibn Khaldūn: The “Mamlukization” of Leadership Organization

A useful synthesis of modern scholarship on the organization of the sultanate’s leadership may be found in the fourteenth chapter of the first volume of The origins of political order by political scientist Francis Fukuyama: The institution of military slavery anchored Muslim power in Egypt and Syria for three hundred years, from the end of the Ayyubid dynasty in 1250 to 1517, when the Mamluk sultanate was defeated by the Ottomans. […] [T]he Mamluks ran a real state, with a centralized bureaucracy and a professional army—indeed, the army was the state, which was both a strength and a weakness. […] Key to its success was the sultanate’s ability to capture fresh waves of new recruits from the Central Asian steppe and from the Byzantine lands to the north and northwest. Some of the recruits were Muslim already, others were still pagan, and others Christian. The process of conversion to Islam was vital to remaking their loyalties and tying them to their new masters. Key too was the fact that the recruits were completely cut off from access to or communication with their families and tribes. As a result of their boyhood training, they acquired a new family, the family of the sultan and the Mamluk brotherhood. […] In addition to the way they were educated, a key to the success of the Mamluks as a political institution was the fact that they were a one-generation nobility. They could not pass on their Mamluk status to their children; their sons would be ejected into the general population and their grandsons would enjoy no special privileges at all. The theory behind this was straightforward: a Muslim could not be a slave, and all of the Mamluks’

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

120

van steenbergen

children were born Muslims. Moreover, the Mamluk children were born in the city and raised without the rigors of nomadic life on the steppe, where the weak died young. Were Mamluk status to become hereditary, it would violate the strict meritocratic grounds on which young Mamluks were selected.3 For Fukuyama, just as for all modern studies that informed Fukuyama’s description, “the institution of military slavery” remained the backbone of power in Egypt and Syria for “300 years,” generating in the process “a real state”; the key to the success of this centralized and bureaucratic military state was the regular supply of new military slaves from the north and their selection for positions of leadership along purely meritocratic principles. Indeed, this powerful late medieval Syro-Egyptian political formation always tends to be imagined as a slave state, uniquely kept together in uniform and continuous ways by a legal military institution that justifies identifying it as the “Mamluk” Sultanate. This uniformity across Egypt and Syria and this continuity between the midseventh/thirteenth and early tenth/sixteenth centuries were secured, according to standard scholarship, by the exceptional combination of a continuous influx of new generations of military slaves (mamlūks), a particular bureaucratic organization that favored mamlūks, and the normative practice of those mamlūks’ political domination. One of the most explicit and straightforward identifications of the late medieval Syro-Egyptian polity along this bureaucratic slave-state line may actually be found in the writings of the late medieval North African scholar, politician, judge, and historian ʿAbd al-Raḥmān Ibn Khaldūn (732–808/1332–1406).4 In the early 780s/1380s, after a very active career in the politics and scholarship of North Africa and al-Andalus, Ibn Khaldūn ended up in Egypt, where he remained an active member of Egypt’s power elites, as a judge and legal scholar, as a courtier and royal envoy, and as a teacher and author, until his death in 808/1406. The most important and impressive product of Ibn Khaldūn’s authorship undoubtedly was his universal history, the Book of warnings (Kitāb al-Iʿtibār). This multivolume literary work includes a famous “first book” (al-kitāb al-awwal)—better known as “the Introduction” (al-Muqaddima)— which presents Ibn Khaldūn’s full-fledged macrohistorical theory of the laws that in his view determined human history, and a second and third book, which 3 Fukuyama, Origins 202–207; summarizing the writings of David Ayalon, Peter Holt, Reuven Amitai, Stephen Humphreys, Linda Northrup, Amalia Levanoni, Jean-Claude Garcin, and Carl Petry. 4 Ayalon, Mamlūkiyyāt.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

121

present an extensive chronicle of the history of West Asia and North Africa from the time of Creation until his own eighth/fourteenth century. Ibn Khaldūn’s remarkable historiographical project was concluded with a presentation of his own autobiographical microhistory, the “Acquaintance with Ibn Khaldūn” (alTaʿrīf bi-Ibn Khaldūn).5 As has been noted on many occasions, Ibn Khaldūn’s explanation of the sultanate’s leadership experience through the legal status of slavery appears in the fifth volume of the standard edition of the Book of warnings. Presenting the history of Islamic West Asia from the fifth/eleventh century until Ibn Khaldūn’s own time, this volume includes the following key passage that summarizes the basic writ of this Mamluk argument in a coherent historical narrative: When the [Abbasid] dawla was drowned in sedentary lifestyles and luxury and donned the garments of calamity and impotence and was overthrown by the heathen Tatars [= the Mongols], who abolished the seat of the Caliphate and obliterated the splendor of the lands and made unbelief prevail in place of belief, because the people of the faith, sunk in self-indulgence, preoccupied with pleasure and abandoned to luxury, had become deficient in energy and reluctant to rally in defense, and had stripped off the skin of courage and the emblem of manhood—then, it was God’s benevolence that He rescued the faith by reviving its dying breath and restoring the unity of the Muslims in the Egyptian realms, preserving its order and defending the walls of Islam. He did this by sending to the Muslims, from this Turkish people and from among its great and numerous tribes, commanders to defend them and utterly loyal helpers, who were brought from the House of War to the House of Islam under the rule of slavery (al-riqq), which hides in itself a divine blessing. By means of slavery they learn glory and blessing and are exposed to divine providence; cured by slavery, they enter the Muslim religion with the firm resolve of true believers and yet with nomadic virtues unsullied by debased nature, unadulterated with the filth of pleasure, undefiled by the ways of sedentary living, and with their ardor unbroken by the profusion of luxury.6

5 Martinez-Gros, Ibn Khaldûn; Fromherz, Ibn Khaldun; Dale, Orange trees; Irwin, Ibn Khaldun; Ibn Khaldūn, Cheddadi (trans.), Livre. 6 Ibn Khaldūn, ʿIbar v, 441 (v, 371) (translation adapted from Lewis, Islam i, 97–99); also in Ayalon, Mamlūkiyyāt 345–346 (but in a more rudimentary translation).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

122

van steenbergen

Just as Fukuyama would suggest six centuries later, Ibn Khaldūn explains here, at the very start of his “report of the dawla of the Turks (dawlat al-turk) who upheld the Abbasid dawla (al-dawla al-ʿAbbāsiyya) in Egypt and Syria,” how in his reading these Turkish slaves appeared as Islam’s saviors in Egypt in the middle of the seventh/thirteenth century and how around them a truly meritocratic military apparatus of power emerged. In this process, Ibn Khaldūn continues in the same passage, “they are elevated in the ranks of the dawla,” “some of them are nominated to sit on the throne of the Sultan and direct the affairs of the Muslims,” “one intake comes after another and generation follows generation,” and “the branches of the dawla flourish with the freshness of youth.”7 In the same passage, Ibn Khaldūn goes on to describe the integrative functionality of this, for him, unique and salvationist historical appearance of military slavery by making explicit reference to the complex Arabic concept of ʿaṣabiyya, mostly rendered in modern studies of Ibn Khaldūn’s writings as social cohesion, solidarity, or group feeling. He states that the institution of military slavery “intensifies solidarity, increases coercive force, and is conducive to a protective kind of cohesion.”8 In fact, the causal connection that is suggested here between ʿaṣabiyya, coercive empowerment, and safety ties this passage directly to the general theory of human history that Ibn Khaldūn famously formulated in the first part of his Book of warnings, the Muqaddima. More generally, this entire passage in the middle of his chronicle situates his understanding of the Syro-Egyptian sultanate entirely within the interpretive and moralizing historical framework that he developed in that Muqaddima. This framework is entirely constructed, as among others Tarif Khalidi has argued, around the analytical problem of “how power is acquired, how it is maintained, and how it is lost.”9 As is well-known, Ibn Khaldūn’s theoretical pursuance of this problem in the Muqaddima is built upon the idea of leadership formation’s iterative movement between the two ideal types of tribal nomadism and urban efflorescence, a movement caused by the waxing and waning of ʿaṣabiyya’s social cohesion and of its corollary, social power, in these tribal and urban contexts, respectively. In fact, the above-quoted passages from volume five of Ibn Khaldun’s chronicle do not just situate this “report of the dawla of the Turks” in Egypt and Syria within the theoretical framework that is explained in the Muqaddima by referring explicitly to ʿaṣabiyya. They also achieve this by participating in the development of a moralizing argument of salvation that 7 Ibn Khaldūn, ʿIbar v, 441–442 (Lewis, Islam; Ayalon, Mamlūkiyyāt). 8 Ibn Khaldūn, ʿIbar v, 441–42 (Lewis, Islam; Ayalon, Mamlūkiyyāt). 9 Khalidi, Arabic historical thought 224. See also Cheddadi, Pouvoir.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

123

is constructed around the opposite processes of integration and disintegration that follow, for Ibn Khaldūn, from the interaction between social power and, respectively, “nomadic virtues” or “sedentary lifestyles.” Interestingly, Ibn Khaldūn suggests that in Egypt and Syria this endless process of social power’s waxing and waning that he described in the Muqaddima, and thus this iterative movement in human history, were uniquely overcome by the legal institution of military slavery and its direct and regularly renewed infusion of nomadic virtues into a sedentary context. By continuously refreshing the empowering cohesion of ʿaṣabiyya and by keeping at bay in this “mamlukizing” manner the centrifugal tendencies of sedentary lifestyles, this institution of military slavery brought about for Ibn Khaldūn, as it were, his own version of “the end of history.”10

2

Dynastic vs. mamlūk Realities and the Complexity of the Sultanate’s Leadership Formation

Despite this paradigmatic centrality of military slavery in various imaginations of the sultanate’s political organization, it is well established by now that for most of the eighth/fourteenth century the position and power of the sultan in Cairo, and thus the claims to power and sovereignty in and beyond Egypt and Syria, were not monopolized by former military slaves only. Rather, the sultanate’s leadership was dominated by one kin group and its many (mamlūk and non-mamlūk) associates, all rallying around the many descendants of the former mamlūk and Sultan al-Malik al-Manṣūr Qalāwūn (d. 689/1290). As a result of this dominance, eighth/fourteenth-century Syro-Egyptian sociopolitical, cultural, and economic history is first and foremost a dynastic history, constituted by the formation and many transformations of the Qalāwūnid royal household and power apparatus. This was certainly never an evident trajectory for the sultanate’s history. However, for diverse reasons—not in the least also the many contingencies of human history—the Qalāwūnid dynastic apparatus and its elites (of both mamlūk and non-mamlūk origins) prevailed into the 780s/1380s, when they were removed in complex and violent contexts by

10

Cf. the famous title of Francis Fukuyama’s The end of history and the last man. But see also Loiseau, Reconstruire x, 148–157, who concludes that the chronicle report of the history of this “dawla of the Turks” that follows this general appreciation of the “divine blessing” of military slavery in Egypt scrupulously follows the default model of social power’s waxing and waning, ending its historical reconstruction in the mid-1390s on a far more negative note than this introduction would suggest (“L’État en phase terminale”).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

124

van steenbergen

a new group of supporters of the former mamlūk and non-Qalāwūnid Sultan al-Malik al-Ẓāhir Barqūq (d. 801/1399).11 Instrumental in that formation of the centennial Qalāwūnid dynastic reign in Egypt and Syria was the long term in royal office of Qalāwūn’s son al-Malik al-Nāṣir Muḥammad, who reigned and ruled in increasingly powerful and centripetal capacities between 693/1293 and 741/1341. It has even been argued that this reign coincided with the construction of “[a] greater interconnectedness between al-Nasir Muhammad’s household, Egypt’s and Syria’s governing elites, the state’s bureaucratic and military structures, normative ideas of social order, and cultural aesthetics than ever before in the sultanate’s history.”12 Both the reality and the perception of this Qalāwūnid cohesion were shattered when Barqūq eventually managed to consolidate his power. However, as is being increasingly well established by now, the former mamlūk Barqūq did so by investing heavily in generating a new “greater interconnectedness” between his own novel royal household, “Egypt’s and Syria’s governing elites, the state’s bureaucratic and military structures, normative ideas of social order, and cultural aesthetics.” Just as in the Qalāwūnid case in the 740s/1340s and beyond, this Barqūqid dynastic cohesion was continued at the turn of the century in the reign of Barqūq’s son and successor al-Malik al-Nāṣir Faraj (d. 815/1412), until its reality and perception also fell to many years of internecine warfare, violence, and all kinds of related historical contingencies.13 Although never again similarly successful, this pattern of radical transformation, empowerment, and elite integration was continued throughout the long ninth/fifteenth century. It manifested most explicitly in the continued grooming for the sultanate of dynastic relatives, as well as in the repeated and mostly violent shattering of longstanding leadership formations when the contingencies of the period’s six main moments of leadership transition—in the mid820s/early 1420s, the early 840s/late 1430s, the late 850s/early 1450s, the mid860s/early 1460s, the early 870s/late 1460s, and the early 900s/later 1490s— never again similarly favored any dynastic succession.14 And in fact, the sultanate’s early days, in the mid-seventh/thirteenth century, had been marked by similarly integrative pursuits and transformative experiences. At that time, these had coalesced around the former mamlūk and Sultan al-Malik al-Ẓāhir

11 12 13 14

See, e.g., Van Steenbergen, Mamluk sultanate; Fuess, Mamluk politics 99–100. Flinterman and Van Steenbergen, Al-Nasir Muhammad 108. Onimus, Maîtres; Igarashi, Establishment; Gardiner, Al-Malik al-Ẓāhir Barqūq. Sievert, Kampf; Loiseau, Reconstruire 200–203; Loiseau, Les Mamelouks 131–132; Mauder, Childless rule.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

125

Baybars (d. 676/1277) and his sons, until their successor Qalāwūn and his supporters shattered any perception of a Ẓāhiriyya cohesion and ushered in a Qalāwūnid century.15 With the sultanate’s integration and reproduction of power elites in Egypt, Syria, eastern Anatolia, the eastern Mediterranean, and the Ḥijāz appearing as all but self-evident, and as a messy process that moved in all kinds of directions and that involved many different agents and agencies, the question remains as to what kept it all together. What made this sultanate appear throughout history as one continuous political entity, which would have remained a coherent, relevant, and stable central leadership formation for longer than most other major Islamic polities before the early modern period? As explained in the preceding section, most of today’s explanations continue to follow in Ibn Khaldūn’s moralizing and teleological footsteps. They respond to this question by characterizing the mamlūk origins of many of these power elites as the unchanging essence and the hard organizational kernel of an autonomous military-bureaucratic organizational structure. However, the sultanate’s diverse power elites are not known to have ever strongly identified with any mamlūk claims.16 Indeed, their leadership experiences and claims were defined by many more factors than any mamlūk origins alone can account for. At the same time, as just demonstrated, neither can the many different household, kin, and dynastic relations of power account for the apparent interconnectedness of those experiences and claims. This is true even though it was especially this type of relation that defined the unity and continued relevance of each of late medieval West Asia’s many other transregional leadership formations, from the Ayyubids to Jingizids, Ottomans, and Timurids. Therefore, the question remains: if not the meritocratic and coercive mechanisms of “military slavery” nor any sustained form of a dynastic identity and loyalty, what made the sultanate’s long history nevertheless appear—and be interpreted then and now—as one of a continuous and structured order of relevant central leadership?

15 16

Northrup, Slave to sultan 75–90; Stewart, Between Baybars and Qalāwūn. See Yosef, The term mamlūk 27 (“There is no evidence that manumitted mamlūks were proud of their slave status. On the contrary, it seems to have been considered degrading and manumitted slaves with aspirations made great efforts to repress their servile past by claiming an exalted origin or by creating marital ties with established families.”).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

126 3

van steenbergen

From Ibn Khaldūn to Timothy Mitchell: ʿAṣabiyya and the Sultanate as a Structural Effect

A useful starting point to pursue this question of the sultanate’s cohesion is in fact a more critical reflection on both Ibn Khaldūn’s notion of ʿaṣabiyya and its resonance with modern theoretical interpretations of the state phenomenon. Just as is true for any other concept that was used by Ibn Khaldūn to construct his theoretical framework, what he meant exactly when using ʿaṣabiyya is not easy to pin down. This flexibility and vagueness of Ibn Khaldūn’s theoretical vocabulary, marked by various neologisms and particular semantic shifts, has been identified as one of the many reasons for the timeless appeal of his theories and writings. In general, however, one could argue that those theories and vocabularies are imbued with a kind of historical materialism that inclines toward a type of natural determinism. Given that, as mentioned, the central problem of the Muqaddima is to understand “how power is acquired, how it is maintained, and how it is lost,” one of the key fragments in that Muqaddima for understanding Ibn Khaldun’s conceptualization of the manifestation of social power in human history, and for seeing how the semantics of his specific vocabulary were interconnected, is the following: “It should be known that royal authority (al-mulk) must be built upon two foundations. The first is coercive force and social cohesion (al-shawka wa-al-ʿaṣabiyya), which finds its expression in the soldiers (al-jund). The second is wealth (al-māl), which supports those soldiers and provides the whole infrastructure needed by royal authority.”17 This fragment opens a theoretical exploration in the Muqaddima of how “disintegration” (al-khalal) befalls any dawla, that is, any constellation of power relationships claiming royal authority (mulk), when interrelated problems befall these military and economic sources of social power. The entire passage thus also informs about the high value that Ibn Khaldūn attributed to the integrative force of ʿaṣabiyya. Power in the Muqaddima is indeed closely related to the construction of kin groups and identities, “since everybody’s pride for his kin and his ʿaṣabiyya is more important (than anything else).”18 For Ibn Khaldūn, lineage (nasab) and ʿaṣabiyya’s social cohesion and solidarity are, in fact, mutually constitutive. At one point the Muqaddima even explains in some detail how “ʿaṣabiyya can only be when interconnected with lineage or something corresponding to it,” and he adds on several occasions that “clientage (al-wilāya) and association through slavery or alliance 17 18

Ibn Khaldūn, Muqaddima iii, 47, 326; Rosenthal (trans.) iii, 45. See more in general Cheddadi, Pouvoir. Ibn Khaldūn, Muqaddima ii, 7, 141; Rosenthal (trans.) ii, 7.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

127

(al-mukhālaṭa bi-al-riqq aw bi-al-ḥilf ) have the same effect [as lineage].”19 Ibn Khaldūn’s ʿaṣabiyya thus invokes social practices defined by relationships of power that, while being social constructs, are of a very personal and patriarchal nature and that serve in particular the successful wielding of violence against outsiders. As is well-known, for Ibn Khaldūn this key quality of ʿaṣabiyya is by necessity strongest in Bedouin environments, where scarce natural resources and strategies of survival favor the emergence of a shared sense of purpose and interest that, if successfully pursued, generates enormous coercive power. In contexts of empowerment and urbanization, Ibn Khaldūn explains, intricate sets of other types of relationships are added to these traditional ʿaṣabiyya bonds. When this happens, inevitably and—in his own wording—“naturally” (ṭabīʿī) these patriarchal relationships come under increasing pressure, they are marginalized, and they eventually evaporate, so that subsequent generations of rulers and elites are forced to replace them with other, less affective, and less effective relationships. Eventually, this loosening of social bonds will lead to the loss of power and the rise of a new ʿaṣabiyya group and its much stronger coercive force. According to Ibn Khaldūn’s Muqaddima, this natural process of patriarchal empowerment may, however, be maximized, and its inevitable disintegration postponed, when other factors manage to expand and strengthen ʿaṣabiyya types of relationships. These factors were for Ibn Khaldūn especially related to “prophecy or sainthood” (nubūwa aw walāya) and to “religious vocation” (al-daʿwa al-dīniyya).20 As suggested above, it was in chapter five of his chronicle, rather than in the Muqaddima, that Ibn Khaldūn added to these factors his view on the specific, divinely ordained appearance of (military) slavery (riqq) in contemporary Egypt and Syria. When considering the abovementioned complex pattern of usurpation, empowerment, and haphazard dynastic integration that was at work in late medieval Egypt and Syria, one could argue that various ʿaṣabiyya groups—constructed around lineage as well as “clientage and association through slavery or alliance”—succeeded each other in the sultanate’s leadership between the seventh/thirteenth and early tenth/sixteenth centuries, without any particular appearance of military slavery (nor for that matter any form of “sainthood” or “religious vocation”) being able to prevent their repeated disintegration. The

19

20

Ibn Khaldūn, Muqaddima iii, 20, 203; Rosenthal (trans.) iii, 18. See also Ibn Khaldūn, Muqaddima ii, 8, 141; Rosenthal (trans.) ii, 8. For the conceptualization of nasab as an inclusive and imagined social construct, see also Ibn Khaldūn, Muqaddima iii, 20, 203; Rosenthal (trans.) iii, 18. Ibn Khaldūn, Muqaddima ii, 27, 166; Rosenthal (trans.) ii, 26; see also Ibn Khaldūn, Muqaddima iii, 4–6, 147; Rosenthal (trans.) iii, 4–6.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

128

van steenbergen

only factor that remained constant, irrespective of any loss of “coercive force and social cohesion” and even of “wealth,” was the sultanate itself, and the entire infrastructure of “royal authority” (mulk) that was constructed around it. Rather than sainthood, religious vocation, or military slavery, it is therefore the particularity of the sultanate itself that invites to be considered another factor that was able to absorb Ibn Khaldūn’s natural forces of social integration and disintegration, and to create, as it were, another layer of empowering ʿaṣabiyya bonds. The most straightforward position to take in any continuation of this theoretical argument would be to return to Fukuyama’s conclusion that “the Mamluks ran a real state, with a centralized bureaucracy and a professional army,” replacing “the Mamluks” with a more representative signifier but accepting as a given this resilient and autonomous agency of the sultanate’s “real state.” Taking this traditional position does not respond to the question, however, of how this apparent resilience and agency were generated or for that matter what signifier would be more appropriate to represent the sultanate’s apparent unity, the continued relevance of its “royal authority” (mulk), and its ʿaṣabiyya effect. As announced in the introduction, in this chapter I choose to pursue a theoretically informed line of inquiry that prioritizes an approach that does not simply accept this “real state” as a given but rather tries to respond to this question of its resilience, agency, and effect. From the middle of the twentieth century onward, more theoretically inspired approaches to political organization and state formation started to emerge in the field of premodern Islamic history. This has happened especially in the study of late medieval Islamic West Asia, where the writings of historians such as Ira Lapidus and John Woods, and more recently Michael Chamberlain and Jürgen Paul, have begun to lead the way to new readings of power and leadership. Each in their own way, these new readings have turned for their analyses to the Weberian concept of traditional patriarchal and patrimonial authority, to the military household, family rule, and patronage, and to mostly urban infrastructures for the local management and organization of resource flows.21 The remainder of the present chapter pursues a line of inquiry that follows from these new readings but that at the same time pushes the need for critical reflection beyond any boundaries of the social world that continue to define them. More specifically, in order to respond to the question of the sultanate’s resilience and its apparent ʿaṣabiyya effect, I turn to meaning, and to the pursuit

21

Lapidus, Muslim cities; Woods, Aqquyunlu; Chamberlain, Knowledge; Chamberlain, Military patronage states; Paul, Lokale und imperiale Herrschaft; Paul, Violence.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

129

of a better understanding of the relationship between processes of meaning making and of leadership organization. One of the reasons for this is that one specific trend in present-day historical state studies prioritizes exactly these discursive processes in its interpretations and analyses of “the state.” It suggests how the unity, uniformity, and agency of any historical state phenomenon appears first and foremost from the successful imagination and representation of a legitimate and coherent order of leadership, rather than from the diverse and messy social realities that always marked the life and times of its leaders. It is this model of the premodern state’s incarnation, not just in “specific organizational structures and mechanisms,” as Pierre Bourdieu argued, but especially in “the form of mental structures and categories of perception and thought,” that I want to explore further in the remainder of this chapter.22 More specifically, I choose to connect Ibn Khaldūn’s late eighth/fourteenth-century ideas about the relationship between “royal authority” (mulk) and ʿaṣabiyya with the analytical model that the contemporary historian and political theorist Timothy Mitchell formulated toward the end of the twentieth century.23 For Mitchell, expanding on Michel Foucault’s notions of “governmentality” and the “capillary” nature of power, any interpretive engagement with “the state phenomenon” should avoid “attributing to it the coherence, unity, and absolute autonomy that result from existing theoretical approaches.” He explains in detail how more worthwhile insights may be gained when the state is considered an effect of practices of power and when at the same time it is acknowledged that that effect is a constituent aspect of those practices. There is a state phenomenon, but its origins and being lie in social practice rather than in any top-down structure. More precisely, this state phenomenon appears for Mitchell from an endless process of structuring abstraction, or reification, that is a function of that practice and of the social power it generates. Mitchell therefore states that “the state needs to be analyzed as […] a structural effect. That is to say, it should be examined not as an actual structure, but as the powerful, metaphysical effect of practices that make such structures appear to exist.”24 This state effect of social practice, as Mitchell argues, consists especially of the creation of the impression of a “two-dimensional” world “that appears to consist not of a complex of social practices but of a binary order: on the one hand individuals and their activities, on the other an inert structure [of ‘the state’].”25

22 23 24 25

Bourdieu, Rethinking 3–4. Mitchell, Limits; Mitchell, Society. Mitchell, Limits 94. Mitchell, Limits 94.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

130

van steenbergen

If we look at the state phenomenon as both an ideational function and a simplifying but effective explanation of social practice, we can also pursue a deeper understanding of that ideational explanation’s capacity to generate an appearance of unity and relevance where complexity and messiness prevail. In fact, following Ibn Khaldūn’s suggestion of the maximization of ʿaṣabiyya practices and relationships of power through external (especially religious) factors, I propose to adopt this specific structural-effect perspective and think of the “royal authority” (mulk) of the sultanate itself as another of these integrative factors.26 The sultanate’s longstanding leadership order should be examined both as a function of a composite variety of ʿaṣabiyya practices and as the shared imagination of a natural, God-given organization of leadership, which effectively absorbed that variety and its many contradictions in a “two-dimensional” and “binary” construct of those who lead and those who are led.

4

Syro-Egyptian Historiographical Practice and the “dawla of the Turks”

In the Syro-Egyptian context, this effective idea of the natural, God-given binary order of the sultanate’s leadership arguably was produced, and reproduced, through all kinds of cultural practices. Following the interpretative model of the state phenomenon as a powerful discursive idea with direct effects, these cultural practices can indeed be considered to operate as communicative micromanagers that performed the sultanate’s leadership organization as “royal authority” (mulk), forcing the social reality of the returning messiness of that leadership organization into this royal authority’s metanarrative of a “binary order” that was topped by a continuous leadership and its Godgiven hierarchies of sovereignty, and that is imagined as “the state” in modern historiography, and as the dawla by Ibn Khaldūn and his contemporaries. Arguably, one of the most effective of these cultural practices was, indeed, the art of literary engagement with the sultanate’s history that Ibn Khaldūn as well as others among the sultanate’s courtiers and scholars practiced in increasingly

26

Considering Ibn Khaldūn’s identification of three types of “royal authority” (al-mulk) (“natural [ṭabīʿī] authority,” serving the ruler’s whims only; “governing [siyāsī] authority,” pursuing worldly public benefit through rational policies; and “the caliphate [al-khilāfa],” pursuing public benefit in this as well as the next world in accordance with “the Lawgiver’s” directions; Ibn Khaldūn, Muqaddima iii, 25, 211; Rosenthal(trans.) iii, 23), one could even argue that it is the integrative qualities of his third, charismatic type that featured most effectively in the background of this imagination.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

131

intense and self-conscious ways27 At the very least, late medieval Arabic historiography’s performative relationship with that complex process of the sultanate’s leadership formation deserves much more and much closer attention, including as one of the alternative ways in which the question of the sultanate’s longstanding cohesion and relevance can be responded to. Interestingly, and unlike what has become the norm in modern scholarship, apart from the afore-quoted famous passage in volume five of his Book of warnings on the salvationist capacity of military slavery, in most of his writings Ibn Khaldūn did not define the political formation of the sultanate as an exclusively mamlūk one. In fact, none of the sultanate’s contemporary historians ever attached much distinguishing value to that mamlūk factor.28 As mentioned before, in his chronicle Ibn Khaldūn identified the sultanate as “the dawla of the Turks” (dawlat al-atrāk) rather than that of the mamlūks. This “dawla of the Turks” concerns a combination of two highly complex terms. As explained above, for Ibn Khaldūn the first component of dawla refers to any constellation of power relationships that claimed royal authority (mulk). More generally, dawla basically carries a chronological and temporal meaning that invokes the combined notions of divine providence and of a period of sovereign rule (mulk) by a specific individual or social group.29 In Ibn Khaldūn’s chronicle the origins of the sultanate’s dawla, of its God-given moment of sovereign rule, are situated in the 640s/1240s, and its narrative runs all the way up to Ibn Khaldūn’s own days, to the year 1394 (“Shaʿbān [7]96”), which is the final year discussed in the chronicle.30 At the start of his report of this dawla’s century and a half of history, Ibn Khaldūn also explains how the second component, “the Turks,” is considered by him a very open social signifier. He clarifies more precisely that “the word ‘Turk’ has become the dominant way of representing all of [the Turks, Turkmen, Armenians, ‘Romans,’ Circassians, etc.], because of the high number and excellence of [the Turkish mamlūks in the elite’s ranks].”31 In fact, all historiographical writings of the period between the later seventh/thirteenth and the early tenth/sixteenth centuries shared in this practice of identifying the sultanate by its “Turkishness” and by its existence since the mid-seventh/thirteenth century.32 Across the substantial variety of authors 27 28 29 30 31 32

Khalidi, Arabic historical thought 182–231. For an Ottoman parallel, see Piterberg, Ottoman tragedy. Van Steenbergen, “Nomen est omen”; Yosef, Term mamlūk 8. Van Steenbergen, Appearances of dawla 54–55; Loiseau, Reconstruire 148–149, n. 17. Ibn Khaldūn, ʿIbar v, 604. Ibn Khaldūn, ʿIbar v, 443. The additions between brackets are taken from the preceding passage in the text of the ʿIbar, to which this one refers. See also Yosef, Dawlat al-Atrāk.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

132

van steenbergen

and texts, however, the precise meaning of this idea of the “dawla of the Turks” tended to vary substantially. One such variation interestingly appeared in the course of the eighth/fourteenth century, when that metanarrative of the “dawla of the Turks” acquired a distinctly Qalāwūnid flavor. This certainly emerges from a panegyric text by a scribe of the court for one of al-Nāṣir Muḥammad’s sons and successors, which proposed and defended its own innovative integration of Qalāwūnid history into the metanarrative of the “dawla of the Turks.” This panegyric indeed did not just explain how “the noble Sultanate remained with the royal house of al-Manṣūr [Qalāwūn] (al-bayt al-sharīf al-Manṣūrī) from Rajab of the year [6]78 (December 1279) until now.”33 It also explicitly integrated this dynastic time into a wider special chronology of “the dawla of the Turks,” which, according to this text, “was only established in the year 643 [1245], and it has continued until this blessed year, the year 743 [1343], the [entire] period [being] 100 years.”34 By referring to the 640s/1240s as a point of origin, the chronological parameters of this author’s dawla clearly aligned with those of Ibn Khaldūn’s in his chronicle “report of the dawla of the Turks (dawlat al-turk) who upheld the Abbasid dawla (al-dawla al-ʿAbbāsiyya) in Egypt and Syria.” Ibn Khaldūn, however, deploys a very different timeline of what happened after the 640s/ 1240s: he identifies three different generations of rulers and elites that populated the sultanate’s dawla between 648/1250 and 796/1394, and he applies his iterative model to explain that by the later eighth/fourteenth century this dawla was in decline because its original mamlūk ʿaṣabiyya had been lost to increasingly degenerate substitutes of social cohesion. As Julien Loiseau concludes, at the turn of the century Ibn Khaldūn thus speculated on the sultanate’s imminent fall, despite the many efforts of his patron, Sultan Barqūq: “it is like the wick of a lamp, which, just before burning out, gives off a flame which makes one believe that it will be kindled again.”35 Despite Ibn Khaldūn’s speculation, the sultanate continued to exist for another century. As mentioned before, in the course of these many years different ʿaṣabiyya groups succeeded each other until the end of the sultanate’s history, but none managed to consolidate their power again for more than one generation. Simultaneously, different groups of historians continued to write the messy history of this discontinuity of social power into the vague metanarrative of the “dawla of the Turks,” and of its structural appearance as representing a natural, God-given, and binary order of the continuous sovereign authority 33 34 35

Van Steenbergen, Qalāwūnid discourse 13. Van Steenbergen, Qalāwūnid discourse 11. Loiseau, Reconstruire 148–153; quote from Ibn Khaldūn’s Muqaddima, on 153.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

133

(mulk) of the sultan and his agents over their subjects. One highly interesting historiographical practice that was pursued to achieve this goal is exemplified by the following representation of a son’s succession of his father on the throne in Cairo in a contemporary court chronicle of Egyptian history: “He is the 37th Sultan among the rulers of the Turks and their descendants in the Egyptian lands, and [he is] the 13th [Sultan] of the Circassians and their descendants. He became Sultan on Wednesday 14 Jumādā l-Ūlā of the year 865 [25 February 1461].”36 As suggested by the number 37, this historian gives this numerical information whenever a sultan’s accession to the throne in Cairo is recorded in his chronicle. Thus, some dozen pages later in the text, we find an even more interesting instance of this practice, when the quick usurpation of this son’s power by the mamlūk military commander Khushqadam (d. 872/1467) is interpreted in the following, similar way: [Al-Malik al-Ẓāhir Khushqadam] is the 38th Sultan among the rulers of the Turks and their offspring in the Egyptian domains, and the first of the “Romans” (Arwām), after there had been 13 rulers of the Circassians and their offspring who had been Sultan, I mean, from the beginning of the reign of al-Ẓāhir Barqūq, who has initially established the dawla of the Circassians. As for those who preceded of the Circassian Turks and the “Romans,” there is much disagreement over them, due to the lack of precision of the historians over this issue and in what has been written by them. [However,] from the reign of al-Malik al-Ẓāhir Barqūq up to the present day, [it is certain] that the first of the Circassians was Barqūq and the first of the “Romans” was Khushqadam, and that they are separated by 81 years to the day, as both assumed the Sultanate on 19 Ramaḍān (i.e., Barqūq on 19 Ramaḍān 784 and Khushqadam on 19 Ramaḍān 865).37 By suggesting a sequence of first 37 and then 38 sultans, these very different accessions to the sultanate—the one a pre-arranged dynastic succession, the other a violent usurpation—are explicitly, directly, and coherently connected by this historian to each other as well as to specific seventh/thirteenthcentury origins of the “dawla of the Turks.” In that century, according to this author, this sequence began with the accession in 648/1250 of the former Ayyubid mamlūk Aybak (d. 655/1257) as “the first of the rulers of the Turks in Egypt,” then in 655/1257 of Aybak’s son ʿAlī (d. 657/1259) as “the second of the

36 37

Ibn Taghrībirdī, al-Nujūm al-zāhira xvi, 218. Ibn Taghrībirdī, al-Nujūm al-zāhira, xvi, 253.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

134

van steenbergen

rulers of Egypt belonging to the Turks,” in 657/1259 of Aybak’s former mamlūk Quṭuz (d. 658/1260) as “the third of the rulers of the Turks in Egypt,” and in 658/1260 of Aybak’s and then Quṭuz’s rival, the former Ayyubid mamlūk Baybars (d. 676/1277), “who is the fourth of the rulers of the Turks.”38 The explicit reference to other numeric sequences of Circassian and even “Roman” (that is, from the Anatolian and southeast European domains of the Ottoman “Sultan of Rūm”) sultans furthermore integrates alternative narratives of ninth/fifteenthcentury specialty and coherence into the metanarrative of that dawla. This construction of specific ruler chronologies clearly parallels what had been done before for Qalāwūnid dynasts and for Ibn Khaldūn’s mamlūk champions. In the final century and a half of the sultanate’s existence, however, this integration was achieved by prioritizing memories of the shared ethnic/geographic origins of different generations and groups of formerly mamlūk sultans.39

5

Concluding Observations

This exploratory essay has subscribed to the argument that any constellation of central power is by default unstable and messy, and it is one of the key pillars of any type of central rule to make this messy constellation appear not only as legitimate but also as a coherent and relevant part of the natural order of things. This was arguably no different for the sultanate that dominated Egypt, Syria, and adjacent regions between the mid-seventh/thirteenth and early tenth/sixteenth centuries. Rather than any more straightforward meritocratic or dynastic idea, however, a heterogeneous, even polysemous, image of sovereign order was successfully generated to create an appearance of more than 250 years of coherence and relevance for the diverse social realities of this sultanate’s different central leaderships. Historiographical interventions such as numbering rulers and applying parallel dynastic, mamlūk, and ethnic chronologies all participated in this creative performance of leadership order and unity across time, space, and historical transformations. In doing so, meritocratic notions of a one-generation slave nobility as well as kin-related

38

39

Ibn Taghībirdī, al-Nujūm al-zāhira vii, 3, 41, 72, 94. A handful of other ninth/fifteenth- and, especially, early tenth/sixteenth-century historians adopted the same historicizing practice of numbering rulers in their chronicles, see, e.g., al-Maqrīzī (d. 845/1442), al-Sulūk iv, 763; Ibn Iyās (d. ca. 930/1522), Badāʾiʿ iii, 324; Ibn al-Ḥimṣī (d. 934/1527), Ḥawādith 285, 293, 336, 344, 403, 449, 465, 482, 499; Ibn Ṭūlūn (d. 953/1546), Mufākahat 56. See also Loiseau, Les Mamelouks 173–205, especially his notion of “the ethnicization of power” on 199.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

135

conceptions of power were integrated into the vague but charismatic contours of a Syro-Egyptian dawla that was defined by seventh/thirteenth-century origins and a vague but distinctive “Turkish” leadership identity. Together with an equally distinctive, and performative, range of infrastructures of power (from the materiality of the Citadel of the Mountain in Cairo to mechanisms of surplus accumulation via iqṭāʿ and waqf, and to the symbolic capital of the Sunni Abbasid Caliphate of Cairo), this powerful combination of specific leadership features and performances continued to generate, as it were, its own highly resilient, integrative, and empowering ʿaṣabiyya effect. As suggested in the handful of examples mentioned above, it did so by gradually shifting discursive priorities from the seventh/thirteenth century’s mamlūk successes to the eighth/fourteenth century’s dynastic realities and subsequently again to the ninth/fifteenth century’s regular usurpations by formerly mamlūk strongmen. This occurred as though in a genuine process of a “mamlukization” of the sultanate’s own ʿaṣabiyya, following in the footsteps that Ibn Khaldūn set in the fifth book of his Book of warnings. Leadership genealogies that focused especially on shared Circassian mamlūk origins and on connections to seventh/thirteenth-century mamlūk sultans and their successes now integrated ninth/fifteenth-century historical narratives and competitive claims to royal authority in the same old notion of the “dawla of the Turks” that had once accommodated the contingent successes of Qalāwūnid dynastic leaders. All this makes for a remarkably effective as well as a remarkably ambiguous idea of central leadership in an imagined “two-dimensional world” of rulers and ruled. This idea only really existed in some concrete, and always distinct, way at the specific moment of the performance and simultaneous abstraction, reification, and even “mamlukization” that made this idea appear as though— despite its dynastic, meritocratic, or other specificity—directly connected to all the other moments of its utterance. Only a tiny sample of these many performative instances of discursive specificity as well as connectivity have been examined here. Yielding surprising insights into the creative adaptability of the idea of the “dawla of the Turks,” this chapter’s examination of this sample encourages above all much further critical exploration of whatever remains of the rich variety of these representations of royal specificity and connectivity. Furthermore, these explorations should not only concern the long and rich cultural practices of the sultanate. Just as Professor Kennedy already suggested for the early Islamic Abbasid Empire and its dire need for cultural relevance, in the wider realm of premodern Islamic history, many creative and highly effective specimens of the representation of sultanates, caliphates, emirates, khanates and similar configurations of central leadership also deserve further exploration. However, unlike in the Syro-Egyptian sultanate’s case, most, if not all,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

136

van steenbergen

of these representations were constructed around dynastic claims to generate Ibn Khaldūn’s interdependence of “royal authority” (mulk) and ʿaṣabiyya. The historical specialty of the Syro-Egyptian sultanate’s success to maintain this interdependence should therefore never be underestimated, even when—or rather, exactly because—its leadership organization and representation were far more complex than any traditional notion of the Mamluk “slave state” suggests.

Personal Note I will always be grateful for—and think back with fond memories of—the hospitality, collegiality, and friendship with which Hugh (and of course also Hilary and their children, especially Xana) welcomed me (and my own family) to St Andrews and Cupar almost twenty years ago now, in the course of 2004. To a young and inexperienced lecturer at St Andrews University’s School of History, it made a big difference to receive support, advice, and trust from a senior colleague whom I until then had above all admired for playing in an entirely different league of scholarship. Our many early morning conversations in Hugh’s red Ford Ka during the daily commute between Cupar and St Andrews in particular have left an imprint that continues to determine my own scholarship until today. When we both moved on from St Andrews, our contact became less regular, but it has always been—and will undoubtedly continue to be—a great pleasure to catch up every now and then, especially at the annual Leeds imc. Thank you for everything, Hugh, and take care!

Bibliography Sources Ibn al-Ḥimṣī, Ḥawādith al-zamān wa-wafayāt al-shuyūkh wa-l-aqrān, ed. ʿU.ʿA. Tadmurī, Beirut 1999. Ibn Iyās, Badāʾiʿ al-zuhūr fī waqāʾiʿ al-duhūr, eds. M. Sobernheim, P. Kahle, and M.M. Muṣṭafā, 5 vols., Wiesbaden 1961, 1972–1975; Istanbul 1931–1936. Ibn Khaldūn, Le livre des exemples, 1. Autobiographie. Muqaddima, trans. A. Cheddadi, Paris 2002. Ibn Khaldūn, Tārīkh Ibn Khaldūn al-musammā kitāb al-ʿibar wa-dīwān al-mubtadā wal-khabar fī ayyām al-ʿArab wa-l-ʿAjam wa-l-Barbar wa-man ʿāṣarahum min dhawī l-sulṭān al-akbar, 7 vols., Cairo 1867–1878. Ibn Khaldūn, Muqaddimat Ibn Khaldūn, ed. Kh. Shahada, Beirut 2001.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

137

Ibn Khaldūn, The Muqaddimah: An introduction to history, trans. F. Rosenthal, 3 vols., New York 1958. Ibn Taghrībirdī, al-Nujūm al-zāhira fī mulūk Miṣr wa-l-Qāhira, ed. I. ʿA. Ṭarkhān et al., 16 vols., Cairo 1963–1972. Ibn Ṭūlūn, Mufākahat al-khillān fī ḥawādith al-zamān, ed. Kh. al-Manṣūr, Beirut 1998. al-Maqrīzī, Kitāb al-sulūk li-maʿrifat duwal al-mulūk, eds. M.M. Ziyāda and S.ʿA. ʿĀšūr, 4 vols., Cairo 1934–1958, 1970–1973.

Studies Ayalon, D., Mamlūkiyyāt: (A) A first attempt to evaluate the Mamlūk military system; (B) Ibn Khaldūn’s view of the Mamlūk phenomenon, in Jerusalem studies in Arabic and Islam 2 (1980), 321–349. Bourdieu, P., Rethinking the state: Genesis and structure of the bureaucratic field, trans. L.J.D. Wacquant and S. Farage, in Sociological theory 12 (1994), 1–18. Chamberlain, M., Knowledge and social practice in medieval Damascus, 1190–1350, Cambridge 1994. Chamberlain, M., Military patronage states and the political economy of the frontier, 1000–1250, in Y.M. Choueiri (ed.), A companion to the history of the Middle East, Chichester 2008, 135–53. Cheddadi, A., Le pouvoir, moteur de l’évolution cyclique des sociétés et de la civilisation, in A. Cheddadi, Ibn Khaldûn: L’homme et le théoricien de la civilisation, Paris 2006, 297–362. Dale, S.F., The orange trees of Marrakesh: Ibn Khaldun and the science of man, Cambridge, MA 2015. Flinterman, W., and J. Van Steenbergen, Al-Nasir Muhammad and the formation of the Qalawunid state, in A. Landau (ed.), Pearls on a string: Artists, patrons, and poets at the great Islamic courts, Seattle 2015, 86–113. Fromherz, A.J., Ibn Khaldun: Life and times, Edinburgh 2010. Fuess, A., Mamluk politics, in St. Conermann (ed.), Ubi sumus? Quo vademus? Mamluk studies—state of the art, Bonn 2013, 95–117. Fukuyama, F., The end of history and the last man, London 1992. Fukuyama, F., The origins of political order: From prehuman times to the French revolution, London 2011. Gardiner, N.D., Al-Malik al-Ẓāhir Barqūq as millennial sovereign?, conference paper, Middle East Studies Association annual meeting, Washington, November 2017. Igarashi, D., The establishment and development of al-dīwān al-mufrad: Its background and implications, in msr 10 (2006), 117–40. Irwin, R., Ibn Khaldun: An intellectual biography, Princeton, NJ 2018. Kennedy, H., The decline and fall of the first Muslim empire, in Der Islam 81 (2004), 4–30.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

138

van steenbergen

Khalidi, T., Arabic historical thought in the classical period, Cambridge 1996. Lapidus, I.M., Muslim cities in the later Middle Ages, Cambridge, MA 1967. Lewis, B., Islam from the prophet Muhammad to the capture of Constantinople, 2 vols., New York 1987. Loiseau, J., Reconstruire la maison du sultan. 1350–1450. Ruine et recomposition de l’ordre urbain au Caire, 2 vols., Cairo 2010. Loiseau, J., Les Mamelouks, xiiie–xvie siècle. Une expérience de pouvoir dans l’Islam médiéval, Paris 2014. Martinez-Gros, G., Ibn Khaldûn et les sept vies de l’Islam, Paris 2006. Mauder, C., Childless rule and the sultan’s son: Muḥammad b. al-Ghawrī and the Mamluk system of succession in early 16th-century Egypt, in T. Trausch (ed.), Norm, Normabweichung und Praxis des Herrschaftsübergangs in transkultureller Perspective, Bonn 2019, 161–185. Mitchell, T., The limits of the state: Beyond statist approaches and their critics, The American political science review 85 (1991), 77–96. Mitchell, T., Society, economy, and the state effect, in G. Steinmetz (ed.), State/culture: State-formation after the cultural turn, Ithaca 1999, 76–97. Northrup, L.S., From slave to sultan: The career of al-Manṣūr Qalāwūn and the consolidation of Mamluk rule in Egypt and Syria (678–689a.h./1279–1290a.d.), Stuttgart 1998. Onimus, C., Les maîtres du jeu: Pouvoir et violence politique à l’aube du sultanat mamlouk Circassien (784–815/1382–1412), Paris 2019. Paul, J., Lokale und imperiale Herrschaft im Iran des 12. Jahrhunderts: Herrschaftspraxis und Konzepte, Wiesbaden 2016. Paul, J., Violence and state-building in the Islamic east, online working paper no. 4, Graduate School Asia and Africa in world reference systems, Martin Luther University, Halle-Wittenberg 2006. Piterberg, G., An Ottoman tragedy: History and historiography at play, Berkeley 2003. Sievert, H., Der Kampf um die Macht im Mamlūkenreich des 15. Jahrhunderts, in St. Conermann and A. Pistor-Hatam (eds.), Die Mamlūken: Studien zur ihrer Geschichte und Kultur. Zum Gedenken an Ulrich Haarmann (1942–1999), Hamburg 2003, 335– 366. Stewart, A.D., Between Baybars and Qalāwūn: Under-age rulers and succession in the early Mamlūk Sultanate, in Al-Masāq 19 (2007), 47–54. Strayer, J., On the medieval origins of the modern state, Princeton 1970. Van Steenbergen, J., Qalāwūnid discourse, elite communication and the Mamluk cultural matrix. Interpreting a 14th-century panegyric, Journal of Arabic literature 43 (2012), 1–28. Van Steenbergen, J., The Mamluk sultanate as a military patronage state: Household politics and the case of the Qalāwūnid bayt (1279–1382), in Journal of the economic and social history of the Orient 56 (2013), 189–217.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

rethinking “the mamlūk state” with ibn khaldūn

139

Van Steenbergen, J., Appearances of dawla and political order in late medieval SyroEgypt: The state, social theory, and the political history of the Cairo Sultanate (thirteenth–sixteenth centuries), in St. Conermann (ed.), History and society during the Mamluk period (1250–1517). Studies of the Annemarie Schimmel institute for advanced study ii, Bonn 2016, 51–85. Van Steenbergen, J., “Nomen est omen”. David Ayalon, the Mamluk sultanate, and the rule of the Turks, in A. Levanoni (ed.), Egypt and Syria under Mamluk rule: Political, social and cultural aspects, Leiden 2021, 119–137. Woods, J.E., The Aqquyunlu: Clan, confederation, empire, Salt Lake City 1976, 21999 (rev. and exp. ed.). Yosef, K., Dawlat al-Atrāk or Dawlat al-Mamālīk? Ethnic origin or slave origin as the defining characteristic of the ruling elite in the mamluk sultanate, in Jerusalem studies in Arabic and Islam 39 (2012), 387–410. Yosef, K., The term mamlūk and slave status during the Mamluk Sultanate, in AlQanṭara 34 (2013), 7–34.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 7

Ibn Khaldūn and the Ḥafṣid Caliphate Allen Fromherz

Why the dismissal of the Ḥafṣid Caliphate? In the classic telling, the caliphate moves from Medina, to Damascus, to Baghdad, to Cairo, and finally to Istanbul. Tunis, which is far to the west of these Middle Eastern capitals, is often left out of the caliphal succession. The famed historian and polymath, Ibn Khaldūn, however, thought otherwise. This chapter analyzes how and why Ibn Khaldūn (d. 808/1406) described the founding of the Ḥafṣid Caliphate in Tunis in the middle of the seventh/thirteenth century. Instead of claiming that the caliphate had transferred from Baghdad to Cairo, he told an alternate history, one in which the Ḥafṣid Caliphate was the legitimate caliphal dynasty recognized by the Sharīfs of Mecca. Ibn Khaldūn wrote more than a century after the founding of the Ḥafṣid Caliphate by Muḥammad al-Mustanṣir in 651/1253. A Ḥafṣid caliph was his patron. This Ḥafṣid ruler, Abū l-ʿAbbas al-Aḥmad ii (d. 796/1394), wished to restore the power of the Ḥafṣids, even to reunite the old Almohad possessions across North Africa. He regularly went to war with dynasties to the west and had ambitious plans, although ones that ultimately failed to bear fruit. His court, according to Ibn Khaldūn’s autobiography, was full of the typical jealous and ambitious ministers and officials trying to gain favor with the ruler. Some of these rivals threatened Ibn Khaldūn’s life and questioned his loyalty to the caliph. With such a patron, and in such a court, there were many political reasons for Ibn Khaldūn to portray the history of the founding of the Ḥafṣid caliph in a particularly positive way. He used history as panegyric, praising the caliph’s ancestor and legitimizing the idea of non-Qurashī caliphs, even as he tried to subtly influence his own caliph and patron.

1

The Ḥafṣid Caliphate in Historical Context

Ibn Khaldūn called the founding of the Ḥafṣid Caliphate, the proclamation of Muḥammad al-Mustanṣir by the “Sharīfs of Mecca” as caliph, the greatest day in the history of the city of Tunis. In the centuries since, however, that proclamation of Muḥammad al-Mustanṣir as the caliph, along with much of the history of the Ḥafṣid Caliphate, has been mostly forgotten by historians, sidelined out

© Allen Fromherz, 2022 | doi:10.1163/9789004525245_008

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

141

of mainstream Islamic history, especially by non-Maghribi writers. The Ḥafṣids, unlike the Umayyads, Abbasids, or even the Ottoman sultans who later claim the title discretely, are often relegated to lists of pretender caliphs. Despite Ibn Khaldūn’s colorful story of how the proclamation of the Ḥafṣid Caliphate came about, and the timing of the declaration in the 1250s ce, the same decade the Abbasids of Baghdad were about to be crushed by the Mongols, the Ḥafṣid Caliphate is regarded by many Muslim and non-Muslim historians alike as an insignificant footnote in Islamic history. The true caliphate, in the standard, widely accepted version of caliphal succession, made its way from Baghdad to Cairo to Istanbul, expiring with the last of the Ottoman sultans. Historians, both Muslim and non-Muslim, mostly adhere to this version of the history of the caliphate, largely agreeing that the spiritual leadership of Islam was transferred not to Tunis but to Cairo and then onward to Istanbul. When the Mongols killed the last Abbasid ruler of Baghdad in this version of caliphal history, one of the Abbasid family escaped to Egypt under Mamluk protection. The first Ḥafṣid caliph in Tunis, al-Mustanṣir, therefore, is recognized even less than the last of the Abbasid caliphs of Baghdad, al-Mustaʿṣim. During the first centuries of the Abbasid Caliphate, the Islamic empire was at its height. After these celebrated Abbasids, however, the caliphate entered a period of decline around the end of the fourth/tenth century. The Sunni Arab caliphs were then reduced to mere puppets of the Buyid sultans: Shiʿi Daylamites from what is now Iran. As Hugh Kennedy remarks, this plunged the Abbasid Caliphate into “two generations of obscurity and impotence in which the ancient office seemed to many an irrelevant relic.”1 Later, as if in a last vigorous rally before death, the Abbasid Caliphate returned to what Kennedy calls a “silver age” during long the reign of Ahmad al-Naṣir. The Fatimids no longer could challenge their claim to the caliphate. Even in the Maghrib, the Almohad caliphs, whom the traveler Ibn Jubayr had once imagined capable of extending their power to Egypt and beyond, had lost authority with their defeat at Las Navas de Tolosa (609/1212) by a combined Christian army in Muslim Spain. Nonetheless, this hopeful silver age of the Abbasids would not last for long. In Muḥarram 656/January 1258 the Mongol conqueror Hülegü killed the last Abbasid caliph of Baghdad. The fall of the Baghdad caliph, however, reverberated as far west in Christendom as France. Jean de Joinville told a story, later taken up by Marco Polo, that the caliph died after Hülegü, the Mongol conqueror of Baghdad, forced the caliph to eat his own treasure. Whatever his personal qualities, al-Mustaʿṣim’s main legacy seems to be how he died, not

1 Kennedy, Caliphate 129.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

142

fromherz

what he did when alive. In various versions he was either drowned, strangled, kicked to death, rolled up in a carpet, or, in the Christian version, suffocated under a mountain of gold. Generations after the fact, the Tunis-born historian Ibn Khaldūn claimed the ignoble end of the Abbasids was, in fact, a predictable outcome. It was another example of his cycle of history and the cost of luxury and decadence. In fact, a competent scholar could see it coming. The third/ninth-century polymath Abu Yūsuf al-Kindī, Ibn Khaldūn said, had predicted the precise end of the Abbasids centuries before. Of course, that book, like all the best books of prediction, was lost. Ibn Khaldūn had many political reasons to describe the end of the Abbasids as foretold by fate, as a warning for future caliphs, such as his patron in the fourteenth century, the Tunis-based Ḥafṣid Caliph Abū l-ʿAbbas. For a major part of his career as a minister, functionary, and scholar, Ibn Khaldūn worked for the Ḥafṣid caliph, thousands of miles away from Baghdad and hundreds of miles from Cairo. This Ḥafṣid Caliphate, centered in Tunis, a thriving Mediterranean port, had existed for over a century. Ibn Khaldūn dedicated his universal history, one of the most important sources for the history of the Berbers and North Africa, to the Ḥafṣid caliph (r. 772–796/1370– 1394) and may have had some hope in him as a new unifier of the Islamic west.2 The caliph, Abū l-ʿAbbas, traced his legitimacy back Muḥammad alMustanṣir. This first caliph of the Ḥafṣids was, according to Ibn Khaldūn, confirmed by the “Sharīfs of Mecca” around 650/1253, even as the Abbasid caliph still lived. Yet the Ḥafṣid Caliphate was not based on some sort of North African or Maghribi exceptionalism. The Ḥafṣids were a successor dynasty to the Almohads, who independently declared their own caliphate. Perhaps surprisingly, however, the Ḥafṣid Caliphate was not based on the Almohad caliphal precedent. Instead, the proclamation of came from the Sharīfs of Mecca, not from an Almohad source. True, the Ḥafṣids also saw themselves as successors of the Almohad caliphs, who followed the mahdī, Ibn Tūmart, the revolutionary Berber preacher who brought his doctrine to the mountain Berbers of the Atlas in the early sixth/twelfth century. However, Muḥammad al-Mustanṣir’s caliphate was legitimated in the east, in Mecca. This was a change from the previous Almohad practice of assuming the caliphate based on the new Almohad doctrine, the new dispensation that had started with the mahdī of the Almohads, Ibn Tūmart. Muḥammad al-Mustanṣir’s powerful father, Abū Zakarīyāʾ,

2 For the life of Ibn Khaldūn, see Ibn Khaldūn, Le livre des exemples, especially the Autobiographie. See also, Fromherz, Ibn Khaldūn.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

143

had broken Tunis away from the later Almohad caliphs in Marrakech due to their betrayal of Almohad doctrine, their embrace of ʿIsā, or Jesus, as the mahdī over Ibn Tūmart. But Abū Zakarīyāʾ did not proclaim a new caliphate, even on Almohad doctrinal grounds. Instead, his son, Muḥammad al-Mustanṣir, facing several rebellions from the so-called “Almohad Shaykhs,” Berber Almohad notables in Tunis, loyal to his father but not to him, reversed course and accepted a caliphal proclamation coming from the east, from Mecca, not from the Almohads. The specifics of Muḥammad al-Mustanṣir’s rise to power and the attempts by the Almohads to rebel against his authority may partially explain this shift from an Almohad-endorsed caliphate to a Meccan-endorsed caliphate. Early in his reign, Muḥammad al-Mustanṣir had sought legitimation from two groups opposed to the Almohads: Andalusi exiles fleeing the Iberian lands conquered by the Christians and the Sufis, leaders of religious brotherhoods that were growing rapidly in popularity at the time. Unlike his father, he was far from a good Almohad. He conflicted regularly with the Hintāta and other Berber/ Almohad “shakyhs” who were attempting to maintain a semblance of Almohad doctrine in Tunis. The lingering Almohads in his court were a problem; expecting his older brother to take over instead, these Almohad Berbers rebelled. They forcefully rejected Muḥammad al-Mustanṣir’s rule. Muḥammad al-Mustanṣir felt compelled to purge the recalcitrant Almohads. In their place, he first secured the following of Andalusi exiles as well as some local Arab and Berber tribes. Finally, he sought the support of the Sufis. In Ibn Khaldūn’s version of events, it was one of these Sufis, an elusive figure called Ibn Sabʿīn (d. 669/1271), who composed the letter from the Sharīfs of Mecca, the letter declaring Muḥammad al-Mustanṣir as the legitimate caliph of the Islamic world. Like many a papal bull in medieval Europe, the legitimacy of the letter, which Ibn Khaldūn called a daʿwa (or “declaration”), written in Mecca and sent to Tunis, is hotly disputed. Even today, among Tunisia’s medievalists there is an active and lively debate about Mamluk versus Ḥafṣid claims to the caliphate. A conference I hosted in Tunis in 2018 on Caliph al-Mustanṣir’s claims led to an active back-and-forth on Mamluk versus Maghribi sources. The Encyclopedia of Islam article for early Islamic and medieval Tunisia, written by the distinguished scholars Mounira Chapoutot-Remadi and Radhi Daghfous, seemed to dismiss Muḥammad al-Mustanṣir’s claim to the caliphate, considering, instead, the Mamluk’s counter claims: “Abū Zakarīyāʾ [the father and Ḥafṣid founder] simply styled himself amir [prince], but his son […] in 1253 assumed the caliphal laqab of al-Mustanṣir. This line of development was, however, stifled by the rise and success of the Mamluks in Egypt, with the installation of an

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

144

fromherz

Abbasid caliph in Cairo.”3 Other Tunisian scholars, such as Brahim Jadla and Salah Baïzig, however, largely take Ibn Khaldūn at his word. They consider the Ḥafṣid Caliphate as validated by Mecca, believed and supported as a legitimate claim to authority over the Islamic umma. Afterall, Tunis was one of the most vibrant and important capitals in the Islamic world. For these Tunisian historians, the Ḥafṣid Caliphate was legitimate, or at least was believed to be at the time. While sources from the actual decade in which the caliphate was proclaimed have issues, and the Mamluk archives (which have their own bias, of course) give scholars pause, the proclamation of Muḥammad al-Mustanṣir as caliph was included by Ibn Khaldūn and was mentioned by other chronologies as well. While written a century or so after the proclamation, Ibn Khaldūn could trace the role of his own ancestors, including a great grandfather who witnessed the public celebration. Even so, the historical legitimacy of Muḥammad al-Mustanṣir’s caliphal claims remain a matter of persistent debate.4 Regardless of whether the caliphal letter from Mecca was forged, it did have a major impact on thirteenth-century Tunis. I would argue it revealed a shift in power dynamics, away from the rural Berber and Arab tribal coalitions that secured Almohad power and toward the new growing power of the urban port of Tunis and its Sufi brotherhoods, which often corresponded with merchant, craftsmen, and trading classes. By accepting the letter from Ibn Sabʿīn, who had only recently been exiled out of Tunis by his own top theologians, al-Mustanṣir favored the Sufis, even legitimating the writings of one from an eccentric or “exaggerated” school, Ibn Sabʿīn. This was a major shift, part of a religioussocial-economic transformation that was overtaking the entire Mediterranean world, the twelfth-century renaissance; the rise in mendicant orders and their powerful influence on rulers was not a European story alone. Al-Mustanṣir’s embrace of the Sufis was strategic. They were popular, much more so than the Andalusi elite. They helped gain the support of the merchant classes and the urban populus, and they proved very important in his defense of Tunis against the French. The caliph’s alliance with the Sufis of Tunis can be compared to King Louis ix’s own embrace of the Franciscans.5 The sixth/twelfth century was an era of new religiosity, a second axial age throughout the western Medi-

3 Chapoutot-Remadi and Daghfous, Tunisia. 4 The letter and the legitimacy of the Ḥafṣid Caliphate’s proclamation was debated vigorously between participants at the conference on the Ḥafṣid Caliphate that convened in Tunis in 2018, “Early Ḥafṣid Tunis: Capital of the caliphate.” The main disagreement seemed to center around on Mamluk sources, which are themselves problematic. See https://aimsnorthafrica​ .org/2017‑aims‑symposium/. 5 Fromherz, Saint and the caliph.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

145

terranean. Rulers were quick to appropriate this new religiosity for political purposes and to burnish their own popularity.6 This was the case when alMustanṣir had the letter from the Sufi Sabʿīn proclaiming him caliph read in front of a large crowd in Tunis in 651/1253.

2

Al-Mustanṣir Proclaimed Caliph

Ibn Khaldūn tells us of the beautiful and festive day in Tunis when al-Mustanṣir, the Ḥafṣid ruler, having suppressed an uprising of Almohad Berbers, as well as having put down the rumblings of his Andalusi exile community, was proclaimed caliph. He was confident of his power. After all, he had several important victories under his belt and filled the coffers of Tunis with great booty after suppressing the rebellion of the Arab Sulaym tribe. He had also spread his rule over the city of Constantine (present-day Algeria) and may have thought there was nothing stopping him from going further, maybe even restoring the old Almohad Empire, or eventually going east to challenge the Mamluks. Proclaimed caliph, there were messages of congratulations, or oaths of allegiance, from cities as far away as Fez, Ceuta, and Valencia. Many in al-Andalus, besieged by Christian raiders, hoped this new caliph, recognized as such in a letter from the Sharīfs of Mecca, would push back the Castilians and the Aragonese, restoring the glories of al-Andalus. The governor of Mecca was named Abū Nāmī. It was he who had recognized the caliphate of Muḥammad al-Mustanṣir. Nonetheless, as Ibn Khaldūn mentions, there were still plenty of disputes in the holiest city of Islam over who was really in charge, who held authority in the sacred mosque. Abū Nāmī came from the most noble family in the world, those of the sharifian descendants […] [was named] Abū Nāmī […] After the era of Saladin […] the authority of the Abbasid caliphs had been restored in Egypt, Syria, and the Ḥijāz, the Sharīfs of Mecca had continued to recognize the legitimacy of [the Abbasids]. The right to command the pilgrims and to administer the city [of Mecca] remained, however, in the hands (of Saladin) who passed it on to his descendants, from whom he passed to their freedmen, as is still seen today. There soon arose fierce disputes between these freedmen and the Sharīfs, and the struggle was still going on when the Tartars came to overthrow the caliphate of Baghdad

6 Fromherz, Near West.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

146

fromherz

and the Ḥafṣid dynasty rose to Africa, strongly supported by the peoples [of the Maghrib].7 The most crucial character in the story was Ibn Sabʿīn, who gained the friendship of the Sharīf of Mecca. Ibn Khaldūn’s description of this mercurial man was not completely favorable. At that time there was living in Mecca a Sufi called Abū Muḥammad b. Sabʿīn. This individual, having left Murcia, his native city, had first gone to Tunis, and, as he was deeply versed in the knowledge of the law and the intellectual sciences, he had displayed the claim of having mastered walking the straight path of Sufism. He even professed some of the extravagant doctrines learned in this school, and he openly taught that nothing exists except God, a principle we discussed in our chapter on exaggerated Sufism.8 Ibn Khaldūn critiqued such “Sufis” who “exaggerated,” but this did not mean he condemned Sufism as a whole. In fact, he desired to practice some form of “true” Sufism himself—one not manipulated by those he considered charlatans. He regularly referred to his father in his autobiography, a man who, unlike Ibn Khaldūn, managed to avoid the trappings and temptations of power and position to pursue a Sufi path. The relationship between Ibn Khaldūn and the Sufis and Sufism has been misinterpreted. Although Ibn Khaldūn points to excesses by Sufis who took advantage of their followers, he was far from being a rabid, “anti-Sufi” qadi. Ibn Khaldūn was profoundly shaped by his own Sufi tendencies.9 Part of the problem, of course, was that there were almost as many interpretations of what was “true” or “false” Sufism as there were Sufis. In any case, Ibn Sabʿīn was exiled from Tunis by the Andalusi scholar Abū Bakr alSakūni, the head of the qadis of Seville, who accused Ibn Sabʿīn of heresy. He left Tunis, fearing for his life, ending up in Mecca. It was before Muḥammad alMustanṣir had established his power over Andalusi exiles, forming an alliance with the Sufis and various other constituencies against the Andalusi clerics and the Almohad Shaykhs. Ibn Sabʿīn was condemned as a criminal, a heretic on the run. “In Mecca, however, the Sufi Ibn Sabʿīn was, ‘taking refuge there, in the inviolable asylum of the mosque, he befriended the Sharīf [Abū Nāmī], lord of the city, and convinced him to join him in recognizing the sovereignty 7 Ibn Khaldūn, Kitāb al-ʿIbār vi, 407. Translations are my own. 8 Ibn Khaldūn, Kitāb al-ʿIbār vi, 407. 9 Ibn Khaldūn, Ibn Khaldūn on Sufism.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

147

of al-Mustanṣir, the sultan of Ifrīqiya. Wishing to gain the favor of this monarch and to find a way to come back [to Tunis], he wrote an epistle in his own hand a letter describing the acceptance [of al-Mustanṣir as caliph].’” The letter could not have arrived at a better time. Having consolidated his power, having increased his popularlity among the Sufis, having little reason to listen to the urgent demands of the Andalusi exiles, and, finally, having broken away from the Almohads, this letter now allowed al-Mustanṣir a new dispensation. This letter from an exiled eccentric, who had only recently been called a heretic, solidified his legitimacy and created a new caliph, endorsed by the Sharīf of Mecca himself. “When the sultan received this letter, he summoned all the dignitaries of the empire, as well as the people, in order that it might be read to them.” The elite scholars of Islam quickly changed their tune on Ibn Sabʿīn. “[After the letter was read] […] the imam of the palace then took the floor. Following a long speech on the admirable style of the letter, he pointed out the excellent impact it was to have on the world, by making known the brilliance and glory of the ruler, as witnessed by the eagerness with which the inhabitants of the Holy City recognized his authority.”10 What a change in opinion there seemed to be about the writings of a man once condemned as a heretic. It was also interesting that a palace loyalist, not al-Sakūni, the Andalusi traditionalist from Seville who had earlier condemned Ibn Sabʿīn and forced him into exile, was chosen to read the letter. The caliphate was a coup, not just for al-Muṣtansir but also for the Sufis. The actual contents of this letter are buried in the Kitāb al-ʿIbar of Ibn Khaldūn, largely neglected by nineteenth-century orientalists. It has very few references to the Ḥafṣid ruler himself, al-Mustanṣir. Instead, it appeared to be an opportunity for Ibn Sabʿīn to get back at his former persecutors. He proclaims a vision of the nature of authority. There are Quranic allusions and even a philosophy of the unity of existence. Why would the enigmatic Sufi write this letter in support of a ruler who signed off on his exile? Ibn Sabʿīn’s letter focused on the intimate link between new seventh/thirteenth-century Sufi ideas and notions of power, authority, and legitimacy in Arabia and North Africa.11 The Sufi Ibn Sabʿīn, in addition to making a political move convenient for Muḥammad al-Mustanṣir, wished to create a political

10 11

For these passages, see Ibn Khaldūn, Kitāb al-ʿIbār vi, 407. For the translation to French, see Histoire des Berbères ii, 111–112. The full text of the proclamation, which is absent from several other editions and is declared to be unworthy of translation by De Slane, can be found in Ibn Khaldūn, Kitāb al ʿIbār x, 407–417.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

148

fromherz

equivalent of his vision of creation and the nature of God.12 His theology of the all-encompassing circle reflected the all-encompassing form of political rule embodied by the caliphate. Also, an alliance between the caliph and various Sufis could free the ruler from dependence on the Andalusi functionaries, such as Ibn al-Abbār, who often had an ambiguous loyalty to the caliph. Ibn Khadūn used the pilgrimage to Mecca as an excuse to escape the court of his Ḥafṣid patron Abū l-ʿAbbas. I wonder if he had Ibn Sabʿīn in mind. It is unclear to what extent Muḥammad al-Mustanṣir actually approved of the exile of Ibn Sabʿīn, or to what extent he authorized his forced pilgrimage to Mecca to protect him from enemies at home. Ibn Sabʿīn, however, died in Mecca and would never return to Tunis. Ḥafṣid claims to the caliphate, made possible by his letter, would continue for centuries, despite the rise of rivals in Egypt and Yemen. At least according to the way Ibn Khaldūn described the caliphate, therefore, it was not a political design that led to al-Mustanṣir’s declaration. Rather, it was the result of one of those rare moments in history that, perhaps, cause the most change over time, the confluence of a rarified ideal (in the form of Ibn Sabʿīn’s notion of the caliphate) and an expedient reality, in the form of alMustanṣir’s need to establish his authority. The contents of the caliphal letter from Ibn Sabʿīn found in Ibn Khaldūn’s narrative deserves much more scrutiny. What also needs clarification, however, is the role of the historian, minister, polymath, and political, religious operative, Ibn Khaldūn. As one of the main sources on the history of the Ḥafṣids, and as a minister for the Ḥafṣid Caliph Abū l-ʿAbbas, who traced his own legitimacy to al-Mustanṣir, Ibn Khaldūn was not without his own motivations, his own reasons to omit or include particular details about the Ḥafṣids. Despite gaining legitimacy from Mecca, Muḥammad al-Mustanṣir faced some new, internal challenges to this authority. The Andalusi exiles, for example, remained restless. They seemed to take his role as caliph seriously. They wanted this caliph to do his caliphal duty, to engage in holy war against the kings of Aragon or Castile. At the same time, the caliph had to balance their demands with the constant rebellions of the Arab bedouin, the dissatisfaction of the Almohad Berbers, the former elite, and even the rise of the Sufis.13 His most ardent critics, however, were the Andalusi exiles. Ibn al-Abbār, the great Andalusi poet, made the mistake of questioning the caliphate of Muḥammad al-Mustanṣir, saying, in biting Arabic verse:

12 13

Cornell, All-comprehensive circle. Salah Baïzig provides a short overview of the proclamation and its political context in his recent article, ʿUlama and the Ḥafṣid caliphate.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

149

ibn khaldūn and the ḥafṣid caliphate

In Tunis there is a foolish tyrant, Who calls himself Caliph.14 The caliph was “so enraged upon hearing this ditty, he ordered the author to be tortured and finished off with a lance […] This corpse of this unfortunate man, his books, his poems and collections […] were burned together in the same fire.”15 Of course, Ibn Khaldūn was familiar with the killing, and even burning, of great poets and scholars. His friend and mentor Ibn al-Khaṭīb was strangled to death in Fez in 776/1374, his body exhumed and burned. Perhaps Ibn Khaldūn was worried for his own skin. He had every reason to placate the Ḥafṣid caliphs with a history that affirmed their claims. Ibn Khaldūn may have avoided certain details about the legitimacy of the Ḥafṣid Caliphate even as he still balanced on a pin in his sympathy for Ibn al-Abbār. Later, Ibn Khaldūn fled Tunis and worked for the Mamluks, who had their own caliphs who were supposed to be legitimate Abbasid successors; one wonders what they might have thought of his legitimation of the Ḥafṣid Caliphate. Obviously, Ibn Khaldūn, despite all the praise of his writing by later historians, cannot be taken only at face value, even when he appears to be telling just a chronological narrative. The complicated relationship between Ibn Khaldūn’s own political calculations and his writings deserve further exploration.

3

Ibn Khaldūn’s Motivations

The late historian Barbara Stowasser, in a thought-provoking article, compared Ibn Khaldūn’s work to that of the Florentine Machiavelli.16 Although they probably knew nothing of one another—time and distance separated them—there are some interesting similarities between the Florentine and the Tunisian, especially the notion of a cycle in the regular rise and fall of dynasties. They both used history as a type of “training manual” for rulers. For Machiavelli, it was the multivolume History of Florence, from the “earliest times” to the death of Lorenzo the Magnificent, that held many hidden lessons about the realities of power, revealed in The prince.17 For Ibn Khaldūn it was his “Universal History,” the Kitāb al-ʿIbar, focused mainly on the story of North Africa, that provided the recipe book for his Muqaddima, which discussed the ingredients 14 15 16 17

Ibn Khaldūn, Kitāb al ʿIbār vi, 419–420. Histoire des Berbères x, 116. Ibn Khaldūn, Kitāb al ʿIbār vi, 419–420. Histoire des Berbères x, 116. Stowasser, Religion and political development. Machiavelli, Machiavelli.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

150

fromherz

of history. The similarities between the two can be explained by the fact that both Machiavelli and Ibn Khaldūn lived in highly turbulent, fractious political environments, mostly dominated by rulers with failed ambitions to restore the Almohad Empire. It is interesting to note Machiavelli’s failure to convince his patrons of the value of his writings, where he attempted to unify the many competing states of Italy against a host of external foes, such as the French and the Holy Roman Emperor. A century before Machiavelli, the Ḥafṣid Caliph Abū lʿAbbas of Tunis was Ibn Khaldūn’s Lorenzo de’ Medici (d. 1492 ce) or Caesar Borgia of Romagna (d. 1507ce). One of Ibn Khaldūn’s overriding hopes was to unify the Maghrib once again, to take on the Christian threat in al-Andalus, going from court to court to find the right leader. He tried politics as a means to that end, only to find himself exiled or thrown in prison, much the same fate as Machiavelli. Was Ibn Khaldūn’s writing really just about providing a critical, well-researched history that does not repeat mindless, ridiculous myths, his claim in his introduction, or is it also a deeper political project, an attempt to reshape and reform a troubled Maghribi and Islamic world under siege? Perhaps he wished to return the vigor, ʿaṣabiyya, and ultimately the legitimacy and unity to the dispirited, divided, and infighting rulers of the medieval Maghrib and Granada. When writing about historical eras and examples distant from his own times, such as the Umayyads and Abbasids, Ibn Khaldūn seemed free to openly critique, categorize, or analyze the actions of past rulers and dynasties. As his history nears his own present, especially the origins of the Marinids, Naṣrids, and Ḥafṣids, by whom Ibn Khaldūn was regularly employed as a minister, a new tone emerges, one that becomes increasingly aware that “audience” mattered. Ibn Khaldūn ended his work for Abū l-ʿAbbas of the Ḥafṣids after being ousted from favor by a jealous, Svengali-like, fellow minister. Abū l-ʿAbbas was the great hope of the Ḥafṣids, and perhaps even for Ibn Khaldūn, but, in the end, he, too, was a disappointment. Nonetheless, the social context of his writings, the relationship between Ibn Khaldūn and his patrons, has yet to be fully explained. Imagine Abū l-ʿAbbas, Ibn Khaldūn’s patron in the 1370s, reading, or having read to him, passages from Ibn Khaldūn’s history. It was the story of his own family and the founding of his ancestral dynasty and caliphate and the first version of Ibn Khaldūn’s book, dedicated to the caliph.18 He might have read how Ibn Khaldūn deftly portrayed the caliphate of al-Mustanṣir as a success, despite the divisions he faced, and also inserted the caliphal proclamation of Ibn

18

Fromherz, Ibn Khaldūn.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

151

Sabʿīn with little explanation of its actual contents. Abū al-Abbas might have also read Ibn Khaldūn’s other writings on the qualifications of the caliph: “1) knowledge, 2) probity, 3) competence, 4) freedom of the senses and limbs from any defect […] there is a difference of opinion concerning a fifth prerequisite, that is, 5) Qurashī descent.” The ambiguity of the last condition may seem to be Ibn Khaldūn’s concession to Abū al-ʿAbbas. The Ḥafṣids, like the Almohad caliphs before them, claimed a sharifian ancestry along with a distinguished line of Berber, Amghar (Berber chief) blood. It is uncertain to what extent these claims to Qurashī descent among the Almohads, let alone the Ḥafṣids, were actually widely believed, especially in the more rarified circles of Kairouan and Tunis. Some scholars, such as Abū Bakr al-Bāqillānī (d. 403/1013), required Qurashī origin. Ibn Khaldūn seemed to be saying Qurashī descent is a problem, however, because it often involves “dropping the pre-requisite of competence.” This lack of competence was seen in the Abbasid Caliphate, based on a Qurashī descent, that had fallen into luxury, disarray, and, eventually, dissolution. Besides, the ʿaṣabiyya of the Qurashīs “had come to disappear and dissolve,” and “non-Arab rulers controlled the caliphs.” Was Ibn Khaldūn signaling that Berber blood was no disqualification for the caliphate? Ibn Khaldūn links the third condition of competence with warfare against nonbelievers. One can think, again, of the relative success of al-Mustanṣir against the Christians in the seventh/thirteenth century and the battles of Abū l-ʿAbbas against the Aragonese and the French in the eighth/fourteenth century. Moreover, “when one considers what God meant the caliphate to be, nothing more needs (to be said) about it. God made the caliph His substitute to handle the affairs of His servants. He is to make them do the things that are good for them and forbid them to do those that are harmful. He has been directly told so. A person who lacks the power to do a thing is never told directly to do it.” He then makes a realist argument for the caliphate, one that emerges not out of religious texts but out of the world, “existence attests to (the necessity of group feeling for the caliphate). Only he who has gained superiority over a nation, or a race is able to handle its affairs. The religious law would hardly ever make a requirement in contradiction to the requirements of existence.”19 What mattered to Ibn Khaldūn was not the conclusions of religious scholars like al-Bāqillānī. Rather, what mattered was the extent to which a leader might be able to unite the heterogeneous groups under his power toward a unified, singular body and project that unity in the form of competence in jihād.

19

The passages cited in the paragraph above are from Ibn Khaldūn, Muqaddimah 158–160.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

152

fromherz

Decades later, Ibn Khaldūn seemed willing to discuss the possibilities of power and Muslim unity with the would-be Muslim world conqueror Timurlane. Although some have seen the meeting between the historian and the world conqueror as a heroic attempt by the Maghribi to dissuade or deceive the invader, it was also a chance for Ibn Khaldūn to meet a new potential unifier of the Islamic world. Timurlane was no Qurashī, but he was, before his defeat by the Mamluks, competent at winning. In the end, Timurlane failed to take Egypt, the doorway to the Maghrib and the hope of a restored al-Andalus. Ibn Khaldūn’s dream of a newly unified Islamic empire that could push against the threat of Christian expansion would only be fulfilled after his death (808/1406) and the restoration of the Ottomans under Mehmed i in 816/1413. Every time he thought he had found the right ruler, the true unifier of the Muslim world, Ibn Khaldūn’s hopes were crushed. He ended up in Egypt, in and out of office as a Mālikī judge, constantly defending himself against political intrigue born out of hopeless disunity and dissolution. Midway upon the journey of his life, Ibn Khaldūn’s patron, the Ḥafṣid Caliph Abū l-ʿAbbas, was a profound disappointment. He did not achieve the glories of earlier Almohad caliphs, let alone the Abbasid caliphs of old. Despite their own late “silver age” in the eighth/fourteenth century, the Ḥafṣid rulers remained in control of a statelet, not an empire like that of the Almohad or Abbasid caliphs. The Andalusi dream of restoration, taking back cities lost to the Christians, a dream Ibn Khaldūn shared with his late mentor Ibn al-Khaṭīb of Granada, seemed further and further from reality. In the last decades of his life, the Mamluks, too, despite their own claims to the caliphate, were a great disappointment for Ibn Khaldūn. Even Timurlane could not unify the Islamic world. Instead, by 808/1406, the last year of Ibn Khaldūn’s life, the hope of a unified caliphate seemed dashed.

Personal Note I was Professor Kennedy’s PhD student at the University of St Andrews from 2002–2006. During my first formal meeting with him, we discussed my topic, the rise of the Almohad Empire, and the importance of obtaining a long list of relevant primary sources. When we were done, he stood up in his yellow babouches, went to his crowded bookshelf, and handed over a mountain of material containing a great deal what I would need for my dissertation. I am ever grateful to Hugh for his generous spirit and dedication to writing history for history’s sake.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

ibn khaldūn and the ḥafṣid caliphate

153

Bibliography Sources Ibn Khaldūn, Histoire des Berbères, trans. W. De Slane, 3 vols., Algiers, 2001. Ibn Khaldūn, Le livre des exemples, ed. and trans. A. Cheddadi, Paris 2002. Ibn Khaldūn, Kitāb al ʿIbār wa diwān al-mubtadaʾ wa-l-khabar fī ayyām al-ʿarab wa-lʿajam wa man ʿāsarahum min thawī al-sultān al-akbar, 7 vols. and index, Beirut 2006. Ibn Khaldūn, Ibn Khaldūn on Sufism: Remedy for the questioner in search of answers (Shifāʾ al-sāʾil li-tahdhīb al-masāʾil), trans. Y. Özer, Cambridge 2017. Ibn Khaldūn, The Muqaddimah, trans. F. Rosenthal, ed. N.J. Dawood, Princeton 2018.

Studies Baïzig, S., The ʿulama and the Ḥafṣid Caliphate, in Journal of North African studies 26 (2021), 625–630. Chapoutot-Remadi, M. and Daghfous, R., Tunisia ii (b), in ei2, x, 644-648. Cornell, V., The all-comprehensive circle: The soul, intellect and the oneness of existence in the doctrine of Ibn Sabʿīn, in A. Shihadeh (ed.), Sufism and Theology, Edinburgh 2007, 31–48. Fromherz, A., Ibn Khaldun, life and times, Edinburgh 2011. Fromherz, A., The Near West: Medieval North Africa, Latin Europe and the Mediterranean in the second axial age, Edinburgh 2016. Fromherz, A., Introduction and summary: The Ḥafṣids and the axial western Mediterranean, in Journal of North African studies 26 (2021), 593–596. Fromherz, A., The saint and the caliph, in Journal of North African studies 26 (2021), 642–653. Kennedy, H., Caliphate: The history of an idea, New York 2016. Machiavelli, N. dei, Machiavelli: The chief works and others, ed. A. Gilbert, Durham, NC 1965. Stowasser, B., Religion and political development, some comparative ideas on Ibn Khaldūn and Machiavelli, Georgetown 1983.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

part 2 Economy and Society

∵

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 8

A Three-Centered System: Aleppo, Damascus, and Cairo in the Age of the Ayyubids R. Stephen Humphreys

The voluminous contributions of Prof. Hugh Kennedy reach across almost the whole Dar al-Islam during its first six centuries.* His work, always characterized by solid erudition, balanced judgment, and lucid prose, has done much to define our current understanding of the political life, military institutions, and urban society of the medieval Islamic world. Within this impressive corpus of scholarship Egypt and greater Syria have always had a significant place. It thus seems appropriate to pursue that dimension of his work by asking how the principal cities of Egypt and Syria anchored and shaped the strange polity constructed by the Ayyubids during their 90-year domination (564–658/1169– 1260) of that region.

1

The Ayyubid Political System

From its beginnings, the Ayyubid regime in Egypt and Syria never constituted a centralized state, let alone an empire, save in a very loose sense. It was constructed as, and down to its end in 658/1260 it remained, a coalition of autonomous principalities, each ruled by a member of the same extended family, the socalled sons of Ayyūb—actually the lineal descendants of two Kurdish soldiers, Ayyūb b. Shādhī (d. 570/1175) and his brother Shīrkūh (d. 564/1169). This coalition could be remarkably cohesive when led by a dominant senior prince or when faced by external threats to the family’s hegemony, but otherwise it was riven by unresolved tensions, which periodically spilled into open conflict. The highly fluid alliances and rivalries that characterized Ayyubid political life were a mix of realpolitik and the tangled relationships within the family.1 * The first version of this paper was presented as a lecture at University College, University of Oxford in the spring of 2015. I wish to express my sincere thanks to Prof. Catherine Holmes for extending this invitation as well as for her generous assistance during my stay in Oxford. 1 The dynamics of Ayyubid political life can be explored in Cahen, Ayyūbids; Humphreys, From Saladin, 15–40 et passim; Eddé, Principauté ayyoubide 33–192. For analyses that differ in significant ways, see Gottschalk, Al-Malik al-Kāmil; Chamberlain, Crusader era.

© R. Stephen Humphreys, 2022 | doi:10.1163/9789004525245_009

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

158

humphreys

Each of the Ayyubid principalities was focused on a single city and its hinterland, and in principle each was hereditary—that is, it was intended to be passed down through the descendants (or sometimes collaterals) of its first ruler. Hereditary succession did in fact hold good, though it was not always smooth, in most of the principalities: Aleppo, Hama, Homs, Cairo, and the extensive territories east of the Euphrates. Damascus, however, traversed a far more troubled path; the city was besieged eleven times between 589/1193 and 658/1260, every time but one by Ayyubid armies. Succession to rule in that city was actively contested throughout the decade following the death of Ṣalāḥ alDīn, and then again in 624–626/1227–1229, 634–635/1237–1238, 636–637/1239, 643/1245, and 648/1250. The struggle for power can sometimes be analyzed as a self-contained system, but the multicenteredness of the Ayyubid system was not merely a political fact; it was a social, economic, and cultural phenomenon as well. Cairo was by far the largest and wealthiest urban complex in the Ayyubid realm, with a population of perhaps as much as 200,000 in this period, but only in Egypt did it play the role of a true primate city. Damascus and Aleppo in particular had long and distinctive histories, for the most part independent of Cairene control, and each had a powerful sense of its own identity. Ayyubid Syria had many medium-size and small cities, and Damascus and Aleppo were the largest, with populations of roughly 50,000 each, and they clearly stood apart as the northern and southern anchors of this cluster. All three cities—Cairo, Damascus, Aleppo—were major centers of artisan manufacturing, regional and longdistance commerce, religious life, and cultural expression. On every level, then, the Ayyubid realm can be understood as a system of three interlocking urban networks. Within this dynamic general system, in turn, each city acted as the autonomous hub of its own regional network.2

2

The Northern Frontier: Aleppo

The workings of the overall system emerges most clearly through a separate examination of its three centers. Among the three cities, Aleppo enjoyed by far the greatest dynastic and political stability. From the moment that Ṣalāḥ alDīn installed his son al-Ẓāhir Ghāzī as nominal prince of the city in 582/1186, the succession passed from father to son without incident until the Mongol

2 Population figures are discussed in Humphreys, Islamic history 247–248; Eddé, Principauté ayyoubide 558–565; Berkey, Culture and society 379–380.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a three-centered system

159

conquest in 658/1260. So solid was the line of succession that for a few years (634–640/1236–1242) Aleppo was governed effectively and with no internal opposition by a regency council headed by Ḍayfa Khātūn, daughter of Sultan al-ʿĀdil (595–615/1198–1218) and the formidable grandmother of a very young prince. A quick look at the map suggests that Aleppo’s interactions with Anatolia, the Mediterranean, and the Jazīra would be at least as important as its links to Damascus, let alone Cairo, and this is in fact the case. A second distinctive feature of Ayyubid Aleppo is that it lay on the edge of a Christian world. Antioch to the west and Cilicia to the northwest were ruled by Christian regimes, Frankish and Franco-Armenian respectively. Within its own borders, moreover, the principality of Aleppo still had a substantial Christian population (both Melkite and Syrian Orthodox), and the same was true of Rum Seljuq Anatolia to its north. In this connection we should recall that Aleppo had become almost a Byzantine client state in the late tenth and early eleventh centuries, and in the early twelfth century it lay within easy reach of Crusader and even Byzantine armies. By Ayyubid times, to be sure, Aleppo no longer had to fear either regional Christian armies or the intrusion of crusading expeditions. After the abortive expedition of Frederick Barbarossa in 1189, no Crusader armies would threaten it again. Even so, it was still closely entangled with its Christian neighbors and subjects. Alone among the three cities, Aleppo needed to confront a powerful Muslim state on its borders, that of the Rum Seljuqs. To be sure, relations were only occasionally hostile—e.g., the effort of Sultan Kay-Kāʾūs (r. 608–616/1211–1220) to exploit Ayyubid preoccupation with the Fifth Crusade in 615/1218—but until their crushing defeat by the Mongols in 641/1243, the Rum Seljuqs were far larger and more powerful than the Ayyubid principality of Aleppo. They remained an important presence even after they were reduced to a Mongol puppet state, if only because of the presence of Mongol garrisons there.3 On the other hand, the relative stability and apparent prosperity that Seljuq rule brought to Anatolia in the early thirteenth century clearly enhanced Aleppo’s growing role as a commercial entrepôt, linking the Mediterranean to the Silk Route as well as to northern Syria and Iraq.4 An increased economic vitality in the region, and perhaps some population growth, may be visible in the partial repopulation of the abandoned villages (the famous “villes mortes”)

3 Cahen, Formation of Turkey 49–71; Eddé, Principauté ayyoubide 89–92, 112–132. 4 Heyd, Histoire du commerce, i, 372–378. For a detailed inventory of routes, products, and commercial relations and practices, see Eddé, Principauté ayyoubide 511–529.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

160

humphreys

of the great limestone plateau southwest of Aleppo.5 Aleppo’s Christian neighbors in Antioch and Cilicia were by now more an opportunity than a threat, serving both as conduits of trade with the Mediterranean and Anatolia and as potentially useful buffer states. The later twelfth and thirteenth century witnessed a great flurry of construction throughout Syria and Egypt, but Aleppo surely produced the most impressive monumental ensembles of the Ayyubid period—the massive citadel, a cluster of madrasas (especially the Ẓāhiriyya and Madrasat al-Firdaws), and the stunning shrine to al-Ḥusayn b. ʿAlī. All these monuments are evidence of a continuing tradition of limestone masonry at the highest technical level, but they are also distinguished by their spacious though not grandiose proportions (apart from the awe-inspiring citadel), their harmonious designs, and their restrained but sophisticated ornament and finish. Aleppo’s architects and masters were clearly in touch with trends elsewhere—Damascus, Anatolia, Iraq, and Iran—but they knew how to create a well-integrated style of their own—an achievement which is in itself a sign of a distinctive cultural identity.6 On the level of intellectual achievement, Aleppo was undoubtedly a significant center of Islamic scholarship, but in this realm it was perhaps not quite in the first tier. It did produce one of the most important Syrian historians of the thirteenth century, Kamāl al-Dīn b. al-ʿAdīm (d. 660/1262), and the impressive range of texts he cites in his monumental biographical compilation demonstrates the extraordinary richness of Aleppo’s libraries. Two other historians of note also resided there. Bahāʾ al-Dīn b. Shaddād (d. 631/1234), the qāḍī of Saladin’s army in the last years of his reign, wrote a famous biography—almost a hagiography—of his patron. During the transition from Ayyubid to Mamluk rule in the 1260s and 1270s ʿIzz al-Dīn b. Shaddād (d. 584/1285; unrelated to Bahāʾ al-Dīn) compiled a comprehensive historical-administrative topography of Syria and the Jazīra. In the religious sciences and other high-prestige areas of study, however, few of Aleppo’s scholars seem to have achieved a wide reputation in the Muslim world. This fact (or apparent fact) is not easy to explain, and I will not speculate here. Possibly it is connected with the awful destruction wrought by the Mongols when they stormed the city in 658/1260, an event that cost Aleppo so much of its accumulated cultural capital. However this may be, Aleppo did not lack capable scholars. Along with scions of local notable families, like Kamāl al-Dīn b. al-ʿAdīm, the scholars in Aleppo’s madrasas included

5 Sourdel-Thomine, Peuplement. 6 Sauvaget, Alep 109–154; Tabbaa, Constructions of power; Mulder, Shrines of the ʿAlids 63–113.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

161

a three-centered system

many first- and second-generation immigrants from Iraq and Iran. Roughly half of the city’s madrasa professors between ca. 1150 and 1260 seem to have belonged to this latter group.7

3

The Keystone of the Ayyubid Confederation: Damascus

In spite of important similarities, Damascus presents a very different picture. It was roughly the same size as Aleppo, though possibly somewhat larger in this period. It too had very close economic links with the Frankish territories to its west—in this case the much reduced but very wealthy Kingdom of Jerusalem, whose capital after 583/1187 was the coastal city of Acre. This polity was a kingdom in name only, since it consisted of a coastal strip about ten miles wide stretching from Ascalon to Beirut. It controlled the city of Jerusalem, on very limited terms, only for a decade (626/1229–636/1239). Even in its reduced dimensions, however, the seaports of Acre and Tyre produced ample revenues to sustain it. Its limited manpower resources meant that it posed no threat in itself to Muslim rule in southern Syria and interior Palestine. Left to their own devices, the kings of Jerusalem usually strove to avoid conflict with their Muslim counterparts. But of course the alluring proximity of the city of Jerusalem, so recently lost, made the kingdom a powerful magnet for the crusading impulses of Europe—to name only the major expeditions, the Fifth Crusade (1217–1221), the Crusade of Frederick ii (1228–1229), two minor but surprisingly effective crusades in 1239–1241, and the grand and tragic Crusade of Louis ix (1249–1254). As a result, and quite against their will, the rulers of Damascus were constantly on the front lines against the Franks, and when not at war, embroiled in Frankish politics.8 Like Aleppo, Damascus was a major commercial entrepot, but it did not have Aleppo’s significant role in the great transcontinental networks that linked Syria to the Mediterranean basin, Anatolia, Iraq, Iran, Inner Asia, and ultimately China. Damascus was, however, the main redistribution point in south and central Syria for imports from the Crusader ports of Acre and Tyre, and via the pilgrimage route it was intimately connected with the holy places of Medina and Mecca. Jerusalem—no small pilgrimage attraction in its own right—

7 Eddé, Principauté ayyoubide 347–415; Sourdel, Professeurs de madrasa. On the life and oeuvre of Ibn al-ʿAdīm, see Morray, Ayyubid notable. 8 For a concise overview of Ayyubid policy toward the Crusades, see Humphreys, Ayyubids, Mamluks, and the Latin East. Entanglements with Acre became especially intense during the decade following the death of al-Kāmil Muḥammad in 635/1238.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

162

humphreys

likewise lay within its economic and political orbit. To this commerce it contributed its own products—fine textiles, metalwork, glass, dried and candied fruits, etc. Still, in this arena Damascus was more a regional than a transcontinental actor. Much of the city’s prosperity flowed from agricultural and horticultural revenues—in the first instance from its magnificent oasis, the Ghuta, covering an area some 20 by 30km, and often portrayed in medieval Islamic literature as one of the four earthly paradises. The important croplands of the Hawran lay close to hand as well. Damascus was most fundamentally set apart by three things: 1. Its strategic location, squarely between Egypt and the major principalities of northern Syria and the Jazīra, inevitably made it the focal point of the endemic struggle for power among the Ayyubid princes. 2. Its concentration of shrines and holy places lent it an aura of sacrality second only to Jerusalem, and indeed linked it to that city. 3. Under Ayyubid rule Damascus arguably became the foremost center of legal and religious studies in the Arabic-speaking lands, and it stood among the leading centers throughout the Islamic world. Damascenes surely would have welcomed a more peaceable existence than they actually led between the death of Ṣalāḥ al-Dīn and the coming of the Mongols, but there was no escaping the consequences of their geographical location. The struggle for power among the Ayyubid princes was an unavoidable fact of life. Between the death of Ṣalāḥ al-Dīn and the Mongol conquest, Damascus was ruled by twelve different princes. Only twice was there a succession from father to son—in 589/1193, when the city passed to Ṣalāḥ al-Dīn’s eldest son al-Afḍal ʿAlī, and in 624/1227, when al-Muʿaẓẓam ʿĪsā was succeeded by his son al-Nāṣir Dāʾūd. Within two years both these successors were tossed aside, each at the hands of an ambitious uncle. These repeated conflicts were not simply a struggle to rule a wealthy and prestigious city and its hinterland but rather embodied a contest between two crucial but incompatible political principles: 1. the need for a strong central authority to hold the Ayyubid coalition together; 2. the intense determination of the regional princes to exercise full autonomy within their own realms, as well as to retain some independence of action within the coalition. Which of these two principles would prevail at any given moment was reflected in, and in fact depended on, the political status of Damascus. Although control of that city was never a sufficient condition for unity (or at least an initiative toward unity) within the Ayyubid system, it was always a necessary one.9

9 The system of autonomous appanages was complex, of course. Syria’s smaller cities and towns were inevitably brought within the orbit of Cairo, Damascus, or Aleppo. Even so, their princes

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a three-centered system

163

The tenacious but almost unspoken principle of regional autonomy flowed from the very real belief that every “major” Ayyubid prince—“major” being a highly malleable term—had a right to a share in the governance of the clan’s vast domains. It was taken for granted that such a share would take the form of rulership over some piece of territory—the possession of an autonomous appanage, we might say. Hence the determination of each prince to guard his rights and prerogatives within his own domain. However, the Ayyubid confederation could survive in a hostile environment —periodic crusades most commonly, but also sporadic but very dangerous irruptions by the Rum Seljuqs, Khwarizmians, and ultimately the Mongols— only if it retained some capacity for unity and common action. That capacity could be supplied only in small part by fluid and unpredictable family loyalties; in the final analysis, it had to be imposed. The power to impose unity might come from a dominating prince within the lineage, almost always someone who was father and patron to all the others. This was in fact the case with Ṣalāḥ al-Dīn (the founder) and his younger brother al-ʿĀdil (who replaced most of Ṣalāḥ al-Dīn’s sons with his own), but after al-ʿĀdil’s death in 615/1218 there was never again anyone who could fill that role. His eldest son al-Kāmil Muḥammad appeared to be on the verge of success, but his sudden death in 635/1238 ended that possibility. In a word, a father of the other key princes could hold the system together, but brothers, cousins, uncles, or nephews found the effort an exhausting and unproductive exercise.10 The only alternative to dominant status within the family was a preponderance of economic and military resources, and in the Ayyubid realm such resources belonged only to the ruler of Egypt, whose revenues (and hence whose standing army) substantially surpassed those available to all the Syrian and Jaziran princes put together.11 Even so, hard power was not enough by itself. Even a ruler as relentless and astute as al-Kāmil (r. 615–635/1218–1238) needed years of struggle and astute diplomacy to achieve a precarious paramountcy, and even this rested less on main force than on a reliable alliance with his younger brother al-Ashraf Mūsā in Damascus. Moreover, al-Kāmil’s achieve-

10 11

jealously guarded their status as members of the ruling family, and if they played their hands skillfully, they had some capacity to choose which major center to align with. Homs sided variously with Damascus and Aleppo, Hama normally with Cairo. East of the Euphrates, Harran might be allied with Aleppo, integrally linked with Damascus, or left to follow its own path. The remote fortress cities in the northern Jazīra, like Mayyāfāriqīn, Amida, and Ḥiṣn Kayfā, were only occasionally dragged into the fray. The preceding analysis is largely derived from Humphreys, From Saladin to the Mongols 10–11 et passim. Humphreys, Emergence of the Mamluk army 74–76.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

164

humphreys

ment evaporated immediately upon his death. His younger son al-Ṣāliḥ Ayyūb (r. 537–547/1240–1249) ultimately muscled his way to direct rule over Egypt and Damascus, while the latter’s own son, al-Muʿaẓẓam Tūrānshāh, clung to the Jaziran fortress of Ḥiṣn Kayfā, but the other Syrian and Jaziran principalities remained outside his orbit. Had Damascus been notable only for its commerce and strategic location, there would not be a great deal more to say about it. But, as we have noted, in many ways Damascus was the religious and cultural heart of the Ayyubid domains. Beginning perhaps with Nūr al-Dīn (who ruled there 549–569/1154– 1174), the city regained a prominence throughout the Islamic world that it had not enjoyed since Umayyad times, five centuries before. This was in part because Damascus and the region around it (not to mention its outer dependencies in Palestine) formed a land of prophets and saints—Jewish, Christian, and Muslim. The very soil was infused with their sacred presence. This fact made it a focus of ziyārāt, the pious visitations to the shrines of holy men or sacred events that were so integral a part of Islamic life for many centuries. It may also be part of the reason—apart from sheer serendipity—for the striking number of Ayyubid princes whose tombs are in Damascus: Ṣalāḥ al-Dīn, his brother and successor al-ʿĀdil, and the latter’s sons al-Kāmil Muḥammad, al-Muʿaẓẓam ʿĪsā, and al-Ashraf Mūsā, to name only the most notable figures.12 Apart from this heritage, Damascus also emerged as a vital center of Islamic learning in Ayyubid times. This was not entirely a new thing; the process had begun in the eleventh century and was intimately connected to the revitalization of Sunni thought throughout the eastern Islamic world in that period. Al-Khaṭīb al-Baghdādī (d. 463/1071), whose immense biographical dictionary almost defined the canon of respectable Sunni scholars down to his own time, spent some years in Damascus and clearly had a powerful impact on the scholars there. The vast Taʾrīkh madīnat Dimashq of Ibn ʿAsākir (d. 571/1176), composed under the patronage of Nūr al-Dīn, was clearly an effort to replicate for his native and heretofore somewhat provincial city what al-Khaṭīb had done for the cosmopolis of Baghdad.13 On a very different level, after al-Ghazālī (d. 505/1111) abandoned his professorial post in Baghdad, he spent some years in seclusion in the eastern minaret of the Great Mosque of Damascus as he recharted his life’s course. 12 13

For a superb synthesis of these issues, see Talmon-Heller, Islamic piety. For the crossconfessional sanctity of holy persons and their shrines, see Meri, Cult of saints. We still await a full-length study of Ibn ʿAsākir, but the contributions in Lindsay, ed., Ibn ʿAsākir, are a very useful first step.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a three-centered system

165

Damascus remained a provincial if active center of religious learning until Nūr al-Dīn’s occupation of the city in 549/1154. But under him and his Ayyubid successors, the city saw an explosion of religious building—e.g., some 80 new colleges of sharia jurisprudence (madrasas) endowed between 549/1154 and the Mongol occupation in 658/1260, along with an impressive array of congregational mosques, Sufi convents, hospitals, and the like. The suburb of Ṣāliḥiyya, founded by a few Ḥanbalī Palestinian refugees under Nūr al-Dīn, blossomed into a major center of religious life and learning under the Ayyubids, and it contains the most impressive architectural ensemble erected during their rule, the grand hospital built by the Kurdish amīr Sayf al-Dīn al-Qaymarī ca. 648/1250. The material and financial infrastructure of religious life created under Ayyubid rule is impressive in its own right, but the literary and scholarly output it supported is even more so. Thirteenth-century Damascus, we might say, was the true heir of Baghdad in defining the Sunni synthesis in the religious sciences.14

4

The Dominating Metropolis: Cairo

We come at last to Cairo, a city profoundly different from all the others. It was in the first place a recent foundation (recent by Middle Eastern standards at least): it began with the garrison encampment of Fustat in 20/641, and the governmental center, al-Qāhira, had been founded only in 358/969—i.e., precisely two centuries before the Ayyubid seizure of power. In sharp contrast to Aleppo and Damascus, it had no real roots in a deeper past, whether Pharaonic, Ptolemaic, or even Roman-Byzantine. Cairo was far larger than its Syrian counterparts, with a population that certainly surpassed that of Damascus, Aleppo, Homs, Hama, and Jerusalem put together. More than that, it was a true primate city, one which dominated politically, economically, and socially the entire Nile Valley. Syria was a land of many cities, Egypt was a land of one. Cairo enjoyed an economic location even more privileged than Aleppo’s, for it could exploit the vast agricultural production of Egypt even while it stood astride the inter14

Humphreys, Islamic history 249–254; idem., Politics and architectural patronage; idem., Women as patrons. Pouzet, Damas, gives a meticulously detailed survey of the city’s religious culture in the Ayyubid and early Mamluk periods. Chamberlain, Knowledge and social practice, analyzes the not always ennobling politics of scholarship in the city. Hirschler, Medieval Damascus, demonstrates the richness of Damascene libraries in the medieval period. In the eyes of contemporaries, Baghdad surely retained a certain primacy due to its size and prestige, and the cultural ties between Damascus and Baghdad remained close, but during the seventh/thirteenth century the scholars of Damascus arguably made a more substantial contribution to the Sunni canon.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

166

humphreys

section of the Mediterranean and Indian Ocean trade networks, both of which were approaching their medieval peaks in the Ayyubid period.15 In view of all this, Cairo’s dominant role within the Ayyubid coalition would seem all but inevitable. However, Cairo’s potential power was compromised (though hardly vitiated) by one fact—viz., that it lay on the southwestern edge of the Ayyubid realm. The ruler of Egypt thus had to reach out from the margins in order to assert some measure of control across the length and breadth of Syria and the Jazīra. The thing could be done, as we have seen, but it was exhausting and required both luck and consummate political skill. In a broader geopolitical framework, Cairo was both sheltered from outside attack and highly vulnerable to it. Cairo was spared the depredations of the Khwarizmians in the early 1240s and the onslaught of the Mongols in 1260 and the four decades following. On the other hand, the two largest crusades of the thirteenth century (in 614–618/1217–1221 and 647–648/1249–1250) were aimed squarely at Egypt and its capital, and the Fourth Crusade (1203–1204) originally contemplated Egypt before its leaders were distracted by the lure of Constantinople. Both the first and second Damietta expeditions were very nearly a catastrophe for the Ayyubids, and no doubt every Ayyubid ruler in Egypt was acutely aware of the close-run struggle for Cairo between Nūr al-Dīn and Amalric of Jerusalem between 558/1163 and 564/1169. From the security of our time and place, we might complacently assert that in all these struggles against the Franks Cairo was in almost a literal sense a bridge too far. But it did not look that way at the time. It is easy to forget what a polyglot place Cairo must have been. One can easily imagine hearing at least a dozen languages in the streets. Aleppo and Damascus had mixed religious and ethnic populations to be sure: Muslims, Christians, and a fair number of Jews; Arabs, Kurds, Rumis (both Greek and Armenian), and Turks (nomadic Turcomans in addition to persons of mamluk origin), along with some Iranians. Cairo had all these, and in greater numbers, but also Berbers, Nubians, Sudanese, and Ethiopians. Cairo was after all an African city, and its most direct links with the wider world were via the Nile Valley—north toward the Mediterranean, south deep into the heart of Africa. The bulk of its black African population would have been slaves, or persons whose parents or grandparents had been slaves. Nevertheless, they were an integral and highly visible part of the urban landscape. The links to North Africa and al-Andalus were quite different—descendants of the Berber soldiers who had come with the Fatimids in 358/969, merchants, religious scholars looking for a more cos-

15

Goitein, Mediterranean society 29–74; Margariti, Aden; Abulafia, Great sea 287–353.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a three-centered system

167

mopolitan milieu than they could find at home, pilgrims on the route to Mecca and Medina, and refugees from the Reconquista after the disaster of Las Navas de Tolosa in 1212. The evidence of the Geniza demonstrates how extensive and tight-knit Cairo’s connections to North Africa were for Egypt’s Jewish community, and this was no less true for Muslim merchants and scholars. Cairo under Ayyubid rule was only beginning its climb to the kind of religious and cultural preeminence it would attain in Mamluk times. As late as the mid-seventh/thirteenth century, at least, Damascus continued to be the more dynamic center of Sunni-oriented learning and literature. But since Cairo had been founded by an Ismaʿili Shiʿite dynasty, and since it was the center of the Ismaʿili “mission” (daʿwa) down to the mid-twelfth century, this is simply what we should expect. On the other hand, even under Fatimid rule Sunnis seem to have remained a large majority of Egypt’s Muslim population, and we see the clear beginnings of a reinvigoration of Sunni learning in the mid-sixth/twelfth century (albeit especially in Alexandria rather than Cairo), well before the Ayyubid coup d’état in 564/1169.16 Ṣalāh al-Dīn undertook a program to establish Sunni religious and educational institutions in Cairo, but he was a warrior sultan with grand ambitions, and his construction program inevitably privileged major military and administrative projects—the vast citadel on the Muqaṭṭam hills and an extensive new band of walls to encompass as much of the urban perimeter as possible. We see a greater emphasis on religio-educational initiatives under his successors al-Kāmil Muḥammad and al-Ṣāliḥ Ayyūb—though again, the list of buildings does not match the sheer number erected in Damascus and Aleppo. The great age of Sunni religious architecture in Cairo would only come with Mamluk rule after 658/1260.17 Cairo had only a limited repertory of formal institutions, however, it had enticements of equal or greater weight—wealth, the visibility offered by a great cosmopolitan city, and closeness to the paramount center of power. Those advantages would become much stronger in Mamluk times, all the more because Aleppo and even Damascus were so vulnerable to Mongol predations in the later thirteenth century, but they were already operative in the reign of al-Kāmil, a highly educated ruler with a real interest in religious and philosophical questions. We thus see an increasing migration of Syrian scholars to Cairo in those years, and of course an increasing proportion of Ayyubid cultural production originating there. Over time Cairo would surely have reduced Aleppo and Damascus to a provincial if respectable cultural and scholarly status in any 16 17

Lev, State and society 133–152. MacKenzie, Ayyubid Cairo; Behrens-Abouseif, Islamic architecture 78–93; Ruggles, Tree of pearls 31–56, 78–101, 111–140.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

168

humphreys

case, but the transition was given a powerful jolt by the Mongol onslaught of 657–658/1259–1260. Among the flood of Syrian refugees heading toward Egypt was a host of scholars. Not really typical, but perhaps emblematic, was Kamāl al-Dīn b. al-ʿAdīm, who bore with him the precious draft volumes and notes for his vast (and apparently never completed) history of Aleppo. He died a short time after reaching Cairo, in 660/1262, but his work somehow survived him and now resides in an Istanbul library. Each of the three principal cities of the Ayyubid realm clearly lived within, and indeed defined, a distinctive world of its own, but each was intimately connected with the other two. In this way they created an interwoven system of political action, economic life, population flows, and cultural activity that stretched not just from North Syria to the Egyptian Delta but from Anatolia to Nubia and beyond, and in the east reaching as far as the Caucasus, Iran, and Central Asia. On an even grander scale, this system linked the Mediterranean basin with the broad world of the Indian Ocean. Almost every major current of Eurasian history in the seventh/thirteenth century flowed through Ayyubid Syria and Egypt and had a powerful impact there. To borrow a familiar image from Islamic historical writing, the Ayyubid lands were a mirror of the age. It is obvious that this tripartite network of urban centers was not always a harmonious one, that it was afflicted with endemic tensions and periods of violent disruption. Nevertheless, even at its worst moments it never failed to function. Because Aleppo, Damascus, and Cairo each had such a distinctive personality of its own, whether political, socioeconomic, or cultural, we inevitably tend to view these cities as separate, almost self-contained entities, each of which is most readily studied in near-isolation from the other two. And indeed that is how scholarship, both Middle Eastern and western, has largely evolved over many centuries. However, it may prove far more rewarding and productive to situate these three cities as dynamic elements within an integrated albeit friction-ridden system, elements whose profile and function within the Ayyubid political system become fully intelligible only when we examine them from that perspective.

Bibliography The published scholarship on these three cities during the twelfth and thirteen centuries is overwhelming, and the original sources (mostly in Arabic and Latin, but also in Old French, Armenian, and Syriac) are voluminous. I have therefore focused on a select number of titles that address the main issues in this essay, and through which readers can find both the sources and literature underlying my discussion.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a three-centered system

169

Abulafia, D., The great sea: A human history of the Mediterranean, Oxford 2011. Behrens-Abouseif, D., Islamic architecture in Cairo: An introduction, Leiden 1992. Berkey, J., Culture and society during the late Middle Ages, in C. Petry (ed.), Cambridge history of egypt, i, Cambridge 1998, 375–411. Cahen, C., Ayyūbids, in ei2, i, 796–807. Cahen, C., The formation of Turkey: The Seljuq Sultanate of Rum, eleventh to fourteenth century, P.M. Holt (ed. and trans.), Harlow (Essex) 2001. Chamberlain, M., Knowledge and social practice in medieval Damascus, 1190–1350, Cambridge 1994. Chamberlain, M., The Crusader era and the Ayyubid dynasty, in C. Petry (ed.), Cambridge history of Egypt, i, Cambridge 1998, 211–241. Eddé, A.-M., La principauté ayyoubide d’Alep, 579/1183–658/1260 (fis 21), Stuttgart 1999. Goitein, S.D., A Mediterranean society: The Jewish communities of the Arab world as portrayed in the documents of the Cairo Geniza, i: Economic foundations, Berkeley 1967. Gottschalk, H.L., Al-Malik al-Kāmil und seine Zeit, Wiesbaden 1958. Heyd, W., Histoire du commerce du Levant au moyen-age, i, Leipzig 1885–1886. Hirschler, K., Medieval Damascus: Plurality and diversity in an Arabic library: The Ashrafiya library catalogue, Edinburgh 2016. Hirschler, K., Medieval Arabic historiography: Authors as actors, London 2006. Humphreys, R.S., From Saladin to the Mongols: The Ayyubids of Damascus, 589/1193– 658/1260, Albany, NY 1977. Humphreys, R.S., The emergence of the Mamluk army, in si 45 (1977), 67–99. Humphreys, R.S., Politics and architectural patronage in Ayyubid Damascus, 1193–1260, in C.E. Bosworth, C. Issawi, R. Savory, and A.L. Udovitch (eds.), The Islamic world: Essays in honor of Bernard Lewis, Princeton 1988, 151–174. Humphreys, R.S., Islamic history: A framework for inquiry, Princeton 1991 (rev. ed.). Humphreys, R.S., Women as patrons of religious architecture in Ayyubid Damascus, in Muqarnas 11 (1994), 35–54. Humphreys, R.S., Ayyubids, Mamluks, and the Latin East in the thirteenth century, in msr 3 (1998), 1–17. Lev, Y., State and society in Fatimid Egypt, Leiden 1991. Lindsay, J.E. (ed.), Ibn ʿAsakir and early Islamic history, Princeton 2001. MacKenzie, N.D., Ayyubid Cairo: A topographical study, Cairo 1992. Margariti, R.E., Aden and the Indian Ocean trade: 150 years in the life of a medieval Arabian port, Chapel Hill 2007. Meri, J.W., The cult of saints among Muslims and Jews in medieval Syria, Oxford 2002. Morray, D., An Ayyubid notable and his world: Ibn al-ʿAdim and his world as portrayed in his biographical dictionary of people associated with the city, Leiden 1994. Mulder, S., The shrines of the ʿAlids in medieval Syria: Sunnis, Shiʿis and the architecture of coexistence, Edinburgh 2014.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

170

humphreys

Pouzet, L., Damas au viie/xiiie s.: vie et structures religieuses dans une métropole islamique, Beirut 1988. Ruggles, D.F., Tree of pearls: The extraordinary architectural patronage of the 13th century Egyptian slave-queen Shajar al-Durr, Oxford 2020. Sauvaget, J., Alep: essai sur le développement d’une grande ville syrienne des origines au milieu du xixe siècle, Paris 1941. Sourdel, D., Les professeurs de madrasa à Alep aux xiie–xiiie siècles d’après Ibn Saddad, in beo 13 (1949–1951), 85–115. Sourdel-Thomine, J., Le peuplement de la region des “villes mortes” (Syrie du Nord) à l’époque ayyoubide, in Arabica 1 (1954), 187–199. Tabbaa, Y., Constructions of power and piety in medieval Aleppo, University Park, PA 1997. Talmon-Heller, D., Islamic piety in medieval Syria: Mosques, cemeteries and sermons under the Zangids and Ayyubids (1146–1260), Leiden 2007.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 9

Informal and Formal Trading Associations in Egypt and Ifrīqiya, 850–1150 Chris Wickham

It is well-known that Islamic law dedicates a large section of its discussions to regulating contracts.* Jurists were highly preoccupied with the issue of usury from the start, and were alarmed by the problem of risk (gharar), both of which they saw as illegitimate. Some of the major law-school founders, such as Mālik b. Anas (d. 179/795), who was particularly influential in Mediterranean lands, banned sales unless all the elements of the sold object were known. For many, unripe dates could not be sold, in case they did not ripen; and some wondered whether even tenancy was lawful, since the uncertainties of crops made leases risky, for landlords or tenants or both. Jurists got around the last issue, which was a good thing really, at least for élites, or else agricultural relations would have entirely broken down. The issue of commercial contracts obviously presented problems, since profit and loss are impossible to predict, and profit at someone else’s expense might well be seen as usurious; all the same, the Prophet Muḥammad had engaged in plenty of commercial activity, which was a key legitimizing starting point. Given that, the model of the commercial contract came to be powerful when legitimizing tenancy as well in Ḥanbalī and some Ḥanafī law. And, given that, it is even less surprising that contract law was a fundamental issue for jurists to confront. Mālik’s al-Muwaṭṭaʾ is an early example of a systematic treatment in the later eighth century, and many writers picked up from it. Mālik and his successors discussed at length the qirāḍ contract, in which a sedentary merchant provides the capital used for a trading expedition and the traveling merchant does the actual work; the eventual profits are split between them.1 This contract is very similar to the

* An earlier version of this paper was given at a conference, “Networks and ties of exchange,” held (remotely) in Leiden in June 2021; I am very grateful to the organizer, Cecilia Palombo, and to the other conference participants, for stimulating and helpful comments. It seemed to me appropriate for a Festschrift for Hugh Kennedy, whom I have known for some forty years, given Hugh’s wide interests, covering (as not many do) the whole medieval Islamic world. 1 See, for a general account of the law concerning qirāḍ (unhelpfully called commenda throughout), Udovitch, Partnership and profit 170–246; for Mālik’s al-Muwaṭṭaʾ as a text, see

© Chris Wickham, 2022 | doi:10.1163/9789004525245_010

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

172

wickham

contract called commenda by the Genoese in the twelfth century, although in the latter the traveling merchant very often contributed to the capital, too (usually but not always a third) and received a higher share of the profit as a result. This division was not specified by Mālik or other early jurists, who indeed were uneasy about the traveling merchant contributing any capital, but texts from slightly later periods do envision such a possibility more concretely, even if, again, without much enthusiasm (one example is the formulary book of Ibn al-ʿAṭṭār from late fourth/tenth-century Córdoba).2 Such contracts helpfully allowed the accumulation of commercial capital by even impecunious traveling merchants, and were widely used in the Mediterranean from the ninth century onward. There is a whole historiography that discusses who invented the commenda first, with some arguing that the qirāḍ influenced Latin practice, others who see the commenda as developing independently, and others again who locate its ultimate origins in imperial Roman societas partnership law, given that this envisages that partners can bring service, including travel, to a partnership rather than money. Some of the arguments here owe more, it has to be said, to rival grand narratives than to the evidence itself, which is weak. To me, however, as also to John Pryor, it is quite plausible that the Roman Empire did have a practice resembling medieval qirāḍs/commendae, even if its surviving law did not discuss one-voyage contracts at all, and that what developed was a long postRoman koinē of similar maritime practice extending to the Latin-, Greek-, and Arabic-speaking Mediterranean alike. Conversely, however, the development of Islamic juristic reasoning about qirāḍs was far more complex than that of any other legal or mercantile community, and the qirāḍ was more similar to the commenda than any of the others were to each other; so, even if the latter was not borrowed from the former, it is very plausible that it was influenced by it.3 Here, the real point is that it was normal that sea-going merchants in the medieval Mediterranean had access to capital, using contracts with very simmost recently Brockopp, Muhammad’s heirs 105–113. For dates, a good survey of the complexities of juristic reasoning is Rittenberg, Gharar in post-formative Islamic commercial law 187–194 and following. For tenancy, see Haque, Landlord and peasant 325–330, 336–337; cf. Johansen, Islamic law; for the complexities of real agrarian life in Iraq in the earliest Islamic centuries, see Kennedy, Landholding and law. 2 For unease among early jurists, see, e.g., Udovitch, Partnership and profit 194–195. For Ibn al-ʿAṭṭār’s Kitāb al-wathāʾiq wa-l-sijillāt, see the translation, Ibn al-ʿAṭṭār, Formulario notarial, n. 31 (209–214). 3 See esp. Udovitch, At the origins; Pryor, Origins; Khalilieh, Admiralty and maritime laws 224– 246, all referring to a wider bibliography.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

informal and formal trading associations in egypt and ifrīqiya

173

ilar terms. The only substantial difference between any of them is that the Jewish ʿisqa, discussed among others by Maimonides in late twelfth-century Egypt (note that he calls it a qirāḍ as well), which Philip Ackerman-Lieberman has shown was frequently used by the merchants documented in the Geniza of the Ben Ezra synagogue in Fusṭāṭ, assigned a share in the risk of the loss of capital to a traveling merchant who lost his creditor’s goods (e.g., to shipwreck or piracy), whereas the qirāḍ theoretically did not (the commenda, in different forms, could do either).4 This, however, seems to me a detail, even if an interesting one, inside a general array of closely analogous maritime legal assumptions. Merchants across the Mediterranean may have spoken different languages, and/or have had different religions, but they borrowed capital and faced trading risks—and, indeed, bought and sold—in much the same way. They also faced the same issues of trust in foreign ports and built up networks of trusted counterparts in the same way; and dealt with fraud and deceit in very similar ways.5 I am not going to discuss the non-Islamic parts of the Mediterranean here, but it is important to remember that what I have to say has relevance outside the Islamic lands, too; we should not fetishize the undoubted differences between Muslim-ruled and Christian-ruled regions when we talk about commerce. Back to qirāḍ, though. We know that it was a concept used in the third/ninthcentury Mediterranean because North African and indeed Andalusi jurists, starting with the famous Mālikī jurist Saḥnūn, who died in 240/854 in Kairouan, the capital of Ifrīqiya, assume its existence in their fatwās about trade. These are collected in a huge early sixteenth-century fatwā collection, al-Wansharīsī’s al-Miʿyār, although some of them at least already appear in the early tenthcentury Kitāb Akriyat al-sufun by Muḥammad b. ʿUmar al-Kinānī, an Andalusi who moved to Alexandria. As fatwās often do, they could envisage very complicated commercial patterns, as when the Annaba jurist al-Būnī (d. 440/1048) discussed the problems occurring when a merchant goes by sea from Sfax to Tripoli with a qirāḍ, and then back with a separate qirāḍ, but is murdered on the way back, although the goods (māl) arrive in Sfax without him—how does it get divided up? Or when the Mahdia jurist al-Māzārī (d. 536/1141) discusses whether the traveling merchant with a qirāḍ who goes to the Ifrīqiyan oasis town of Tozeur, a land route, has the right to engage in commerce on his own account (which he does).6 The point here is: these contracts are not in 4 Ackerman-Lieberman, Business of identity 156–171, 191–193. 5 See Goldberg, Trade and institutions 139–164, 177–184, for trust. 6 I have used for al-Wansharīsī the abbreviated but still very extensive translation by Vincent Lagardère in Histoire et société; for the fatwās cited, viii.161, viii.95; v.1 for Saḥnūn; for alKinānī, see Khalilieh, Admiralty and maritime laws 273–330, esp. 288–293.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

174

wickham

themselves the object of dispute or theorization, as in Mālik’s treatise; they are the assumed background to much more detailed conflicts. They are taken for granted as part of the landscape of trade. So are, though slightly less often in these texts, formal partnerships (shirka)—the word itself is common enough in these fatwās, but it otherwise normally denotes an agrarian contract, seen here, too, as a commercial association—but again al-Māzārī cites one for a merchant engaging with the Ifrīqiya-Sicily trade route. In Spain in the same period, Ibn al-Ḥājj (d. 737/1135) similarly envisages a shirka in which a traveling merchant goes to the Maghrib and dies, but one of his associates (sharīk) cannot find the written contract which would allow him to claim his goods against other creditors (he loses).7 It may be that every qirāḍ assumed a shirka here, but I think it is likely that shirkas were generally longer-lasting agreements for commercial and other activities. So, jurists routinely took it for granted that merchants in the central Mediterranean, and westward to al-Andalus, had written deeds of association and/or contracts for the one-off transport of commercial goods. This is true no matter how many of these fatwās were legal concoctions to make a specific juristic point—which many are; but, in most of the cases I have cited, I think they are too specific for that. And, I would equally stress, the sometimes perplexing claims by historians that merchants in the Islamic world relied on reputation as fair dealers and not on formal law to enforce contracts is falsified by every fatwā, for nearly every one is issued by a legal expert who has to have been appealed to by a plaintiff (a word often used in the texts) as at least a quasi-formal legal act. Shelomo Goitein belonged to the long tradition of writers about Islamic societies who thought that most things about them were informal. So, in his classic discussion of the Egyptian Geniza’s commercial letters he stressed informal ṣuḥba partnerships, “companionship” as he called them. Conversely, Goitein did certainly recognize that shorter-term formal shirka or khulṭa partnerships were common, too, and he discussed qirāḍs, which were all one-off contracts, at some length. Goitein thought that the Muslim version of the qirāḍ, in which the traveling merchant was not responsible for losses, was much more common than the Jewish-law version, in which he was coresponsible; AckermanLieberman has corrected that, as I said. But it was clear to Goitein that these contracts were absolutely normal in the Jewish merchant environment of Fusṭāṭ and Alexandria, and numerous qirāḍ agreements of different kinds (not all for international commerce) survive in the Geniza. They could involve multiple people, and a single merchant could have multiple contracts, thus poten-

7 Lagardère, Histoire et société v.93, v.308.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

informal and formal trading associations in egypt and ifrīqiya

175

tially generating just the sort of complex and also transnational legal problems that fatwās envisaged.8 This general picture was a decade ago sharpened considerably by Jessica Goldberg, who showed that ṣuḥba partnerships were rather more formal than Goitein envisaged, involving an agreement between two merchants to act as each other’s agents, a quasi-legal agreement that had to be—or at least often was—ended with oaths. She showed that fully formalized commercial associations were called khulṭa and argued that shirka was a more general term that could designate either a khulṭa or a qirāḍ.9 It may well be that this shows a more complex picture than that known to central Mediterranean jurists; for a start, neither ṣuḥba nor khulṭa are found in the Ifrīqiyan fatwās. But it may also be that the Geniza simply shows, up close, how merchants really behaved, and how they coped with multiple challenges in their day-to-day commercial activities, and that jurists were either slightly more separate from that world (although jurists could certainly trade themselves) or else were more schematic in their legal vision. It must also be stressed that the Geniza letters concern quite large-scale merchants, whether they operated internationally or internally in the Nile delta of northern Egypt. (They seldom went south of Fusṭāṭ, except to the great flax entrepôt of Būṣīr some 100km south of the capital, and less often to one or two other similar centers to the south of that, such as Ashmūnayn.) The great Geniza letter collections of the eleventh century, of Ibn ʿAwkal and Nahray b. Nissīm, show operations on a rather larger scale than those of any documented Genoese or Venetian merchant after 1100.10 So it would not be surprising that they wished to—they had to—use quite complex sets of commercial arrangements; big dealers always do. But the way they dealt was, all the same, despite differences, very analogous indeed to the world that the Ifrīqiyan fatwās presumed. This reassures us both that the world of the fatwās was not just invented by jurists out of their legal readings, and that the Jewish merchants in the Geniza letters dealt according to rules that worked for Jews and Muslims alike, a question that has bothered some writers but, as should be clear, and will be even clearer in a moment, does not bother me.

∵ The Geniza is highly fascinating, and it is not be surprising that many scholars spend their careers focused on it. On the other hand, the Geniza is also not 8 9 10

Goitein, Mediterranean society 164–179. Goldberg, Trade and institutions 123–143. For Geniza merchant ranges, see esp. Goldberg, Trade and institutions. I will discuss the relative scale of Egyptian and Italian trading in a book in press.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

176

wickham

the only set of records of commercial activities from the Egypt of the fourth to sixth/tenth to twelfth centuries. There are many collections in Arabic script, made by Muslims (and indeed also Christians) rather than Jews, from tenthto twelfth-century Egypt. They largely consist of letters and tax receipts, and mostly, like the Geniza, were thrown away rather than archived; but they present some slightly different issues. These collections are very incompletely edited indeed; only some six percent of the surviving texts from the Arab conquest up to 1300 have any form of edition at all. And, even more striking, those that have been edited for the period after 900 or so are little used by historians, even Geniza specialists, even though they are not hidden away, and are indeed by now mostly available online in the wonderful and very user-friendly Arabic papyrology database run by Andreas Kaplony at the University of Munich.11 These letters do not for the most part come from Fusṭāṭ-Cairo; they come from Middle Egypt and broadly concern the area between the Fayyūm and Asyūṭ, some 200km further up the Nile. I will mention three sets here. In the Fayyūm, a cache of third/ninth-century papyri found in the regional center of Madīnat al-Fayyūm shows the establishment of a formal partnership, shirka, in 250/864 between two merchants from the city and one from Fusṭāṭ to buy Fayyūmī cloth for sale in the capital, and then the development of that partnership, above all through the eyes of one of the two local merchants, Jaʿfar Abū Hurayra, whose archive (this time, probably a real archive) this was.12 Several dated texts from the late 860s and 870s show Abū Hurayra buying advance sales of cloth from local weavers. Most of the archive is undated letters from the same period; some are from Abū Hurayra to his father and other family members, or their letters back about small-scale sales in cloth, grain and other food, and occasional jewelry—here the sales are for both commerce and family needs, and sometimes display family tensions, for Abū Hurayra thinks his father’s second wife Sayyida is an unbeliever, presumably but not certainly a Christian (almost all these letters are, however, between Muslims). Many others, however, are from Abū Hurayra’s partners in Fusṭāṭ; these show that he was sending variable quantities of cloth to the capital on a weekly basis, on donkeytrains, to merchants who did their best to sell at good prices (and sometimes did not sell if prices were too low), returning the money in purses of 8 to 16 dīnār at a time, medium-scale sums, to Abū Hurayra and his associates, and who would also take commissions from him for personal acquisitions. Abū 11 12

The Arabic papyrology database is available at https://www.apd.gwi.uni‑muenchen.de/​ apd/project.jsp. I cite all the texts that follow using its abbreviations. For 250/864, P.Marchands i 1; P.Marchands publishes (or will publish, as it is so far incomplete) the whole Abū Hurayra archive.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

informal and formal trading associations in egypt and ifrīqiya

177

Hurayra and his family and associates, some of whom lived in the quarter of the cloth-merchants (bazzāzūn) by the great mosque of Madīnat al-Fayyūm, were thus middlemen for the systematic sale of local linen cloth to the capital. The scale of these sales was consistent, even if in no way matching that of the major Geniza merchants.13 A second relatively well-documented location is the medium-sized town of Ashmūnayn, about 100km south of the Fayyūm. From here come a number of interesting letters from the tenth and eleventh centuries, specifically about exchange, which I will summarize. A fourth/tenth-century letter to a Muslim politely notes that a silk garment has not come and that the sender will pay for it only when it does; another to an associate in the town discusses sending weapons, which are expected from Kayl, a Christian; another refers to a grainseller who also sells papyrus (in a very late attestation of that writing material), rosewater, and saffron; another discusses the transport by Christian boatmen of flax and perhaps wheat out of villages in the northern part of Ashmūnayn’s district on a very large scale (in one citation, 800 bales of flax). In the fifth/eleventh century, we find two letters to a Muslim in the city, Abū l-Qasīm Hishām: in one, he is accused of incorrect accounting when he sends lupins (lupin seeds are edible); in the other, he is asked to buy 200 irdabbs (about 18,000 liters) of cheap barley in Ismant, north of the city, and 100 dīnār-worth of sugar—again a very substantial amount—because Fusṭāṭī merchants are rumored to be coming to Ashmūnayn and prices are about to go up. Another letter, from a village north of the city, reassures the sender’s associates that he will not break his ties with them, even though Ashmūnaynī merchants came to him and offered more money. A legal document from 1028, with an Ashmūnayn provenance, discusses the unblocking of holding charges for two eunuch slaves and the payments to the different people who have been feeding them while the deal, which involved a cloth-merchant’s daughter, went through.14 Most of these texts, unlike those for the Fayyūm, are one-offs; one cannot even imagine building a picture of any individual merchant from them, or of the relative importance of any type of product in Ashmūnayn and its district. They provide, however, a guide to the sense, even if not the scale, of activity

13

14

Advance sales: P.Marchands i 2–9. For the family letters, see in general P.Marchands ii; ii 3 for Sayyida (and her birth family) as unbelievers. For letters to and from merchants in the capital, P.Marchands iii and v,1. Money purses with specific sums: v,1 2, 3, 5, 7 (the highest, 44 dīnār), 8, 11. Quarter of the cloth merchants: e.g., v,1 4. See, in general, Younes, Textile trade. Respectively, P.Cair.Arab. 307; P.Heid.Arab. ii 62; P.KarabacekPapiergeschichte, page 80; P.Vind. Arab. i 2, 19 with 36, 20; Chrest.Khoury i 25.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

178

wickham

there. They show that commercial dealers in Ashmūnayn often dealt in many types of goods at once, even though they were often so specialized in their basic identities—cloth-merchants dealing in slaves, grain-sellers dealing in papyrus. They attest, as do documents from the whole of Egypt, to a capillary money economy and to a widespread use of wage labor. And—this seems to me particularly significant—the commercial letters are strikingly similar to the sorts of ways Geniza merchants wrote to each other. I have already noted that Jewish merchants from Fusṭāṭ came to buy goods in this area; so, evidently, did those of other religions (this being the reason why I have stressed religious affiliation here, whenever I can be sure of it). And the mercantile environment they would have found matched that of the capital in many ways. Ashmūnayn merchants were more integrated into the countryside than were those whose records survive in the Geniza, but that is doubtless simply the result of the fact that their city was much smaller—the countryside was only half an hour away, even on foot. They also transacted in goods that, as is generally recognized, Jews did not, or not much: grain, slaves, and arms. But it is equally worth noting that flax and cloth, which Jews did specialize in, are prominent in this non-Jewish documentation, too. At a provincial level, then, the evidence from Ashmūnayn reflects the world of our Jewish sources well, although usually (even if not always) on a smaller scale. Finally, let us look briefly at the evidence from a more heterogeneous set of around 150 published commercial letters from the period 900–1200, which show rather more small-scale transactions. These often have no geographical markers at all, so we cannot pin them down, although most, even though not all, are almost certainly from Middle Egypt; theirs was a local and regional circuit, focused frequently enough on Fusṭāṭ (even if very often at second hand), but not often beyond. Here is a sample of such letters, again summarized. From the fourth/tenth century: “please reply to my letter; I have clothing for several named people; ask him to send the turbans if he has bought them, or else to get them by boat from Damietta; I’m going to Mecca, send me the money there.” In a letter from Idfū in the far south: “I’m sorry your slave-girl is ill, but I sold her to you healthy, and you still owe me the price; I’ve sent you perfumes worth 3 dīnār, please sell them for me; please send me the sycamore wood you promised, and buy for me the acacia firewood and ricin oil for the lamp, 2 dīnār-worth of each.” In another: “the delay with the flax means that it has been nibbled by mice, so we couldn’t leave it—it will go to Jīza by boat; I will be away, without leaving the barley for the children; get me the small dye cask, the large shawl, the provisions bags and the small garment.” In another: “I send a big packet with chickpeas, beans, dates and fruits, plus all my pieces of clothing; let me know they have arrived.” In another, a very tense text: “I keep asking

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

informal and formal trading associations in egypt and ifrīqiya

179

you about the profit for the seeds and plants, and about our accounts, but you don’t reply; we need to send the dyes to our patron; send me the inkpot, serving plate, nets, mats, and scales; get oil and sell it for me.”15 From the fifth/eleventh century: “it isn’t true that I left your brother sleepless before my door—there’s a doorkeeper; anyway, he bought the grain before it was winnowed, against my instructions, and he has run off with the 100 dīnār.” In another: “I haven’t sent you the flax, as you know Abū Sahl has a grudge against you, but he is here in the Bahnasā district now, so I could try to get it from him and send it to you.” In another: “have you bought a dīnār-worth of salt from the Chinese man (alṢīnī)? If not, do so and send it.” In another, to a merchant currently in Fusṭāṭ: “Abū Bakr still has not settled my commission for the shawls; when I sent you the cotton, I forgot one bag—Jamāl will bring it—get it dyed; I also sent two chests plus locks, sell them and give the outstanding money to the plasterer’s wife.”16 From the sixth/twelfth century, for which we have, as usual for Egypt, fewer texts: “please send me two pieces of linen cloth, rough weave, Persian style, and one good linen coat, and give the cutter a deadline.” In another: “take the 35 dirham which the shipman brings and buy me tamarind and pepper.”17 These texts again recall the Geniza in their style, very closely, although the actors in almost all of them are much smaller-scale operators, not only smaller than the Geniza merchants but even than dealers like Abū Hurayra. But there is one important difference. They are much more informal. Arabic words for partnership of different types, shirka and khulṭa, qirāḍ, or ṣuḥba, are very rarely used in our published Arabic-script documents; and words for partners, such as ṣāḥib (again a common word in Jewish texts), related to ṣuḥba, or sharīk, related to shirka, or the less common qarīn, are used far more often in texts in their nonspecific meanings of friend, spouse, holder, boss, or colleague. To be precise: sāḥib/aṣḥāb and qarīn seem to be used for business partners in our period in Arabic-script texts only a dozen times; one of these, which also contains the only citation in Arabic-script texts from this period of the word ṣuḥba, which is so common a concept in the Geniza, in fact concerns Jewish dealers in large part, so does the only Arabic-script citation from this period of the word qirāḍ. Muqāraḍa, a related word, appears only once as well. Another ṣuḥba-related word, aṣḥābuna, the frequent word for a community of Jewish merchants in the Geniza, does not appear at all; and ibḍāʿ or biḍāʿa, canvased as Muslim alternatives to ṣuḥba, only appear (as biḍāʿa) meaning “merchandise.” Shirka appears as meaning a formal commercial partnership in 15 16 17

P.Berl.Arab. ii 66; P.RagibEdfou 1; P.Vind.Arab. i 12, 15, 26. P.Vind.Arab. i 18, 21, 22, 23. P.Heid.Arab. ii 39; P.Vind.Arab. i 60.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

180

wickham

what may be the first Abū Hurayra text, from 250/864, as I said, but otherwise only twice more; its near-synonym, khulṭa, does not appear with this meaning at all. Muʿāmala, an unwritten contract in the Geniza, simply means “relationship” in the Arabic-script texts, although once or twice at least it refers to a commercial relationship that is unspecified. Nor is this simply a question of a different vocabulary. No other contract or partnership words appear in their place.18 So we have here a paradox. Far from the Geniza letters showing a predominance of informal relations, as Goitein thought, they actually show a world more clearly marked by legal and quasi-legal agreements than do the sources for Muslims and Christians in Egypt. But this does not mean that there is any significant religious difference here. The fatwās from the central and western Mediterranean, which make much of qirāḍ, and a certain amount of shirka, too, make an argument of that type impossible to sustain. What we are looking at is scale. There are some large-scale operators in the Arabic-script texts from Egypt, it is true, but most of the letters I have been discussing are for commercial activities on a much smaller level, a few dīnār each. Abū Hurarya was a bigger fish than most, and it is not chance that the only surviving formal shirka contract among these texts was for him and his family, but in the rest of his letters, too, we are witnessing simpler deals, and not much in the way of problems of agency, which are so much a feature of the Geniza letters—indeed, members of Abū Hurayra’s family lived at both ends of his fairly straightforward trade route, from the Fayyūm to Fusṭāṭ. The range of slightly different formalized structures, which is very clear in the Geniza, was simply not necessary in this sort of environment. The standard words in English for formal trading are often ordinary words that have gained a specialized meaning, like “company”; so were many of the parallel words in medieval Arabic, such as ṣuḥba. Merchants picked up words like this, and as their trading grew in scale, they became formalized. Of the

18

Ṣāḥib/aṣḥāb and qarīn seem to be used for business partners in our period in Arabic-script texts in only P.SijpesteijnProfit (a. 901); P.Khalili i 8; cpr xvi 16; P.Heid.Arab. ii 14, iii 46; P.Cair.Arab. 392; P.Berl.Arab. ii 33; P.Vind.Arab. i 7, 16, 35, 45–46. For ṣuḥba, P.Heid.Arab. iii 46. For qirāḍ, P.GenizahCambr. 77. For muqāraḍa, P.Berl.Arab. ii 38. For aṣḥābuna, see Goldberg, Trade and institutions 38–39, 139–143 and passim. For ibḍāʿ or biḍāʿa, see most recently Cohen, Maimonides and the merchants 72–73. For muʿāmala (cf. Goldberg, Trade and institutions 155–157), see P.Marchands ii 40; P.Hamb.Arab. ii 45; and Chrest.Khoury i 41 (the least generic); the only texts that refer to some form of a commercial relationship. Sharīk (partner) is used above all to mean “co-owner” in the Arabic-script corpus, outside religious contexts (see below). Shirka after 250/864 reappears with the plausible meaning of “commercial partnership” in P.Cair.Arab. 291; and cpr xvi 22.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

informal and formal trading associations in egypt and ifrīqiya

181

words I have mentioned, perhaps only qirāḍ had a single meaning in the sources I have looked at; here, the jurists probably made it into a legal word par excellence early on. Even shirka, which indeed means “company” today, meant any kind of association, including the co-ownership of land, in the period we are looking at; sharīk, for its part, is in Arabic-script texts most often used to characterize God, who “has no associate.” These, and even more so the others, only gained their formalized commercial meanings slowly. I guess that, if we had commercial documents from Baghdad from the ninth and tenth centuries, we might find them already used by the great merchants of the time; indeed, they may be used in Iraqi narrative or normative texts I have not looked at. But they belong to the world of large-scale trade. This is the world that the Geniza tells us the most about. To reiterate, the merchants documented in it were larger-scale operators than anyone in twelfth-century Italy, and indeed I have only found one merchant operating on as large a scale anywhere in medieval sources for this period, the amīr Amīnaddīn at the end of the twelfth century, who traded between Cairo and Damascus. We have three letters from his subordinates that do not use contract words but, again reflecting scale, are among the few Arabic-script texts to use aṣḥāb to mean “business partners.”19 It was also the world the Ifrīqiyan fatwās theorized, as of course the jurists thought in formal terms, even if their mental structure was slightly simpler. But smaller traders, who operated more locally, normally with the help of family members as well, did not need the same degree of structure. And that smaller scale was also, of course, the scale on which most commerce operated, in Egypt or anywhere else, in every period. Informality was in that sense, too, by no means an Islamic characteristic, but rather a characteristic of face-to-face communities, including commercial ones, everywhere.

Bibliography Sources The Arabic Papyrology Database, https://www.apd.gwi.uni‑muenchen.de/apd/project​ .jsp. Ibn al-ʿAṭṭār, Formulario notarial y judicial andalusí [Kitāb al-wathāʾiq wa-l-sijillāt], trans. P. Chalmeta and M. Marugán, Madrid 2000.

19

P.Vind.Arab. i 44, 45–46 (aṣḥāb).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

182

wickham

Studies Ackerman-Lieberman, P.I., The business of identity, Stanford 2014. Brockopp, J.E., Muhammad’s heirs, Cambridge 2017. Cohen, M.R., Maimonides and the merchants, Philadelphia 2017. Goitein, S.D., A Mediterranean society, vol. 1, Berkeley 1967. Goldberg, J.L., Trade and institutions in the medieval Mediterranean, Cambridge 2012. Haque, Z., Landlord and peasant in early Islam, Islamabad 1977. Johansen, B., The Islamic law on land tax and rent, Beckenham 1988. Kennedy, H., Landholding and law in the early Islamic state, in J. Hudson and A. Rodríguez (eds.), Diverging paths?, Leiden 2014, 159–181. Khalilieh, H.S., Admiralty and maritime laws in the Mediterranean Sea (ca. 800–1050), Leiden 2006. Lagardère, V., Histoire et société en occident musulman au Moyen Age, Madrid 1995. Pryor, J., The origins of the commenda contract, in Speculum 52 (1977), 5–37. Rittenberg, R.M., Gharar in post-formative Islamic commercial law, Phd diss., University of Pennsylvania 2014. Udovitch, A.L., At the origins of the western commenda, in Speculum 37 (1962), 198–207. Udovitch, A.L., Partnership and profit in medieval Islam, Princeton 1970. Younes, K.M., Textile trade between the Fayyūm and Fustāt in the iiird/ixth century according to the Banū ʿAbd al-Muʾmin archive, in A. Regourd (ed.), Documents et histoire: Islam, viie–xvie siècle, Geneva 2013, 313–334.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 10

Good Governance in Theory and Practice: Comparing Abū Yūsuf’s Kitāb al-Kharāj with Papyri Petra M. Sijpesteijn

The caliph (al-Rashīd), may the Almighty God support him, wishing to keep injustice away from his subjects and promote their wellbeing, asked me to produce a comprehensive book on land taxes, tithes, alms, poll-taxes, etc., and other things that require supervision and action.1

∵ Muslim rule was given shape at the caliph’s court, in the army headquarters, and in chanceries, but also on traders’ ships, behind the desks of historians and jurists, and in the offices of city planners, local administrators, and taxcollectors.2 In his wide-ranging studies, over many years, Hugh Kennedy has shown us that these myriad social and cultural facets, operating at the microand the macrolevel, from top down and bottom up, need to be studied together to understand the workings of the caliphate, both at its highest intellectual level and in its humdrum day-to-day implementation. This contribution examines, in that tradition, the interaction between policy made at the caliph’s court and its execution on the ground through the lens of good governance, in particular in fiscal policy and administration.

1 Abū Yūsuf, Kitāb al-Kharāj 3. Author’s translation. E. Fagnan translated the text into French (Abou Yousof Yaʿkoub, Le livre de l’impot foncier). B. Lewis has translated the introduction into English (Islam), and A. Ben Shemesh provides an English translation of excerpts (Abū Yūsuf’s Kitāb al-Kharāj). 2 This work was supported by the European Research Council under grant number 683194. I would like to thank Peter Webb, Caterina Bori and Cecilia Palombo, with whom I discussed some of the points raised in this chapter, I would also like to thank Mehdy Shaddel for his thoughtful comments on an earlier version of this work. Any remaining mistakes are, obviously, my own.

© Petra M. Sijpesteijn, 2022 | doi:10.1163/9789004525245_011

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

184

sijpesteijn

Abū Yūsuf (d. 182/798), whose words are cited above, presents taxes as one of the ruler’s essential tools for creating a flourishing society. And the effectiveness of that tool depends upon a properly instituted and attentively run tax-collection system. Thus, when the caliph asked him to write a book on how taxation should be organized, Abū Yūsuf not only included detailed legal information about fiscal categories, rates, and calculations, but he also paid much attention to the morals and deportment of a proper fiscal administration: how tax collectors should behave, taxpayers be treated, their grievances addressed, and, above all, how rulers should make sure that the instruments in place to do so are functioning well and the conditions of a successful system realized. How were the ideas developed by Abū Yūsuf in his Kitāb al-Kharāj (Book on taxation), which were clearly an attempt to constitute a coherent, integrated fiscal policy, implemented on a day-to-day basis? To what extent were these prescriptions, produced at the court and directed at the caliph, made real on the ground throughout the Muslim empire? To address this question I start with an analysis of the main points of Abū Yūsuf’s advice. These will then be compared with three papyri from Egypt, dating from the second/eighth to third–fourth/ninth–tenth century, that deal with tax collection and other aspects of the administration in order to explore the relationship between the theory of good governance and the practice of effective government.3

1

Abū Yūsuf’s Three Principles of Just Rule

Having been trained in Kufa by the jurist Abū Ḥanīfa (d. 150/767), Abū Yūsuf moved to Baghdad to work in the capital’s judiciary sometime after his teacher’s death. He is considered one of the three founders of the Ḥanafī school of law.4 Caliph Hārūn al-Rashīd (r. 170–193/786–809) appointed him the (first) chief qadi of Baghdad. He also asked him to produce a statutory work of government, which became known as the Kitāb al-Kharāj (Book of taxes). The book consists of 28 questions that the caliph allegedly posed to Abū Yūsuf, together with the jurist’s answers, preceded by an introduction addressing the caliph.

3 See Bosworth, who stated that mirrors for princes “often reflect a theoretical rather than a practical state of affairs”; Early Arabic 25. This study is part of a larger project connecting ideas of social justice as expressed in mirrors for princes and other Arabic literary works as a paradigm to understand Arabic request letters preserved on papyrus; Sijpesteijn, Righting wrongs ch. 4 Theory and practice of providing social justice. 4 Melchert, Formation.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

185

Kitāb al-Kharāj is the first work of government by a religious scholar and signals a significant step toward bringing the religion of Islam and pre-Islamic Persian and Greek ideas on government together.5 Abū Yūsuf includes a large number of ḥadīths to support his assertions concerning just government and fair fiscal management, as traditions became a more important source of authority.6 For him, the main purpose of government is to serve the common welfare (al-maṣlaḥa), with the ruler being responsible for deciding how this is to be achieved.7 In laying out the assumptions and norms of good governance, Abū Yūsuf draws up rules for the conduct of the ruler and the workings of the polity. Despite its title, and in line with the foundational and comprehensive nature of its ambitions, Abū Yūsuf’s treatise covers much more than taxes, dealing also with diverse matters of administration, including the treatment of prisoners and non-Muslims. The introduction falls in the category of advice literature, using explicit terms known from this genre, such as “I recommend you” (ūṣiyuka).8

5 Other works of wisdom literature offering advice to rulers were produced at the caliphal court, with which Abū Yūsuf’s introduction can be compared. ʿAbd al-Ḥamīd al-Kātib’s (d. 132/ 750) letter dating to 129/746–747 on behalf of Caliph Marwān (r. 127–132/744–750) to his son ʿUbayd Allāh has been identified as the earliest piece of advice literature. See also the extensive letter that Ṭāhir b. al-Ḥusayn (d. 207/822) wrote to his son ʿAbdallāh in 206/821 (cf. Bosworth, Early Arabic). For these early waṣīyas (testaments), see Marlow, Hierarchy 118–128. More extensive works of advice literature or mirrors for princes should include Ibn alMuqaffaʿ’s (d. 139/756) Risāla fī l-ṣaḥāba, which contains recommendations for the caliph and his interactions at court. A more systematic work, including information on the organization of the (fiscal) administration, was written by Qudāma b. Jaʿfar (d. ca. 337/948), entitled Kitāb al-Kharāj wa-ṣināʿat al-kitāba (cf. Heck, Construction). The genre of Adab al-kātib, explored for example by Ibn Qutayba (d. 276/889), also contained advice on good governance, see Gutas, Classical Arabic; Qutbuddin, Arabic oration 409–413. In these works the need to apply a fair fiscal policy and closely supervise fiscal agents figures as well. In addition, there are some additional works dedicated to taxation. These are, however, mainly collections of traditions on taxation without providing much discussion or additional information. See, for example, Kitāb al-Kharāj by Yaḥyā b. Ādam (d. 203/818) and Kitāb al-Amwāl by Abū ʿUbayd (d. 224/838). 6 “And I wrote down for you sound ḥadīths with which you can incite and which support what you would like to be done, God willing” (wa-katabtu laka aḥādīth ḥasana fīhā targhīb wataḥḍīḍ ʿalā mā saʾalta ʿanhu mimmā turīdu al-ʿamal bihi in shāʾa allāh). Abū Yusūf, Kitāb al-Kharāj 6. 7 Heck, Role. 8 Abū Yusūf, Kitāb al-Kharāj 5. Echoing the “testaments” in which a ruler offered counsel (waṣīya) to his successor or appointee. See above, n. 5. The Kitāb al-Kharāj is, however, generally not included in discussions of advice literature. At best it is considered among the works on taxation that function as didactic treatises; R. Sellheim and D. Sourdel, Kātib.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

186

sijpesteijn

In this introduction Abū Yūsuf describes his foundational values and recommendations for a ruler seeking to serve the public good by establishing a just and flourishing society. These recommendations inform the whole book but are most explicitly formulated in the introduction, which is the focus of this study as well. Abū Yūsuf ascribes an important role to the ruler in creating a virtuous (i.e., pleasing to God) society, which prospers and benefits everyone living in it. God appoints the ruler to look after his subjects (al-umma wa-l-raʿīya), investing him with tremendous power. Abū Yūsuf emphasizes the greatness of this task by repeating it in four different ways in one sentence: “God made you their shepherd, entrusted you with them, visited them on you and appointed you over their affairs.”9 The potential rewards are great, but so too are the consequences should he fail.10 The core value guiding the ruler in the execution of this commission is piety (al-taqwā), for only work that God approves of is effective, and it is to God that the ruler is ultimately answerable.11 A recurring theme is that of the ruler appointed by God as a guardian of His dependents. The ruler works for God, “like a shepherd who renders service to his master.”12 In other words, the ruler is only temporarily responsible for the people in his realm. He merely works under contract, deputizing for their true and permanent Lord.13 Although not their ultimate master, the shepherd is nevertheless responsible for his flock while they are under his care. A shepherd who is negligent and allows his sheep to suffer or even perish will be held to account, paying perhaps even with his own life. A shepherd who, on the other hand, makes his flock thrive (aṣlaḥa) is lucky and will be generously rewarded

9 10

11 12

13

Abū Yusūf, Kitāb al-Kharāj 3. “O prince of the believers, God, may He be thanked, invested you with tremendous power, whose rewards are great and whose punishments are most severe” ( yā amīr al-muʾminīn inna allāh wa-lahu al-ḥamd qad qalladaka amran ʿaẓīman thawābuhu aʿẓam al-thawāb waʿiqābuhu ashadd al-ʿiqāb). Abū Yusūf, Kitāb al-Kharāj 3. “Do not waste what God invested you with of the affairs of this umma and flock” ( fa-lā tuḍayyunanna mā qalladaka allāh min amr hādhihi al-umma wa-l-raʿīya). Abū Yusūf, Kitāb al-Kharāj 3. A long passage stretching over two pages in the printed edition reminds the ruler of the inevitability of “answering” to God; Abū Yusūf, Kitāb al-Kharāj 3–4. “For shepherds (i.e., rulers looking after people) are working for their Lord in the way that a shepherd works for his master” (inna al-ruʿāt muʾiddūn ilā rabbihim mā yuʾaddī al-rāʿī ilā rabbihi). Abū Yusūf, Kitāb al-Kharāj 3. Cf. “I council you, O prince of believers, to preserve what God entrusted to your care and to protect that over which God has appointed you as shepherd” (ūṣiyuka yā amīr al-muʾminīn bi-ḥifẓ mā istaḥfaẓaka allāh wa-riʿāya mā istarʿāka allāh). Abū Yusūf, Kitāb al-Kharāj 5. The sense of deputy is also conveyed in the form of address, “God’s shepherd” (rāʿī allāh), used in panegyrics to caliphs, such as the one where al-Farazdaq (d. 110/728 or 112/730) addresses al-Walīd (r. 86–96/705–715) thus; Crone and Hinds, God’s caliph 5, 9.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

187

by God.14 The ruler is thus charged by God to watch over His people and, while personally liable to Him for their safety and well-being, is also able to draw upon God’s help in the fulfilment of his mandate.15 The image of the ruler as a shepherd entrusted by God with the affairs of his subjects, so that they benefit from his leadership and care, is one that Abū Yūsuf continues to use throughout his introduction.16 As will become clear, it is also popular in the papyri. Looking after people on God’s behalf requires hard work and constant attention,17 as well as the ability to withstand criticism.18 The rewards, however, when the ruler acts properly are significant. In this context, Abū Yūsuf introduces the second, perhaps even the main, pillar on which 14

15 16

17

18

“The wasteful shepherd is liable for what perished while in his care […] but the one who makes (the flock) prosper is happiest because of that and God will reward him many times more than that which he provided to Him” ( fa-inna al-rāʿī al-muḍayyiʿ yaḍmanu mā halaka ʿalā yadayhi […] wa-idhā aṣlaḥa kāna asʿad min hunālika bi-dhālika wa-wafāhu allāh aḍʿāf mā wafā lahu). More sentences with the same sense follow: “Thus watch out that you do not lose any of your flock so that their Lord will redeem their compensation from you, and will cause your loss through what you lost of your reward”; “and what you did for the benefit of those whose affairs God entrusted to you will be to your credit and what you lost of it will be to your debit. So do not forget to look after the affairs of those whom God entrusted to you and you will not be forgotten. Do not be negligent of them and of that which improves their position and you will not be forgotten”; “An unjust shepherd causes the destruction of the flock.” Abū Yusūf, Kitāb al-Kharāj 5. “I ask God, O prince of the believers […] that He not abandon you to yourself in any of your responsibilities.” Abū Yusūf, Kitāb al-Kharāj 5. It also occurs in the traditions quoted at the end of the introduction synonymous with rulers. E.g., Abū Yusūf, Kitāb al-Kharāj 12, 14. The ruler as shepherd in temporary charge of the owner’s flock similarly appears in later advice literature, although not as elaborately and unapologetically. Cf. in Ṭāhir’s letter to his son ʿAbdallāh, “Know that you have been placed in your governorship as a custodian of valuables, a watchman and a shepherd; the people in your realm are only called ‘your flock’ (raʿīyatuka) because you are their shepherd (rāʿiyuhum) and their overseer.” Bosworth, Early Arabic 37–38. “Day and night you are to work for a large group of people whom God entrusted to you” (qalladaka amr hādhihi al-umma fa-aṣbaḥta wa-amsayta wa-anta tabnī li-khalq kathīr qad istarʿāhum allāh). Abū Yusūf, Kitāb al-Kharāj 3. That an ideal ruler works constantly also comes up in some of the traditions that Abū Yūsuf includes at the end of the introduction. Abū Bakr (r. 11–13/632–634) describes himself: “I did not fall asleep and dream, nor did I make assumptions and err.” And tells his successor, ʿUmar (r. 13–23/634–644): “God is entitled to ask you to finish at night what you did not get to during the day and vice versa.” Abū Yusūf, Kitāb al-Kharāj 11. Cf. “I urge you, O prince of the believers, to guard what God entrusted you and the subjects whom God put under your protection, and to do so taking into account only Him and work only for Him in this respect” (ūṣīka yā amīr al-muʾminīn bi-ḥifẓ mā istaḥfaẓaka allāh wa-riʿāyat mā istarʿāka allāh wa-an lā tanẓura fī dhālika illā ilayhi wa-lahu). Abū Yusūf, Kitāb al-Kharāj 5. Abū Yūsuf encourages the ruler “not to fear anyone’s rebuke concerning (what you do for) God” (wa-lā takhaf fī allāh lawma lāʾim); Kitāb al-Kharāj 4.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

188

sijpesteijn

a ruler’s success in God’s eyes rests, namely, how his subjects experience his rule: “On the day of judgement, the happiest shepherd before God is the one who has made his flock happy.”19 In other words, people need a ruler to look after them, to make them prosper, but they are also the arbiters of whether he has been successful in doing so. They can speak up to the ruler, knowing that their opinion carries weight at the final reckoning.20 So how can the ruler make sure that his subjects flourish, satisfying both God and the flock entrusted to him? Abū Yūsuf has the answer. If the key to prosperity is good governance, the key to good governance is first and foremost justice—upholding what is right and opposing what is unjust—especially in matters related to taxation.21 This means that rulers have to establish what taxes people are liable for and to clarify any confusion concerning their (fiscal) obligations.22 In addition, they should support justice by ensuring that bound19

20

21

22

Fa-inna asʿad al-ruʿāt ʿinda allāh yawm al-qiyāma rāʿin saʿadat bihi raʿīyatuhu. Abū Yusūf, Kitāb al-Kharāj 3. The admonition is repeated at the end of the introduction as a saying that ʿUmar b. al-Khaṭṭāb used in a letter to Abū Mūsā; Abū Yusūf, Kitāb al-Kharāj 14–15. Abū Yūsuf is, however, very clear that subjects should not aim to remove or replace a ruler no matter how unjust he might be, however, but bear his rule and obediently await his punishment by God; Kitāb al-Kharāj 9, 10, 14. “Let justice prevail in matters that God appointed you over and entrusted you with, if even for one hour only” ( fa-aqim al-ḥaqq fīmā wallāka allāh wa-qalladaka wa-law sāʿa). Abū Yusūf, Kitāb al-Kharāj 3. “[A]and I hope, that if you act according to what is evident in it (this book), that God will provide you with abundant fiscal income without having to resort to oppression of any Muslim or non-Muslim” (wa-innī arjū in ʿamalta bimā fīhi min al-bayān an yūfira allāh laka kharājaka min ghayr ẓulm muslim wa-lā muʿāhad); “the people’s prosperity depends on […] removing oppression and injustice from them caused by uncertainty about their duties” ( fa-inna ṣalāḥahum bi […] wa-dafʿ [rather than rafʿ as the edition has?] al-ẓulm ʿanhum wa-al-taẓālum fīmā ishtabaha min al-ḥuqūq ʿalayhim. Abū Yusūf, Kitāb al-Kharāj 6). A ruler’s obligation to uphold justice and combat injustice also figures prominently in the traditions quoted at the end of the introduction. The main topic of these traditions is surely the need for any Muslim to lead a pious life in anticipation of the upcoming Last Day. The first traditions cited that relate to governance, however, discuss that just rulers will be rewarded at the last judgement, while unjust rulers can expect nothing good from it. This theme is prevalent throughout; see Abū Yusūf, Kitāb alKharāj 8–9, 10, 11, 12–13, 14–15. As an exemplary caliph, the major (or only) praiseworthy characteristic listed for ʿUmar b. ʿAbd al-ʿAzīz (r. 99–101/717–720) is that he “addressed people’s complaints of injustice” (radda al-maẓālim ilā ahlihā) to such an extent that it became his only concern, overshadowing care for himself; Abū Yusūf, Kitāb al-Kharāj 16. “God in His largesse and mercy made rulers (holders of authority) His representatives on earth. He gave them insight to clarify to the people what is obscure to them in matters concerning them and to elucidate that which they find confusing concerning their obligations” (wa-inna allāh bi-mannihi wa-raḥmatihi jaʿala wulāt al-amr khulafāʾ fī arḍihi wa-jaʿala lahum nūran yuḍīʾu li-l-raʿīya mā uẓlima ʿalayhim min al-umūr fīmā baynahum wa-yubayyinu mā ishtabaha min al-ḥuqūq ʿalayhim). Abū Yusūf, Kitāb al-Kharāj 5. “And

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

189

aries are respected23 and that people receive what they are entitled to through rigorous and transparent processes.24 A ruler should treat everyone equally— those who are close to him as well as those who are distant.25 Looking for models to emulate past practices of righteous people (al-qawm al-ṣāliḥūn) is therefore of the utmost importance.26 Conversely, an unjust ruler, or one who relies on untrustworthy and corrupt counsel, brings his community into ruin.27

23

24 25 26

27

to remove from them confusion about what they assume are their obligations” (wa-dafʿ ʿanhum [rather than rafʿ as the edition has?] […] wa-l-taẓālum fīmā ishtabaha min alḥuqūq ʿalayhim). Abū Yusūf, Kitāb al-Kharāj 6. I understand this passage as relating directly to the topic of the book identified in its title: taxes. Other scholars have used this passage to argue for Abū Yūsuf’s insistence that the caliph relies on and governs communally with ʿulamāʾ, Zaman, Caliphs 17. Or conversely, that caliphs have religious insight through which they can make decisions independent of religious scholars on matters of religion, El-Hibri, ʿUmar 763–764. Iqāmat al-ḥudūd. Abū Yusūf, Kitāb al-Kharāj 5, 6. Following Intisar Rabb’s clarification of the terms, I apply here a broader interpretation of “upholding boundaries” than the usual translation, “impose ḥudūd punishments.” Cf. in Lewis’s translation: “enforcing the penalties for offenses as laid down in the holy law”; “enforcement of the legal penalties [ḥudūd].” Lewis, Islam 155. Cf. Rabb, Doubt. Radda al-ḥuqūq ilā ahlihā bi-l-tathabbut wa-l-amr al-bayyin. Abū Yusūf, Kitāb al-Kharāj 5. Wa-aʿjal al-nās ʿindaka amr allāh sawāʾan al-qarīb wa-l-baʿīd. Abū Yusūf, Kitāb al-Kharāj 4. “Reviving the customs which righteous people maintained is the most important thing for reviving customs is a good thing, something that remains and does not abide” (waiḥyāʾ al-sunan allatī sannahā al-qawm al-ṣāliḥūn aʿẓam mawqiʿan fa-inna iḥyāʾ al-sunan min al-khayr alladhī yaḥyā wa-lā yamūtu). Abū Yusūf, Kitāb al-Kharāj 5. Zaman interprets sunan al-qawm al-ṣāliḥīn more specifically as the practices of past caliphs, especially the rāshidūn and ʿUmar b. ʿAbd al-ʿAzīz, whose deeds and opinions are quoted throughout the book; Zaman, Caliphs 17. See, for example in the traditions at the end of the introduction, those where Abū Bakr urges ʿUmar as he appoints him as his successor to “preserve my admonition,” emphasizing that he himself followed “the way of the one who preceded me”; Abū Yusūf, Kitāb al-Kharāj 10–12. The recommendation to examine practices of past generations for recommendable practices can be found in other mirrors. See in Ṭāhir b. al-Ḥusayn’s (d. 207/822) letter to his son ʿAbdallāh (d. 230/844): “Take careful note of everything which you observe in this present world, and consider the examples of your predecessors, the rulers and leaders of past centuries, and of nations now disappeared from the earth.” Bosworth, Early Arabic 40. In ʿAlī’s (r. 35–40/656–661) letter to Mālik b. al-Ḥārith al-Ashtar al-Nakhāʿī (d. 38/658): “Do not abandon a beneficial custom (sunna ṣāliḥa), which has been maintained by good men before you, has secured concord (ulfa), and worked for the welfare of the common people; and do not introduce a new custom which violates in any way the ancient traditions of justice that have been established before you, lest the reward go to him who founded them and you be guilty of sin by destroying part of them.” Marlow, Hierarchy 124 n. 45. Wa-jawr al-rāʿī halāk li-l-raʿīya wa-istiʿānatuhu bi-ghayr ahl al-thiqa wa-l-khayr halāk li-lʿāma. Abū Yusūf, Kitāb al-Kharāj 5.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

190

sijpesteijn

To summarize: Abū Yūsuf’s three main principles of good governance in relation to taxation start with the observation that a ruler is appointed by God and is accountable to Him for his actions (“efficacy in work is by God’s will,” fa-inna al-quwwa fī l-ʿamal bi-idhn allāh).28 Secondly, the ruler’s relationship with his subjects is that of a shepherd and his flock: he is a temporary guardian of his people on behalf of their Lord. The ruler has to care for his subjects and is personally liable for their well-being: if they flourish he will be rewarded, if they suffer he must carry the blame. And thirdly, it is by establishing and maintaining a just society that a ruler creates the conditions for his subjects to thrive. In terms of (fiscal) administration this means instituting clear rules, ensuring their proper execution, and minimizing abuses by employing honest and upright agents. In short, a just ruler creates a just society, in which his people can flourish, ensuring his own salvation as well as theirs.29 These principles are echoed in the language of papyrus letters written to and by government officials at the same time in Egypt, to which we will turn next.

2

Providing Justice to the Flock

The three documents from Egypt, with which Abū Yūsuf’s principles of sound governance will be compared, date from the second/eighth to fourth/tenth centuries. They are all written on papyrus. One letter is exchanged between government officials, the other two are written to a government agent by individuals outside the administration. One originates in the Fayyūm oasis, another in the Upper Egypt region of Asyūṭ. The provenance of the third is unknown, but it most likely also comes from the villages outside the capital Fusṭāṭ in Middle Egypt, Upper Egypt or the Fayyūm oasis. The first letter dates to the second/eighth century and was sent to the Fayyūm.30 Al-Layth son of Muḥammad and Muḥammad son of ʿAbdallāh, not known outside this papyrus, write to a fiscal agent in the Fayyūm. Egypt’s gov28 29

30

Abū Yusūf, Kitāb al-Kharāj 3. In this sense I follow more closely Paul Heck’s interpretation (Abū Yūsuf) that Abū Yūsuf’s priority is that the ruler is just, rather than that of Qasim Zaman (Caliphs), who assumes that Abū Yūsuf’s main point is to call for a ruler who relies on religious scholars, or Tayyib El-Hibri (ʿUmar), who argues that Abū Yūsuf considers a just society to be the result of the rulers being in possession of special religious insights and knowledge. Later sources such as Ibn Taymiyya (d. 728/1328, in al-Siyāsa al-sharʿīya fī iṣlāḥ al-rāʿī wa-l-raʿīya) also refer to the interdependent well-being of ruler and ruled as shepherd and flock. I would like to thank Caterina Bori for bringing this to my attention. Diem, Arabische Briefe des 7, no. 26.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

191

ernor (amīr) has made the two men responsible for the taxes in the Fayyūm and related districts (ll. 3–4 wajjahanā ʿalā kharāj kūrat al-Fayyūm wa-sāʾir aʿmālihā), and they write in response to a complaint from the oasis’s inhabitants about some local tax collectors. The two men have been ordered to gather all local tax collectors together and examine what has to be done concerning the complaint. The addressee of the letter, apparently one of the collectors accused by the taxpayers, is told to “come to us as soon as you set eyes on our letter, not delaying a single minute” (ll. 9–11 fa-idhā [naẓa]rta fī kitābī hādhā fa-iqdam ʿalaynā [wa-lā] tataʾakhkhar sāʿatan in shāʾa allāh). The letter opens and closes with the absolute professional minimum of one blessing for the addressee. Unlike most private and business letters from the period, including those exchanged between government agents, which generally discuss multiple unrelated topics together in one epistle, this letter, like the other two discussed below, deals with one subject only.31 Singling out one topic in this way obviously puts emphasis on it and heightens the urgency of the message. Despite the letter’s brevity, however, al-Layth and Muḥammad are careful to exploit the formality of the language for a certain rhetorical effect, in particular to underline the social hierarchy at work and the distance between themselves and the addressee. The two refer to themselves as having been appointed by the governor to manage fiscal affairs in the district, whereas the addressee fulfills a subordinate function. Their appointment might be temporary, directed at resolving the complaint at hand, but it might also be permanent, undermining the authority and independence of the addressee and his colleagues. The hierarchical relationship between the senders and addressee is confirmed by the use of an imperative in their request formula, which is only softened by the conditional “God willing.” No blessings for the addressee have been added when he is addressed in the letter’s body, emphasizing the curt tone of the letter. The letter’s language also connects directly to the concerns that Abū Yūsuf raises in his admonition to Caliph al-Rashīd, even using the same language that occurs in the introduction to the Kitāb al-Kharāj. Al-Layth and Muḥammad write that the amīr appointed them over fiscal affairs in the Fayyūm and that “he urged us to be intimately informed about the subjects (al-raʿīya), to make them flourish […] and to remove from them the damaging conditions that have been mentioned to me” (ll. 5–6 awṣānā bi-taʾalluf al-raʿīya, wa-istiṣlāḥihim wa- […] wa-dafʿ al-muʾīdhiyāt allatī dhukirat fīmā jāʾanī ʿanhum). Al-Layth and Muḥammad refer to the subjects as a “flock.” In the cascade of perfect moral

31

Grob, Documentary 90.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

192

sijpesteijn

governance, the amīr, himself an agent of the caliph, has instructed his representatives to secure the well-being of the subjects entrusted to their care (now that they have been made responsible for the organization of their tax payments) and to remove anything that stands in the way of achieving this. The procedure described in the letter also corresponds precisely to a concern that Abū Yūsuf broaches in relation to the need to provide due care for the caliph’s subjects regarding the proper organization of their tax payments. That is to say, the emphasis on the ruler’s responsibility to be just and to promote justice, for which trustworthy and righteous agents are essential. The procedure that figures in this letter is the well-known process of handling complaints about abusive officials by a state supervised office, known as maẓālim.32 Mainly studied as an instrument of the highest government echelons, namely the caliph’s court, similar procedures were obviously in place lower down in the administration as well.33 The maẓālim is an important instrument with which rulers can correct abuse and further the cause of social justice in their realm. In this case an unnamed group of taxpayers in the Fayyūm had apparently complained to the governor about the local tax collectors. The amīr delegates two officials, the senders of our letter, al-Layth and Muḥammad, to look into the matter and take whatever measures are necessary. As a first, fact-finding step, al-Layth and Muḥammad call in the tax collectors, including the addressee, apparently one of the tax collectors in question. That this process of examining accusations of abuse by government officials is part of a larger project of safeguarding taxpayer welfare and addressing mistreatment by state agents is clear from the letter’s formulation quoted above. The second letter, containing a complaint to an amīr by monks from a monastery in Asyūṭ about the behavior of certain of the amīr’s officials and an extortionate local administrator, is another example of the maẓālim procedure.34 The letter can be dated to the third/ninth century. The monks describe themselves as “weak and poor” (ll. 9; 20; 21 wa-naḥnu ḍuʿafāʾ wa-masākīn), relying not on trade, agriculture, or any other form of income, but dependent on alms. When the amīr’s agents passed by their monastery, the monks explain, they received them hospitably, as they treat all guests, but after the officials had

32

33

34

Amedroz, Mazalim; Darling, History 79–80; Lev, Administration 202–228; Nielsen, Maẓālim; Sourdel, Vizirat ii, 640–648; Tillier, Maẓālim; van Berkel, Abbasid maẓālim; van Berkel, Embezzlement. Tillier examined the role of qadis in provincial maẓālim courts (Qāḍīs). I include a discussion of maẓālim procedures in the papyri in my forthcoming book, Sijpesteijn, Righting ch. 2. Vanthieghem, Violences.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

193

eaten, they seized the monks and brought them to the house of the local administrator. The monks were beaten and received a hefty fine of several dīnār.35 They sold their clothes and the donkey they relied upon for raising water from the well, which they had received as a donation, but still the monks were not able to raise enough money to pay the fine imposed upon them. They now ask the amīr to take action against the perpetrators, the local officials and their leader, the head of the district (ll. 16–17 hāʾulāʾi al-aʿwān wa-ṣāḥibihim al-wālī fī al-nāḥiya), who has been taking monthly payments from the monks amounting to three dīnār per year, which they are also unhappy about.36 Again both the language and the procedure described in the letter point to the concerns Abū Yūsuf raises in his advice to the caliph. This letter, too, features a process for complaints against government officials handled by a state office, the above-described maẓālim. It documents one step earlier in the process, involving a complaint to the governor formulated as a petition using common expressions and polite request formulae.37 The duty incumbent upon the amīr to address abuse at the hands of his own appointees is raised explicitly in the letter: the monks “seek protection from God and the amīr’s justice against his agents and their leader, the district’s governor,” who has been extorting money “unjustly, as a hostile and unfair act.”38 As in the previous letter, the contents of the letter are confined to the complaint at hand. The opening and closing encomia are much more extensive in this letter, and throughout the letter any reference to the addressee, the amīr, is accompanied by elab35 36

37

38

ll. 11–12 akhadhūnā bi-ajmaʿinā wa-sāqūnā ilā dār al-amīr fa-ḍarabanā ḍarban mubarriḥan wa-gharimūnā jumlat danānīr. ll. 18–19 “we pay him every month a quarter dīnār with a total of three dīnār per year, which oppresses us, and constitutes a hostile act against and unfair treatment of us” (wanaḥnu nuʾaddī ilayhi fī kull shahr rubʿ dīnār min thalātha danānīr sanatan ẓulman lanā wa-ʿudwan wa-taʿassufan). L. 4 “we inform the amīr” (nuʿlimu al-amīr); ll. 15–17 “for we rely on God and the amīr, may God prolong his life and ameliorate his position, and if the amīr would be so kind, through his fairness, kindness and justice, for we seek refuge with God and the amīr’s justice to help against these agents and their leader, the district’s manager” ( fa-naḥnu bi-llāh wa-ʿadl al-amīr aṭāla allāh baqāʾahu wa-adāma ʿizzahu fa-in yatafaḍḍala al-amīr biinṣāfihi wa-faḍlihi wa-ʿadlihi fa-in nastaghīru bi-llāh wa-bi-ʿadl al-amīr fī hāʾulāʾi al-aʿwān wa-ṣāḥibihim al-wālī fī al-nāḥiya); l. 21 “if the amīr, may God prolong his life, could please look into our case” ( fa-in rāʾa al-amīr aṭāla allāh baqāʾahu an yanẓura fī amrināʾ); ll. 27– 28 “and let the amīr, may God prolong his life, do something about our complaint being so kind, and being rewarded, God willing” ( fa-raʾy al-amīr aṭāla allāh baqāʾahu fīmā anhaynā ilayhi min amrinā mutafaḍḍilan maʾjūran). For the expression “show me the amīr’s opinion” (ʾari raʾya al-amīr) to formulate polite requests, see Khan, Historical development. Ll. 16–17 fa-innā nastaghīru bi-llāh wa-bi-ʿadl al-amīr fī hāʾulāʾi al-aʿwān wa-ṣāḥibihim alwālī fī al-nāḥiya; l. 19 ẓulman lanā wa-ʿudwan wa-taʿassufan.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

194

sijpesteijn

orate blessings. This fits the purpose of the letter and its rhetoric of placing the petitioned (the addressee) in an elevated position vis-à-vis the humble and hard-pressed petitioners. The letter uses some of the same expressions and words that are attested in Abū Yūsuf’s introduction to his treatise as discussed above. The amīr is appointed by God to dispense justice to his flock. That God is closely involved is clear: “God wants (l. 24 aʿādhā) the amīr, may God prolong his life and make his might endure, to ensure that no one falls victim to an injustice (l. 25 an yaẓlima aḥadan) during his (i.e., the amīr’s) blessed reign and good fortune for all his subjects (literally: “for his entire flock,” l. 25 ʿalā jamīʿ raʿīyatihi).” The amīr’s responsibility to provide justice is invoked throughout the letter. The monks “fled to God and the amīr’s door […] because of what we know of his (i.e., the amīr’s) fairness and the justice with which he treats his subjects (literally: his flock, raʿīyatihi), whether they are close to or far away from him.” The emphasis on treating all subjects justly, whether they are closely connected or unrelated, echoes Abū Yūsuf’s admonition almost word-for-word.39 The monks’ emphasis on their weakness and poverty is echoed in several of the traditions that Abū Yūsuf quotes at the end of his introduction. The forceful advice ascribed to Caliph ʿAlī (r. 35–40/656–661) to his tax collector to make sure that taxes are never so burdensome that the taxpayers’ clothes, food, or any of their working animals have to be sold is strikingly close to the monks’ claims to have had to sell their clothes and their donkey.40 The final letter contains another complaint directed at a government official, this time a wālī, a local administrator. It is dated to the third–fourth/ninth– tenth century. The sender acts as the representative of the community of a local mosque. He complains about the prayer leader (and muezzin?) at the mosque, who has introduced unauthorized additions to the call to prayer and has brought a jar of date wine into the mosque.41 The congregation is so upset that they cannot bear to look at him when he stands in front of them to lead the prayer. The sender urges the wālī to remove the man immediately. In this letter another aspect of a ruler’s responsibility vis-à-vis his subjects is treated, namely, his obligation to ensure that they are properly guided in religious matters and not led astray by poor religious leadership. It takes us

39

40 41

Ll. 13–14 fa-harabnā ʿalā wujūhinā hāribīn ilā allāh wa-ilā bāb al-amīr aṭāla allāh baqāʾahu lammā balaghanā min inṣāfihi wa-ʿadlihi alladhī yassarahu li-raʿīyatihi adnāhum waaqṣāhum. See the emphasis on treating subjects equally, both those who are close to the caliph and those who are more distant discussed above (n. 25). Abū Yusūf, Kitāb al-Kharāj 15–16. Rāġib, Lettres no. 5, ll. 11; 14.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

195

away from Abū Yūsuf’s direct fiscal concerns but compares well with his recommendations in a general sense. Abū Yūsuf does express a concern for moral guidance, moreover, when he warns that a ruler who deviates (from God’s path) causes his subjects to deviate.42 God has appointed the governor to encompass this very purpose: “God, thanks be to Him, made you our governor” (ll. 4–5). And since “prayer is the pinnacle of our religion” (l. 7 wa-l-ṣalāt hiya ʿuẓm dīninā), the stakes are high, which explains the sender’s request that the wālī take urgent action. Indeed, in these matters the governor is for his subjects (literally: his “flock,” l. 5 li-raʿīyatihi) like a father (mithla al-wālid), but also a shepherd, “guarding our affairs and safeguarding our position” (l. 7 li-annaka rāʿī umūrinā wa-munqidh aḥwālinā). Conversely, the subjects leave all matters relating to the proper conduct of prayer to the judgement of the governor, “so that you can do what has to be done” (ll. 6–7 li-taqūma mā yajibu). The extensive opening encomia, as in the previous letter a product of the letter’s function as a petition, consist entirely of references to the risks the governor runs in dealing with the case. If he handles it well, he acts according to the responsibilities inherent in his job description. If he handles it poorly, he risks sinning and wandering from God’s path. They echo Abū Yūsuf’s statement that being a ruler comes with great potential reward, but also great potential punishment, and his wish that God help the caliph to discharge the tasks He has imposed on him. The sender writes: “May God honor you through a successful execution of your work through obedience to Him, and may He preserve your religion for you, protect you from any sin that removes you from God, the Praised and Exalted” (ll. 2–3). The ruler’s responsibility to ensure his appointees behave justly and to address them if they do not underpin the sender’s motivation for writing his letter. The wālī’s responsibility for his appointees, in this case those in charge of religious affairs, is spelled out by the letter’s sender when he writes: “No one is more entitled than you to be in charge of and guard the main mosque […] no one (except you) is to supervise who enters it as muezzin or as prayer leader in Ramaḍān or any other month. (You are in relation to our needs concerning our local mosque) like the amīr (i.e., governor) when he acts for the people of Fusṭāṭ in relation to their main mosque.”43 On the one hand, the sender thus emphasizes the wālī’s sole power over what happens in the mosque and who serves the community there. On the other hand, he points out that with this power comes a responsibility to ensure that those whom the wālī appoints live 42 43

“Do not deviate, for (then) your flock deviates” (wa-lā tazigh fa-tazīghu raʿīyatuka). Abū Yusūf, Kitāb al-Kharāj 3. Ll. 8–10.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

196

sijpesteijn

up to their duties. By ensuring this, the wālī acts properly as someone looking out for the community’s well-being, in this case by seeing to it that their religious caretakers are qualified and acting appropriately. And the degree to which the wālī avails himself of this task, the sender explains, is what the wālī is accountable to God, Who will judge whether the wālī has executed his task well. By informing the wālī of his appointee’s improper behavior and suggesting the wālī undertake action by removing him, the sender is, moreover, he writes, not merely doing his duty but also doing the wālī a favor, since the situation threatens to imperil the wālī’s standing before God. Underlying this “service” provided to the wālī, however, also lies an obvious threat that echoes Abū Yūsuf’s text. The wālī’s subjects are unhappy, the sender writes, they cannot stand the look of the prayer leader. They have, the sender seems to suggest, asked him to approach the wālī about the situation. If the wālī deals with the unsatisfactory situation by removing the prayer leader, he can look forward to God’s judgement as fearlessly as those whom God placed under his care and will consequently be happy. If he does not, however, his subjects, in the person of the letter’s sender and the community he represents, will surely convey their dissatisfaction to their ultimate Lord, with negative consequences for their temporal caretaker, the wālī.

3

Conclusion

Undoubtedly, the caliph’s first interest in obtaining a work on taxation was to increase the yield of the land taxes.44 Abū Yūsuf’s “book on taxes” consequently contains much practical information on what taxes should be paid, by whom, and over what possessions. Successful and profitable tax collection, however, as Abū Yūsuf explains, ultimately depends on good governance. Abū Yūsuf’s Kitāb al-Kharāj indeed deals with many topics other than taxation, as he discusses what good governance entails. There are, however, three themes that stand out in his introduction in relation to what good governance means on a practical level. They concern the ruler having been appointed by God as a temporary caretaker of his subjects. Like a shepherd answerable to the flock’s owner for any sheep that are harmed, the ruler is responsible to God for any of his subjects who suffer under his rule. It is this prospect of being liable when

44

“I hope, that if you act according to my explanations, God will increase for you the yield of your land tax.” Abū Yusūf, Kitāb al-Kharāj 6.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

197

the accounts are settled that Abū Yūsuf emphasizes. The ruler can ensure his flock’s flourishing by providing justice and fighting injustice, especially by relying on just and trustworthy lieutenants. Rather than rising up against a ruler who does not abide by these recommendations, Abū Yūsuf emphasizes, subjects should remonstrate against the shortcomings in his behavior. The ruler should be open to criticism and should put in place mechanisms to address his operatives’ corrupt or abusive behavior. Subjects, on the other hand, should bear a bad ruler with equanimity, safe in the knowledge that their verdict on his rule will be tallied at the last judgement. The three papyri discussed above display a similar image of expectations of just rule among inhabitants of the Islamic empire, showing some of the ways in which disadvantaged subjects, rather than being forced to take the law into their own hands, could turn to government officials for redress against abusive and unjust behavior by administrators. The expectation that the ruler and his representatives ensure a just society by themselves, behaving properly and by addressing complaints of improper behavior, is the basis on which the rationale of all three letters rests. Indeed, the three letters show the office of maẓālim and the process of raising complaints about corrupt or abusive officials in action at the level of the province and its districts.45 The ruler as shepherd protecting his people against enemies and ensuring security and plenty is a common motif in ancient Near Eastern, Egyptian, and Biblical rhetoric.46 The papyri also make extensive use of the terms “flock” (raʿīya) and “shepherd” (rāʿī). The idea of the shepherd being accountable to the flock’s ultimate owner, God, which is of great concern to Abū Yūsuf and is also present in the papyri, is a more specific elaboration. It emphasizes the ruler’s responsibility in adhering to God’s purpose, celebrating God’s beneficence toward rulers whose reigns increase the well-being and happiness of their subjects, and warning against God’s wrath for those whose reigns do not.47 Although just rulers being pleasing to God appear in other administrat-

45

46 47

Literary sources speak of a maẓālim court in Fusṭāṭ during al-Maʾmūn’s second reign at the beginning of the third/ninth century (Tillier, Histoire 33), but our documents show the practice to have been in place already before this. Rudman, Commissioning 524–526; Sperl, Islamic kingship 24. This differs from Abbasid panegyrics praising caliphs as realizing a resurgence of life and prosperity under their rule by offering protection and nourishment to their subjects and whose reigns are marked by God’s approval and support. In other words, due of course to the genre of the panegyric, the caliph is depicted as acting according to God’s will by providing justice and nourishment, while Abū Yūsuf and the papyri appeal to the ruler to act thus, appealing to him with potential rewards and threatening him with punishment. Cf. Sperl, Islamic kingship 31, 34.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

198

sijpesteijn

ive recommendations, it is the combination of a shepherd being responsible to the flock’s Lord that is striking in these four texts. The image of the ruler as a shepherd nourishing and protecting his flock, so that every sheep is happy and flourishes, is not otherwise so present in contemporary Arabic literature on statecraft. It is definitely explored by no one else as extensively as by Abū Yūsuf, although, as Caterina Bori pointed out to me, the imagery reappears in Mamluk scholars such as al-Nuwayrī (d. 733/1333) and Ibn Taymiyya. I was unable to examine how their symbolism works exactly and whether they connect this in the same way that Abū Yūsuf does to God’s overlordship. Abū Yūsuf’s Kitāb al-Kharāj and the papyri were products of a shared system of ideas about good governance and just rule as a precondition of a healthy and prosperous society. Although many of these ideas appear in a general way in many periods and places throughout Near Eastern history, some specificities— such as the God-appointed ruler temporarily in charge of God’s creatures and answerable to Him for their well-being—seem to have been especially popular at the time these texts were produced. Relying on just officials was, moreover, of direct and special concern in the domain of tax collection, where abuse was easy and occurred often, as the papyri tell us.48 Although this chapter does not intend to argue for the direct influence of Abū Yūsuf’s work on the way requests to the authorities to deal with abusive officials were formulated, it does mean to show how literary sources produced at court in Baghdad can be brought into conversation with documents that reflect the reality of daily life and the reality of governance in a far-away province like Egypt.49 In other words, the officials and scribes composing papyrus letters in Egypt did not do so with Abū Yūsuf’s Kitāb al-Kharāj open on their desk or even on their shelf, but the moral claims made in the papyrus letters and Abū Yūsuf’s admonitions to the caliph all came out of a shared culture of the moral responsibilities of leadership.

48 49

See the warnings of the governor Qurra b. Sharīk to his officials against abuse of taxpayers and corruption, Papaconstantinou, Rhetoric. Although Abbasid caliphs desired to increase direct control over the provinces of the empire in legal, political, and financial matters, their ambitions were often thwarted by political conditions at the capital; Tillier, Histoire 32. There is no indication that Abū Yūsuf’s text circulated in Egypt, although there is no evidence that it was not available in the province either.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

good governance in theory and practice

199

Bibliography Sources Abū Yusūf, Kitāb al-Kharāj, Beirut 1399/1979.

Translations Abou Yousof Yaʿkoub [Abū Yusūf], Le livre de l’impot foncier (kitâb el-kharâdj), trans. E. Fagnan, Paris 1921. Abū Yusūf, in Islam from the prophet Muhammad to the capture of Constantinople, trans. B. Lewis, i, New York and Oxford 1987, 151–170. Ben Shemesh, A. (trans.), Abū Yūsuf’s Kitāb al-Kharāj, vol. 3 of Taxation in Islam, London 1969. Diem, W., Arabische Briefe des 7. Bis 13. Jahrhunderts aus den Staatlichen Museen Berlin, Berlin 1997. Rāġib, Y. Lettres arabes i, in Annales Islamologiques 14 (1978), 15–35, pls. vi–xi. Vanthieghem, N., Violences et extorsions contre des moines dans la région d’Assiout: Réédition de P.Ryl.Arab. ii 11, in Journal of Coptic studies 18 (2016), 185–196.

Studies Amedroz, H.F., The mazalim jurisdiction in the Ahkam Sultaniyya of Mawardi, in jras (1911), 635–674. Berkel, M. van, Abbasid maẓālim between theory and practice, in M. Tillier (ed.), Le pluralisme judiciaire dans l’Islam premoderne (beo 63), Beirut 2014, 229–242. Berkel, M. van, Embezzlement and reimbursement: Disciplining officials in ʿAbbāsid Baghdad (8th–10th centuries a.d.), in International Journal of Public Administration 34 (2011), 712–719. Bosworth, C.E., An early Arabic mirror for princes: Ṭāhir dhū l-yamīnayn’s epistle to his son ʿAbdallāh (206/821), in jnes 29 (1970), 25–41. Crone, P., and M. Hinds, God’s caliph, Cambridge 1986. Darling, L.T., A history of social justice and political power in the Middle East: The circle of justice from Mesopotamia to globalization, London and New York 2013. El-Hibri, T., ʿUmar b. al-Khaṭṭāb and the Abbasids, in jaos 136 (2016), 763–783. Grob, E., Documentary Arabic private and business letters on papyrus: Form and function, content and context, Berlin and New York 2010. Gutas, D., Classical Arabic wisdom literature and scope, in jaos 101 (1981), 49–86. Hallaq, W.B., From regional to personal schools of law? A reevaluation, in Islamic law and society 8 (2001), 1–26. Heck, P.L., The role of law in ʿAbbasid political thought: From Ibn al-Muqaffaʿ (d. 139/ 756) to Qudāma b. Jaʿfar (d. 337/948), in J.E. Montgomery (ed.), Occasional papers of the School of Abbasid Studies, Cambridge, 6–10 July 2002, Louvain 2004, 83–110

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

200

sijpesteijn

Heck, P.L., The construction of knowledge in Islamic civilisation, Leiden 2002. Heck, P.L., Abū Yūsuf, in G. Böwering et al. (eds.), The Princeton encyclopedia of Islamic political thought, Princeton 2013, 15–16. Khan, G., The historical development of the structure of medieval Arabic petitions, in bsoas 53 (1990), 8–30. Lev, Y., The administration of justice in medieval Egypt: From the 7th to the 12th century, Edinburgh 2021. Marlow, L., Hierarchy and egalitarianism in Islamic thought, Cambridge 1997. Melchert, C., Formation of the Sunni schools of law, 9th–10th centuries, Leiden 1997. Nielsen, J., Maẓalim, in ei2, vi, 933–935. Papaconstantinou, A., The rhetoric of power and the voice of reason: Tensions between central and local in the correspondence of Qurra ibn Sharīk, in S. Procházka, L. Reinfandt, and S. Tost (eds.), Official epistolography and the language(s) of power (Papyrologica Vindobonensia Österreichische Akademie der Wissenschaften 8), Vienna 2015, 267–281. Qutbuddin, T., Arabic oration: Art and function, Leiden 2019. Rabb, I., Doubt in Islamic law: A history of legal maxims, interpretation and Islamic criminal law, Cambridge 2014. Rudman, D., The commissioning stories of Saul and David as theological allegory, in Vetus Testamentum 50 (2000), 519–530. Schacht, J., Abū Yūsuf, in ei2, i, 164–165. Sellheim, R., and D. Sourdel, Kātib. i. In the caliphate, in ei2, iv, 754–757. Sijpesteijn, P., Righting wrongs: Justices and redress in the early Islamic empire, forthcoming. Sourdel, D., Le vizirat ʿabbāside, 2 vols., Damascus 1959–1960. Sperl, S., Islamic kingship and Arabic panegyric poetry in the early 9th century, in Journal of Arabic literature 8 (1977), 20–35. Tillier, M., Histoires des cadis Egyptiens, Cairo 2012. Tillier, M., Qāḍīs and the political use of the maẓālim jurisdiction under the ‘Abbāsids, in C. Lange and M. Fierro (eds.), Public violence in Islamic societies, Edinburgh 2009, 42–66. Tillier M., The Maẓālim in historiography, in A.M. Emon and R. Ahmed (eds.), Oxford handbook of Islamic law, Oxford 2018, 357–380. Zaman, M.Q., The caliphs, the ʿulamāʾ and the law: Defining the role and function of the caliph in the early ʿAbbāsid period, in Islamic law and society 4 (1997), 1–36.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 11

A Matter of Trust: On Some Principles of Governance in the Letters of Qurra b. Sharīk Arietta Papaconstantinou

The Greek and Arabic letters sent by Qurra b. Sharīk, governor of Egypt from 90/709 to 96/714, to Basileios, administrator of the district of Aphrodito in Middle Egypt, are justly famous among students of early Islam.1 The first publications of them were accompanied by translations into English or German, while the edition of the bulk of the Greek texts, now in the British Library, was preceded by a long introduction that carefully reconstructed from their content the logistics of the administration and taxation of the province under the Umayyads.2 Since then, other aspects of the letters have been discussed at some length, most prominently their bilingual nature and what that can reveal about Umayyad linguistic policies,3 or the implications of some of the letters for early judicial practices in Egypt.4 The letters are part of a larger group of documents that were originally kept at the offices of the district of Aphrodito and are often referred to as the Aphrodito archive. This includes documents produced by the office, such as registers and memoranda, and documents received by it from the villages under its jurisdiction and from the governor’s office in Fusṭāṭ. The latter group consists mainly of Qurra’s letters to the local administrator, Basileios. These are concerned with a variety of issues, the most common being resource extraction of various forms, and they are usually occasioned by some form of perceived mismanagement on the part of the recipient. Some years ago I gave an overview of the different types of accusations brought by Qurra against Basileios and argued that despite the extraordinary rhetoric of power and blame displayed in the letters, the governor could not carry out his threats without the risk of compromising the precious local knowledge represented by this local administrator—and, presumably, others

1 2 3 4

Translations are from Bell, Translations, with many terminological modifications. Bell, Introduction to P.Lond iv; Bell, Administration of Egypt. Richter, Language choice; Richter, An unseren Herrn. Tillier, Dispensing justice; Tillier, L’invention du cadi.

© Arietta Papaconstantinou, 2022 | doi:10.1163/9789004525245_012

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

202

papaconstantinou

like him.5 I also noted, as had Nabia Abbott and H.I. Bell, that Qurra’s letters showed a sustained concern for equity and justice, but I did not push the question further. Here I would like to give this long-repeated claim some more depth by looking a little more closely at the vocabulary of the letters relating to the qualities of the good official and then offering some thoughts on its broader significance. This topic has been the subject of countless conversations with Hugh, and I hope these brief and preliminary analyses will provide the foundation for more.

1

God, Justice, the Faithful, and Their Commander

Qurra’s letters display a strong consciousness of a direct line from God through the Commander of the Faithful to himself and further down to his officials. Apart from the standard invocation in the bismillāh, God often also appears in the body of the letters as the one who is ultimately behind the requests made of Basileios. It is “by God’s command” (κελεύσει Θεοῦ) that Qurra intends to find out about church dues,6 to send inspectors to Aphrodito,7 to pay those who will have helped find fugitives,8 to figure out monetary fraud,9 and to hear about malpractice by his officials.10 The Nile flood happens by God’s command,11 and so does the reception of the revenue,12 or the evils that befall those who act badly.13 His letters are sent by God’s command.14 God also blesses the harvest,15 an essential part of the province’s revenue. The reason Basileios was sent to his post, Qurra says, is “to fear God and uphold his faith, and obtain what is due to the Commander of the Faithful” (ἀνύσαι τὸ δίκαιον τοῦ Ἀμιραλμουμνιν).16 He and his subordinates should act “fearing God and upholding justice and equality” (φυλάττοντας τὸ δίκαιον καὶ 5 6 7 8 9 10 11 12 13 14 15 16

Papaconstantinou, Rhetoric of power. P.Lond. iv 1363, 10 (4 Sept. 710). P.Ross.Georg. iv 11, 3 (13 June 710); P.Ross.Georg. iv 6, 25–26 (23 January 710); P.Lond. iv 1344, 16 (after 2 April 710). P.Ross.Georg. iv 2, 13 (17 June 710). sb xx 15102, 16 (19 February 709). P.Lond. iv 1349, 22 (14 January 710). sb x 10458, 7 (19 May 710). P.Ross.Georg. iv 1, 26 (2 April 710); P.Lond. iv 1370, 17 (3 November 710); P.Lond. iv 1464, 5–6 (709–714). P.Ross.Georg. iv 2, 17 (17 June 710). P.Lond. iv 1384, 33 (709–714). P.Lond. iv 1380, 13 (1 June 710). P.Lond. iv 1380, 10–11 (1 June 710).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a matter of trust

203

τὴν ἰσότητα),17 and their assessment of taxes and fines should follow those principles.18 Qurra hopes “in God” that Basileios will be a one of those good officials.19 “For the serviceable (χρήσιμος) official collects without negligence what is due to the Commander of the Faithful with care and good will” (ἀκαταφρονήτως συνάγ̣ει τὸ δίκαιον τοῦ Αμιραλμουμνιν μετὰ κυβερνήσεως καὶ καλοθελείας), “not losing or destroying anything,” says Qurra.20 In any case, God knows if even a miliaresion has been withheld.21 God also knows whether the lists of fugitives drawn up by the local office are complete22 and will denounce any misuse of inheritance to Qurra (ἐμβάλοι ἡμῖν).23 The importance of justice and equality is repeatedly stressed by Qurra. In one of his letters he uses the term ἀπροσωπολήμπτως (impartially), which is rare in the papyri but used in patristic literature as one of the qualities of God as judge.24 Other letters insist on fairness and clemency (ἐπιήκεια)25 or guidance and assistance (χειραγωγεία).26 The individual who was in power locally was ultimately, through multiple delegations, doing the work of God himself, and had to be not only just but also accessible to all so as to fulfil the expectations for justice of the population. In one letter Qurra explicitly tells Basileios that the official’s duty is to receive without hesitation “the petitions of all the people of his district” (τὰς προσελεύσεις πάντων τῶν τῆς διοικήσεως αὐτοῦ), “and with the fear of God to administer to each his due.”27 Further in the same letter the injunction is repeated: “devote yourself to the people of your administrat17 18 19 20 21 22 23 24

25 26 27

P.Lond. iv 1345, 1–2 (1 January 710). sb iii 7241, 17–19 (7 January 710); P.Lond. iv 1356, 8–9; 13–14; 36 (15 April 710); P.Lond. iv 1345, 26–27 (1 January 710); sb iii 7241, 17–19 (7 January 710). P.Lond. iv 1361, 6 (31 July 710). P.Lond. iv 1349, 19–21 (14 January 710). P.Lond. iv 1380, 19–20 (1 June 710). P.Lond. iv 1343, 16 (30 December 709). P.Ross.Georg. iv 15, 10 (ca. 30 January 710). P.Lond. iv 1356, 33–34 (15 April 710): σχεῖν τὸν φόβον̣ τοῦ Θεοῦ πρὸ ὀφθαλμῶν καὶ [ἀπροσωπ]ολήμπτως διαστεῖλαι τὸν ῥηθέντα μοιρασμόν (to have the fear of God before their eyes and to carry out said division impartially). In various forms, the term appears 81 times in the Thesaurus linguae Graecae, and the overwhelming majority of these refers to God or Christ in their judicial function. It seems to have been a favorite term for Cyril of Alexandria and John Chrysostom in particular. P.Lond. iv 1349, 8 (14 January 710); P.Lond. iv 1394, 22 (709). P.Lond. iv 1349, 8 (14 January 710); P.Lond. iv 1375, 10 (1 May 710). P.Lond. iv 1356, 7–9 (15 April 710). Bell had translated this as “receiving the representations of all the people in his district.” The term προσελεύσεις, however, had a more technical sense by that time and is used both in Justinian’s Novels and in later texts to describe “access to emperor as suppliant, hence petition” (Lampe, Lexicon s.v.). There is little doubt this is how it is used here in the context of official hearings of constituents’ complaints.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

204

papaconstantinou

ive district, to hearing what they say and to judge each according to what is just, not shutting yourself up but allowing them [access to you].”28

2

Competence and Trust

If the promise of justice and fairness co-opted the consensus of the population at large, what held the power pyramid itself together was loyalty and trust. In order to be trusted, the official had to show himself competent and useful.29 He also had to surround himself with trustworthy and competent subordinates. That duo of virtues (ἱκανὸς καὶ πιστός) comes up repeatedly in the letters, especially when Qurra instructs Basileios to find competent and trustworthy people to carry out given tasks at the level of the villages of the district, or to convey requisitioned goods to Fusṭāṭ or Alexandria.30 Trustworthy men, as Ian Forrest has shown in detail for the later Middle Ages,31 were usually men of means in local communities. This is explicitly stated by Qurra in a long letter regarding requisitions for the army.32 To begin with, well-off guarantors are to be chosen to cover costs in case anyone tries to evade their duties (l. 3–4). Once the goods have been collected, Qurra gives instructions on their expedition: “The supplies ordered from your administrative district you are to send off in full, appointing in charge of them serviceable and trustworthy men of means (εὐπόρους), testifying [lac.] on behalf of other trustworthy men, and they should not be deceivers or fraudsters” (12–16). Any money sent in lieu of supplies in kind must be sent “by your trustworthy agent with instructions to pay it over in full” (l. 38–39). The entire operation was underpinned by local guarantors in each village who were also “men of means” (l. 43–44).

28 29 30

31 32

P.Lond. iv 1356, 12–15 (15 April 710). P.Lond. iv 1337, 18 (10 September 709); P.Lond. iv 1343, 11; 19 (14 January 710); P.Lond. iv 1348, 11 (13 January 710). P. Lond. iv 1367, 10 (13 October 710); P.Ross.Georg. iv 5, 8–9 (709–714); P.Ross.Georg. iv 2, 8 (17 June 710); P. Lond. iv 1367, 10 (13 October 710); P.Lond. 4 1343, 35 (30 December 709); P.Lond. 4 1371, 10 (710–711); only “trustworthy”: P.Lond. 4 1366, 8 (28 September 710); P.Lond. 4 1375, 12 (1 May 710); P.Lond. 4 1392, 11 (711); P.Lond. 4 1404, 15 (709–714). In one case we find πιστοὺς καὶ εἰδήμονας (trustworthy and expert): P. Lond. 4 1356, 17 (15 April 710). Forrest, Trustworthy men. sb iii 7241 (7 January 710).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

205

a matter of trust

The trust put in well-off villagers was certainly a way to co-opt those local elites, but it was also a risk for them as they were made responsible for others’ defaults. In many ways, this high-risk, high-gain game rested on the same principle as tax farming and seems to have been very efficient in maintaining the social order, at least as seen from above. Another quality of trustworthy men was literacy. When instructing Basileios on how to register fugitives and those who hid them, he suggests sending “a competent and trustworthy person who is able to write” to meet “the commissioners for the fugitives” (ἐπικειμένους τῶν φυγάδων) and draw up the register with them.33 It is clear from papyri throughout the late antique period that those able to write in villages had the precious advantage of being aware of, and often in control of, other peoples’ affairs. Literacy offered on a plate the status of mediator with the powers that be and was therefore an important social asset. Perceived trustworthiness tended to go hand in hand with such symbolic capital. In a previous study I suggested that perhaps the main reason why Qurra—or any governor—needed people like Basileios was because they held information he could not easily obtain without them.34 This was related to a request for the names of the men who returned from the raids on North Africa, where Qurra very candidly admitted he had no idea who they were or how many of them there were: “We do not know the number of the sailors who returned home to your administrative district of those who went out to the raiding fleet of Africa with ʿAṭāʾ b. Rāfiʿ, whom Mūsā b. Nuṣayr dispatched, and of those who remained in Africa.”35 Basileios was therefore to find out and report on the subject. The importance of local knowledge is acknowledged by Qurra in another letter, once again about fugitives: “Now on receipt of this letter send three competent men who know the places where they find refuge (γινώσκοντας τοὺς τόπους ἐν οἷς προσφεύγουσιν), putting at their head a competent and trustworthy man.”36 It is noteworthy that the men who know where the fugitives went had to be supervised by someone trustworthy, as their own credentials were probably not so good—just like police informers today, presumably.

33 34 35 36

P.Lond. 4 1333, 5–9 (25 December 708); duplicate: P.Lond. 4 1332 (same date). Papaconstantinou, Rhetoric of power, 274–275. P.Lond. iv 1350, 3–16 (29 January 710). P.Ross.Georg. iv 2, 6–8 (17 June 710).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

206 3

papaconstantinou

Bad Officials

This entire edifice of the virtues and duties of the good official had its flip side, the bad official (κακὸς ὑπουργός).37 Negligence (ἀμέλεια) was seemingly the most common problem, and Qurra never tires of asking Basileios not to neglect such and such duty and not to allow his subordinates to be neglectful themselves. Yet although neglect reflected badly on an official, it does not seem to have been the worst offence. Unsurprisingly perhaps, it was transgressions of the code of conduct outlined above that were the most contentious. Acting with partiality by showing “sympathy” or “antipathy” to groups or individuals during the assessments of fines, taxes, or special levies angered Qurra enormously, as it undermined the population’s trust in the justice of the state. He put into place a system of inspectors who were specifically instructed to check whether such assessments were done in a way that was just to everyone. Sympathy and antipathy are too present to be simple transgressions, and they point to a more systemic issue. Part of it was probably the traditional networks of patronage at work, but another part was straightforward corruption. Qurra says so explicitly, even using the term δωροδοκία (bribe).38 Again, this is a relatively rare term in papyri, appearing in particular in the second century in a petition by Dionysia to the prefect, and in the text of an edict promulgated under Trajan that prohibited bribes in legal procedure.39 Intriguingly, among the virtues of God as judge found in patristic texts there is a passage by John Chrysostom that combines impartiality and imperviousness to bribes as virtues of the Judge and Examiner (ὁ ἀπροσωπόληπτος δικαστὴς, ὁ ἀδωροδόκητος ἐξεταστὴς).40 Bad officials also disobey (παρακοή) the governor’s requests and display contempt (καταφρόνησις) for him, as well as for their own soul.41 They are also lazy gluttons: “and we did not send you there to idle away in gluttony (καὶ γὰρ οὐκ ἀπεστείλαμέν σε σχολάσαι εἰς τὸ φαγονῖν),” complains Qurra, once again using the imagery of a vice with strong patristic overtones.42

37 38 39 40 41 42

P.Lond. iv 1349, 37 (14 January 710). P.Lond. iv 1337, 21 (10 September 709). P.Oxy. ii 237, 7 (ad 186) and P.Oxy. xxxvi 2754, 7 (111) respectively. John Chrysostom, Sermon 2 on penitence, pg 60.701 lines 68–69. P.Lond. iv1345, 36 (1 January 710); P.Berl. Frisk 6, 6 (26 January–24 February 710). P.Lond. 4 1380, 9 (1 June 710).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

207

a matter of trust

4

Timeless Virtues of the Imperial Official

Such virtues, vices, and complaints refer to the very real logistics of running an empire, and the real situations one could encounter when dealing with mid-level officials. However, they also represent a received version of moral language that comes in a direct line from the cumulative Mediterranean and Middle Eastern koine that preceded it. Comparable letters were sent to Ptolemaic officials by their superiors, with injunctions to do what was useful to the king, to provide justice for all and not to be partial, and to assess taxes according to everyone’s means.43 Similarly, Assyrian official correspondence insists on vigilance and assiduousness, criticizes negligence, and implies subordinate officials wanted to look good in the eyes of their superiors.44 This is not the place to make sweeping claims on the functioning of preindustrial empires, but the structures of such entities are very recognizable in this case, as in many others. In order to rule by consensus, local elites had to be brought into the power pyramid, while guaranteeing justice and fairness to the population was seen by the imperial center as the only way to keep conflict at bay. These were rather contradictory expectations, however, as even-handedness went against the natural tendency of local elites, hence the existence of elaborate centralized systems of justice that allowed individuals to bypass them—at least in principle.45 The cornerstone of the entire edifice was trust. As this was a pyramid, however, like loyalty, trust could only go one way: vertically. Alternative horizontal trust networks were discouraged, and when possible dismantled, as they could only foster disloyalty—or at least divided loyalty. To paraphrase Max Weber, imperial centers claimed the monopoly of legitimate trust. To come back to Qurra and his favorite virtues, arguably they were part of the gradual construction of Umayyad imperiality, manifesting itself through an ideology of power largely based on the Roman model that preceded it. Even though the notion of citizen in the Roman sense had disappeared, the discourse casting the imperial center as the guarantor of equity and justice for the inhabitants remained. Coming before the theoretical texts that eventually surfaced in the Islamic tradition, Qurra’s letters provide an interesting early glimpse into this ideology on the ground.

43 44 45

Armoni, Amtliche Ermahnungsbriefe, 134–136. Baker and Groß, Doing the king’s work, 84–85, 79. These dynamics are well described in Rustow, The lost archive, 55–66 and passim; and the essays in Morris and Scheidel, Dynamics of ancient empires.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

208

papaconstantinou

A technology of power among many, the ideology underpinning the image of the good official was at the center of Roman imperial legitimacy since the inception of the empire and the maintenance of its attachment to Republican tropes. The notion, like that of ruling through soft power by co-opting “trustworthy” locals, became central in Umayyad governance, and arguably constituted a key part of its claim to imperial continuity in formerly Roman lands.

Bibliography Sources Papyri are cited according to the Checklist of editions of Greek, Latin, Demotic, and Coptic papyri, ostraca, and tablets (permanent link: https://papyri.info/docs/checklist). John Chrysostom, Sermon 2 on penitence (“Dominica tyrophagi de poenitentia et eleemosyna. Sermo ii”), in J.-P. Migne, Patrologia graeca 62, Paris 1862, col. 699– 706.

Studies Armoni, C., Amtliche Ermahnungsbriefe aus dem hellenistischen Ägypten, in S. Procházka, L. Reinfandt, and S. Tost (eds.), Official epistolography and the language(s) of power, Vienna 2015, 133–138. Baker, H.D., and M. Groß, Doing the king’s work: Perceptions of service in Assyrian royal correspondence, in S. Procházka, L. Reinfandt, and S. Tost (eds.), Official epistolography and the language(s) of power, Vienna 2015, 73–90. Bell, H.I., Translations of the Greek Aphrodito papyri in the British Museum, in Der Islam 2 (1911) 269–283, 372–384; 3 (1912) 132–140, 369–373; 4 (1913) 87–96; 17 (1928) 4–8. Bell, H.I., Two official letters of the Arab period, in Journal of Egyptian archaeology 12 (1926), 265–281. Bell, H.I., The administration of Egypt under the Umayyad khalifs, Byzantinische Zeitschrift 28 (1928), 278–286. Forrest, I., Trustworthy men: How inequality and faith made the medieval church, Princeton 2018. Lampe, G.W.H., A patristic Greek lexicon, Oxford 1961. Morris, I., and W. Scheidel (eds.), The dynamics of ancient empires: State power from Assyria to Byzantium, Oxford 2009. Papaconstantinou, A., The rhetoric of power and the voice of reason: Tensions between central and local in the correspondence of Qurra ibn Sharīk, in S. Procházka, L. Reinfandt, and S. Tost (eds.), Official epistolography and the language(s) of power, Vienna 2015, 267–281.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

a matter of trust

209

Papaconstantinou, A., Fusṭāṭ and its governor: Administering the province, in A cosmopolitan community: Muslims, Christians and Jews in old Cairo, Chicago 2015, 43–47. Richter, T.S., Language choice in the Qurra dossier, in A. Papaconstantinou (ed.), The multilingual experience in Egypt from the Ptolemies to the ˁAbbāsids, Farnham 2010, 189–220. Richter, T.S., “An unseren Herrn, den allberühmten Korra, den herrlichsten Gouverneur, durch Dich, glorreichster Herr Basilios, Pagarch von Djkow mit seinen Gehöften.” Verwaltung und Verwaltungssprachen Ägyptens im 8. Jh. nach den Qurra Papyri, in F. Feder and A. Lohwasser (eds.), Ägypten und sein Umfeld in der Spätantike. Vom Regierungsantritt Diokletians 284/285 bis zur arabischen Eroberung des Vorderen Orients um 635–646. Akten der Tagung vom 7.–9. 7. 2011 in Münster, Wiesbaden 2013, 121–138. Rustow, M., The lost archive: Traces of a caliphate in a Cairo synagogue, Princeton 2019. Tillier, M., L’invention du cadi: La justice des musulmans, des juifs et des chrétiens aux premiers siècles de l’Islam, Paris 2017. Tillier, M. Dispensing justice in a minority context: The judicial administration of Upper Egypt under Muslim rule in the early eighth century, in R. Hoyland (ed.), The late antique world of early Islam: Muslims among Jews and Christians in the east Mediterranean, Princeton 2013, 133–156.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 12

Calculating the Population of Samarra Alastair Northedge

One of the most striking of Hugh Kennedy’s article titles has always been ‘The feeding of the five hundred thousand: Cities and agriculture in early Islamic Mesopotamia.’ Evidently, from the beginning one of his major interests has been the Abbasids and their cities, and in this case how to feed them from a distance. The question of how populations were fed, although the main objective of Kennedy’s study, is not the part of the question dealt with here, but rather, how many of them there were to feed. This is still a work in progress, and review of the variables may lead to a better approximation in the future. The urban population of Iraq was certainly very large under the early Abbasids, outstripping capitals of antiquity such as Ctesiphon.1 Its expansion falls into two phases: first, the settlement of the Arab tribes in the amṣār of Kufa and Basra, where large numbers were settled as fighters (muqātila) in the years after the conquest; and second, the expansion of the Abbasid capitals of Baghdad and Samarra, which was partly a consequence of the settlement of the army in the capital and partly, in the case of Baghdad, a product of the economic expansion of the city, which lasted from the foundation in the 760s until the decline of Abbasid Iraq in the first half of the fourth/tenth century. In the case of Kufa and Basra, the numbers in around 660 were of the order of 60,000–80,000 fighters, according to the reports. In the governorship of Ziyād b. Abīhi, there were 80,000, with 120,000 dependents, in Basra. That period may have been brief, as many departed for the conquests and did not return. Although it is not possible to confirm these figures from archaeological evidence, it is true that the outlines of the two amṣār can be detected on the ground in corona satellite imagery from the 1960s.2 Basra was walled in 155/771–772 1 This was the view of Adams, Land behind Baghdad 62–63, 72. However, Ctesiphon may have been bigger than supposed, as there were unwalled suburbs that have not been taken into account. It is regrettable that no thorough modern publication of Sasanian Ctesiphon exists at present. 2 Declassified corona satellite imagery can be downloaded from the US Geological Survey (https://earthexplorer.usgs.gov/) for free if the image has already been scanned, the original imagery being taken on film. Most of the useful imagery was taken between 1966 and 1972.

© Alastair Northedge, 2022 | doi:10.1163/9789004525245_013

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

211

calculating the population of samarra

figure 12.1

The site of Basra, showing the limits of the early Islamic city

at the time of its maximal expansion, and the wall line has been identified (fig. 12.1).3 Kufa was not walled but surrounded by a fosse (khandaq) under Caliph al-Manṣūr (fig. 12.2).4 Its maximal area can be seen from the line of a canal, which was located perhaps on the line of the khandaq, and delimited

3 Pellat, al-Baṣra. corona image ds1035–1040da019, dated 23/9/1966. 4 Northedge, Umayyad and Abbasid urban fortifications.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

212

figure 12.2

northedge

The site of Kufa showing the limits of the early Islamic city

the early Islamic city, larger than 1960s Kufa.5 The areas are: Kufa, 1,375 ha, and Basra, 2,200 ha, approximately. Both were large cities by any standards, even if somewhat smaller than the early days when amṣār were effectively temporary camps. By contrast, we know very little about early Islamic Baghdad, the city that existed before the return from Samarra and before the construction of the Dār al-Khilāfa on the east bank of the Tigris. We do not even know for cer-

5 corona image ds1039–2088df064, dated 28/02/1967.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

213

calculating the population of samarra

tain where the Round City was located precisely, though some Iraqi scholars say they do know its location under modern-day Baghdad. It is evident that one cannot know the population of Baghdad, apart from estimates from the texts. There is no other source, as the early Abbasid city has long been built over. The most recent serious estimate is that of Lassner.6 According to Lassner’s estimate, based on the text of al-Khaṭīb al-Baghdādī, the city area was of the order of 70km2 (6,965 ha). Although Baghdad is generally known only from literary sources, there is reason to believe that the statements preserved by the Khaṭīb (d. 1071) on the dimensions of the city are more or less accurate thus justifying the claims of native Baghdadis that their city was more spacious than all others. One account gives a figure of 53,750 jarib; two others indicate 43,750 jarib—the lower estimate apparently having been obtained from a survey undertaken in the time of al-Muwaffaq (ca. 890). It is significant that the figures cited in this account are not only expressed in terms of jarib but in linear measurement as well. The West Side was found to be 250 habl in length and 105 in width making for a total of 26,250 jarib. The east side was 250 habl in length and 70 in width making for a total of 17,500, and for the entire city 43,750 jarib. Simple arithmetic computation reveals that the figures correspond, and one square habl = one jarib […] the known value of the jarib is 1,592sq. meters. The fact that the dimensions of the city as obtained in three different accounts, and according to three different metrological systems (habl, jarib, mil), are in agreement, may be taken as an indication of their accuracy.7 Lassner then suggests that this means a population of the order of 280,000 and offers that it could be up to twice as much. A simple length-times-width calculation can hardly be very precise, as Baghdad was not a rectangle; in fact, this type of calculation is roughly in accordance with the size of Samarra, that is, Baghdad at its maximum was physically somewhat larger than Samarra, which has 58km2 of built-up area.8 Samarra evidently never reached the level of the population of Baghdad, as its economy was entirely based on the support of the state, and the environ6 Lassner, Topography 158–60. Lassner, Massignon; Lassner, Notes; Lassner, “Habl.” 7 Lassner, Topography 157–158. 8 Northedge, Remarks.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

214

northedge

mental situation was poorer than that of Baghdad. The caliph moved there, because it was not Baghdad, and settled there with his army and the court, which were both financed through taxation. Merchants and other businessmen moved there to sell their products to the state. The factor that makes the difference is that the archaeological remains are preserved. It is quite an exceptional case: there is no other large ancient city in the world whose remains have been preserved so well.9 And indeed, much of the plan can be recovered, though, as with all archaeological material, the preservation is incomplete. Even so, it is not possible to know what the population of Samarra really was. There was no census of the population, as occurred in China under the Sui and T’ang dynasties.10 De la Vaissière, in a study of Central Asian populations, was able to reach a degree of precision with regard to eastern Turkestan (Xingjian) and their Turkish populations, but was limited to approximations for the western side.11 Certainly there were surveys, as discussed by al-Qāḍī, but they were mainly concerned with registering populations and land for tax purposes.12 However, it is possible to advance the question, even if one cannot reach a satisfactory conclusion. Estimates of medieval populations commonly depend upon a calculation of population by density and area. For example, the work of Dijandjgava on Fusṭāṭ and Cairo relies on textual estimates and very generalized statements from surviving cities whose old centers furnish evidence of early modern times, presumed to be similar to the density of medieval cities.13 Samarra is a very different case, as it was much more spread out, with a population spread visible in the archaeology, as well as being mentioned in the texts. “The people spread out in their construction at Surra Man Raʾā more than they had done at Baghdad, and they built broad residences.”14 In this sense, Samarra is atypical for a medieval city, though it may have been quite similar in style to Abbasid Baghdad. The issue of Abbasid city versus medieval city was tackled

9

10 11 12 13 14

That is why we embarked on recording the remains, see Northedge, Historical topography; and Northedge and Kennet, Archaeological atlas. Much today has been destroyed. The idea was to recover as much as possible from old air and satellite imagery (mainly 1953) so that we could know as much as possible about what the city had been like. Bielenstein, Census; Pulleyblank, Registration. De la Vaissière, Early medieval. Al-Qāḍī, Population census. Dijandjgava, Ways of estimating. Yaʿqūbī, 263.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

calculating the population of samarra

215

by Lassner in his article ‘Massignon and Baghdad: The complexities of growth in an imperial city’, where Massignon is presented as explaining Baghdad as a typical walled medieval Islamic city, whereas in the early days it evidently was not. However, Baghdad did finally become such a typical walled city in the later period, from the seventh/thirteenth century onward.

1

The Urbanism of Samarra

Evidently, when al-Muʿtaṣim’s city Surra Man Raʾā was founded in 221/836, the main city was laid out with the Dār al-Khilāfa in the north and a main avenue running south from the gate of the palace for over 7 km, with a single bend adjacent to the first congregational mosque, now lost, where the markets underneath the center of the modern city were also placed (figs. 12.3–4).15 This avenue was later called Shāriʿ Abī Aḥmad and was subsequently supplanted as the main avenue by the Shāriʿ al-Aʿẓam. To one side or the other of the avenue were placed cantonments (qaṭāʾiʿ) of the lesser military, such as Waṣīf, the Farāghina, the Maghāriba, the Jund, and others, while major figures and their followings were placed outside the main city, the Usrūshaniyya under alAfshīn Khaydar b. Kawūs al-Usrūshanī outside the village of al-Maṭīra to the south, and the Turks under Ashnās outside the small town of Karkh Fayrūz to the north. This initial urban plan was similar, but not identical, to that of Baghdad. There was no fortified Round City; rather, the Dār al-Khilāfa was placed at the head of the avenue, and the mosque in the center of the city. It was indeed the model of the rabaḍ of al-Ḥarbiyya with its blocks of houses for the military. The idea of the fortified royal city was abandoned until later in the fourth/tenth century.16 Nearly all the military cantonments (see fig. 12.5), and the subsequent caliphal city of al-Mutawakkiliyya, were built on the same model of a palace for the leader,17 from which an avenue led away, with the larger houses of the major figures on the avenue, and blocks of houses laid out behind (see fig. 12.6).

15 16 17

Northedge, Topography, ʿAskar. Northedge, Umayyad and Abbasid urban fortifications; Northedge, Early Islamic urbanism. The leader was not necessarily the war chief. Ashnās, for example, never led the Turks in war but was the civil governor of the cantonment, as was Ḥarb, the governor of al-Ḥarbiyya in Baghdad.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

216

figure 12.3

northedge

The site of Samarra, showing the extent of the ancient remains

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

217

calculating the population of samarra

figure 12.4

The division of the archaeological site into zones

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

218

figure 12.5

northedge

The disposition of the military cantonments

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

219

calculating the population of samarra

figure 12.6

Al-Mutawakkiliyya, showing the classic layout of a cantonment (qaṭīʿa) at Samarra

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

220

northedge

The list of cantonments (see fig. 12.5): 1. The cantonments of al-Dūr, defined as Area U in the archaeological evidence (Northedge, Historical topography fig. 78). 2. The cantonments of al-Karkh, situated outside the pre-Islamic town of Karkh Fayrūz and identified as Area F (Northedge, Historical topography fig. 76). 3. Area G, identified as al-Wazīriyya (Northedge, Historical topography fig. 61, 146). 4. The cantonments of al-Jawsaq, Area X (Northedge, Historical topography fig. 60, 144–146). 5. The “old” cantonment of Waṣīf, situated in Area H. 6. The cantonment of the Maghāriba, sited on the Shāriʿ al-Khalīj in Area J (Northedge, Historical topography, fig. 80). 7. The cantonments of the Jund and Shākiriyya, situated in the avenues (Area J). 8. The new cantonments of the east side (Shāriʿ al-Askar and Shāriʿ al-Ḥayr al-Jadīd) (Northedge, Historical topography fig. 81). 9. The cantonment of al-Afshīn at al-Maṭīra (Area K) (Northedge, Historical topography fig. 82). 10. The cantonment of Balkuwārā (Area R) (Northedge, Historical topography fig. 84). Evidently these cantonments cover the whole period, not just the initial foundation. Then, subsequently, as the city began to expand, a greater and greater civilian population began to settle in the central area and took over parts of the central cantonments. In particular, the political Turks wanted to be near the caliph. Large mansions were built near the Dār al-Khilāfa, and much of the military was moved out, including the Turks of Waṣīf, who moved to the Soghdian cantonment of al-Maṭīra. In addition, the new congregational mosque was built by al-Mutawakkil “in the direction of al-Ḥayr,” and some important figures, in particular the kuttāb, heads of the dīwāns, wished to live near the new mosque, as indeed the archaeological evidence shows continued to be the case long after the abandonment of Samarra by the caliphs. In addition, there was the new caliphal city al-Mutawakkiliyya, which was built and abandoned in two years 245/859–247/861 (see fig. 12.6). The usefulness of this short-lived construction is that it shows in fully preserved detail what the caliph of the time would want, although conceivably it was never fully finished.18 18

For example, the zone around the Abu Dulaf mosque, where one would expect the mar-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

calculating the population of samarra

2

221

Approaches to the Population

If the calculation of the population from the surface area is not going to work in this case, one can ask what other methods can be used. 2.1 The Mosques of Abbasid Samarra In the Muslim context, one traditional approach is to calculate the size of the mosque or mosques.19 In the earlier Abbasid period, we are still in the period when there was a single congregational mosque in a city, where the entire adult male population was supposed to hear the khuṭba on a Friday. In fact there were four jawāmiʿ in Samarra. First, the unfinished mosque of al-Rashīd’s octagonal city al-Mubārak, construction of which was abandoned in 180/79620 and was never used. Second, al-Muʿtaṣim’s mosque, which has now disappeared, was abandoned as too small according to al-Yaʿqūbī. Third, al-Mutawakkil’s mosque, with the Malwiya, which continued to be the congregational mosque from 237/851–852 until some point probably in the sixth/twelfth century. And fourth, the mosque of Abu Dulaf, which was used for prayer in 246/860 and the following year and then said to have been abandoned, although it probably continued to be used by the inhabitants of the longer-term town of al-Māḥūza (Area C).21 Thus, the mosque of al-Mutawakkil is the only one whose area is useful. There was a considerable question of access, as many of the outlying districts were too far for people to travel for the Friday prayer, and certainly the Turks of al-Karkh were probably not expected to travel to the mosque of al-Mutawakkil, as there was a mosque in the courtyard of the palace of Ashnas, where the Turks assembled for a letter of complaint made to al-Muhtadī in 256/870.22 However, the pattern of prayer typical of later times was not yet established: there are no mosques for local quarters. If one did not pray in the jāmiʿ, one prayed in the masjid of one’s chief, as in the Umayyad period. Architects’ guides today for the design of mosques allot 0.85 m2 for the individual at prayer, 60–80cm width for sitting shoulder to shoulder, and 1.2 m for fore and back to enable the person to make the prostration. Probably it is better

19 20 21 22

kets to have been located, is empty and largely unbuilt, apart from a row of shops on the avenue. The market would, at the time of the occupation of the city, have been open-air and not yet built up. E.g., Dijandjgava, Ways of estimating. Northedge, Historical topography 93–94. Northedge, Historical topography 216. Al-Ṭabarī, Taʾrīkh iii, 1,799.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

222

northedge

table 12.1

Areas of the Samarra mosquesa

Congregational Mosque of al-Mutawakkil (H3) Inner Mosque 37,440 m2 Inner Mosque plus roofed ziyādas 66,468 m2 Outer Enclosure 165,886 m2 Abu Dulaf Mosque (T1) Inner Mosque Outer Enclosure

29,110 m2 124,536 m2

Mosque of Ashnas (F1.1)

11,383 m2

a Details on the individual buildings can be found in Northedge and Kennet, Archaeological atlas, passim.

to say 1m2. One could presume then that the internal area of the mosque corresponds to the maximum envisaged by the architects, evidently not necessarily the number who regularly attended. There were, of course, also some unusable areas in the mosque. It is unlikely that the outer uncovered enclosures were ever intended to be used for prayer, but the roofed ziyādas outside the mosque certainly were. A modern-day equivalent can be seen in the Mosque of the Prophet in Madina, which in the version of the mosque in the 1980s had ziyādas very similar to those of Samarra. There is also the question of the reception capacity of the muṣallās intended for the festival prayers, where the male population would have been expected to attend the festival prayers. There were five in Samarra. One is located behind the mosque of al-Mutawakkil, evidently intended for the same public as the mosque, and one to the east of the Dār al-Khilāfa. Then there is one apparently intended for the Turks of al-Karkh and al-Dūr, and one or two for al-Mutawakkiliyya. All are substantially larger than their corresponding congregational mosques. These figures suggest that the architects were expected to accommodate up to approximately 60,000 adult males in the mosque of al-Mutawakkil, but it would be surprising if as many as 30,000 ever prayed in the Abu Dulaf mosque, as the mosque was not in reach of so many, though there was at least one day when people crowded in to present the caliph with petitions.23 More likely, the 23

Al-Ṭabarī, Taʾrīkh iii, 1,453–1,454. Northedge, Historical topography 223.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

223

calculating the population of samarra table 12.2

Y19 Y6 U337 T1042 S11

Areas of the muṣallas at Samarra

Muṣallā adjacent to the Congregational Mosque Muṣallā east of the Dār al-Khilāfa Muṣallā of the Turks in al-Karkh Muṣallā of al-Mutawakkiliyya Muṣallā east of the Raṣāṣī canala

197,47 m2 122,57 m2 200,25 m2 182,82 m2 210,75 m2

a The public for whom S11 was intended is uncertain. It could have been intended for the inhabitants of al-Muḥammadiyya, built earlier than al-Mutawakkiliyya, or as a second muṣallā for the new caliphal city.

architects provided more space than necessary, for reasons of prestige. And the same is true of the muṣallās. 2.2 Houses and Their Occupants In 2001, Derek Kennet published a paper entitled ‘The form of the military cantonments at Samarra and the organisation and size of the Abbasid army’, the latter part of which was largely based on calculations from archaeological evidence.24 The chapter aimed to calculate the size of the army, not the general population of Samarra, and was naturally restricted to discussing the role of the male members of the army. The point of departure for calculating the population density was the size of the cantonment at Balkuwārā, which was settled with 12,000 Arabs, Ṣaʿālīk,25 and others by al-Mutawakkil under the nominal command of his son al-Muʿtazz, a young child at the time.26 The number of houses in the Balkuwārā cantonment seemed to correspond with the number of 12,000 Arabs, if the figure of an ʿarāfa of 10–15 men accommodated in a unit with four rooms and a stable at Malatya in 140/757–758 under al-Manṣūr were used, as cited in al-Balādhurī’s Futūḥ:27 “There was built for the jund that they settled there for each ʿarāfa two lower bayts and two upper bayts above them and a stable (and the ʿarāfa is 10 persons to 15 men).” The unit would correspond to a small house at Balkuwārā. However this would imply that the soldiers were single, unmarried men, as 10–15 men with their wives and families could not be accommodated in four rooms. While it

24 25 26 27

Kennet, Form. Ṣaʿālīk (vagabonds) were tribal Arabs, perhaps those dropped from the dīwān 25 years earlier, or their descendants. Northedge, Historical topography 171. Kennet, Form 171. Al-Baladhurī, Futūḥ 187.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

224

northedge

table 12.3a

Typology of buildings relative to the question of residence

Total Block Blocks House Mansion 1 Mansion 2 Mansion 3 Mansion 4 Mansion 5 Palace

Group of small Abbasid houses, commonly in a rectangular block. Area of unclear block construction, which would include several “block” units. Site excavated as a “house” by the German Expedition or the Directorate-General of Antiquities. Very large residential building other than that defined as a palace. Large residential building greater than 8000 m2. Residential building 5,000–8,000m2. Residential building 2,000–5,000m2. Individual residential building less than 2,000 m2. Monumental residential building identified as belonging to a member of the Abbasid family.a

1,874 82 43 16 96 54 46 12 23

a The terminology defining a palace corresponds to the definition given in Northedge, Palaces.

is likely that some soldiers were unmarried and could be accommodated in a barracks-like environment, it is probable that many were married and would require more space, there being no process of retirement from the dīwān,28 and each family may have included more than one generation of soldiers. As there are only 1,596 cantonment houses, there probably never were 12,000 at Balkuwārā—either some were quartered elsewhere, or the hoped-for recruitment of 12,000 was not achieved, or the figure included entire families. In any case, al-Balādhurī’s expression for the allocation of 10–15 men sounds more like a block of houses than an individual building: the word bayt at this time normally indicates a multiroom apartment rather than a single room. For example, al-Muntaṣir, when walī ʿahd, had a bayt in al-Jaʿfarī, a secondary residence unlikely to have been small. In addition, many of the house blocks include a yard at the end for a stable. In fact, when the cataloguing of the inventory of buildings took place,29 an approximate typology was made, covering the entire site.

28 29

Kennedy, Armies. As published in Northedge and Kennet, Archaeological atlas.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

225

calculating the population of samarra table 12.3b

Buildings that might or might not be residential

Building

Any archaeological definable building remains, not oth- 1,860 erwise specified. Modern structures and ancient tells are excluded. Irregular complex of archaeological building remains, 182 apparently belonging to a single unit, but not with an identifiable residential plan. Regularly planned building with complex plan. 121

Complex

Complex building

In fact there has been a tendency, when thinking of Samarra in army terms, to count only the masses of small houses in each cantonment (qaṭīʿa), but the society was an integrated one, from lord to vassal. The populations of the larger houses must have been considerable. Kennet concentrates on the small houses, which were most often organized in blocks of 6–25 houses (see fig. 12.7), sometimes with an additional communal yard that could have been used for a stable. Although the houses of each block were commonly the same, house sizes in different blocks were varied, ranging from around 95m2 to 1,300m2.30 The house would normally be of a single story of four–seven rooms on one or two sides of a courtyard, built of earth. Although only one case of such a house being excavated is known to the author (M8, possibly excavated by the German Expedition in 1911–1913 at al-Iṣṭablāt), no details of how it was lived in are known—the walls were simply cleared. The only comparative evidence comes from the Umayyad houses of the Amman Citadel, destroyed with their contents by earthquake a century earlier, probably in 131/749.31 The two houses excavated, one in 1949 by Lankester Harding and the other in 1976–1977 by the author under the direction of C.M. Bennett, were each composed of seven rooms around a courtyard, with a latrine and cistern. Though architecturally different from Samarra, being built of stone, it was obvious that they were inhabited by the entourage of the unidentified Umayyad prince who built the palace and citadel, quite possibly the jund al-Balqāʾ mentioned in the arrest of the Abbasid Ibrāhīm al-Imām in 130/747. The house excavated in 1949 on the museum site may have been inhabited by a single familial unit, but the Area C house excavated in 1976–1977 seemed to

30 31

Kennet, Form. Northedge, Contents.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

226

northedge

figure 12.7

Examples of blocks of cantonment houses

have two hearths, which suggests two family units (see fig. 12.8: rooms B and F). This could represent two separate individuals in the dīwān put in the same house by the administration, with their families or not, or it could be that the eldest son of the family was married and required separate accommodations. Evidently it is impossible to say how many lived in a cantonment house. They could have been used for one or more families, including male descendants recruited to the dīwān, or for a group of unmarried men. However, the design is oriented toward the settlement of families, as is also indicated by the caliphal policy of buying Turkish slave girls to marry to the Turks. Hilāl al-Ṣābiʾ, in Rusūm dār al-khilāfa, has an entertaining text on how to calculate the average population of a house, which he unfortunately combines with the number of baths to reach 96 million as the total population of Baghdad:32 “We find that one house may contain 20 people, while another may contain two, three, or more or less than that. However if we desire a middle and reasonable course, which leaves no room for doubt, we take half the twenty, and double the three, and by adding these we have sixteen. When we take half the sixteen, we come up with eight people, including men and women, old and

32

Dijandjgava, Ways of estimating, gives an alternate source in Ibn al-Faqīh for parts of this story.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

227

calculating the population of samarra

figure 12.8

Plan of Building B at Amman, destroyed in 131/749 northedge, contents

young.”33 Probably the upper figure of 20 refers to a multigeneration Baghdad house, as has survived at least until recently in al-Karkh and Ruṣāfa. While this is absolutely no basis for calculating household size, it is not a bad hypothesis; eight could have been the average household size. The figure from the Chinese census in Xinjiang at the time was five.34 The total number of identifiable small or cantonment houses is 19,571, including identifiable apartments in the palaces. The division of small houses into the different areas is given in Table 12.4. The overall number represents the maximum of all periods, including the brief occupation of al-Mutawakkiliyya and al-Jaʿfarī, when Surra Man Raʾā was said to have been emptied. If we subtract them, the number of houses at maximum occupation is 16,428. 33 34

Al-Ṣābiʾ, Rusūm dār al-khilāfa 22. De la Vaissière, Early medieval.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

228

northedge

table 12.4

A B C D E F G H J K L M Q R T U X

Total number of small houses by zone

Al-Jaʿfarī Sub-palaces of al-Jaʿfarī Al-Māḥūza Shaykh Walī/Karkh Fayrūz

199 10 328 126 393 Qaṭāʾiʿ al-Karkh 3,287 Al-Wazīriyya 621 Central City, northern half 659 Central City, southern half 4,623 Qaṭāʾiʿ al-Afshīn 1,645 Al-Maṭīra unidentifiable Al-Iṣṭablāt (al-ʿArūs) 315 Al-Musharrahāt (al-Shāh) 253 Balkuwārā 1,599 Al-Mutawakkiliyya 2,934 Qaṭāʾiʿ al-Dūr 1,496 Qaṭāʾiʿ al-Jawsaq 1,083 Total 19,571

If we were to hypothesize a household size of eight, this would give a population size of 156,560. Excluding al-Mutawakkiliyya as double counting, that would mean a maximum of 131,400. If a household size of five, the figure would be 97,855, and without al-Mutawakkiliyya, 82,140. 2.3 The Elite and Commanders It is likely that that the elite housing was not merely a nugatory addition to the population of the city, bearing in mind the numbers of the household and servants who lived on site (see fig. 12.9). There were 289 palaces and other houses of the elite, among whom could be counted the palaces of the Abbasid family, the commanders of the army, the kuttāb (secretaries) responsible for the dīwāns, and other leaders, such as the Twelver Imams transplanted to Samarra and the Tahirid Isḥāq b. Ibrāhīm, who maintained a house at Samarra.35 A portion of the structures simply denominated “building,” 1,860 in total, may also have been residences.

35

Northedge, Historical topography 189.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

229

calculating the population of samarra

figure 12.9

Examples of elite housing from al-Mutawakkiliyya

The typology was set out as: table 12.5

Typology of elite housing

Type

Description

Total Hypothesized inhabitants

Palace

Monumental residential building identified as belonging to a member of the Abbasid family.a

23

500

a The terminology defining a palace corresponds to the definition given in Northedge, Palaces.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

230

northedge

table 12.5

Type

Typology of elite housing (cont.)

Description

Total Hypothesized inhabitants

Mansion 1 Very large residential building other than that defined as a palace. Mansion 2 Large residential building greater than 8,000m2. Mansion 3 Residential building 5,000–8,000m2. Mansion 4 Residential building 2,000–5,000m2. Mansion 5 Individual residential building less than 2,000m2. House Site excavated as a “house” by the German Expedition or the Directorate-General of Antiquities. Equivalent to Mansion 4 or 5.

table 12.6

300

96 54 46 12

250 100 50 30

43

30

Zonal division of elite housing

P A B C D E F G H J K L M R T U X

16

Al-Jaʿfarī Subpalaces of al-Jaʿfarī Al-Māḥūza Shaykh Walī/Karkh Fayrūz

M1 M2 M3 M4 M5

2 3

Qaṭāʾiʿ al-Karkh Al-Wazīriyya 1 Central City, northern half 11 Central City, southern half Qaṭāʾiʿ al-Afshīn Al-Maṭīra Al-Iṣṭablāt (al-ʿArūs) 1 Balkuwārā 1 Al-Mutawakkiliyya Qaṭāʾiʿ al-Dūr Qaṭāʾiʿ al-Jawsaq Totals 20

1

4 5 1

2 7 1 18 17 4

1 2 1 16

3 1 34 2 3 96

2

4 1 1 1 9 7 1 1 4 16 7 2 55

1

9 1 1

2

1 19 15

10

47

12

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

calculating the population of samarra

231

One can hypothesize a rather greater number of inhabitants than for the smaller housing, between 30 and 300 a house. If this were so, the elite housing would add around 48,300 to the population. If, as with cantonment housing, we exclude al-Jaʿfarī and al-Mutawakkiliyya, as they were intended replacements of Surra Man Raʾā but only served for a short time, there are 160 elite residences, perhaps serving a population of 32,000.

3

Conclusion

Adding the cantonment housing and the elite housing together, we arrive at a total of 204,868 with al-Mutawakkiliyya, or 164,034 without it, that is, at the period of maximum expansion under al-Mutawakkil between 232/847 and 245/859. If a household size of five, it would be 146,155, or 114,750 without alMutawakkil’s folly of his last years. The size of the congregational mosque, suggested to be capable of accommodating a maximum of 60,000 adult males, is not out of line with this sort of calculation. There is no way of knowing how close these figures are to the truth, but indeed it could be elaborated by further closer analysis of the data, adding more allowance for destroyed areas. But it does seem perhaps not far from the truth.

Bibliography Sources al-Balādhurī, Kitāb Futūḥ al-buldān, ed. de Goeje, Leiden 1866. al-Ṣābiʾ, Customs of the caliph’s palace (Rusūm dār al-khilāfa), trans. E. Salem, Beirut 1977. al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, eds. M.J. de Goeje et al., 16 vols., Leiden 1879–1901. al-Yaʿqūbī, Kitāb al-Buldān (bga 7), ed. de Goeje, Leiden 1892.

Studies Adams, R.M., Land behind Baghdad, Chicago 1965. Bielenstein, H., The census of China during the period 2–742ad, in Bulletin of the Museum of Far Eastern Antiquities 19 (1947), 125–163. Creswell, K.A.C., Early Muslim architecture, vol. 2, 1st. ed., Oxford 1940. De la Vaissière, E., Early medieval Central Asian population estimates, in jesho 60 (2017), 788–817.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

232

northedge

Dijandjgava, G., Ways of estimating population numbers in medieval Islamic cities as exemplified in the case of Fustāt Cairo, in Al-Masāq 5 (1992), 65–69. Durand, J.D., Historical estimates of world population: An evaluation, in Population and development review 3 (1977), 253–296. Pellat, Ch., al-Baṣra, ei2, i, 1,085–1,087. Kennedy, H., The financing of the military in the early Islamic state, in The Byzantine and early Islamic Near East, iii, ed. A. Cameron, Princeton 1995, 361–378. Kennedy, H., The armies of the caliphs: Military and Society in the early Islamic state, London 2001. Kennedy, H., The feeding of the five hundred thousand: Cities and agriculture in early Islamic Mesopotamia, in Iraq 73 (2011), 177–199. Kennet, D., The Form of the military cantonments at Samarra and the organisation and size of the Abbasid army, in A medieval Islamic city reconsidered (Oxford studies in Islamic art 14), C. Robinson (ed.), Oxford 2001, 157–182. Lassner, J., Notes on the topography of Baghdad: The systematic descriptions of the city and the Khaṭīb al-Baghdādī, in jaos 83 (1963), 458–469. Lassner, J., The “habl” of Baghdad and the dimensions of the city: A metrological note, in jesho 6 (1963), 228–229. Lassner, J., Massignon and Baghdad: The complexities of growth in an imperial city, in jesho 9 (1966), 1–27. Lassner, J., The topography of Baghdad in the early Middle Ages, Detroit 1970. Lassner, J., The shaping of Abbasid rule, Princeton 1980. Lawrence, D., et al., Long term population, city size and climate trends in the fertile crescent: A first approximation, in PLoS one 11 (2016): e0152563. doi:10.1371/journal .pone.0152563. Le Strange, G., Baghdad during the Abbasid caliphate, London 1900. Northedge, A., Archaeology and new urban settlement in early Islamic Syria and Iraq, in Studies in late antiquity and early Islam, vol. 2: Settlement patterns in the Byzantine and early Islamic Near East, G.R.D. King and A. Cameron (eds.), Princeton 1994, 231– 265. Northedge, A., The palaces of the Abbasids at Samarra, in A medieval Islamic city reconsidered (Oxford studies in Islamic art 14), C. Robinson, Oxford 2001, 29–67. Northedge, A., The historical topography of Samarra (Samarra studies 1), London 2005. Northedge, A., Remarks on Samarra and the archaeology of large cities, in Antiquity 79 (2005), 119–129. Northedge, A., Umayyad and Abbasid urban fortifications in the Near East, in Die Grenzen der Welt, Arabica et Iranica ad honorem Heinz Gaube, L. Korn, E. Orthmann, and F. Schwarz (eds.), Wiesbaden 2008, 39–64. Northedge, A., The contents of the first Muslim houses: Thoughts about assemblages from the Amman citadel, in Proceedings of the 7th international congress on the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

calculating the population of samarra

233

archaeology of the ancient Near East, ii, R. Matthews and J. Curtis (eds.), Wiesbaden 2012, 633–661. Northedge, A., ʿAskar al-Muʿtaṣim: The central city of Samarra, in Beiträge zur Islamischen Kunst und Archäologie, iv, J. Gonnella (ed.), Wiesbaden 2014, 38–62. Northedge, A., Early Islamic urbanism, in A companion to Islamic art and architecture, vol. 1: From the Prophet to the Mongols, F.B. Flood and G. Necipoglu (eds.), New York 2017, 155–176. Northedge, A., and D. Kennet, Archaeological atlas of Samarra (Samarra studies 2), 3 vols, Oxford 2015. Pulleyblank, E.G., Registration of population in China in the Sui and T’ang Periods, in jesho 4 (1961), 289–301. al-Qāḍī, W., Population census and land surveys under the Umayyads (41–132/661–750), in Der Islam 83 (2008), 341–416. Susa, A., Rayy Sāmarrāʾ fī ʿahd al-khilāfa al-ʿabbāsiyya, 2 vols., Baghdad 1948–1949. Töllner, H., Die Türkische Garden am Kalifenhof von Samarra, ihre Entstehung und Machtergreifung bis zum Kalifat al-Muʿtamids, Bonn 1971. Torres Balbás L., Extensión y demograffa de las ciudades hispanomusulmanas, in si 3 (1955), 37–59.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 13

Flour for the Caliph: Watermills in the “Land behind Mosul” Cristina Tonghini

1

Introduction

Our knowledge of much of the northern Mesopotamian region in the aftermath of the Arab conquest in the first half of the seventh century is still rather fragmentary; as is well-known, the written record pertaining to the first decades is virtually absent and becomes more substantial in the Abbasid period. In recent decades, archaeological research has been able to provide new data that allow us to understand the transformations in the settlement and land use of some areas, such as the Syrian Jazīra. Historian Hugh Kennedy has achieved a masterly integration of written and archaeological documentation into a coherent illustration of how the expanding demand for agricultural products for the sustenance of the growing cities of Iraq was satisfied thanks to significant investments in the productive hinterland of the region.1 An important element in the productive network of Mesopotamian agriculture is undoubtedly the management of water resources and of the related productive activities, such as digging canals and the construction of infrastructures for milling cereals. Archaeological research can detect traces of these mills and provide a decisive contribution to the reconstruction of the ancient landscape. The presence of mills testifies to a willingness and ability to invest in the agricultural economy of a given area, since their construction is a costly operation; it reflects the cultivation of cereals; it provides an indication of settlement density, though not necessarily limited to the immediate vicinity. In fact, mills can supply the need for flour of even quite distant areas, as is indicated in written sources: in the Abbasid period, the hinterland of Mosul supplied Baghdad with flour; in the Ottoman period, certain areas specialized in milling cereals and provided for the needs of a regional network (see below).

1 Kennedy, Feeding.

© Cristina Tonghini, 2022 | doi:10.1163/9789004525245_014

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

235

The thorniest problem concerning mills is the difficulty of attributing the traces preserved on the territory to a precise chronological horizon. Consequently, their documentary value in reconstructing the settlement and production dynamics of a given territory is limited, to say the least. The results of the “Land behind Mosul” project have made it possible to acquire solid chronological data that allow these key elements to be used in the reconstruction of territorial history. A synthesis concerning the early Islamic period elaborated on the basis of the reconnaissance data and the study of mills is provided in this paper.

2

The “Land behind Mosul” Archaeological Project

The “Land behind Mosul: Settlement, landscape and material culture of the Islamic period in northern Iraq” project constitutes a specific research chapter within the “Land of Nineveh archaeological project” (from here on LoNAP). LoNAP was launched in 2012 as a wide-ranging, multidisciplinary research project on landscape archaeology, with the aim of exploring settlement dynamics, land use, and resource management in the northern hinterland of Mosul, from prehistory to the early twentieth century;2 “Land behind Mosul” focuses on the long Islamic period, from the Arab conquest to the end of the Ottoman period (seventh to the early twentieth centuries). The LoNAP survey covers an area of 3,000sq. km in present northern Iraqi Kurdistan, east of the Tigris River, with the foothills of the Zagros Mountains to the north, the Eski Mosul Dam Lake to the southwest, and the Navkur Plain and the Bardarash region to the southeast (see fig. 13.1). Before the last decade, this area, as was the case with a large part of Iraqi Kurdistan, had only marginally been touched by archaeological research, with the exception of monumental sites from the Bronze Age period, such as the aqueduct of Jerwan and the rock reliefs of Kinnis and Maltai.3 As to the archaeology of the Islamic period, this was mainly

2 Morandi Bonacossi, Back to Assyria; Morandi Bonacossi, Land of Nineveh; Morandi Bonacossi, Italian research; Morandi Bonacossi, Water for Assyria; Morandi Bonacossi, Water for Nineveh; Morandi Bonacossi, Creation; Morandi Bonacossi, Funerary landscapes. On the methodology of field work, see in particular Morandi Bonacossi and Iamoni, Landscape and settlement 13–14; Morandi Bonacossi, Creation 57–59. The final publication of the results is currently in preparation. For the Islamic period, see Morandi Bonacossi and Tonghini, Heartland of empires forthcoming. 3 For an overview of earlier research in the LoNAP area, see Morandi Bonacossi and Iamoni, Landscape and settlement 9–10.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

236

figure 13.1

tonghini

General map of the LoNAP survey area a. savioli (sources: esri, airbus ds, usgs, nga, nasa, cgiar, n robinson, nceas, nls, os, nma, geodatastyrelsen, rijkswaterstaat, gsa, geoland, fema, intermap and the gis user community)

concerned with urban sites, and only fairly recently has the rural-scape of the entire region become the focus of field investigations.4 Although the area of the LoNAP survey is not strictly included within the borders defined by the Euphrates and Tigris Rivers, it can still be regarded, from several points of view, as part of so-called greater Mesopotamia.5 More specifically, as far as the long Islamic period is concerned, most of the survey area fell under the administration of Mosul, the main urban entity of eastern Jazīra from the early Islamic period.6

4 McPhillips and Wordsworth, Landscapes; Rousset, L’archélogie islamique; Tonghini and Vezzoli, Islamic period settlement 483–484. 5 Kennedy, Feeding 177 and note 4, already extends this concept well beyond the borders of “traditional Mesopotamia.” 6 Robinson, Empire and elites. Outside of the urban entities, trying to match the written sources with the rural settlements identified through archaeological survey work often offers poor rewards, see the various contributions in Morandi Bonacossi and Tonghini, Heartland forth-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

237

In order to recompose the history of settlement in the area, “Land behind Mosul” relied on the archaeological documentation gathered by the LoNAP survey, which was integrated with documentation derived from the study of the written sources. Both kinds of documentation exhibit significant limitations in the composition of the history of settlement. As far as the archaeological evidence is concerned, the major difficulty is the chronological assessment of the various sites identified in the survey; more specifically, the lack of a regional chrono-typology of reference for the pottery finds constitutes the main obstacle to a precise definition of the chronological horizon. The situation is gradually improving thanks to a greater degree of interest in the archaeology of the Islamic period than the region ever experienced in the past and a number of new field projects. However, it should be noted that a large number of these new projects consist of surveys, while excavation of the buried deposits is still limited.7 The chronological assessment of the sites identified in the LoNAP project thus relied on the types found in the region (northern Bilād al-Shām, Iraq), which are well-known and securely dated in specialist literature.8 Types attested at single-phase sites have also been considered in the chrono-typology of the LoNAP project as far as the Islamic period is concerned. For the purpose of settlement dating, the long Islamic period was broadly divided into three main horizons, following the classification proposed by Whitcomb:9 early Islamic period (seventh–tenth centuries), middle Islamic period (eleventh–fifteenth centuries), late Islamic period (sixteenth–twentieth centuries). The material culture of the latest of these, the late Islamic period, is especially difficult to assess in the light of the documentation available today.10 It goes without saying that these broad chronological horizons make it practically impossible to pick up significant fluctuations in settlement patterns and

7

8

9 10

coming, and in particular the chapter by Melčàk. It is with the Ottoman period that the written and material documentation can be more easily integrated. Usta in Tonghini and Usta, Mesopotamian waterscapes 23–29. For recent research projects in the region, see Kopanias and MacGinnis, Archaeology. For landscape archaeology specifically concerned with the Islamic period, see Nováček et al., Medieval urban landscape. For excavations of Islamic period deposits, see Nováček, Zangiperiod architecture. Tonghini and Vezzoli, Islamic period settlement. Excavations at the site of Gomel within the framework of the LoNAP project made it possible to recover an assemblage of the early Islamic period, excavated from single-phased dumps, see Tonghini, Pottery. Whitcomb, Islamic period. Simpson, Analysing the recent past; Tonghini and Vezzoli, Islamic period settlement 488– 489.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

238

tonghini

long abandonment phases. An initial assessment of settlement evolution in the broad terms mentioned above has been summarized in a recent work, and the present paper relies on these data; a more conclusive picture is currently being processed and will be presented in the final publication of the LoNAP project’s results.11 As to the written documentation, when it comes to the rural world, it is often very difficult to match the written sources with the archaeological evidence, and to identify on the ground the few sites mentioned in the texts. The vitality of the rural world can be reflected by the revenues it provides, but again, it is in most cases very difficult to relate these data to a specific territory. It is not until the Ottoman period that the written and material documentation can be more easily integrated.12 The results of the archaeological survey show that the area has been extensively occupied for the whole of the Islamic period, with a density that can vary over time but also in relation to the physical environment; in fact, the LoNAP area comprises a variety of geographically distinct terrains, such as the wellwatered upland and valley-floor plains and the piedmont valleys. The present chapter will focus more specifically on the early Islamic period to show how the archaeological evidence gathered in the context of the “Land behind Mosul” project is contributing to a better understanding of the rural-scape in line with the picture offered in the written texts.

3

Settlement in the Early Islamic Period

The area of the LoNAP survey, as mentioned above, features a variety of physically distinct zones: located in Transtigridian northern Iraq, it comprises the Zagros piedmont in the north, with upland plains and valley-floor plains in the south and in the southeast (see fig. 13.2). The whole area is watered by numerous streams, seasonal as well as permanent, with tributaries of the Tigris (the Rubar Dohuk, the Bandawai Stream, the Rubar Dashqalan, and the Khosr River) and the Upper Zab (the Gomel and Al-Khazir Rivers).

11 12

Tonghini and Vezzoli, Islamic period settlement. A number of historians have been working on the written records: Vittorio Berti (Università degli Studi di Padova), Syriac sources; Simone Cristoforetti (University Ca’ Foscari of Venice), Persian sources; Miroslav Melčák (Akademie věd České republiky), medieval Arabic sources; Onur Usta (Canakkale Onsekiz Mart University), Turkish Ottoman sources. See also Tonghini and Usta, Mesopotamian waterscapes, for a good example of integrated research.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

239

figure 13.2

LoNAP survey, settlement distribution map: Sites attributed to the Islamic period and watermills a. savioli

flour for the caliph: watermills in the “land behind mosul”

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

240

tonghini

The climate, with a mean annual rainfall of 450–600 mm (300–450 mm in drought years), allows dry-farming agriculture. The shallow soils of the western upland plains are suitable for staple crops, wheat being the most widely spread, while the deep soils of the valley-floors, such as the well-watered Navkur Plain to the west of the survey area, support the cultivation of rice and a variety of fruits and vegetables, as well as cereals.13 The rich agricultural productivity of this area is a long-durée feature, periodically intensified significantly by investment programs carried out by the various political entities that came to control it over time. More specifically, a vast system of canals was put into place in the neo-Assyrian period, increasing agricultural production in order to sustain a large population—reflected in the boom of settlements recorded by archaeological research—and to generate agricultural surplus. Water transportation also benefitted from the massive extension of the water network.14 With the fall of the neo-Assyrian Empire this man-made water-network system probably fell out of use,15 but settlement size, with periodic variations, reflect a degree of continuity of agricultural productivity, although it never reached the volume of the neo-Assyrian period.16 The role of the region as the “granary of the empire” seems to have been resumed in the early Islamic period. The evidence from the written sources, as well as documentation derived from archaeological research, points in this direction. As mentioned above, when it comes to the rural world, it is always very difficult to anchor the scanty information provided by the texts to specific territories and sites, and the LoNAP area constitutes no exception. General trends

13 14 15

16

Morandi Bonacossi, Funerary landscapes. See a soil map of the area in Morandi Bonacossi, Water for Nineveh fig. 3. Morandi Bonacossi, Water for Nineveh; Morandi Bonacossi, Creation. The neo-Assyrian canal network does not seem to have been re-used in the Islamic period, unlike what happened in the Syrian Jazīra, see Berthier, Peuplement rural; see also Morandi Bonacossi, personal communication and (ed.), Settlement and land use forthcoming. No traces of canals pertaining the Islamic period have been identified in the area of the LoNAP survey so far. For an overview of the various issues related to water management, see Wilkinson and Rayne, Hydraulic landscapes. On settlement patterns in the post-Assyrian period, see Morandi Bonacossi and Iamoni, Landscape and settlement; Gavagnin, Iamoni, and Palermo, Land of Nineveh; Palermo, Filling the gap. For a regional overview, see Simpson, From Tekrit to the Jaghjagh. For a preliminary chart of settlement chronology in the LoNAP area, see Morandi Bonacossi, Creation fig. 3.5. A final assessment on the documentation derived from the LoNAP survey is currently in preparation; for the post-Assyrian periods, see Morandi Bonacossi and Tonghini, Heartland forthcoming.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

241

can, however, be sketched for Mosul and its hinterland on the basis of a number of studies published so far.17 Robinson effectively demonstrates how the Umayyad family of the Marwanids made important investments in the area of Mosul; the creation of this new urban entity on the western bank of the Tigris required a hinterland capable of generating an adequate surplus to sustain the massive building program of the princes. The productivity of agriculture in the area was therefore a major concern for the ruling class; the building of watermills on the Tigris reported in the written text is a clear indication of their policy in this respect.18 As had been the case with neo-Assyrian Nineveh, the Transtigridian plains of northern Kurdistan constituted the productive cereal-growing hinterland of Mosul. The role of the Mosuli agricultural hinterland did not diminish with the establishment of the Abbasid Caliphate, but it became even more crucial for the sustenance of a growing population in the province of Iraq, a consequence of the impressive evolution of urban entities such as Kufa and Basra and even more so with the foundation of the new capital, Baghdad. Not only was Mosul backed by a productive hinterland, but it occupied a strategic position for transportation on the Tigris River.19 The crucial role of Mosul as an important food supplier for Baghdad seems to be well-known, and even Caliph al-Manṣūr is reported to have been aware of it.20 As to the archaeological documentation that specifically concerns the Mosuli hinterland investigated by LoNAP, the lack of documentary series of reference for the Islamic period, and, more specifically, of a coherent chronotypology for the pottery produced and consumed in the area, does not currently allow us to compose a history of settlement other than in extremely broad terms.21 Partial processing of the finds shows that 396 out of a total of 586

17

18

19

20 21

The fundamental work for the area is undoubtedly Robinson, Empire and elites. A more specific study of the LoNAP area is currently in preparation, conducted by M. Melčàk, and will appear in Morandi Bonacossi and Tonghini, Heartland forthcoming. Robinson, on the basis of what is available from the written texts, gives a detailed account of the transformation of the area from the Arab conquest to the early Abbasid period and demonstrates how the creation and growth of Mosul relates to Marwanid policy; Robinson, Empire and elites in particular 36–40, 77–86. Robinson, Empire and elites 86 and note 183. For a discussion on the relevance of river transportation in the heart of the early Abbasid Caliphate see Kennedy, Feeding 179– 181. Robinson, Empire and elites 86 and note 186; on the strategic position of Baghdad at the center of the communication network of the period, see Kennedy, Feeding 189–190. Figures are very enlightening in this respect: 4,846 ceramic shards can be associated with the Islamic period; 72% of these cannot be ascribed to a more specific chronological hori-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

242

tonghini

surveyed sites with surface finds can be attributed to the Islamic period; of these, 117 can be ascribed to the early Islamic period (seventh–tenth centuries), 123 to the middle Islamic period (eleventh–fifteenth centuries), 63 to the late Islamic period (sixteenth–early twentieth centuries), and 93 to the Islamic period without further specifications (see fig. 13.2).22 In light of the dating tools available today, and in consideration of the fact that the LoNAP chronological assessments are based on surface finds, further differentiations within these broad chronological horizons have been avoided. To be more specific, as far as the early Islamic period is concerned, we regarded our diagnostic tools as too meager to differentiate conclusively between settlements from the seventh–eighth centuries and those from the ninth–tenth centuries.23 As far as this period is concerned, therefore, the interpretation of the data that can be achieved today flattens any variations in the region’s settlement history into a single broad chronological horizon. The written texts, as mentioned above, refer to important investments of the Umayyad Marwanids in the rural hinterland of Mosul, thus implying a change from the late Sasanian period.24 The archaeological data from the LoNAP survey in part confirms this trend, with 103 sites occupied in the Sasanian period and 117 in the early Islamic period. However, as already noted, these figures are partial, and as far as the Islamic period is concerned, one must bear in mind that there are 93 sites from an unspecified Islamic period, suggesting that the number of sites ascribed to the three better-defined horizons may be a down-

22

23 24

zon, a more refined chronological indication being identifiable only for the remaining 28 %, of which 42 % can be assigned to the early Islamic period (seventh–tenth centuries), 42 % to the middle Islamic period (eleventh–fifteenth centuries), and 16% to the late Islamic period (sixteenth–early twentieth centuries). See Tonghini and Vezzoli, Islamic period settlement, for a full discussion of the limits of the pottery reference typology and its construction. See also Müller-Wiener, Islamic pottery. These figures are those already discussed in Tonghini and Vezzoli, Islamic period settlement; they are not conclusive, since field work was still in progress. For the same reasons, no indications of the size of settlements are offered here. The final processing of data will appear in the final publication of the results of the project, see Morandi Bonacossi and Tonghini, Heartland forthcoming. For a discussion of the diagnostic types employed, see Tonghini and Vezzoli, Islamic period settlement 486–487. The rural hinterland of Mosul must have retained a good degree of productivity in the Sasanian period, and its administration system was maintained in the early Islamic period. See Robinson, Empire and elites 90–108. For a preliminary discussion of the Sasanian period in the LoNAP area, see Gavagnin, Iamoni, Palermo, Land of Nineveh 154–155; Palermo, Filling the gap. For the final publication, see Palermo in Morandi Bonacossi and Tonghini, Heartland forthcoming.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

243

ward estimate. Still more significant is the fact that 53 out of 117 sites ascribed to the early Islamic period were not occupied in the Sasanian period and are thus new foundations.25 In the light of all these limitations, and in spite of them, the picture that appears from the archaeological record shows that the area of the LoNAP survey was densely settled in the early Islamic period, matching what is indicated in the written texts. Other sets of data can be used to further enhance this picture, such as those related to milling installations, which are discussed below.

4

Land Use and Resource Management in the Early Islamic Period: Evidence from Watermills

Further evidence for the assessment of settlement patterns, land use, and resource management in the rural-scape may derive from the study of the watermills. Those identified in the LoNAP area belong to the horizonalwheeled type, widely spread where water streams feature low volumes; from an architectural and technological point of view they show similarities to a Near Eastern/Mediterranean/Levantine variety26 that has been studied in presentday Jordan, Lebanon, Cyprus, and Iran.27 They are located beside water streams, on a slope, and they are characterized by a tower-shaped structure, the socalled drop tower.28 This is pierced by a chute, fed by a channel; the water is forced through the chute and exits from an outlet placed at the bottom of the drop-tower façade. The jet activates the horizontal waterwheel installed in a subterranean chamber at the base of the drop-tower; this wheel transmits power to the grinding devices located in the mill room above the wheel chamber.29 The drop tower is generally the more carefully built part of the mill installation, in order to withstand water pressure, and it is this part that survives today in most cases. 25 26 27

28 29

Nováček and Wood, Monastic landscape, on the rise of settlement in the early Islamic period. Schriwer, Water and technology especially 81. Selected references from a rich bibliography on the subject are: Harverson, Watermills in Iran; Gardiner and McQuitty, A water mill; Greene, Water mills; McPhillips et al., Jawz Valley; McQuitty, Water-mills; McQuitty, Harnessing the power; Neely, Sasanian period drop-tower gristmills; Schriwer, Water and technology. This element is also referred to as “penstock,” or using the Hebrew term, aruba; Avitsur, On the history; Schriwer, Water and technology 4–5. For a detailed description of the functioning of this type of watermill, see Harverson,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

244

tonghini

Mills reflect a demand for processed cereals, and therefore, their number is proportional to the intensity of settlement and to the scale of agricultural production. Because building milling structures is a costly operation and tends to be part of a wider program to shape the waterscape and to manage its potentials, their presence reveals the effect of a central power policy at a local level.30 The written sources indicate that in the early and middle Islamic periods they generally belonged to the state, to wealthy pious institutions, and to monasteries; only in much later times did they become community enterprises.31 The main difficulty in using the evidence from the mills when composing the history of settlement, land use, and resource management relates to their assignment to a chronological period. In general, and especially up to the most recent centuries, they did not have a residential function, and surface finds are generally absent. Our understanding of cereal processing and distribution modes in general is still too limited to attempt an association between specific mills and specific settlements. Although it is reasonable to see them located on the edge of settlements, as demonstrated by an archaeological and ethnographic survey in Iran,32 this cannot be assumed to be the rule, and it may vary over time. Evidence from the written sources from the late Ottoman period shows that at some point the mills of Wādī Bāndaway became the grinding center for a vast area that goes as far as al-Qosh, 7km to the east.33 Studies conducted in Jordan estimate the mill-settlement distance as a radius of 10–14 km.34 On the basis of these considerations, it became clear that specific dating elements had to be identified in order to assess the evidence from the LoNAP survey; moreover, watermills had never been systematically studied in the Kurdistan region before, and this new study could not count on regional comparative elements. Thus, as a first step, a typology of the building features of the structures identified in the area was established; subsequently, in-depth investigations were carried out at selected sites, such as site 124 in the Wādī Bāndaway area (see fig. 13.2). Wādī Bāndaway is a perennial water stream fed by a number of karst springs located along its course. From the Zagros foothills

30 31 32 33 34

Watermills in Iran; McQuitty, Water-mills; McQuitty, Harnessing the power; Neely, Sasanian period drop-tower gristmills; Schriwer, Water and technology 5–6. Wilkinson and Rayne, Hydraulic landscapes. Rogan, Reconstructing water mills; Schriwer, Water and technology 30–39; Tonghini and Usta, Mesopotamian waterscapes. Harverson, Watermills in Iran 155. Tonghini and Usta, Mesopotamian waterscapes 25–29. Gardiner and McQuitty, A water mill 31; Schriwer, Water and technology 63.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

245

it runs south to reach the Tigris River; a remarkable concentration of watermills has survived on its banks. Site 124 comprised two groups of mills, with a total of five different drop towers, located on a gentle slope overlooking the wadi (see figs. 13.3–4). The stratigraphic analysis of the masonry revealed that they were the result of different building phases; the buried deposit was excavated in part to give a better definition of the building sequence and of the technical characteristics of the structures (see fig. 13.5). Radiocarbon dating of the mortar made it possible to ascribe the earliest element of the sequence, drop tower 1, to a 602–774 time-span, the second half of the seventh century being the highest calibrated probability.35 This constitutes the earliest date for this type of mill in the medieval Middle East.36 Only drop tower 1 has been conclusively ascribed to the earliest period, although the masonry typology of drop tower 5 shows certain similarities that suggest a close chronological horizon, but more conclusive evidence is necessary on this point. If this is confirmed, the two drop-towers would have formed a chain fed by the same canalization system. Continuous re-use of the milling complex, with transformations, rebuilding, additions, and enlargements, have concealed the features of the early layout. The shape of the drop tower does not differ significantly from those built in later periods; on the contrary, the building technique and materials are quite distinctive (see figs. 13.4–5). The drop tower is founded on the natural bedrock; the face of its front is built with two types of stone, a limestone (available in the area) and a conglomerate (which constitutes the natural ground). Both types of stone are hewn or roughly squared; the blocks are rather large (H 38–42cm) and are laid in even, horizontal, and parallel courses. The blocks of the quoins are no more accurately squared than the others, but an attempt at alternating longer and shorter stones can be 35

36

For a full description and discussion of the stratigraphic analysis, with a summary of the excavation results of the mills at site 124, see Tonghini and Usta, Mesopotamian waterscapes table 4 in particular for radiocarbon dating. Doubts about the dating of the watermill on the Deh Luran plain in Iran published by Neely (Sasanian period drop-tower gristmills) have already been expressed by the present writer, see Tonghini and Usta, Mesopotamian waterscapes n. 41. Unlike what it is generally reported by surveyors, a large amount of pottery was found at this site, including a number of shards embedded in the masonry and floors; Neely, Sasanian period drop-tower gristmills 243–244. The date 680 ± 150 was obtained with Optically Scanned Thermoluminescence on one of the shards. In the present writer’s view, the possibility that the whole pottery assemblage is residual cannot be entirely discarded. The horizontal-wheeled watermills in Bilād al-Shām, excavated by Genequand (La meunerie hydraulique) and ascribed to the Umayyad period are of a different type: they are fed by an oblique chute, without the drop tower.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

246

figure 13.3

tonghini

General map of site 124, with the two groups of mills e. reali, l. tarducci

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

figure 13.4

247

Aerial view of site 124, upper group. From the left: drop tower 2, drop tower 1, and mill house bu7 a. savioli

noted (see fig. 13.5). The mortar is of very good quality and shows fine inclusions of charcoal intentionally added to ensure hydraulic properties to the mortar; charcoal is not macroscopically visible in the mortar of later periods. Archaeological evidence indicates that the mill complex—or part of it— remained in use until the early decades of the twentieth century, with possible interruptions.37 As far as the early Islamic period is concerned, the evidence that emerged at site 124 seems to parallel the information in the written sources. Although a more conclusive study of these will only be available with the final publication of the results of the LoNAP project,38 a relevant piece of information has already been discussed by Robinson: this concerns the installation of river mills in the hinterland of Mosul as early as the beginning of the eighth

37 38

For a detailed discussion of the sequence of occupation, see Tonghini and Usta, Mesopotamian waterscapes. See in particular the work of V. Berti, Monastic behaviours, on the Syriac sources and that of M. Melčàk, Land behind Mosul, on the medieval Arabic sources in Morandi Bonacossi and Tonghini, Heartland forthcoming.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

248

figure 13.5

tonghini

Site 124, drop towers 1 (right) and 2 (left): Rectified photomosaic of the façade from the southwest (e. reali, l. tarducci); archaeological characterization c. tonghini

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

249

century as part of the investment policy of the Umayyad Marwanid family.39 The sources also indicate that river mills constituted a prominent feature of the whole of the Tigris valley in the ninth and tenth centuries, “from Amid to Baghdad.”40 There is also a crucial record dated 139/756 that suggests that milling activity was managed as a caliphal monopoly, and only later is there evidence of private watermills.41 The sources do not provide a description of these mills, and therefore, an association with the horizontal-wheeled mills remains conjectural. However, the dating of the earliest mill at site 124 seems to confirm the general trend sketched on the basis of the written sources. The investments made by the Marwanid family in Mosul had to rely on the economic surplus generated by agriculture, and were thus paralleled by investments in the ruralscape; the installation of watermills in Wādī Bāndaway seems to reflect this policy and may be a further indication of the agricultural productivity of the area. So far, the archaeological evidence from site 124 has not provided direct evidence of continuity of use in the ninth and tenth centuries; however, the possibility that the milling complex of site 124 remained in use in the Abbasid period is quite substantial. With the foundation of the new capitals, Baghdad and then Samarra, and the growth of the population registered in Iraq, the area of Mosul became crucial to providing the south with foodstuff; not only cereals but also flour seems to have been transported from Mosul down to Baghdad.42 It is hard to imagine that any functioning watermill would have been abandoned at the time of such a substantial demand for cereal and flour. As to settlement patterns, as it happens with the rest of the LoNAP survey, newly founded settlements have been identified in the area of Wādī Bāndaway: three of the newly founded sites belong to the early Islamic period, and seven more can be ascribed to an unspecified Islamic period.43 As discussed above, the data from the survey do not allow settlements from the second–third/seventh– eighth centuries to be differentiated from those from the third–fourth/ninth– tenth centuries, but they illustrate a certain vitality in settlement patterns in the early Islamic period as a whole, with a slight increase in the number of sites from the Sasanian period. Further evidence is clearly necessary to identify patterns of continuity and change in the two early Islamic phases.

39 40 41 42 43

Robinson, Empire and elites 79, 84–85. Robinson, Empire and elites 85. Robinson, Empire and elites 84–85, 85 n. 178. Robinson, Empire and elites 84–85; Kennedy, Feeding 190. Tonghini and Usta, Mesopotamian waterscapes fig. 3 and table 1.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

250 5

tonghini

Conclusion

Recent archaeological research from the “Land of Mosul” archaeological project has contributed new elements to our understanding of settlement, land use, and resource management of the area in the early Islamic period—and beyond. The new data match the information provided in the written sources and make it possible to integrate the use of both kinds of documentation. The archaeology not only confirms that the grain production and processing in the hinterland of Mosul was essential to feed the “five hundred thousand”44 of Baghdad but also that an investment strategy in the agriculture of the region had already started in the previous decades with the Marwanid caliphs of the Umayyad family. In the future, the progress of archaeological research in the area and, more specifically, the spread of stratigraphic excavations will undoubtedly allow a better chronological definition of diagnostic tools than is possible at present. At the same time, the dating of mill installations pursued in the context of the “Land behind Mosul” project45 will contribute significantly to the reconstruction of the landscape in the long Islamic period.

Bibliography Avitsur, S., On the history of the exploitation of water power in Eretz-Israel, in Israel exploration journal 10 (1960), 37–45. Berthier, S. (ed.), Peuplement rural et aménagement hydroagricoles dans la moyenne vallée de l’Euphrate. Fin viie–xixe siècle, Damascus 2001. Berti, V., Monastic behaviours and anthropic environment from Syriac sources: On the steps of Rabban Hormizd, in D. Morandi Bonacossi and C. Tonghini (eds.), Heartland of empires: Settlement and land use in the eastern Tigris plains of northern Iraq from the Hellenistic to the Ottoman period (iamkri 2), forthcoming. Gardiner, M., and A. McQuitty, A water mill in Wadi el Arab, north Jordan and water mill development, in Palestine exploration quarterly 119 (1987), 24–32. Gavagnin, K., M. Iamoni, and R. Palermo, The land of Nineveh archaeological project: The ceramic repertoire from the early pottery Neolithic to the Sasanian Period, in basor 375 (2016), 119–169.

44 45

Kennedy, Feeding. A sampling campaign of the mortar to radiocarbon date a number of mills is planned in the 2021 season.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

251

Greene, J.A., The water mills of the ʿAjlun-Kufranja valley: The relationship of technology, society and settlement, in Studies in the history and archaeology of Jordan 5 (1995), 757–765. Genequand, D., La meunerie hydraulique au début de l’époque islamique (viie–viiie siècle) au Proche-Orient: un état de la question, in L. Jaccottey and G. Rollier (eds.), Archéologie des moulins hydrauliques, à traction animale et à vent, des origines à l’époque médiévale, Besançon 2016, 507–528. Harverson, M., Watermills in Iran, in Iran 31 (1993), 149–177. Kennedy, H., The feeding of the five hundred thousand: Cities and agriculture in early Islamic Mesopotamia, in Iraq 73 (2011), 177–199. Kopanias, K., and J. MacGinnis, The archaeology of the Kurdistan region of Iraq and adjacent regions, Oxford 2016. McPhillips, S., and P.D. Wordsworth (eds.), Landscapes of the Islamic world, Philadelphia 2016. McPhillips, S., Harnessing hydraulic power in Ottoman Syria: Water mills and the rural economy of the Upper Orontes Valley, in S. McPhillips and P.D. Wordsworth (eds.), Landscapes of the Islamic world, Philadelphia 2016, 143–166. McPhillips, S., A. Meier, J. Slim, and J. Bradbury, The Jawz Valley: Reconstructing an Ottoman “waterscape” in Mount Lebanon, in Levant 51 (2019), 201–218. McQuitty, A., Water-mills in Jordan: Technology, typology, dating and development, in Studies of the history and archaeology of Jordan 5 (1995), 745–751. McQuitty, A., Harnessing the power of water: Watermills in Jordan, in H.-D. Bienert (ed.), Men of dikes and canals: The archaeology of water in the Middle East (OrientArchäologie 13), Rahden and Westfalen 2004, 261–272. Melčàk, M., Land behind Mosul as reflected in Arabic sources: An overview of scholarship and sources, in D. Morandi Bonacossi and C. Tonghini (eds.), Heartland of empires: Settlement and land use in the eastern Tigris plains of northern Iraq from the Hellenistic to the Ottoman period (iamkri 2), forthcoming. Morandi Bonacossi, D., Back to Assyria: Cities, villages and canals in the Land Behind Nineveh, in The ancient Near East today 2 (2014), online. Morandi Bonacossi, D., The Land of Nineveh Archaeological Project. Assyrian settlement in the Nineveh hinterland: A view from the centre, in J. McGinnis, D. Wicke, and T. Greenfield (eds.), The provincial archaeology of the Assyrian Empire, Cambridge 2016, 141–150. Morandi Bonacossi, D., Italian research in the Nineveh region between archaeological investigation and cultural heritage protection and management, in L.P. Petit and D. Morandi Bonacossi (eds.), Nineveh, the great city: Symbol of beauty and power (Papers on archaeology of the Leiden Museum of Antiquities 13), Leiden 2017, 98– 103. Morandi Bonacossi, D., Water for Assyria: Irrigation and water management in the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

252

tonghini

Assyrian Empire, in L.P. Petit and D. Morandi Bonacossi (eds.), Nineveh, the great city: Symbol of beauty and power (Papers on archaeology of the Leiden Museum of Antiquities 13), Leiden 2017, 132–136. Morandi Bonacossi, D., Water for Nineveh. The Nineveh irrigation system in the regional context of the “Assyrian triangle”: A first geoarchaeological assessment, in H. Kühne (ed.), Water for Assyria, Wiesbaden 2018, 77–115. Morandi Bonacossi, D., The creation of the Assyrian heartland: New data from the “Land behind Nineveh,” in B.S. Düring and T. Stek (eds.), The archaeology of imperial landscapes: A comparative study of empires in the ancient Near East and Mediterranean world, Cambridge 2018, 48–85. Morandi Bonacossi, D., Funerary landscapes in the Land of Nineveh: Tracking mobile pastoralists in the Transtigridian Piedmont of northern Iraq, in D. Lawrence, M. Altaweel, and G. Philiph (eds.), New agendas in remote sensing and landscape archaeology in the Near East: Studies in honour of Tony J. Wilkinson, Oxford 2020, 41– 62. Morandi Bonacossi, D. (ed.), Settlement and land use in the eastern Tigris plains of northern Iraq from the prehistory to the Iron Age (iamkri 4), forthcoming. Morandi Bonacossi, D., and M. Iamoni, Landscape and settlement in the eastern upper Iraqi Tigris and Navkur plains (northern Kurdistan region, Iraq): The Land of Nineveh Archaeological Project, seasons 2012–2013, in Iraq 77 (2015), 9–40. Morandi Bonacossi, D., and C. Tonghini (eds.), Heartland of empires: Settlement and land use in the eastern Tigris plains of northern Iraq from the Hellenistic to the Ottoman period (iamkri 2), forthcoming. Müller-Wiener, M., The Islamic pottery, in P. Pfälzner and P. Sconzo, The eastern Ḫabur archaeological survey in Iraq Kurdistan: A preliminary report on the 2014 season, in Zeitschrift für Orient-Archäologie 9 (2016), 10–69 (44–45). Neely, J.A., Sasanian period drop-tower gristmills on the Deh Luran Plain, southwestern Iran, in Journal of field archaeology 36 (2011), 232–254. Nováček, K., M. Melčák, L. Sarková, and N.A.M. Amin, Medieval urban landscape in northern Mesopotamia, Oxford 2016. Nováček, K., Zangi-period architecture in Iraqi Kurdistan: Medrese Qubahan at Amedi (’Amadíya), in Zeitschrift für Orient-Archäologie 4 (2011), 176–210. Nováček, K., and Ph. Wood, The monastic landscape of Adiabene in the first centuries of Islam, in Journal of Islamic archaeology 7 (2020), 1–20. Palermo, R., Filling the gap: The upper Tigris region from the fall of Niniveh to the Sassanians, in K. Kopanias and J. MacGinnis (eds.), The archaeology of the Kurdistan region of Iraq and adjacent regions, Oxford 2016, 277–296. Robinson, Ch.F., Empire and elites after the Muslim conquest: The transformation of northern Mesopotamia, Cambridge 2000. Rousset, M.-O., L’archélogie islamique en Iraq: Bilan et perspectives, Damascus 1992.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

flour for the caliph: watermills in the “land behind mosul”

253

Rogan, E.L., Reconstructing water mills in late Ottoman Transjordan, in Studies of the history and archaeology of Jordan 5 (1995), 753–756. Schriwer, C., Water and technology in Levantine society, 1300–1900: An historical, archaeological and architectural analysis (bar international series 2728), Oxford 2015. Simpson, St. J., From Tekrit to the Jaghjagh: Sasanian sites, settlement patterns and material culture in northern Mesopotamia, in K. Bartl and S.R. Hauser (eds.), Continuity and change in northern Mesopotamia from the Hellenistic to the early Islamic period, Berlin 1996, 87–126. Simpson, St. J., Analysing the recent past: The archaeology of death, pastoralism, pots and pipes in the Ottoman Jazīra and beyond, in Al-Rafidan 32 (2011), 1–47. Tonghini, C., The pottery from the early Islamic dumps, in D. Morandi Bonacossi, H.A. Qasim, C. Coppini, K. Gavagnin, E. Girotto, M. Iamoni, and C. Tonghini, The Italian-Kurdish excavations at Gir-e Gomel in the Kurdistan region of Iraq: Preliminary report on the 2017 and 2018 field seasons, in Mesopotamia 53 (2018), 1–96 (10–13). Tonghini, C., and V. Vezzoli, The Islamic period settlement in Iraqi Kurdistan: Results from the land of Ninive archaeological project, in Proceedings of the 11th international congress on the archaeology of the Ancient Near East, vol. 2, Wiesbaden 2020, 483–494. Tonghini, C., and O. Usta, Mesopotamian waterscapes: Watermills in the “Land behind Mosul,” in Mesopotamia 56 (2021), 1–32. Wilkinson, T.J., and L. Rayne, Hydraulic landscapes and imperial power in the Near East, in Water History 2 (2010), 115–144. Whitcomb, D., The Islamic period as seen from selected sites, in B. MacDonald (ed.), The Southern Ghors and Northeast Araba archaeological survey, Sheffield 1992, 112– 118.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 14

Bedouin, Bandits, and Caliphal Disappearance: A Reappraisal of the Qarāmiṭa and Their Success in Arabia Peter Webb

The Qarāmiṭa are a group best-known today for the extent to which their memory is maligned.* Originating from a nebulous network of grassroots uprisings in third/ninth-century Iraq, a Qarāmiṭa-ruled state emerged in fourth/ tenth-century eastern Arabia, and their expansionist ambitions and violent tactics soon made enemies everywhere—from Egypt and Syria to Iraq and Yemen. Leaving little written history of their own, their memory was predominantly carried by their many foes who uniformly reviled them; sectarianism also entered the rhetoric as the Qarāmiṭa’s pseudo-Ismaʿili creed was denounced as heretical by Shiʿi and Sunnis alike, and hence a pall of deviation and sacrilege infuses most premodern discussions of Qarāmiṭa history. The biases of the premodern anti-Qarāmiṭa sources rationalize the movement’s success as an upshot of caliphal weakness, the gullibility of the peasant and Bedouin rabble who joined the Qarāmiṭa, and a mixture of cunning manipulation and terrifying violence employed by Qarāmiṭa leaders.1 Yet this obscures recognition that the Qarāmiṭa created a viable state that, at its height, had the military capacity to threaten Damascus, Cairo, Mecca and the Iraqi countryside simultaneously. And their achievement was durable: among the fourth/tenth-century regional state-building efforts across the Middle East following the weakening of the central caliphate, the Qarāmiṭa’s rule in eastern Arabia was one of the more stable and long-lasting. Modern-era scholarship has revisited the movement, but with mixed results for elucidating the practical and political aspects of the Qarāmiṭa state. For example, considerable Arabic writing in the 1970s and 1980s pursued an * This chapter’s genesis began as a paper I gave at the International Medieval Congress, Leeds, July 4, 2016. I would like to begin by thanking Ann Christys for inviting my presentation, and Matthew Gordon and Harry Munt for sharing material in ensuing discussions. 1 Detailed sources include the various references in al-Ṭabarī, Tārīkh; Ibn al-Jawzī, al-Muntaẓam; and the continuous narrative summary in al-Maqrīzī, Ittiʿāẓ i, 173–238; see also sources collected in Zakkār, al-Jāmiʿ. The prevailing premodern literary “othering” of the Qarāmiṭa is analyzed in El-Cheikh, Women 66–75.

© Peter Webb, 2022 | doi:10.1163/9789004525245_015

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

255

intriguing new spin that described the Qarāmiṭa as pioneers for equality (musāwāt) and justice (ʿadāla), opposing the greed and iniquity of the Abbasid caliphs and Turkic warlords of the fourth/tenth century.2 This, however, was a rather transparent exercise to posit the Qarāmiṭa as “Islam’s first socialists,”3 putative forerunners of 1970s’ socialist Arab governments, hence these works ultimately tell us more about the intellectual architecture of twentiethcentury Arab socialism than the realities of the fourth/tenth-century Qarāmiṭa. Such interpretations were aptly critiqued by Suhayl Zakkār as “anachronizing” (ʿaṣarnat ḥarakat al-Qarāmiṭā),4 and Zakkār’s approach focused on sectarianism, suggesting that the Qarāmiṭa identity and their spread related to their particular creed. Zakkār echoed the scholarly trend led by Wilfred Madelung, Farhad Daftary, and Heinz Halm, whose rigorous scrutiny of Qarāmiṭa doctrines helpfully unpacked old biases and generalizations of Sunni heresiography.5 Yet this scholarship skews the interpretation of Qarāmiṭa politics toward a confessional bent, prompting the conceptualization of the Qarāmiṭa as a theological venture where different groups developed different creeds as the basis of their core mobilization and differentiation from each other and the surrounding Sunni and Shiʿi states. Such notions perpetuate the problematic paradigm that presumes the primacy of faith driving Middle Eastern sociopolitical formation, and there is accordingly a need to “materialize” the Qarāmiṭa, to conceptualize them beyond theocracy and explore other factors that contributed to their specific success in Arabia. Hugh Kennedy, when he was my PhD supervisor, prompted my interest in the Qarāmiṭa as a means to address practical questions about how fourth/ tenth-century Iraqi Hajj pilgrims interacted with Arabia given the threat of Qarāmiṭa raids, and how fear of the Qarāmiṭa influenced Iraqi litterateurs’ conceptualizations of Arabia. Since Hugh also first advocated the interpretation of Qarāmiṭa attacks as economically minded efforts to expand revenue sources,6 for Hugh’s Festschrift, I pursue the practical considerations of the Qarāmiṭa regime, questioning why their state building was particularly successful in Arabia. For answers, it is valuable to switch our gaze from the intricacies of doctrine and focus upon the interests and social contexts of the Qarāmiṭa’s largest

2 3 4 5

Ghālib, al-Qarāmiṭa 7–8. The thesis of al-Walī, al-Qarāmitạ. Zakkār, al-Jāmiʾ 5. Blois, Abu Saʿidis 19 makes the same critique. Madelung, Das Imamat 48–52, 113–114; Madelung, Fatimiden und Bahrain; Halm, Die Söhne Zikrawaihs; Daftary, Ismāʿīlīs 122–124. Blois (Abu Saʿidis) takes a similar approach, epitomizing the state’s history (14–15) and focusing on theology (15–19). 6 Kennedy, Desert 24–25.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

256

webb

Arabian constituency: the Bedouin of central and eastern Arabia. Adopting a third/ninth-century Arabian perspective opens new insights into the movement’s success.

1

Arabia and Qarāmiṭa States

Qarāmiṭa rule in eastern Arabia (then called al-Baḥrayn) began with Abū Saʿīd al-Jannābī’s emergence in al-Qaṭīf from 276/889–890, his establishment of a capital in al-Aḥsāʾ, and his reduction of the regional seat of caliphal authority in Hajar in 289/902. Al-Jannābī’s precise relationship with contemporary Qarāmiṭa groups in southern Iraq was fluid and sometimes tendentious, and by the time of his assassination in 301/913–914, his state was operating independently of most other Qarāmiṭa, but it was secure and had a stable succession plan: al-Jannābī’s eldest son, Abū l-Qāsim Saʿīd, assumed power and for a decade maintained quite cordial relations with the Abbasid caliphs. Concurrently, another group, led by Zakarawayh b. Mihrawayh, emerged in the northern Arabian desert near Kufa and spread revolts among Bedouin in Syria and also Najd, approaching the territory of al-Jannābī. Zakarawayh’s actions were belligerent—between 293–294/906–907 he despoiled Iraqi Hajj pilgrim caravans and ruthlessly attacked settlements in Syria and Iraq. Zakarawayh was killed in 294/907, but violence returned to Najd in 311/923 when a younger son of al-Jannābī, Abū Ṭāhir Sulaymān, became ruler of the eastern Arabian Qarāmiṭa: he savaged pilgrim caravans and advanced across Arabia, sacking Mecca in 317/930 and removing the Black Stone to al-Aḥsāʾ. A large payment induced Abū Ṭāhir’s brothers to return the Black Stone in 339/951, and while Qarāmiṭa aggression thereafter abated, it did not cease: in 373/983 they attacked Basra, and they occupied Kufa in 375/985. From that point onward, however, the Qarāmiṭa began a territorial decline, yet maintained control in eastern Arabia, administering al-Baḥrayn with relative stability until the late fifth/eleventh century.7 Looking beyond the sensationalism of murdering pilgrims and sacking Mecca, al-Jannābī’s successors maintained a stable state for two centuries, and local populations were of crucial importance: the Bedouins of al-Baḥrayn and Najd were willing and fairly reliable allies from ca. 290–360/902–970. Scholars have expressed varied explanations of the Qarāmiṭa’s Arabian popularity, including the following: 7 The fifth/eleventh century state in al-Baḥrayn is described in detail in the celebrated travelogue of Nāṣir-i Khusraw, who visited in 443/1051.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

257

(a) Bedouin always craved freedom, and Qarāmiṭa egalitarianism provided for this need;8 (b) Bedouin were Arab partisans and opposed the Turco-Persian direction of the Iraqi Abbasid Caliphate;9 (c) Bedouin were illiterate marauders with a proclivity to fall for false messiahs and prophets;10 or (d) Bedouin were opportunistic raiders who joined the Qarāmiṭa to win a share of the booty.11 The difficulty with such views is that they lack specificity: they reduce Bedouins to a “presence” that responds with a limited, stereotypical repertoire of options to follow and fight. Since the nature of the Qarāmiṭa regime was not a recurring feature of Arabian politics, and since the Bedouin violence of the late third/ninth and fourth/tenth centuries reached unprecedented levels of severity and scope, the specificities of the Qarāmiṭa-era revolts indicate something more significant at play. The Qarāmiṭa’s Arabian state formation must have stemmed from unique drivers specific to their particular context. Fourth/tenth-century Arabia was indeed a region in crisis: there was widespread insecurity, settlements were abandoned, Medina received defensive walls in the late third/ninth century,12 and few outsiders ventured into Arabia, such that the region largely falls out of Arabic written history. The chaos resonates well with impressions of Qarāmiṭa violence, but it cannot simply be the case that the Qarāmiṭa availed themselves of the vacuum that was lawless Arabia to establish their state, since their rise and alliances with Bedouin occurred prior to Arabia’s eclipse. Furthermore, Arabia is a diverse region, and it was the population in Najd who supported Zakarawayh and crucially enabled al-Jannābī’s successors to expand outward onto the pilgrim routes and threaten Iraqi interests. Via a longer view of history, we apprehend that the mid-third/ninth century was a particularly remarkable moment for Najd, during which considerable transitions occurred right upon the eve of al-Jannābī and Zakarawayh’s arrivals.

8 9 10 11 12

Ghālib, al-Qarāmiṭa 7–8. Al-Rāshid, Pilgrim Road 83. Blachère and Pellat, al-Mutanabbī, ei2. Kennedy, Prophet, 286–287; Landau-Tessaron, Arabia 446. Munt, Construction 238.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

258 2

webb

Najd and Its Political Economy in Early Islam (to 132/750)

Arabic chronicles offer scant indicators to reconstruct the historical situation of Najd in the mid-third/ninth century, since the chronicles say little about central and eastern Arabia over the two centuries following the wars of apostasy (al-ridda) during the caliphate of Abū Bakr (11–12/632–633). Narratives about the nascent Muslim state’s campaigns against secessionist tribes in Najd and into eastern Arabia (al-Baḥrayn and al-Yamāma) project an image that the early caliphate needed to pacify the region in the immediate aftermath of the Prophet’s death, but after the initial hiccup, central Arabia was controlled. To paraphrase Hayden White, the chroniclers’ subsequent silence induces readers to assume “nothing happened”13 thereafter: Bedouin life continued without anything of wide significance transpiring. The “pacific Najd” image is challenged, however, by scrutinizing sources beyond the chronicles. There are attestations in literary accounts of luṣūṣ and ṣaʿālīk bandits,14 and even apparent attacks on pilgrim caravans by an obscure outlaw, Mālik b. al-Riyab.15 This might be explainable as low-level violence and criminal opportunism that persists in any physically challenging region, such as the desert of Najd, remote from large centers of population, but the early Islamic state was involved in power consolidation, yet via official efforts that largely escaped the attention of chroniclers. Though long predating the Qarāmiṭa, aspects of this early stage of caliphal inroads into Najd society reveal an instructive foil to the early Abbasid-era context, which then brings us to the Qarāmiṭa. Central to early caliphal statecraft in Najd was the creation of governmentcontrolled pasturing preserves (ḥimā, pl. aḥmāʾ), which spread from the edges of the Ḥijāz into the best grazing grounds of Najd around Ḍariyya during the reigns of the first caliphs.16 The preserves were intended to provide sufficient grazing for the state’s large stocks of horses and camels used in the initial conquests, and afterward, the preserves enabled the state and elite functionaries to maintain the animal wealth raised in taxes. From the local Najdī perspective, the ramifications of this pasture appropriation by the state, as a matter of practical reality and as revealed anecdotally, seem impactful. The desert upland of Najd offers relatively limited lush pasturelands, and given the nexus of camel ownership and social status among the Bedouin, 13 14 15 16

White, Content of the form 11. See a compilation of bandit details in ʿAṭwān, al-Shuʿarāʾ al-ṣaʿālīk 171–208. Al-Iṣfahānī al-Aghānī ixx, 163. A helpful list of the main aḥmāʾ is compiled by al-Samhūdī, Wafāʾ iv, 72–101.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

259

there would be natural competition over access to reliable pasture. The declaration of a ḥimā to monopolize good land accordingly was a strong assertion of authority: the ḥimā creator prevents rivals from using the land and compels them to acquiesce, fight, or migrate away. Arabic literary accounts of pre-Islam suggest that certain strongmen declared proprietary aḥmāʾ for their herds,17 and the Prophet is remembered as declaring a ḥimā over Naqīʿ and perhaps al-Rabadha, too,18 indicating express continuation of this local form of sovereignty assertion. The early caliphate’s claim to pasture rights was thus an assertion of sovereignty in a way the locals would understand, and it had further social consequences. From Najdī perspectives, the caliphate manifested as an acquisitive organism: it demanded taxes and it appropriated some of the best lands, and when evaluating Najdī-caliphal relations in this period, a core issue is to investigate what the caliphate gave the Najdīs in return. State exploitation of subject areas is a staple feature in the history of governance, but this usually operates within a locally defined norm of reciprocity that is sufficient to appease the locals and prevent their resentment turning into violence.19 As a consequence, howsoever sympathetic Bedouin groups may have been to Islam, the nascent caliphate’s expanding aḥmāʾ would restrict the locals’ grazing opportunities, and could be expected to have stirred tensions and prompted some form of negotiation that would shape the dynamic of Najdīcaliphal relations. The early caliphs do appear sensitive to the possibility of Najdī discontent and revolt, as statements ascribed to ʿUmar (r. 13–23/634–644) reveal willingness to admit some nonstate herds into the ḥimā, lest large herd owners set their animals loose in agricultural estates, and lest smaller-scale Bedouin raise angry petitions to the caliph.20 An Umayyad-era anecdote suggests that when ḥimā vegetation was sufficiently lush, a public decree would announce access for private herders,21 but while some ḥimā expansion, such as ʿUthmān’s (r. 24– 35/645–656) absorption of new wells into the ḥimā of Ḍariyya, was achieved by purchasing rights from the Ḍabīna tribe,22 other indications suggest that some administrators more jealously guarded the preserves at the expense of locals, 17 18 19

20 21 22

The most celebrated pre-Islamic ḥimā declarer is Kulayb b. Rabīʿa, a leader of the Wāʾil and associated with assertions of sovereignty, see Ibn Nubāta, Sarḥ 92–93. al-Samhūdī, Wafāʾ iv, 72, 78. Much research on state appropriation and reciprocity is conducted in the context of early modern and colonial societies; Scott, Moral economy 180–192 sets out the theories, and their lessons seem sufficiently fundamental to help us consider premodern cases, too. Wakīʿ al-Ṭarīq, 346; al-Samhūdī, Wafāʾ iv, 78–79. Al-Samhūdī, Wafāʾ iv, 76–77. Al-Samhūdī, Wafāʾ iv, 87.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

260

webb

converted water resources surrounding the reserves into agricultural estates for themselves, or created new personal aḥmāʾ without caliphal permission.23 Complaints against greedy administrators feature in early Muslim-era Arabian poetry, such as the verse of Kaʿb b. Maʿdān of the Ashqar: One does one’s best to keep one’s own, But the administrators are wolves on your land.24 It also appears that some ḥimā essentially became their administrators’ private fiefs, handed down from one generation to the next.25 Caliph al-Walīd b. ʿAbd al-Malik’s (r. 88–98/705–715) maternal uncle, Ibn Khulayd al-ʿAbsī, reportedly made constructions for his own use in the territory of the Ghanī around the ḥimā of Ḍariyya; evidently to the considerable chagrin of the Ghanī, as they are said to have destroyed his installations when the Abbasids replaced Umayyad rule.26 Elite Ḥijāzī families’ control over Najdī territory reveals another aspect of land appropriation from local hands. For the early caliphate, maintenance of the aḥmāʾ was evidently important enough to merit continuous official involvement. The benefits for the caliphate are clear: the aḥmāʾ provided means to preserve state wealth as represented in the large herds raised from taxation, which would need concentrated pasture, and the aḥmāʾ enabled some physical presence in the otherwise underadministered Najd, where the caliphate’s government (sulṭān) did not establish major centers. Accordingly, most aḥmāʾ remained officially administered throughout the Umayyad period via the governor of Medina who appointed a deputy (ʿāmil pl. ʿummāl) to maintain them.27 Even if elite Ḥijāzī families such as the Zubayrids and the descendants of ʿUthmān were entitled to run particular aḥmāʾ, the formal relationship with gubernatorial power was officially upheld. Alongside the deputies, the government also adopted a pragmatic approach vis-à-vis the locals: while Najdīs technically lost control over ḥimā land, some were nonetheless incorporated by appointment as “rangers” (ḥuwwāṭ) to maintain security, and local tribal leaders were promoted to head the ranger groups.28 Co-opted clans thus could continue to use the pasture pre-

23 24 25 26 27 28

Al-Samhūdī, Wafāʾ iv, 78, 84, 87–88. Al-Jāḥiẓ, al-Bayān iii, 358. Al-Samhūdī, Wafāʾ iv, 76, 84. Al-Samhūdī, Wafāʾ iv, 94. Al-Samhūdī, Wafāʾ iv, 94. Al-Samhūdī, Wafāʾ iv, 87, 94.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

261

serves, but references to local intrusion into the aḥmāʾ29 and the engagement of the ḥuwwāṭ in skirmishes against would-be intruders reveals that not all groups were equally incorporated.30 It also appears that complications over pasture access could have serious repercussions, including violence, petitions to the government for justice, and migration of groups who lost control of land.31 Beyond the aḥmāʾ, the Umayyads did not take significant further measures to integrate Najd into their realm. In contrast to the Umayyads’ extensive contemporaneous development of the Ḥijāz,32 and their attention to maintaining communication between their Syrian holdings and their homelands near Mecca via the northern Ḥijāz and Wādī Sirḥān, no material efforts were expended to develop communication or waystations in Najd along the routes towards Iraq.33 In terms of the negotiated balance between caliphal interests and Najdī locals, a “norm of reciprocity”34 may have been expressed in part by curbs on the caliphate’s exploitative intervention: the Umayyads did not sequester all good Najdī pasture—reportedly, pasture around Fayd was granted to the Ṭayyiʾ outright, presumably as a means to appease them.35 Land appropriations in central Najd were significant, however, and the early caliphal system was overall prejudicial to the Najdīs. Locals lost some of their best pasturelands to the sulṭān while receiving very little infrastructure or integration into the government in return. Scattered ʿummāl maintained a limited administrative presence, but most government was run in faraway Medina, and thus the Najdīs were largely left to their own devices to resolve quotidian issues, and low-level conflicts could continue to smoulder unless one of the parties actively sought to involve the sulṭān. The system logically engendered latent instability: good land was appropriated, while the government was otherwise largely absent. References to outlawry in Umayyad-era Najd thus could be read as acts of resistance to the appropriating and nondistributing government,36 and there are helpful the-

29 30 31 32 33

34 35 36

Al-Samhūdī, Wafāʾ iv, 76. Al-Samhūdī, Wafāʾ iv, 87–88. Al-Samhūdī, Wafāʾ iv, 87–88, 92. See Munt, Trends; and Munt, Caliphs. Al-Rāshid, Darb Zubayda 37–44 discusses the Umayyad work building up the Hajj infrastructure, however, closer analysis of the sources suggest that their attention was actually light, and in Najd, particularly scant, see Webb, Power. A term borrowed from Scott, Moral economy 167–176. Al-Samhūdī, Wafāʾ iv, 102. One thinks, though necessarily loosely, of the Hobsbawmian-style “social bandits” whose banditry results from a changing status quo attendant upon the rise of the caliphate

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

262

webb

oretical parallels with the “social bandit” figure most thoroughly theorized by Hobsbawm. He posits that social bandits emerge in conditions where the status quo is disrupted in ways that cause new hardships on the locals, and where central power has relatively weak control.37 The early caliphate in Najd can indeed be read as a change to the status quo and transformations of structures of power and norms of legitimacy, and the incidents of small-scale banditry in Najd—where central authority was weak—without triggering full-scale revolts suit Hobsbawm’s conceptions rather well.38 If the Umayyad-era outlaws were indeed taking a measure of self-help against the state, there were scarce tangible assets against which they could direct their violence since the state did not create much infrastructure. It seems that intertribal relations were the most operative local politics, while the latent instability caused by intergroup jockeying over reduced pasturing opportunity was diffused by local stakeholders allied to the sulṭān as ḥuwwāt/rangers. Accordingly, the anecdote of al-Qāsim b. Jundab, a leader of the Fazāra who reportedly neither concerned himself with the Hajj, visited settlements, nor traveled beyond his own lands, seems a reasonable representative of much Najdī experience under the Umayyads.39

3

Early Abbasid Najd: the “Darb Zubayda”

Whether or not the Umayyad system would have been sustainable over a longer term became moot since the geopolitical shift engendered by the rise of the Abbasid Caliphate in Iraq engendered immediate transformations in Najd’s significance in the eyes of officialdom—an important new significance that has eluded comment in modern studies. In order for Iraqis to travel to Mecca

37 38

39

(Hobsbawm sets out conditions for social banditry and the weakness of central power in Bandits 9–17). Substantiating this particular angle of banditry as protest, however, needs careful consideration, especially given the limitations of Hobsbawm’s theory in practice. Hobsbawm, Bandits 9–17, 24–26. Note that Hobsbawm expressly excluded Bedouin from his analysis (Bandits 20), though the rationale for this is hasty, and it appears Hobsbawm was not sufficiently familiar with Arabic material to identify the parallels. He was, moreover, particularly keen to focus his analysis on peasant societies to posit social banditry as resistance to capitalism. The Umayyad-era Najd is accordingly far from Hobsbawm’s world, and while his theories have been as widely critiqued as they have been admired, some of his lessons do appear to help aspects of understanding Arabian banditry and should be further tested. We return to Hobsbawm in the final section of this chapter. Al-Samhūdī, Wafāʾ iv, 91.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

263

and Medina, they had to journey across the length and breadth of Najd, and because the early Abbasids were particularly keen patrons of the Hajj as a kind of procession to demonstrate their legitimacy, it became paramount for them to maintain communication pathways through Najd and ensure peaceable relations with Najdī tribes.40 The practical manifestation of Najd’s theoretical importance took the form of official attention right from the outset of the Abbasid era. The caliphs sponsored unprecedented investment in Najdī infrastructure, building roads, way stations, storehouses, wells, water channels, and cisterns to facilitate pilgrim traffic. Although known today as “Zubayda’s Way” (Darb Zubayda), after Zubayda bt. Jaʿfar, wife of Caliph al-Rashīd (r. 170– 193/813–833) and patron of many infrastructural works, much of the intensive construction was undertaken before Zubayda: it began with the first Abbasid, al-Saffāḥ (r. 132–136/750–754),41 considerable work was ordered by al-Manṣūr (r. 36–58/754–775), and al-Mahdī (r. 158–169/775–785) was especially active.42 The cumulative effect was that within 40 years of the Abbasid takeover, the 800-mile track from Iraq to the Ḥijāz was furnished with over 50 waystations, 1,200 wells, and some 90 reservoirs/cisterns (birka/ḥawḍ) for storing rain and groundwater, also complemented by extensive canal networks channeling water to convenient points for pilgrim travelers.43 The work required continual maintenance, too, as wells can run dry without upkeep: even Mecca’s Zamzam had shortages in the Hajj of 190/806, prompting Zubayda to sponsor full-scale redigging.44 While the infrastructure naturally eased Iraqi urbanists’ transit through Najd, the continuous work also had an impact on Najd itself, which had a momentous local effect that scholarship has yet to fully explore. The food stores, markets, and access to water that the pilgrims could avail themselves of once a year were available year-round to the locals, providing unprecedented food and water security. Also, the tremendous manpower needed to under-

40

41 42 43 44

The prospects of the active need to foster security in Najd perhaps bear connection to events in eastern Arabia at the outset of the Abbasid period: in 151/768 (or 152/769) the locals of Yamāma killed the governor of Caliph al-Manṣūr, and further insurrections continued during al-Mahdī’s reign. After considerable effort, al-Mahdī successfully defeated the rebels, and it can be appreciated that officialdom did not relish the prospects of constant security actions to maintain the Hajj roads. He ordered the construction of new milestones and fire signals, as well as the construction of rest houses/forts (qusūr) for pilgrims (al-Ṭabarī, Tārīkh vii, 465). Al-Ṭabarī, Tārīkh viii, 136. Textual and archaeological evidence for the works are detailed in al-Rāshid, Pilgrim road, and significantly updated in al-Rāshid, Darb Zubayda. Al-Yaʿqūbī, Tārīkh ii, 428–429.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

264

webb

take the 50-year infrastructure projects involved locally raised labor,45 offering equally unprecedented work opportunities and further food security for Najdīs. Further roads, including an official postal route (barīd), were constructed to cross Najd, equipped with officials, entailing a much denser presence of the sulṭān in the region as well. In sum, the Abbasid attention to Hajj road building both provided for the Najdīs and integrated them into regularized interaction with the government. Al-Mahdī is also associated with the end of the aḥmāʾ system: al-Samhūdī (d. 911/1506) observes that during al-Mahdī’s caliphate the aḥmāʾ were “let open” (ubīḥat),46 and the evidence of family control over aḥmāʾ also seems to cease contemporaneously (i.e., in the third quarter of the second/eighth century).47 The coincidence of the abandonment of aḥmāʾ and the undertaking of intensive infrastructure projects suggests changes in official interaction with Najd and a shift from the largely one-sided system of the Umayyads to reciprocal arrangements with manifest benefits for the locals. As a practical matter, the Abbasid concessions can be seen as easy for the state to make: the Abbasids had little need for Arabian animal wealth; their vast incomes from Iraqi agriculture relegated the aḥmāʾ to comparative fiscal insignificance, and they could accordingly be relinquished, especially given the benefits in doing so. The abandonment must have had the positive effect of appeasing Najdīs by returning the best pastures to free grazing,48 and since nomadic groups were already well-stocked via the expanding new markets and storehouses, and wellwatered given the massive expansion of wells and reservoirs, the return of ḥimā pastureland would have been a further boon. We can appreciate that numerous Najdī groups were sufficiently employed and supplied thanks to the Darb Zubayda so as to obviate the urgency of competition over pasture, and hence on a wide scale, Najdīs reaped the benefits of a surplus, opening the scope for an unprecedently felicitous chapter in Najdī-sulṭān interactions in the late

45

46 47 48

Some arduous infrastructural work, such as paving parts of the desert track and the manual carrying of water was performed by slaves owned by the Abbasid caliphal families (e.g., Wakīʿ, al-Ṭarīq 59; al-Ṭabarī, Tārīkh viii, 539), but these appear to be exceptions to the wider-scale levy of locals variously attested in the sources; see al-Ṭabarī, Tārīkh viii, 539; al-Yaʿqūbī, Tārīkh ii, 448. Al-Samhūdī, Wafāʾ iv, 83. Al-Samhūdī, Wafāʾ iv, 83, 89. Bedouin pride in grazing at Ḍariyya is articulated in an anecdote of an exchange between a Bedouin and al-Mufaḍḍal b. Isḥāq, a relatively obscure philologist of the generation before al-Aṣmaʾī (d. 213 or 216/828 or 831), i.e., active in the early Abbasid period (al-Samhūdī, Wafāʾ iv, 99–101).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

265

second/eighth century, perhaps ultimately intended by the caliphs to provide security across Najd without requiring large garrisons to police the long and exposed Hajj roads. The greater sharing of resources between officialdom and Najdī locals, and the construction of pilgrim infrastructure alongside local camping sites are also discernible in geographical texts. For examples, at the way station of al-Nuqra the al-Fazāra occupied the surrounding steppe (bādiya) while “Quraysh and merchants” owned the town (ḥāḍira).49 The Banū Muzayna and the Quraysh co-owned al-Rawḥāʾ;50 the way station at Fayd was apparently under direct local control, as the town’s rulers (wulāt) were said to be from the Banū Nabhān.51 The indications reveal increased co-option of locals and balances of power established between local and caliphal/Qurayshi groups expressed through land usage. Viewed from a Najdī perspective, the stark contrast between the Abbasid energies and the dearth of Umayyad-era activity need be stressed. The experience of previous generations of Najdīs accustomed to the distant and theoretical power of the sulṭān was rapidly overturned under the early Abbasids into a scene of regular, more cooperative, and certainly more beneficial contact. Also unlike the Umayyads, the first Abbasid caliphs performed multiple pilgrimages during their reigns with large entourages, and thus the period of 132–193/750– 809—the rise of the Abbasids to the end of al-Rashīd’s caliphate—witnessed an extraordinary physical presence of the sulṭān in Najd alongside unparalleled availability of food and water.52 While Abbasid energies can be appreciated as positively effecting Najdī populations and abetting security, follow-up consequences also bear consideration, particularly two issues regarding expectations of justice and ramifications of charity. Concerning justice, it is noted that the Abbasid caliphs projected their legitimacy in venerable Mesopotamian terms as water givers and justice dispensers,53 and in the context of this chapter, the Abbasids explicitly intertwined the Hajj into that traditional ideal-leader legitimization. Early Abbasidera poetry evidences the connection through a sobriquet of the Abbasid house as “The tribe of the water-giver for pilgrims” (Banū Sāqī al-Ḥajīj), a hearkening of the Abbasids’ ancestor Abū Ṭālib’s right to provide water to Hajj pilgrims

49 50 51 52 53

Wakīʿ, al-Ṭarīq 76. Wakīʿ, al-Ṭarīq 201–203. Wakīʿ, al-Ṭarīq 62. For lists of pilgrimages led by the caliphs in person, see al-Maqrīzī, al-Dhahab; for an analysis of their political significance, see Webb, Power. Discussed in Darling, History; and Sperl, Islamic kingship.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

266

webb

on the eve of Islam. A verse by Marwān b. Abī Ḥafṣa (d. 182/798–799) praising al-Mahdī exemplifies the rhetoric’s employment to underscore the caliph’s legitimacy and righteousness: A new Caliph from the tribe of the Water-giver for Pilgrims, Upon his face the light of truth shines witness.54 Putting the rhetoric into practice, the manifold wells that the caliphs ordered dug into Najd seem physical manifestations of the discourse, and to this point, it is noteworthy that those wells were known locally by the name of the patron,55 emphasizing the nexus between the person of the ruler and dispensation of water, invoking the paradigm of just provision. For the Najdīs the permanent presence of wells bearing caliphal names was also augmented via access the locals could enjoy to the caliph’s person during those multiple caliph-led pilgrimages. Access to the ruler afforded opportunities to seek justice, further embedding legitimate connection between ruler and ruled, as illustrated in an anecdote in al-Masʿūdī’s Murūj al-dhahab of a petition and act of caliphal clemency.56 Further references to caliphal charity for populations on the Hajj routes via public distributions and lavish gifting on procession57 suggest that the repeated Hajj pilgrimages of Abbasid caliphs did involve the typical performative acts of just rulership, and the Darb Zubayda in its totality may be readable in this light as dually (a) a practical project for Iraqi pilgrim transit, and (b) a venture of symbolic value addressed both to the caliph’s Iraqi subjects whose pilgrimages were fundamentally facilitated by caliphal patronage, and to the Najdī constituency who were integrated into the sphere of the rulers’ justice via raising expectations of access to justice, including water supply. Explicitly, too, the performance of the Hajj conjoined with charity is a trope of Abbasid-era praise poetry that constructs the image of the caliph as divinely aided succour for his subjects, such as the second/eighth-century poet and poetry narrator Dāwūd b. Razīn’s praise of al-Rashīd:

54 55

56 57

Marwān b. Abī Ḥafṣa, Shiʿr 37. See Shiʿr 54 for the phrase’s use in praise of al-Rashīd. For further discussion of water provision as power legitimization, see Zadeh, Early Hajj 55–60. Consider, for example, the repairs at Zamzam identified with Umm Jaʿfar, after Zubayda, see al-Fākihī, Akhbār iii, 155; along the Hajj routes, the naming of wells after caliph sponsors is widely evidenced in Wakīʿ, al-Ṭarīq. Al-Masʿūdī, Murūj iv, §§ 2,514–2,517. For examples, see al-Yaʿqūbī Tārīkh ii, 407; al-Dīnawarī Akhbār 562; al-Ṭabarī Tārīkh viii, 243; Ibn al-Jawzī, al-Muntaẓam iii, 207; al-Maqrīzī, al-Dhahab §96.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

267

Light shines in every land with Hārūn, The straight path is his just method. He is an Imam: his work directed to God, His concerns are for Jihad and Hajj. Men’s eyes are blinded by his light When graced by his serene appearance. In God’s protection, Hārūn the Generous Gives the hopeful more than they ever hoped.58 Contemporary poetry reveals that the gift-giver image spread also to the high echelons of elite around the caliph, for example a woman from the Banū Kilāb praised the Barmakid vizier and governor Jaʿfar b. Yaḥyā (d. 187/803): I went past al-ʿAqīq, and its people Were complaining that their spring had no rain. But a rain-less spring is really no worry, If Jaʿfar is giving gifts!59 While the elaborate displays of charity as demonstration of legitimate leadership was a cornerstone of the performative aspects of caliphal rule and highestlevel elite identity, ordinary pilgrims were also enjoined by Islam to make distributions when traveling on Hajj, and Nadji locals benefitted from these smaller-scale distributions, too. Anecdotally, such distributions are attested, for example, in a story about a poet contemporary with the caliphates of alMahdī and al-Rashīd, Abū l-ʿAtāhiya, who encountered a Bedouin on the road to Mecca: We saw a Bedouin in the shade of a milestone with a short woollen cloak which either covered his head and left his feet sticking out, or covered his feet and left his head bare. Abū l-ʿAtāhiya asked him: “Why did you chose to live in this barren wasteland instead of good fertile land?” The Bedouin responded: “Well, if God had not made some of his creation content to live in rough land, then the good land would become too crowded!” “So how do you live?” Abū l-ʿAtāhiya asked.

58 59

Al-Ṭabarī, Tārīkh viii, 243. Ibn Khallikān, Wafayāt i, 330.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

268

webb

“From you Hajj pilgrims—you come by and we accept charity from you, and then you return and we accept more charity.” “But we only make the journey during one part of the year, how do you live the rest of the time?” The Bedouin looked down, and then said: “By God, I don’t know—I can only say [paraphrasing Quran 65:3] we are given sustenance from where we do not expect it, more than we make a living from where we expect it!” Abū l-ʿAtāhiya turned and recited the following poem: Oh seeker of worldly things! Leave the world to your opponents. Why do you bother with transient pleasure? Isn’t the shade of this milestone enough?60 Express association of Iraqi Hajj pilgrims with sustenance also materializes in Bedouin poetry composed about way markers and fire beacons that dotted the tracks through Najd. One poet contemplates the beacon at Umm Khurmān, where the Kufan and Basran Iraqi Hajj roads met: Oh Umm Khurmān, raise up your flame Do you see the men leading she-camels? Your fire has long been extinguished, Are you asleep, or do you need wood?61 A second poem, by a Bedouin from the Hudhayl, addressed the same beacon: Oh Umm Khurmān, send up a flaming light The porridge and flour have run out!62 The verses indicate local reliance on trade from the Hajj caravans: the lighting of the fire tower emerges as a metaphor for the impending Hajj, the consequent influx of travelers, and hence the replenishment of cereals. In wishing for the beacon to be ignited, the poets allegorically express yearning for Iraqi pilgrim charity and eloquently demonstrate the integration of Iraqi and Arabian economies a millennium ago. In 186/802, al-Rashīd performed the penultimate of his many pilgrimages, and on this occasion, in the presence of a great courtly entourage summoned 60 61 62

Al-Iṣfahānī, al-Aghānī iv, 86. Yāqūt, Muʿjam al-buldān i, 251. Yāqūt, Muʿjam al-buldān i, 251.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

269

to Mecca, he announced his succession plan, duly sanctified with inscriptions placed in Mecca and oaths sworn by his two successor sons in the Sanctum. Two years later, in 188/804, al-Rashīd led the Hajj again, reportedly covering the entire distance on foot;63 and then, remarkably, after this climax of procession and piety, the denouement of the caliphal pilgrimage was abrupt. Never again would an Abbasid caliph perform the Hajj, and never again would a caliph make an in-person appearance along the Hajj roads. Najd, after benefitting from over 50 years of constant attention from the sulṭān, intensive building, and extensive food and water provision, fell from the caliphal gaze as the third/ninth century dawned. The long-term effects of the several-generations’ worth of provision for Najdīs and their growing expectation of and, indeed, likely dependency on charity was met with near complete cessation of Darb Zubayda construction and maintenance. The stark contrast caused significant issues for generations of third/ninth-century Najdīs.

4

The Waning Darb Zubayda

Al-Rashīd’s last pilgrimage did not instantly signal complete cessation of the Darb Zubayda, since pilgrimage by the general Iraqi populace continued, and al-Rashīd’s wife Zubayda and Caliph al-Maʾmūn engaged in an intrafamily competition via sponsoring water-provision projects in Mecca in 211–212/826– 827.64 Zubayda also continued performing the Hajj after al-Rashīd’s death, but sources do fall silent on infrastructural projects in Najd. Reasons for the stoppage naturally stem from the Fourth Fitna (193–211/809–827), the protracted and disastrous internecine struggle between al-Rashīd’s cosuccessors, al-Amīn (r. 193–198/809–813) and al-Maʾmūn (r. 198–218/813–833), whose infighting resulted in the sack of Baghdad and devastation in the Iraqi countryside, with seriously deleterious consequences for Iraqi agriculture and economy, and which spread violence across the caliphate, including into Arabia. In 200/815 ʿAlid claimants rebelled against the Abbasids both in Mecca and in Zubāla, a station on the Hajj road in Najd; the ʿAlid aspirants mustered Bedouin auxiliaries, and almost simultaneously, a whole tribal group in Najd, the Banū Nabhān, rose in sympathetic revolt.65 Al-Maʾmūn’s forces defeated the threats in Najd and Mecca by 202/817; however, al-Maʾmūn remained in his eastern Iranian strong-

63 64 65

Al-Maqrīzī, al-Dhahab § 95. Al-Fākihī, Akhbār iii, 152–155. Al-Iṣfahānī, Maqātil 541.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

270

webb

hold, only returning to Iraq in 204/819, and throughout the remainder of his reign, to 218/833, he never ventured into Arabia. The end of al-Rashīd’s caliphate and the reigns of al-Amīn and al-Maʾmūn, viewed from a Najdī perspective, translate into a 30-year absence of the caliph’s presence. There had been no caliphal pilgrim processions with excessive charity, nor were more wells dug, or old wells expanded or vigilantly maintained. Given the symbolism of the caliphal patronage along the Darb Zubadya, this absence would have symbolic significance of seemingly abrupt abandonment. There were also practical ramifications. The water infrastructure of cisterns and canals was vulnerable to silt and sand and needed regular maintenance, yet there is no indication, amid the Fourth Fitna crisis, of substantial, or indeed any, work undertaken in Najd. From the perspective of the Bedouin rebels in Zubāla in 199/815, we behold a group who had witnessed more than a decade of relative neglect, which may have abetted local receptiveness to new claims of the local ʿAlid rebel to provide a more just rule for local interests. Al-Maʾmūn’s successor, al-Muʿtaṣim (r. 218–227/833–844) likewise is ascribed no works in Najd. This can be rationalized given that al-Muʿtaṣim’s reign is famed for the construction of the new palace city in Samarra and the exponential growth of his slave armies purchased from the eastern borders of the caliphate. The extensive fiscal outlay to build a new private army and construct a new private palace city was incurred against the backdrop of declining Iraqi agricultural yields, hence there would be limited surplus for restoring infrastructure damaged during the Arabian unrest and the lengthy neglect after the Fourth Fitna. As for the Hajj, al-Muʿtaṣim’s involvement was limited, too: he is said to have richly embellished a canopy over the Zamzam well in Mecca in 220/835, again performing the “ruler/water-giver” role,66 but he made no plans to make the pilgrimage himself, and one anecdote suggests he planned for his slave army to make their Hajj around his new palace in Samarra instead.67

66 67

Al-Fākihī, Akhbār ii, 75–76. The anecdote is only reported by the later geographer al-Muqaddasī (d. 378/988), Aḥsān 122–123; see discussion in Gordon, Breaking 65–66. While there are grounds for inferring an anti-Abbasid polemic in the account, it is not, on its face, part of an overt negative narrative. Al-Muʿtaṣim’s efforts to enforce the strict loyalty of his personal guard are well-attested, and when read in light of the dearth of his Hajj infrastructural work, there may be more to al-Muqaddasī’s anecdote than previously considered. Northedge, Qubbat 78–79 also considers al-Muqaddasī’s story to have a factual background, and cogently evaluates it in light of extant archaeological evidence. While Northedge does not consider al-Muʿtaṣim’s structure as an outright replacement of Mecca, I invite still broader analysis that interprets the project within the wider context of al-Muʿtaṣim’s overall lack of attention to Hajj infrastructure and the rationale behind his expansive private army in Samarra.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

271

By al-Wāthiq’s caliphate (r. 227–232/842–847), therefore, Najd had experienced two diametrically opposed faces of the Abbasid Caliphate. The energy of the first 50 years was followed by 40 years of virtually no attention, and while the Hajj continued with the infrastructure of the early Abbasids coasting along, the lack of upkeep, in particular for water resources, would eventually lead to shortages, and this is precisely what transpired. Unlike the previous three caliphs, al-Wāthiq’s court did manifest interest in the Hajj: al-Wāthiq’s brother Jaʿfar led the Hajj in 227/842 (the highest-ranking prince to do so for a generation), and his mother traveled that year, too (though she died en route),68 and while al-Wāthiq declared intention to make the Hajj himself in 231/846, by this time, circumstances had made it impossible. Water resources were at a crisis level, and al-Ṭabarī reports that a draught of water during the pilgrimage in 228/843 rose to 40 dirham (perhaps a 600 % increase since al-Rashīd’s reign), and food was also scarce: a measure of bread reportedly rose to one dirham.69 Iraqi pilgrims at least could return to the relative prosperity of their urban environments after that arduous Hajj, whereas the Najd’s Bedouin, on the other hand, had no such escape. Their voice is not heard in Iraqi chronicles, but they were clearly the most disadvantaged constituency under the new state of affairs as the life-yielding infrastructure created in Najd during the previous generations began to crumble. The new realities of Abbasid Iraq, where wealth centralized in Samarra and less reached the Hajj roads, tempered the charitable relationship between the sulṭān and Bedouin, and ordinary Iraqi pilgrims were now both poorer and (given the food and water scarcities) more focused on their own survival before considering charity to Bedouin. Unfortunate coincidences of the calendar added to the misery, as Hajj in the late 220s/early 840s occurred in the summer, entailing that pilgrim caravans faced crossing Najd in the hottest conditions. Perhaps the most jarring aspect of the “lean” years for Najd was their proximity to the “good” first 50 years of the Abbasids. While the Umayyad-era Najd had been left largely to its own devices, such conditions were no longer the expected status quo of lived experience for Najdīs of early Abbasid times. The 50-year period when the sulṭān provided work and food, and when the caliph provided water, would have been fresh in Bedouin communal memory: Umayyad times were a distant memory, if recalled at all, whereas all would be aware of the better days, and psychologically speaking, frustration at the new state of affairs is a natural outcome. Given the reported scarcity of water

68 69

Al-Ṭabarī, Tārīkh ix, 123. Al-Ṭabarī, Tārīkh ix, 124.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

272

webb

and food, the symbolic disappearance of the caliph thus coincided with a slip in resource availability below subsistence levels; such combination of nutritional lack with moral indignation is noted in studies from other contexts as a situation with high potential to transform discontent into actual violence.70 Demographically, the problems were quite probably exacerbated given that the extensive charity, security, water provision, and integration with government works over the three generations of the early Abbasids would have combined to provide for an increase of the Najdī population beyond the natural limits of the land. Consecutive generations of Najdīs with increased opportunities for pasturing herds, and with unprecedented food and water security, would have experienced reduced child morality and perhaps even extended life expectancy for the elderly. Thus, the hungry and morally indignant Najdīs would also have been more numerous than usual, and by the period of al-Wāthiq’s reign, the first generation of Najdīs to have lived entirely outside of caliphal attention to the Darb Zubayda would have been of military age (i.e., in their 20s). The overflow of the potential for tension is graphically evidenced. In 231/845 the Banū Sulaym, on the edges of the Ḥijāz near Medina rose in revolt, defeating and killing the commander of a force sent against them by the governor of Medina, and fanning a wave of violence into Najd, including revolts by the Hilāl, Fazāra, Ghaṭafān, Murra, and Numayr.71 Within a year, al-Wāthiq’s generals had defeated the restive tribal groups, but a vicious cycle had commenced. Bedouin raids on government storehouses and markets may have alleviated short-term privation, but they damaged the already teetering infrastructure and also frightened would-be pilgrims (i.e., future sources of charity), and thus the increasingly desperate Bedouin made the underlying problems increasingly severe. The effects were already evident in the 232/847 Hajj, as water shortages became even more acute: a draught of water reached several dīnār, and al-Ṭabarī reports that pilgrims died of thirst in the intense summer heat around al-Rabadha.72 Al-Wāthiq’s response attests to our hypothesis that a root cause of the Bedouin violence was indeed the collapsing infrastructure system, since alWāthiq’s efforts in the wake of violence, beyond dispatching soldiers, was to arrange major repairs in Najd. In 230/845 he appointed two superintendents, ʿUmar b. Faraj over the track from Kufa, and Isḥāq b. Ibrāhīm b. Abī Kham-

70

71 72

Noting that rebellion is in fact a rare expression of discontent, Scott (Moral economy 114–156) considers the tipping points in agrarian contexts. Notwithstanding the different contexts, there are fundamental parallels to Najd in the mid-third/ninth century. Al-Ṭabarī, Tārīkh ix, 129–133, 146–150. Al-Ṭabarī, Tārīkh ix, 150.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

273

īsa for the Basran route, and in the next year, al-Wāthiq proposed to make the Hajj himself, but water was still insufficient and so more funds were allocated for the repair of some 20 wells, the replacement of milestones, and easing of the track.73 In the wake of the sudden flare of widespread violence, the actions of the government reveal an understanding that infrastructural conditions needed remedy, but in 232/847 al-Wāthiq died, and the fate of Najd was left in the hands of his successor al-Mutawakkil (r. 232–247/847–861). Unlike al-Wāthiq, al-Mutawakkil does not appear concerned to make the Hajj as a procession for his legitimate rulership, but al-Mutawakkil did continue infrastructure work in Najd, though on a reduced scale—most references concern the repair of wells, which, in light of the water shortages, represents the most urgent work, and those which connect to the paradigm of justice, too.74 One set of work was near the way station of al-Maslāḥ, where water was reportedly of poor quality (māʾ ghalīẓ), and repaired wells were apparently shared with the local Banū Sulaym;75 given their revolt in the previous decade, it is suggestive that the sulṭān well understood that restive groups should be appeased with attention to infrastructure. However, al-Mutawakkil’s reign had fiscal restraints of its own, and he committed lavish funds to expand palaces near Samarra, hence there were limits to the amounts allocable to Najd. After this period, the official approach to managing the pilgrimage shows signs of increasing reliance upon militarization, as the Hajj superintendent’s strategy came to involve employment of the Shākiriyya, a constabulary of several hundred Turkic soldiers to accompany the Hajj caravan for security.76 These were not a police force for Najd, but rather an ad hoc accompaniment for Hajj caravans. The caravans apparently needed such security, for example under ʿAbdallāh b. Sulaymān the Shākiriyya warded off Bedouin raids during the 251/865 Hajj.

5

The Darb Zubayda’s Nadir: The Late Third/Ninth Century

The prospects for reviving the early Abbasid infrastructure in Najd took a decided downturn with the assassination of al-Mutawakkil and the rapid deterioration of caliphal authority in Iraq throughout the third quarter of the

73 74 75 76

Wakīʿ, al-Ṭarīq 286–287. Wakīʿ, al-Ṭarīq 73, 95, 98. Wakīʿ, al-Ṭarīq 90, 98. Al-Ṭabarī, Tārīkh ix, 293.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

274

webb

third/ninth century. In this period, references to repairs to the road are scant,77 while mentions of superintendents of the Hajj roads and Najd become sporadic and less organized. Names of officials in the mid-third/ninth century are recorded, but with considerable flux, suggestive of instability. The superintendent in 259/873 was killed by Bedouin raiders. In 260/874 Muḥammad b. Abī l-Sāj was appointed in charge of repairs, pilgrim security, and as governor of Mecca, but by 270–271/884 the road had a new superintendent, Aḥmad b. Muḥammad al-Ṭāʾī, and then it appears the post was vacant until 293–294/906. While the appointment of superintendents indicates the technical allocation of funds for improvements in Najd, the superintendents are also recorded as embezzling them, and competition over the post heightened as it became an opportunity for private enrichment.78 Not surprisingly, conditions on the road spiraled downward: al-Ṭabarī reports that the 257/871 Hajj caravan turned back at the Najdī way station al-Qarʿā on account of water shortage, though he adds that those pilgrims who did push on alone reached Mecca.79 Again, underlying the economic hardship, which evidently affected Najdī communities in this period, we encounter a group of 50 Bedouin from alThaʿlabiyya joining the Shākiryya for cash,80 and for so long as there was money to pay, the sulṭān could co-opt some military-aged Bedouin men into applying their energies to guard pilgrims, but this represents a dangerous militarization of the desert, and in the later third/ninth century, the caliphate did not have the same military resources to spare as al-Wāthiq had enjoyed when quelling the unrest in 230–231/845–846. As a result, several ʿAlid rebels appeared in the Ḥijāz to claim Mecca, such as Ismāʿīl b. Yūsuf, who mustered a “conglomeration of Bedouin” (lafīf min al-aʿrāb)81 and engaged in a protracted conflict against the authorities, even attacking pilgrims performing the Hajj in 251/865.82 Records of Bedouin raids on caravans and outposts in Najd also increase. In 259/878 the Banū Asad attacked pilgrims near al-Mughītha and killed the Hajj road superintendent. In 268/882 various Bedouin groups attacked pilgrims returning from Hajj between Tūz and Samīra. In 285/898 the Ṭayyiʾ attacked the returning pilgrims at al-Ajfar.83 And in 286/899 the Shaybān marauded Iraq itself, and when

77 78 79 80 81 82 83

See summaries in al-Rāshid, Darb Zubayda 54–55. The corruption of Ibn Abī l-Sāj is detailed in al-Ṭabarī, Tārīkh ix, 371–372. Al-Ṭabarī, Tārīkh ix, 501. Al-Ṭabarī, Tārīkh ix, 293. Al-Yaʿqūbī, Tārīkh ii, 498. Al-Ṭabarī, Tārīkh ix, 346–347; al-Iṣfahānī, Maqātil 669, 719; a slightly different version in al-Yaʿqūbī, Tārīkh ii, 498–499. Al-Ṭabarī, Tārīkh x, 67.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

275

pursued by a large force, they retreated into Arabia where the central authority could take no punitive measures.84 The Ṭayyiʾ launched further attacks on pilgrims in 287/900, 292/906, and 293/907.85 Economic aspects of the unrest surface during the caliphate of al-Muqtadir (r. 295–320/908–932), whose governor ʿAlī b. ʿĪsā arranged for the employment of 5,000 tribesmen of the Asad to serve as a regiment of Hajj guards for the 315/927 pilgrimage, indicating Najdī willingness to cease raiding in return for employment. The large number involved further reveals the extent to which the decades of unrest had militarized the Hajj route. Sources give the sense that the late third/ninth-century Bedouin were essentially unchecked along the former Darb Zubayda as there were no permanent guards for the vital wells, reservoirs and way stations. The Kitāb al-Ṭarīq, a detailed account written at end of the third/ninth century about the Hajj roads offers remarkable testimony about the number of former amenities—way stations, storehouses, and fortifications—which it identified as in ruins (kharāb), and wells and cisterns it describes as ruined and dry.86 While the text does note good availability of water and supplies in relation to six locations, including “new wells” alongside older early Abbasid-era constructions,87 and while there is specific mention that “people had afterwards returned” to a fortification at Maʿdin al-Qurashī destroyed in the ʿAlid uprising of Aḥmad b. Ḥasan b. Jaʿfar in 273/886–87,88 there is no reference to any works of patronage after al-Mutawakkil, and the available resources are four-fold less than references to ruins. The continuously maintained Darb Zubayda of the previous century was largely a memory, and Hajj pilgrims approached Mecca with trepidation, in columns bristling with arms, across a hostile and unfamiliar desert.

6

The Rise of the Qarāmiṭa: Theoretical Help

The rise, the sensational violence, and the rapid success of the Qarāmiṭa in Najd coincides precisely with the spike in raids ascribed to Najdī tribes in the 280s–290s ah. Though the Iraqi historiographical sources portray the violence as “raids” and “plunder,” these incidents resemble the category of “small wars”

84 85 86 87 88

Al-Ṭabarī, Tārīkh x, 71–72. Al-Ṭabarī, Tārīkh x, 74, 24, 137. Wakīʿ, al-Ṭarīq 41–110 details the route through Najd to Medina, expressly noting 23 ruins of former constructions. Wakīʿ, al-Ṭarīq 67, 70, 82, 87, 107, 108. Wakīʿ, al-Ṭarīq 78.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

276

webb

developed in contemporary theoretical scholarship: seemingly marginal and spontaneous uprisings that have little place in grand historical narratives, but contemporary historians realize they intersect with much more serious issues in the process of the evolution of societies. In the Najdī case, the rise of the Qarāmiṭa, which permanently cut central Arabia from caliphal control, did catch the attention of Iraqi historians, but it seems facile to interpret the Qarāmiṭa as a sudden game-changing force. Rather, they can be read as a step up in the escalation of the previous generation, and this entails that understanding the Qarāmiṭa’s rise requires a reappraisal of those many raids and uprisings as more than just disturbances to order and security (as their scattershot portrayal in Iraqi historical narratives makes them appear). They suggest a 30-year sequence of “small wars,” which, upon the rise of the Qarāmiṭa, finally caught wider attention. Most scholarship on these kinds of smouldering conflicts with seemingly random smatterings of violence has focused on conflict in agrarian societies in the early modern world, but the aims are similar to those of this chapter: we confront the problem of reconstructing the history of peoples who left scant written testimony of their own, yet whose acts of violence transcend mere crime, as they merit being revaluated as forms of resistance and revolt. A common theme in analyzing such events is the lack of participant testimony, leaving historians to “infer what the participants thought from what they did”;89 and consequently, recourse to theory becomes helpful. In the specific case of Najdī violence, the Bedouin’s voice is indeed difficult to perceive. Glimpses into Bedouin interactions on the Darb Zubayda via the poems and anecdotes noted above are all mediated through the literature of Iraqi narrators who were little concerned with writing accounts of Najdī social issues. Moreover, once Najd became more dangerous in the later third/ninth century, urbanite scholars lost almost all access to the region, effectively closing our access to local sentiment, and Iraqi source impressions of the Bedouin fall into facile archetypes of bandits and troublemakers.90 We are therefore unlikely to find “smoking gun” evidence in Iraqi texts to prove the reasons for 89 90

Scott, Moral economy 144–145. Some fourth/tenth-century scholars who attempted the Hajj did interact with Bedouin, and the experience of the lexicographer al-Azharī is instructive: he records his capture when attempting the Hajj and enslavement among the Hawāzin for several years. As a philologist, however, al-Azharī left us no record of Bedouin social history, but rather relished his enslavement for the opportunity it offered him to hear “pure” desert Arabic in practice, al-Azharī, Tahdhīb i, 21. His writings thus preserve much of Bedouin speech as a matter of linguistics, but, unfortunately for the purposes of this chapter, nothing of Bedouin thought.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

277

the Najdīs’ actual discontent, but the levels of violence testify that forces greater than straightforward criminality, greed, and brigandage were at play. We have noted that the Ghanī directed their violence toward a piece of Umayyad infrastructure that had infringed their interests, and in Najdī attacks on the eve of the Qarāmiṭa’s rise, the Hajj was their primary target. The actions of the Qarāmiṭa’s Najdī constituency thus suggests a hostility that boiled over into violence and had specific targets in mind that had symbolic connection to their perceived causes of discontent. Modern theoretical work invites interpretation of such hostility as a meaningful historical act, howsoever rudimentary and/or “pre-political”91 it may seem, and with a consideration of the potential of applying modern theory to the Najdī context, we close this chapter. Key theoretical contributions have been noted above, particularly Hobsbawm’s “social bandit” and Linda Darling’s “circle of justice.” We may imagine some of the early restive tribal leaders corresponding to Hobsbawm’s model: small-scale and local acts of self-help in societies with a surplus population and suffering an economic crisis that entails widescale pauperization, with the intent to restore an older, more secure order.92 The Qarāmiṭa, on the other hand, fit less easily: they did not emerge from Najd (Hobsbawm considers the social bandit as hailing from within the aggrieved population), and their longlasting achievement and considerable territorial control raise them a structural level or two above the usual categorization of social bandits.93 Likewise, Darling’s thesis applies to Najd only so far, inasmuch as Najdīs were neither subject to tax burden, nor were they dependent on Abbasid military for security. Abbasid water provision and access to the caliph are indeed at the core of the “just ruler” archetype set out by Darling, but the Bedouin were not so structurally bound within the circle of justice as were agrarian or urban populations, who needed the state’s provision of the military to maintain prosperity. The disappearance of caliphal attention would entail that the Bedouin no longer were given the opportunity to view the caliphs as legitimate authority, but the level of hostility they expressed suggests deeper factors than simply a rupture in the circle of justice. The particularities of the Nadji situation do share fruitful parallels with Scott’s explanations of agrarian violence based on the theory of “moral economy.” Though subsistence food production, state exploitation, and taxation exchange between agrarian communities and the authorities are the primary ingredients of the revolts that Scott analyzes, his model, which attempts to 91 92 93

Scott, Moral economy 124. Hobsbawm, Bandits 24–30. Hobsbawm, Bandits 106–111 explores the limits of social banditry’s organization and scope.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

278

webb

explain why violence only occurs at particular moments (as opposed to all the time) facilitates broader application. At the core of Scott’s theory is the idea that subject populations develop a tradition of a morally articulated relationship with the authorities, and that the maintenance of this tradition curbs violence. If, however, structural change overturns the tradition, and if the conjunction of that change also leads to extreme privation and other crises, violence as a form of “self-help” becomes a viable option for the subjects, and the most extreme manifestations of such violence can coalesce around religiously defined movements, with particular indignation articulated against the authorities’ abandonment of the former traditions of the moral arrangement.94 This chapter has endeavored to demonstrate how the rise and fall of the Darb Zubayda can be read as a cataclysmic structural change that induced Najdī hostility and finally culminated in their willingness to support the Qarāmiṭa. Whereas the Umayyad-era Najdīs had existed on the very periphery of the sulṭān’s control and suffered a degree of exploitation, the first three generations of the Abbasids rewrote the relationship, effectively canceling Najdī fiscal obligations and providing for Najdī food and welfare on a historically unprecedented scale. Never before (nor since, until the modern era) would a state invest so heavily in the infrastructural transformation of Najd into a well-stocked and effectively-watered region. While the Darb Zubayda is heralded as a great engineering achievement of the Abbasids, the social effect of such infrastructure in Najd is an overlooked angle, given the paucity of Najdī sources. Over the course of three generations Najdī populations experienced caliphal authority as an overwhelmingly net giver of resources and security, which enabled Najdī populations to grow beyond the natural resources of the land. I suggest also that the continual charity created a tradition of a moral relationship between Najdī and Iraqi, the type that became part of the “memory” of the local in Scott’s sense and creates standards of justice.95 When the caliphs ceased giving ca. 190/805, the infrastructure still maintained the locals, but in a declining capacity, and by ca. 230/845 there was a congruence of a real water and food shortage, with the first generation of essentially unprovided-for Nadjis reaching military age. The world their fathers remembered—when the caliph made personal appearances, when water and food were mostly secure thanks to infrastructure and charity, and when locals were employed in the quotidian functions of the Darb Zubayda infrastructure—had passed. In their place, Najdīs experienced effective abandonment, lost expectations of justice, and had a hungry, over-swelled

94 95

Scott, Moral economy 140–152, 166–181. Scott, Moral economy 175–179.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

279

population. From their perspective, the failure of justice entailed they could legitimately seek redress. The Najdīs’ form of self-help took the most basic and violent forms of pillaging and raiding, and it was crushed, but Caliph alWāthiq did also organize elaborate efforts to restore the old moral order. Scott would note that such an outcome is precisely the aim of the morally indigent hostility, and the curb in violence for nearly two further decades suggests its success. Since al-Mutawakkil’s successors spent nothing on Najdī infrastructure and merely afforded military accompaniment for Hajj pilgrims, the moral issues of perceived caliphal misfeasance returned. The sulṭān had again ceased to perform its obligations as a result of profound changes in Iraqi politics and economy that entailed widescale structural changes in how the Abbasids could govern. Marginal groups in Najd were likely unaware of the reasons, but they experienced the structural change in their everyday lives. The 40-year funding of the Darb Zubayda and the several years’ blip of al-Wāthiq and alMutawakkil’s repairs were fundamentally unsustainable from the later third/ ninth century on, and there was no possibility of the old moral order being restored. Unfortunately for the Najdīs, the generous charity of previous generations became an expectation and a dependency for swelling Najdī populations, and they would therefore seek new patrons to restore the “traditional” moral economy. The Qarāmiṭa, with their religious message and promise of economic relief at the expense of Iraqi Hajj pilgrims, would have naturally appealed. However, their success was not guaranteed—history is not so neat. In the early stages, Zakarawayh reportedly encountered difficulties trying to rally support among the Tamīm and the Ṭayyiʾ in 289/902, even though the Ṭayyiʾ were, at that point, beginning their raids against the central authority.96 However, when Zakarawayh gained strength some years later and could present himself as representing a more established sulṭān in his own right, Najdī groups did join, and certainly when Abū Ṭāhir al-Jannābī appeared a decade later with a full-fledged state, Najdī acceptance was markedly greater. The Qarāmiṭā, once established as a state, were also an administrator of justice, and there is evidence that Najdī groups specifically appealed to the Qarāmiṭa in their disputes. The people of Najdī Ḍariyya sought Qarāmiṭa aid in their protracted conflict with the people at al-Rabadha, on the borders of Najd and the Ḥijāz, and Qarāmiṭa assistance was decisive, reducing al-Rabadha and forcing its surviving population to flee.97 And fiscally, the Qarāmiṭa also created a new form of economy 96 97

Al-Ṭabarī, Tārīkh x, 94–95. Al-Samhūdī, Wafāʾ iv, 83–84.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

280

webb

for Najd: after sacking Mecca in 317/930, the Qarāmiṭā permitted the Hajj to resume in 320/932–933 in return for a heavy protection tax (maks) levied on Iraqi pilgrims, reportedly of five dirham per pilgrim and seven dīnār per palanquin.98 The extent to which the Qarāmiṭa shared such funds with local Najdīs remains to be probed, and some Najdīs appear to have continued to apply a mechanism of self-help of their own, as attested by an ʿAlid leader, Abū Jaʿfar al-Mūsawī from Zubāla, who, in the late 330s ah used to attack pilgrims and demand protection money, before becoming the officially sanctioned leader of the Hajj in 340/952.99 The Hajj also remained dangerous: during the pilgrimage in 366/977, a Buyid princess performed a lavish pilgrimage with much charity, though her brother was killed on the same Hajj; she distributed the value of his blood money as ṣadaqa charity.100 But despite the dangers, the degree of stability obtained by Iraqi Hajj caravans paying heavy protection tax does suggest that the Qarāmiṭa-run Najd was more stable than during the previous generation. In conclusion, the effort to “materialize” the Qarāmiṭa reveals a stark lack of material comfort in Najd on the eve of the Qarāmiṭa’s rise. Najdī populations were struggling with a “post-Darb Zubayda” structural change in their economy, in their relations with authority, and in the very survival of their populations. Najdī hunger naturally could be interpreted as the result of caliphal abandonment of the traditions established by the early Abbasids, and the locals would be amenable to the aggressive policies of the Qarāmiṭa, which promised material relief from their suffering by attacking the infrastructure created by the locals’ erstwhile providers. Sectarian divides between the Qarāmiṭa and the Iraqi pilgrims may have aided the construction of difference and the fury of violence against Hajj caravans, but equally it would seem that the ferocity of attacks against pilgrims also represented the final unwinding of Najdī experience with the Darb Zubayda and the fallout from the collapse of their former integration with the sulṭān.

Bibliography Sources al-Azharī, Tahdhīb al-lugha, ed. M.A. Mukhaymir, 12 vols., Beirut 2004. al-Dīnawarī, al-Akhbār al-Ṭiwāl, ed. I.M. al-Ḥājj ʿAlī, Beirut 2001. 98 99 100

Al-Hamadhānī, Qiṭaʿ 317–318. Al-Hamadhānī, Qiṭaʿ 320. Al-Hamadhānī, Qiṭaʿ 322–323.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bedouin, bandits, and caliphal disappearance

281

al-Fākihī, Akhbār Makka, ed. A.M. Ibn Duhaysh, 6 vols., Beirut 1994. al-Hamadhānī, Qiṭaʿ tārīkhiyya min Kitāb ʿUnwān al-siyar fī maḥāsin ahl al-badw wa-lḥaḍar, ed. S. al-Hājirī, Tunis 2008. Ibn al-Jawzī, al-Muntaẓam, ed. S. Zakkār, 12 vols., Beirut 1995. Ibn Khallikān, Wafayāt al-aʿyān, ed. I. ʿAbbās, 8 vols., Beirut 1969. Ibn Nubāta, Sarḥ al-ʿuyūn, ed. M. Abū al-Faḍl Ibrāhīm, Cairo 1964. al-Iṣfahānī, Kitāb al-Aghānī, eds. A.A.A. Muhannā and S. Jābir, 27 vols., Beirut 1992. al-Iṣfahānī, Maqātil al-Ṭālibiyyīn, ed. A. Ṣaqar, Cairo, n.d. al-Jāḥiẓ, al-Bayān wa-l-tabyīn, ed. A.S.M. Hārūn, 4 vols., Cairo 2003. al-Maqrīzī, al-Dhahab al-masbūk fī dhikr man ḥajja min al-khulafāʾ wa-l-mulūk, ed. and trans. J. van Steenbergen, Leiden 2016. al-Maqrīzī, Ittiʿāẓ al-ḥunafāʾ bi-akhbār al-aʾimmat al-khulafāʾ, ed. A. Sayyid, 4 vols., Cairo 2016. Marwān b. Abī Ḥafṣa, Shiʿr Marwān ibn Abī Ḥafṣa, ed. H. ʿAṭwān, Cairo 1982. al-Muqaddasī, Aḥsan al-taqāsīm fī maʿrifat al-aqālim, ed. M.J. de Goeje, Leiden 1906. al-Masʿūdī, Murūj al-Dhahab wa Maʿādin al-Jawhar, ed. C. Pellat, 7 vols., Beirut 1966– 1979. al-Samhūdī, Wafāʾ al-wafāʾ bi akhbār Dār al-Muṣṭafā, ed. Q. al-Sāmarrāʾī, 5 vols., London 2001. al-Ṭabarī, Tārīkh al-rusul wa-l-mulūk, ed. Abū al-Faḍl Ibrāhīm, 11 vols., Cairo 1960–1969. Wakīʿ, Kitāb al-Ṭarīq, ed. A.A.b. N. al-Wuhaybī, Riyadh 1999–2000. al-Yaʿqūbī, Tārīkh al-Yaʿqūbī, 2 vols., Leiden 1882. Yāqūt, Muʿjam al-buldān, 7 vols., Beirut 1993.

Studies ʿAṭwān, H., al-Shuʿarāʾ al-ṣaʿālīk fī ṣadr al-Islām wa-l-ʿaṣr al-umawī, Beirut 1970. Blachère, R., and Ch. Pellat, al-Mutanabbī, in ei2, http://dx.doi.org.ezproxy.leidenuniv​ .nl:2048/10.1163/1573‑3912_islam_COM_0821. Blois, F., The “Abu Saʿidis” or so-called “Qaramatians” of Bahrayn, in Proceedings of the seminar for Arabian studies 16 (1986), 13–21. Cheikh, N. el-, Women, Islam and Abbasid identity, Cambridge, MA 2015. Daftary, F., The Ismāʿīlīs: Their history and doctrines, Cambridge 2007. Darling, L., A history of social justice and political power in the Middle East, London 2013. Ghālib, M., al-Qarāmitạh: Bayna al-madd wa-al-jazr, Beirut 1979. Gordon, M., The breaking of a thousand swords, Albany 2001. Halm, H., Die Söhne Zikrawaihs und des erste Fatimidische Kalifat (290/903), in Die Welt des Orients 10 (1979), 30–53. Hobsbawm, E., Bandits, London 2000. Kennedy, H., The desert and the sown in eastern Arabian history, in I.R. Netton (ed.), Arabia and the Gulf, London 1986, 18–28.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

282

webb

Kennedy, H., The Prophet and the age of the caliphates, London 1986. Landau-Tesseron, E., Arabia, in C. Robinson (ed.), The new Cambridge history of Islam, vol. 1, Cambridge 2010, 395–477. Madelung, W., Fatimiden und Bahrain qarmaten, in Der Islam 34 (1959), 34–88. Madelung, W., Das Imamat in der frühen ismailitischen Lehre, in Der Islam 37 (1961), 43–135. Munt, H., The construction of Medina’s earliest city walls: Defence and symbol, Proceedings of the seminar for Arabian studies 42 (2012), 233–44. Munt, H., Trends in the economic history of the early Islamic Ḥijāz, in jsai 42 (2015), 201–248. Munt, H., Caliphs, the economy, and political separatism in the Hijaz, in H. Kennedy and F. Bessard (eds.), Land and trade in early Islam: The economy of the Islamic Middle East 750–1050ce, Oxford forthcoming. Northedge, A., The Qubbat al-Ṣulaybiyya and its Interpretation, in P. Baker and B. Brend (eds.), Sifting Sands, Reading Signs: Studies in Honour of Professor Géza Fehérvári, London 2006. al-Rāshid, S., Darb Zubaydah: The pilgrim road from Kufa to Mecca, Riyadh 1980. al-Rāshid, S., Darb Zubaydah: Ṭarīq al-ḥajj min al-Kūfa ilā Makka al-mukarrama, Riyadh 1993. Scott, J., The moral economy of the peasant: Rebellion and subsistence in Southeast Asia, New Haven 1976. Scott, J., Weapons of the weak: Everyday forms of peasant resistance, New Haven 1985. Sperl, S., Islamic kingship and Arabic panegyric poetry in the early ninth century, in Journal of Arabic literature 8 (1977), 20–35. al-Walī, T., al-Qarāmitạh awwal hạraka ishtirākīya fī al-Islām, Beirut 1981. Webb, P., Power and money on the Hajj: Connections between caliphate and pilgrimage in early Islam, in H. Kennedy and F. Bessard (eds.), Land and trade in early Islam: The economy of the Islamic Middle East 750–1050ce, Oxford forthcoming. White, H., The content of the form, Baltimore 1980. Zadeh, T., The early Hajj: Seventh–eighth centuries ce, in E. Tagliacozzo and S. Toorawa (eds.), The Hajj: Pilgrimage in Islam, Cambridge 2016, 42–64. Zakkār, S., al-Jāmiʿ fī akhbār al-Qarāmiṭa, 2 vols., Damascus 1987.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 15

Zinā and muḥṣanāt in the Quran Richard Kimber

The Muslims did not need the Quran to tell them that adulterers should be stoned to death, but its failure to do so was troubling to the jurists.1 Various more or less transparent devices were contrived in order to repair the omission. A stoning verse was said to have been recited at one time despite having been omitted from the muṣḥaf, the Quranic text in its final form as Muslims remembered and transmitted it.2 The Quranic allegation that previous recipients of the divine revelation had tampered with it, requiring the Quran to restore the true version, was given substance in a report that the Jews had tried to conceal from Muḥammad the stoning verse in the Torah.3 “The book of God,” in accordance with which the Prophet is said to have actually imposed the death penalty, was invoked as an alternative to the inelastic muṣḥaf.4 A prophetic dictum laying down death by stoning was said to have been delivered following the characteristic indications that the Prophet was about to receive a revelation.5 What Muslims could find in the Quran was a condemnatory attitude to an evidently sexual offence and a technical term for the offence, zinā, that could be taken to include adultery. That the Quranic penalty for the offence was flogging and not stoning was a complication requiring some means of distinguishing such apparently less serious cases from those meriting the death penalty. The means found was the criterion of iḥṣān, a term derived from the Quranic muḥṣanāt occurring several times in the text and thus appearing to provide additional Quranic evidence for the stoning penalty.6 The Quranic term zinā thus came to bear an elaborate exegetical construction covering all irregular sexual relations, including fornication between unmarried partners as well as marital 1 “While no authorities of the formative period seem to have doubted that the Prophet stoned fornicators, there seems to have been some initial doubt as to the precedential value of his decisions to do so.” Azam, Sexual violation 75 n. 30. 2 Mālik b. Anas, Muwaṭṭaʾ ii, 824. 3 Mālik b. Anas, Muwaṭṭaʾ ii, 819. 4 Mālik b. Anas, Muwaṭṭaʾ ii, 822. The caliph ʿUmar is also said to have declared “Stoning is in the Book of God.” Mālik b. Anas, Muwaṭṭaʾ ii, 823. 5 Muslim b. al-Ḥajjāj, Ṣaḥīḥ iii, 1316–1317. 6 Mālik b. Anas, Muwattaʾ ii, 823.

© Richard A. Kimber, 2022 | doi:10.1163/9789004525245_016

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

284

kimber

infidelity.7 The discussion that follows will suggest a different and perhaps simpler interpretation of zinā, and following John Burton and Harald Motzki it will revisit the meaning of muḥṣanāt in the Quran.

1

Zinā

The term zinā occurs just once in the Quran, in Q 17:32. “Do not,” the believers are told, “go near to zinā, it is an abomination and an evil way.” This verse is part of a longer passage, Q 17:22–39, where the believers are also told, in a manner somewhat reminiscent of the Ten Commandments, not to associate any other god with God, to serve only God, not to commit murder or infanticide, to respect their parents, and to be charitable and honest. The verb zanā, to commit zinā, occurs twice, once in Q 25:68 and once in Q 60:12. The context of Q 25:68 is Q 25:63–73, which provides a summary of approved conduct for the faithful similar to that in Q 17:22–39. Those who serve God, we are told, are “those who do not pray to any other god beside God, do not commit murder, and do not commit zinā.” There is a warning that those who commit these sins face eternal punishment in hell, but also a reminder that repentance is possible for those who believe and reform, that God is forgiving and merciful, and that reformed sinners perhaps above all can look forward to God’s reward. The context of Q 60:12 is Q 60:10–12. Here the believers are instructed to admit to their community women, even married women, who as a genuine act of faith have left their homes among the unbelievers in order to join them. The Prophet is instructed to accept such women’s allegiance “on condition that they do not associate anything with God, and do not steal, and do not commit zinā, and do not commit infanticide.” Again, these are similar requirements to those in the two previous passages. Behind the reference to theft rather than murder is perhaps an assumption that women are less likely than men to commit the graver offence, but quite capable of the lesser. The terms zāniya and zānī, a woman and a man respectively who engage in zinā, occur three times each in the same short passage, Q 24:2–3. The verses that immediately follow concern accusations of what is evidently sexual misconduct, but without any further reference to zinā. There are, as we will see, other passages in the Quran that deal with sexual misconduct, but in none of these, or anywhere else in the Quran, is zinā explicitly mentioned again. 7 Hina Azam has shown in detail how rape was brought into the category of zinā. Azam, Rape 449–459; Azam, Sexual violation 98–104.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

285

Q 24:2–9 reads as follows: The zāniya and the zānī, flog each one of them a hundred lashes. Let not pity for them affect you in respect of God’s law, if you believe in God and the Last Day. Let a party of the believers witness their torment (ʿadhāb). (2) The zānī shall marry none but a zāniya, or a heathen. None shall marry a zāniya but a zānī, or a heathen. To the believers, that is forbidden. (3) Those who accuse (women who are) muḥṣanāt8 and then fail to produce four witnesses, flog them eighty lashes and accept no testimony from them ever again. Those are the wicked ones. (4) Except those who repent afterwards and reform. God is forgiving and merciful. (5) Those who accuse their own spouses and have no witnesses other than themselves, the testimony of any one of them shall be that he testify four times by God that he is telling the truth. (6) The fifth shall be that God’s curse will be upon him if he is lying. (7) It shall avert from her the torment (ʿadhāb) that she testify four times by God that he is lying. (8) The fifth shall be that God’s wrath will be upon her if he is telling the truth. (9) This passage, with its flogging penalties for zinā and for uncorroborated accusations of what is evidently associated with zinā but not explicitly so called, contains two of the four penal rulings that are found in the Quran. The other two are both in Q 5:33–39. Q 5:33–34 reads as follows: The requital of those who make war on God and his messenger and engage in subversion in the land is annihilation, or crucifixion, or amputation of their hands and feet on opposite sides, or exile from the land.9 That is an abject fate for them in this world, and in the next world they shall have a great torment. (33) Except for those who repent before you have them in your power. Know that God is forgiving and merciful. (34) Q 5:38–39 reads as follows: The man who steals and the woman who steals, cut off their hands in requital for what they have brought upon themselves, as an exemplary 8 The meaning of muḥṣanāt is discussed below. 9 “The Umayyads took this verse to be about rebels.” Crone, Medieval Islamic political thought 230. It is taken in that sense here.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

286

kimber

punishment from God. God is mighty and wise. (38) Whoever repents after his crime and reforms, God will accept his repentance. God is forgiving and merciful. (39) These four rulings, on zinā, on false accusations of sexual misconduct, on rebellion and subversion, and on theft, are the whole of the Quran’s explicit penal legislation. Three, but only three, of the four rulings provide for repentance and, by implication, salvation for the malefactor in the life to come. The same three all end with the refrain that God is forgiving and merciful.10 The exception among the four is the ruling on zinā in Q 24:2–3, which makes no mention of repentance or of God’s forgiveness and mercy and urges the believers to have no pity on the culprits. Moreover, the restriction on the subsequent marriage of a zānī or zāniya effectively excludes them from the community of believers. Their ultimate fate must be that of the condemned rebel in Q 5:33. Q 25:63–73 does refer to the possibility of repentance even for those who have prayed to other gods, or committed murder or zinā, but the context suggests that repentance here is repentance by conversion to Islam. “God will change their evil deeds for good ones” is more probably an inducement to the unbelievers than carte blanche for the faithful. For the believers zinā is effectively apostasy, and flogging is their public expulsion from the community.11 A second notable omission from the ruling on zinā in Q 24:2–3 is any mention of the four witnesses who are required to sustain an allegation of misconduct against women who are muḥṣanāt, or the equivalent fourfold testimony in the case of a man’s allegation against his own wife. The same requirement for four witnesses occurs in Q 4:15, which again concerns what is evidently sexual misconduct by women of the household (wa-llātī yaʾtīna l-fāḥisha min nisāʾikum), and again avoids any use of the term zinā or its derivatives. Since Q 24:6–9 deals specifically with suspected wives, it seems reasonable to take Q 4:15 as referring to daughters.12 Even in the unlikely event of four witnesses to their suspected sexual misconduct being sought, let alone found, the verse has no specific penalty comparable to those in the four penal rulings discussed above. Confinement “until death takes her,” which the verse does specify, is not

10 11

12

Q 24:23 is less forgiving toward the false accuser. The same two concepts, whoring and apostasy, are intimately linked in Hebrew scripture, where whoring after foreign gods, expressed with the verb zānāh, is a frequent metaphor. Examples are at Ex. 34:15–16; Lev. 20:5; Dt. 31:16; Jg. 2:17, 8:27, 8:33; Ezek. 20:30; Hos. 1:2, 4:12; 1 Chr. 5:25. Q 65:1 implies that a wife who committed adultery would be divorced and expelled immediately from the home, rather than being confined.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

287

a penalty but a safeguard against what might otherwise be the fate of an unmarried girl who brings dishonor upon her family. Scandal may be avoided by her confinement at home, but not by her murder at the hands of a male relative. Even a life sentence is not as harsh as it is perhaps deliberately made to sound. God, the verse concludes, reserves to Himself the right to “find a way.”13 Taking Q 4:15 as referring to unmarried daughters suggests in turn taking Q 4:16 as referring to unmarried sons. The masculine dual in wa-lladhāni yaʾtiyānihā could refer to a man and woman who indulge in al-fāḥisha, but in the context here there seems no reason to avoid the obvious suggestion of juvenile homosexual experimentation.14 The mild and pragmatic ruling that follows, “chastise them, and, if they behave, leave them alone,” together with the absence of any demand for four witnesses, suggests a relaxed attitude to a common enough practice when boys have no access to girls. Verse 16 ends with yet another reminder that repentance is possible and that God forgives, a reminder perhaps intended both for the girls of verse 15 and for the boys of verse 16. They are young and know no better. Verses 17–18, by contrast, warn their elders that deathbed repentance after a life of debauchery may well be too late. The Quranic legislation evidently distinguishes between two categories of sexual misconduct.15 The explicit use of the term zinā is restricted to only one of the two categories, the other being referred to by the term al-fāḥisha. Of the two, zinā is blatant, requires no search for witnesses, is subject to a drastic penalty, admits of no repentance, and condemns the culprits to the ultimate fate of the unbelievers. It is evidently an intolerable public evil that the Quranic legislation is designed to deter and suppress. By contrast, in the domestic context, it is not so much sexual misbehavior itself as the suspicion or accusation of misbehavior that is the object of Quranic concern. The effect of its rulings is not primarily to punish, but to protect, wives from their husbands and girls (and boys) from their families. There is perhaps even a recognition that the sexual temptations of the young are a fact of life that it is useless to try to suppress by penal law. In their case, the legislation hardly amounts to such. It seems rather a tentative reform of existing custom, an attempt to draw formerly private rights at least partially under the purview of public law by the

13

14 15

The reservation provides an obvious Quranic hook on which the exegetes can hang their ḥadd, the dual penalty of flogging or stoning according to the criterion of iḥṣān; al-Ṭabarī, Jāmiʿ viii, 74. Al-fāḥisha is used to describe the homosexual practices of Lot’s people; Q 7:80–81, Q 27:54– 55, Q 29:28–29. “That fāḥisha = zinā is a mere exegete’s assumption.” Burton, Meaning 73.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

288

kimber

practically impossible requirement to produce four witnesses to any suspected misbehavior. The same safeguard, together with a serious deterrent penalty, applies to the presumably public accusations of misbehavior condemned in Q 24:4, and, in the variant form of four sworn testimonies, to a husband accusing his wife in Q 24:6–9. Zinā is thus a narrowly drawn category of sexual misbehavior kept separate, as far as possible, from the sexual indiscretions, real or suspected, of otherwise respectable women of the community. It is proposed here that zinā, as used in the Quran, means specifically prostitution, and that zāniya, like its Hebrew cognate zônah, means a prostitute.16 Her notoriety makes the safeguard of four witnesses to any allegation unnecessary. She is a woman of no household and there is no one but herself to be offended by insulting aspersions on her character. The ease and regularity with which her services are both bought and sold makes repentance unlikely for either party to the transaction. The rebel who surrenders, the thief with the physical evidence of his punishment, the man once flogged for a reckless insult, none of these have ceased to be Muslims. All, if penitent, may be restored to good standing in the community. But there is no place in the community, or in paradise, for a prostitute. But neither is there a place for her abuser. In Q 24:2, the zāniya is mentioned before the zānī, in contrast to Q 5:38, where the male sāriq precedes the female sāriqa. The prostitute is typically a more noticeable, or at any rate more noticed, public nuisance than her client. Rather less typically, the Quran is determined that her client deserves the same opprobrious designation and the same fate as hers and gives him second place.17 It is also likely that the exhortations to avoid zinā in Q 17:32 and Q 25:68 are directed especially at male believers, with the implication that the responsibility for preventing the evil is primarily theirs.18 The promise to refrain from zinā required of unaccompanied muhājirāt, on the

16

17

18

Examples are at Gen. 38:15; Dt. 23:19; Jos. 2:1; Jg. 16:1; 1 Kg. 3:16; Jer. 2:20; Ezek. 16:31; Hos. 4:14; Mic. 1:7; Pr. 6:26, 7:10. Robertson Smith saw a form of sexual relations in pre-Islamic Arabia as “no better than prostitution” and took it to be the original object of the Quranic prohibition on zinā, for which he noted the Hebrew cognate. Montgomery Watt noted the same pre-Islamic practice; Azam, Sexual violation 57–58, 68. The formally equivalent Hebrew zôneh is not found in scripture in the sense of a prostitute’s client. Masculine forms of the verb zānāh (zānū, yizneh, yiznū, zôneh, zônīm) occur only in contexts where the people of Israel are condemned for apostasy. Men are occasionally reproached for resorting to prostitutes, or warned against it, but in other language; Hos. 4:14; Pr. 29:3. Q 17:32 warns the believers that zinā is also a fāḥisha, suggesting of itself that they needed to be told. “Zinā prior to Islam does not seem to have been criminalized, punished, or regarded as a sin.” Azam, Sexual violation 68.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

289

other hand, must recognize that being unaccompanied and without any independent means of support they may well be at risk of falling into prostitution. The two categories of sexual misconduct, referred to respectively as zinā and al-fāḥisha, are not kept entirely apart. In the case of a wife suspected by her husband, it is not directly stated what follows in the perhaps unlikely event of her declining to exculpate herself. There is, however, an implication that she would be liable to the same corporal torment (ʿadhāb) as that prescribed for the convicted zāniya.19 Q 4:25 provides that a female slave who marries and then commits a fāḥisha is liable to half of the free wife’s ʿadhāb, whereas according to Q 33:30 a wife of the Prophet who commits a fāḥisha is liable to double the ʿadhāb. Yet in none of these cases is the culprit explicitly condemned for zinā, and al-fāḥisha continues to be used here and in two other references to wives’ adulterous behavior, Q 4:19 and Q 65:1. The Quran’s careful distinction is in marked contrast with Hebrew scripture, where zānāh and its derivatives are used indiscriminately to designate sexual promiscuity as well as prostitution.20 Muslim tradition, as is well-known, does the same, designating all sexual misbehavior as zinā, with the effect of changing a protective into a punitive regime for domestic sexual misadventures, while paradoxically extending the protection of four witnesses to the actual prostitute and her client.

2

Muḥṣanāt

The term muḥṣanāt occurs eight times in the Quran, four times in Q 4:24–25, twice in Q 5:5, once in Q 24:4, and once in Q 24:23. All occur in the general context of sexual relations between men and women. The verb aḥṣana occurs in the same context in Q 21:91 and Q 66:12, where Mary is said (aḥṣanat farjahā) to have kept herself sexually inviolate.21 In Q 4:24, fa-idhā uḥṣinna is used in sense

19 20 21

Q 24:2 and Q 24:8. Examples are at Gen. 38:24; Dt. 22:21; Lev. 21:9; Ezek. 16:15–17; Hos. 3:3, 4:13, 14. Two other occurrences of aḥṣana (in the imperfect) are in different contexts but still show the same basic meaning of making invulnerable: Q 21:80 and Q 12:48. Harald Motzki extrapolates from the use of aḥṣanat in Mary’s case to describe her as muḥṣina, preparing the way for his proposed reading of muḥṣināt instead of the conventional muḥṣanāt in his argument to follow. Although the precise term muḥṣina (or muḥṣana for that matter) does not appear in the Quran, Motzki provides it with the meaning of a sexually continent woman and proposes its masculine equivalent, muḥṣin, as a sexually continent man; Motzki, Wal-muḥṣanātu 194–196, 205. It might be objected that Motzki is using muḥṣina as if from an intransitive verb, whereas in Mary’s case aḥṣanat needs the direct object farjahā to give the necessary sense. W. Montgomery Watt’s argument, that since muḥṣinīn is

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

290

kimber

of “when they (slave women) become muḥṣanāt.” This basic sense of inviolability is present in all these various uses of muḥṣanāt,22 but they show, in addition, the term’s ability to take on more specific meanings from its various particular contexts. In Q 24:4 al-muḥṣanāt are those women whom the Quran seeks to protect from accusations of sexual misconduct by means of a high standard of proof and a deterrent penalty for the accuser. The preceding verses have just settled the case of prostitutes, whom one can evidently denounce with impunity and without needing to produce witnesses. Muḥṣanāt must now mean respectable women, that is to say, women who enjoy precisely what prostitutes do not, the protection of a household and a presumption of chastity that is now reinforced in law. But just as context suggests a logical transition from the prostitutes of verses 1–3 to respectable women in verse 4, the transition to what follows in verses 6–9 concerning women of a man’s own household suggests that the muḥṣanāt of verse 4 should be taken to mean specifically the respectable women of other men’s households, whether they be married or not. The first occurrence of muḥṣanāt in Q 4:24 admits the same interpretation. The preceding verse, Q 4:23, lists women who are forbidden (ḥurrimat ʿalaykum) to the believers. Precisely what is forbidden is not specified, though it has been taken to mean marriage.23 Yet Hebrew scripture has nearly the same list of prohibited women, and the prohibition there is clearly on incestuous relations within the household.24 The Quranic list of prohibited women in Q 4:23 is mirrored in Q 24:31, with a corresponding list of the men before whom a woman may appear uncovered. They are consequently the men with whom women may live in the same household without being married to them, men who have physical access to them and before whom the women are not muḥṣanāt in the sense of belonging to another man’s household. What makes them inviolable in their own home is the law as laid down in Q 4:23.

22 23 24

always read as an active it is “most likely” that mḥṣnāt should be also, seems rather light; Muhammad at Medina 390. For al-Ṭabarī, muḥṣana is clearly a passive from a transitive verb, meaning allatī muniʿa farjuhā bi-zawj, or put differently, aḥṣana l-rajulu mraʾatahu fa-huwa yuḥṣinuhā iḥṣānan; Jāmiʿ viii, 165. That Mary managed this by herself is no doubt evidence of her special virtue. Motzki concedes that a passive reading, muḥṣanāt, could reflect a social reality, but only that of pre-Islamic tribal conditions and not the more enlightened moral order of the Quran; Motzki, Wal-muḥṣanātu 203–204. “All Qurʾānic instances of the root iḥşān involve the meaning of ‘to guard, protect, store or shelter.’ ” Burton, Meaning 74. Al-Ṭabarī, Jāmiʿ viii, 141, 167; Watt, Muhammad at Medina 280; Paret (trans.), Der Koran 62; Motzki, Wal-muḥṣanātu 210, 211. Lev. 18:6–18.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

291

Q 4:24 extends the prohibition expressed by ḥurrimat ʿalaykum in verse 23 to al-muḥṣanāt mina l-nisāʾ. As in Q 24:4–6, the specific sense of muḥṣanāt is given here by the context. Once Q 4:23 is taken to prohibit not marriage but incestuous sexual relations with women of one’s own household, Q 4:24 can be seen to prohibit irregular sexual relations with women under the protection of other households.25 The exception that follows, illā mā malakat aymānukum, can then be taken to mean, as it usually does, slave women, and in this context, slave women of other households.26 Slave women come into the category of al-muḥṣanāt only insofar as their owner and householder chooses to spare them unwanted sexual molestation, whereas the protection the Quran gives free women of his household is absolute. A man can sell, lend, or prostitute his slave women for others to abuse,27 but not his daughters, since Q 4:24 forbids them to his neighbors.28 Given that wa-l-muḥṣanātu mina l-nisāʾ in Q 4:24 is still grammatically dependent on ḥurrimat in the preceding verse, the separation of the two verses is striking yet clearly intended, since the end of Q 4:23 is marked with the refrain inna llāha kāna ghafūran raḥīm. The separation serves two purposes. It makes clear that the exception to the prohibition on sexual relations, illā 25

26

27

28

The corresponding passage in Leviticus adds similarly to the long list of sexually prohibited near kin a prohibition on sexual relations with one’s neighbor’s wife; Lev. 18:20. Al-Ṭabarī’s exegesis from here on proceeds on the implicit understanding that the prohibition carried over from verse 23 is no longer on marriage but on sexual intercourse, whether with other men’s wives or with any chaste woman beyond the four one may marry, a prohibition, as some reports say explicitly, on zinā; Jāmiʿ viii, 151, 158, 160. It seems the exegetes could envisage irregular sexual relations outside the family, but not within it. Instead of a protection for women against incestuous abuse in the home, a protection presumably as necessary then as it is now, the exegetes saw only a relatively anodyne prohibition of marriage to mothers, sisters, daughters, etc, a practice one can hardly imagine to have been prevalent. “Mā malakat aimānukum—a Qurʾānic cliché for fatayāt.” Burton, Meaning 68. Motzki shows conclusively from contextual evidence that mā malakat aimānukum refers to slaves, presumably female, whom their owners were free to abuse for their own sexual gratification; Motzki, Wal-muḥṣanātu 205–10. Motzki, Wal-muḥṣanātu 199. Q 24:33 discourages such abuse and shows sympathy for the indignity and loss of physical autonomy (taḥaṣṣun) that its victims must endure, but it stops short of a definite and presumably unacceptable prohibition. Hebrew scripture affords daughters the same protection; Lev. 19:29. Muslim exegetes who took al-muḥṣanāt in verse 24 to mean married women had to discover exceptional circumstances where sexual intercourse with married women would not be prohibited and found that married women taken captive in war must have been intended. Others solved the problem by taking al-muḥṣanāt to mean chaste women (al-ʿafāʾif ) and mā malakat aimānukum to mean a man’s wives; al-Ṭabarī, Jāmiʿ viii, 151–160; Motzki, Wal-muḥṣanātu 213–214.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

292

kimber

mā malakat aymānukum, applies only to al-muḥṣanātu mina l-nisāʾ and not to the list of female relatives in the preceding verse. It also underlines a difference in the nature of the prohibition as it applies, respectively, to the women of verse 23 and those of verse 24. In verse 23, the prohibition on incestuous relations is absolute. Verse 24 by contrast opens the way to lawful sexual relations with al-muḥṣanāt through marriage, laying upon men who marry not only the requirement to endow their brides financially but also, muḥṣinīn, to provide them with the continuing protection of a new household.29 Muḥṣinīn is contrasted with musāfiḥīn, seeking casual extramarital sexual relations, the object of the prohibition with which the verse began.30 In the next verse, Q 4:25, (al-)muḥṣanāt occurs three times. There is also one occurrence of the passive verbal form uḥṣinna. The verse anticipates that some men will not have the necessary means to marry al-muḥṣanāti l-muʾmināt and in their case recommends marriage to the Muslims’ believing female slaves ( fatayātikumu l-muʾmināt), with the permission of their owners. Since in this context al-muḥṣanāt are distinguished from female slaves, the term here takes on the specific sense of free women of other households.31 As prospective marriage partners, they are also necessarily unmarried. The addition of al-muʾmināt to both categories is presumably an assurance that the faith of a marriage partner matters more than her free or slave status. Once properly married, slave women, too, are now described as muḥṣanāt.32 As seen in verse 24, they were muḥṣanāt, in the basic sense of belonging to another man’s household, even before being married. Once married, the pro-

29

30

31

32

Muḥṣan(āt) here is equivalent to the term ḥarām as used by the jurists: “The fuqahāʾ […] usually describe an unmarried woman as being ḥarām to all men, in that she is unavailable to them sexually. As such, she is not the property of any other human being. Once married, she continues to be ḥarām to all except the one man who possesses the right to enjoy her.” Azam, Sexual violation 39 n. 53. Motzki takes the striking separation of wa-l-muḥṣanāt in verse 24 from the verb ḥurrimat in verse 23 to suggest that this first part of verse 24, up to wa-uḥilla lakum, is an interpolation ill fitted to the context; Wal-muḥṣanātu 210–211, 215–216. The present explanation is offered as an alternative. Here, Motzki again takes muḥṣināt, as he reads it, to mean virtuous women, but since the context makes it clear that women who are not slaves are meant, he needs to explain why slave women might not be regarded as virtuous. This he does with the evidence in Q 24:33 that slave women were liable to be prostituted by their owners, and only some of them, perhaps converts from Christianity or Judaism, would care enough for their virtue to resent that abuse; Motzki, Wal-muḥṣanātu 199–200, 213. Motzki, again reading muḥṣināt here, changes the requirement the verse lays upon husbands to provide shelter and protection to the slaves they marry into a duty laid upon their wives to become virtuous; Motzki, Wal-muḥṣanātu 200.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

293

tection they previously enjoyed, or not, at the discretion of their owner is superseded by the absolute protection of the law. The meaning of muḥṣanāt now extends beyond the physical protection of a household to include freedom, not from slavery but from the indignity of imposed sexual intercourse.33 With that freedom comes a measure of responsibility. When slave women become muḥṣanāt through marriage (idhā uḥṣinna), they also become liable to a penalty if guilty of a fāḥisha, a penalty specified as half of the ʿadhāb to which al-muḥṣanāt are liable.34 As applied in this context, uḥṣinna to slave women and al-muḥṣanāt to free women, muḥṣanāt comes to mean specifically married women. Al-muḥṣanāt occurs twice in Q 5:5. The context is a permissive ruling allowing the believers to marry al-muḥṣanāt mina lladhīna ūtū l-kitāb, here meaning women, necessarily unmarried, of a Jewish or Christian household, just as they can marry al-muḥṣanāt mina l-muʾmināt, unmarried women of a Muslim household. They must, as ever, be muḥṣinīna ghayra musāfiḥīn, committed to providing the protection of a new household and not engaging in mere fornication. John Burton, following Richard Bell, has proposed that muḥṣanāt should be understood as “women living in the shelter and protection of the household,” with Bell adding “whether married or not.”35 Here it is proposed to refine that definition in the sense of “women living in the shelter and protection of another man’s household,” the sense found in Q 24:4 and Q 4:24.36 The term can apply to “both slave and free women and both Muslim and non-Muslim women and both married and unmarried women.”37 Which it applies to in any given instance depends on the context: In Q 4:24 the term includes all the women of another household, without distinction. So, too, in Q 24:4, though there it has the added implication of respectable women as opposed to prostitutes. In Q 4:25 al-muḥṣanāti l-muʾmināt are free unmarried Muslim women; unmarried Muslim slave women become muḥṣanāt on marriage; and free married women are also muḥṣanāt. In Q 5:5 al-muḥṣanāt mina l-muʾmināt are again unmarried Muslim women, presumably also free women, as in Q 4:25. Al-muḥṣanāt mina 33 34

35 36

37

Motzki makes this point in connection with Q 24:32; Motzki, Wal-muḥṣanātu 200. An unchaste free wife offends both her husband and her father, both of whom may demand satisfaction under the law if they are denied private justice. An unchaste slave, on the other hand, has no legally recognized family of birth and offends only her husband. Burton, Meaning 73–74. Bell, The Qurʾān i, 72, as quoted in Motzki, Wal-muḥṣanātu 192. Al-muḥṣanāt is thus a category whose membership depends on the observer’s point of view. If women of other households are muḥṣanāt to you, the women of your own household are muḥṣanāt to everyone else. Burton, Meaning 68.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

294

kimber

lladhīna ūtū l-kitāb are unmarried Jewish or Christian women. As Burton has shown, without the need for iḥṣān as a Quranic criterion for the application of the stoning penalty, much of the difficulty surrounding the term muḥṣanāt falls away.38 In an ideal sense, al-muḥṣanāt comes almost to mean simply women.

3

Conclusion

The Quran shows a determination to prevent what is seen as the evil of prostitution, an evil ranked with murder and the worship of false gods. To that end it addresses itself in the first place to men, on whom it lays the primary responsibility of self-restraint in refraining from this as from other grave sins. That failing, the Quran provides for outright suppression of prostitution by means of severe, summary, and peremptory justice applied equally to both parties to the transaction. To be a prostitute, or to use a prostitute, is to cease to be a Muslim, with the inevitable consequence at the end of time. The Quran is aware that women are vulnerable to accusations of sexual impropriety. While not indifferent to the possibility of actual impropriety, and determined that sexual relations should be enjoyed within marriage, the Quran shows no interest in any legislative means of suppressing the ordinary sexual misdemeanors that occur even in respectable society.39 As with prostitution, it is to men that the Quran delivers its prohibitions on extramarital sexual relations, including both incestuous relations and those with women of other households. So far as women are concerned, the Quran’s overriding concern is to protect them, both married and unmarried, from the consequences of suspected and perhaps even actual sexual indiscretions. A demand for four witnesses, an improbably high level of proof backed up with a serious deterrent penalty for failing to find them, protects women from accusations from outside their household. Married women accused by their own husbands are protected at the same level in the form of a quadruple oath, and provided with a counter38

39

Motzki suggests that it may have been in order to establish the criterion of iḥṣān for the application of the stoning penalty that the passive readings muḥṣanāt and uḥṣinna were imposed, as he sees it, upon the text. The passive implies the presence of a second party to confer the status of iḥṣān, and Motzki quotes al-Ṭabarī’s opinion that this means a husband, in order to satisfy the requirement of marriage if a woman is liable for the stoning penalty. This perhaps overlooks the fact that the passive reading muḥṣanāt also occurs in contexts where the women concerned are clearly not married; Motzki, Wal-muḥṣanātu 204–205 and n. 37. The reference in Q 24:23 to al-muḥṣanāti l-ghāfilāti l-muʾmināt suggests a recognition that even the virtuous can be caught off guard.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

zinā and muḥṣanāt in the quran

295

balancing defense in the same form. Girls are protected within their families by the requirement for four witnesses, and protected from murder in the name of family honor even if the witnesses are found. No legal sanction is laid down for any such woman’s actually proven sexual misdeed. The Quran is aware that a legal sanction exists for married women who commit adultery, but will no more than allude to it in passing. That the penalty is flogging can be inferred from the three passing allusions. There is no trace of a stoning penalty, a penalty not only absent from the letter of Quranic law but entirely alien to its spirit.

Personal Note From our time at St Andrews I have many reasons to be grateful to Hugh, and I thank Letizia and Maaike for the opportunity to share in this tribute to a good friend and wise counsellor.

Bibliography Sources Mālik b. Anas, al-Muwaṭṭaʾ, ed. M.F. ʿA. al-Bāqī, 2 vols., Cairo 1951. Muslim b. al-Ḥajjāj, al-Ṣaḥīḥ, ed. M.F. ʿA. al-Bāqī, 5 vols., Cairo 1955–1956. al-Ṭabarī, Jāmiʿ al-bayān ʿan taʾwīl āy al-Qurʾān, eds. M.M. Shākir and A.M. Shākir, 24 vols., Cairo 1374/1955.

Studies Azam, H., Rape as a variant of fornication (zinā) in Islamic law: An examination of the early legal reports, in Journal of law and religion 28 (2013), 441–466. Azam, H., Sexual violation in Islamic law: Substance, evidence, and procedure, Cambridge 2015. Burton, J., The meaning of “ihsan,” in jss 19 (1974), 47–75. Crone, P., Medieval Islamic political thought, Edinburgh 2012. Motzki, H., Wal-muḥṣanātu mina n-nisāʾi illā mā malakat aimānukum (Koran 4:24) und die koranische Sexualethik, in Der Islam 63 (1986), 192–218. Paret, R. (trans.), Der Koran, Stuttgart 1996. Watt, W.M., Muhammad at Medina, Oxford 1977.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

part 3 Abbasids

∵

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 16

Muslim Nostalgia: Longing for the Abbasid Past in the Mamluk Era Robert Irwin

After the Mongols had extinguished the Abbasid Caliphate in Baghdad, an alleged survivor of the Abbasid family arrived in Cairo and was installed by the Mamluk Sultan Baybars as Caliph al-Mustanṣir. Then, after al-Mustanṣir died in an ill-fated, indeed almost suicidal, attempt to retake Baghdad from the Mongols in 660/1261, another alleged member of the Abbasid family was acclaimed in Cairo as Caliph al-Hākim. Thereafter, al-Hākim and his successors were effectively prisoners in the citadel of the sultans, and they were brought out only on ceremonial occasions. The Ottoman Sultan Selīm’s conquest of Egypt in 923/1517 brought this lineage of shadow caliphs to an end.1 Very few ʿulamāʾ in Mamluk Egypt and Syria wished for a restoration of the Abbasids to real power, and the learned ʿālim Jalāl al-Dīn al-Suyūṭī (849– 911/1445–1505), who produced a treatise on the excellence of the Abbasids, al-Asās fī faḍl Banī l-ʿAbbās, was exceptional in doing so (as he was exceptional in so many things). He championed the prerogatives of the Cairo caliphs, and he was unreasonably confident that the caliphate would last until the coming of Jesus or the mahdī. It seems clear that his partisanship for the Cairo caliphs was heavily influenced by his father’s and his own personal links with the caliphs al-Mustakfī and al-Mutawakkil. Al-Suyūṭī argued that the Ayyubid sultans had been more virtuous than the Mamluk ones, in part because they had submitted themselves to the caliphs. He also presented al-Manṣūr and Hārūn al-Rashīd as among the greatest rulers in Islamic history, for they had preserved the Islamic community from heresy and impiety. In the Tārīkh al-khulafāʾ he compared the glorious caliphate of Hārūn al-Rashīd to wedding festivities.2 But the subsequent large-scale entry of Turks into caliphal armies was, in al-Suyūtī’s view, a politico-religious disaster and, later, the Buyids, like the Mamluks, kept the caliphs prisoners. Despite its title, al-Suyūṭī’s Ḥusn al-muḥādarat fī akhbār Miṣr 1 On the Abbasid Caliphate in Egypt, see Ayalon, Studies; Little, Religion; Holt, Some Observations; Thorau, Lion of Egypt 110–119; Amitai, Fall; Hassan, Longing; Bannister, Abbasid Caliphate. 2 El-Hibri, Abbasid Caliphate 76.

© Robert Irwin, 2022 | doi:10.1163/9789004525245_017

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

300

irwin

wa-l-Qāhira (The excellent lectures on Egypt and Cairo) was more a history of the Cairo caliphs than of Mamluk Egypt in general. But even al-Suyūṭī did not envisage the return of the Abbasids to real political power, and while he looked back to the golden age of Baghdad, he did not actually expect its return. Nevertheless, he wished the caliphs of his own time to oversee and defend the religious authority of the qadis in opposing Mamluk injustices and their infractions of the religious law, for wherever there was a caliph, there was knowledge and religion. The caliph conferred luster on Egypt, and Egypt had inherited from Baghdad the status of the Islamic state par excellence. Even so, al-Suyūṭī reluctantly accepted that the armed defence and rule of Egypt and Syria by the Mamluks was a necessity.3 Although there could be no question of a restoration of an Abbasid Caliphate with real political power, nevertheless there is evidence in the Mamluk period of an antiquarian veneration for the way court rituals, scholarly enquiries, and much else had been conducted under the Abbasids. There was also nostalgia for that heyday of Arab political and military power as it had been conjured up by chronicles and literary anthologies. “Nostalgia is a sadness without an object, a sadness which creates a longing that of necessity is inauthentic because it does not take part in lived experience. Rather, it remains behind and before that experience. Nostalgia, like any form of narrative, is always ideological: the past it seeks has never existed except as narrative, and hence, always absent, that past continually threatens to reproduce itself as a felt lack.”4 Nostos is Greek for return to home, and algos means suffering. Nostalgia is an unappeased yearning to return home. When nostalgia takes the form of a flight from the present and is directed toward the past and perhaps toward real or fancied memories of that past in the form of some kind of golden age, then nostalgia can generate invented traditions. Nostalgia can play a powerful role in shaping a society’s image of itself. G.M. Young, in the opening pages of Victorian England, asks what History was about. “And the conclusion I reached was that the real, central theme of History is not what happened, but what people felt about it when it was happening: in Philip Sidney’s phrase, ‘the affects, the whisperings, the motions of the people’; in Maitland’s ‘men’s common thought of common things’; in mine ‘the conversation of people who counted.’ ”5 Pre-Islamic poetry was suffused with personal nostalgia for lost youth, deserted campsites, past sexual encounters, and perilous journeys. The most famous and moving qaṣīda on these themes was composed by Imruʾ al-Qays in the sixth 3 Garcin, Histoire. 4 Stewart, On longing 23. 5 Young, Victorian England vi.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

301

muslim nostalgia

century ce. The prominence of nostalgia in the opening retrospective nasīb of the qaṣīda has received a great deal of critical attention in modern times. But, with the exception of literary studies of jāhilī poetry, I think that little has been written about the theme of nostalgia in medieval Islamic culture. But nostalgia for the Abbasid era (most often in an idealized form) was pervasive in the Mamluk era. This is most obvious in the poetry of the period. But it is also evident in chancery manuals, political essays, furūsiyya treatises, manuals on ingenious mechanical contrivances, the revived cult of the dandy, and those select stories in The thousand and one nights that can be confidently dated to the Mamluk period. In the Mamluk period, furūsiyya manuals and treatises on mounted combat and on warfare more generally drew heavily from much earlier Abbasid treatises on the subject, most of which have since been lost. Because those Abbasid treatises in turn often drew upon earlier Sasanian or Greek manuals on warfare, much of what came to be reproduced in the Mamluk treatises had symbolic value rather than offering practical guidance.6 Though some early studies have suggested that the renewed production of furūsiyya manuals under the Ayyubid and Mamluk sultans was a fresh response to the Crusader threat, al-Sarraf’s research has established that this was not so. Instead, treatises that actually originated in the Mamluk era constituted only a small part of the genre.7 The other part consists of pre-Mamluk, mainly Abbasid treatises, copied and reused by the Mamluks as manuals and basic references. Al-Sarraf also shows that that the majority of Mamluk furūsiyya treatises, which did not always acknowledge their sources, were themselves largely based on Abbasid furūsiyya literature and its sources.8 Al-Sarraf has gone on to argue that furūsiyya’s origins are to be found in eighth-century Abbasid Iraq and the training of the Khurasani horse archers. Traditions, terminology, and accounts of the use of the wooden hand-bow were adapted to apply to the composite recurved bow used in the Mamluk period. Overwhelmingly, the treatises on arms and armor copied in the Mamluk period originated in the Abbasid era. The Mamluks produced no such treatises. The literature on hunting and on polo was similarly modeled on Abbasid productions. Though the Mamluk treatises were often beautifully illustrated, the illustrations tended to be decorative rather than instructive. The impulse behind the production of illustrated manuscripts dealing with ingenious mechanical contrivances seems to have been similarly antiquarian. The Abbasid caliphs sponsored ingenious mechanical contrivances that 6 On Abbasid military treatises, see Kennedy, Armies 111–114. 7 Al-Sarraf, Mamluk furūsīyah literature; cf. Al-Sarraf, Furūsiyya literature. 8 Al-Sarraf, Mamluk Furūsīyah literature 143.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

302

irwin

worked and were designed to impress. Thus, the Byzantine embassy of 305/917 was escorted to “the celebrated House of the Tree. In the centre of the court was there was an artificial tree in the middle of a large, apparently natural pool. The branches of the tree were mostly made of silver, though some were of golds, and there were gold and silver birds of many species perched on the twigs. At certain times the trees swayed, the leaves rustled and the birds would sing.”9 Caliph al-Maʿmūn and his successors employed the Banū Mūsā, three brothers who were specialist mathematicians and engineers and who produced a treatise on automata, Kitāb al-Ḥiyal (Book of ingenious devices).10 The Mamluk sultans exercised no such scientific patronage, even though treatises on automata continued to be produced and edited. Thus, Kitāb Fī maʿrifat alḥiyal al-handsīya (Book of knowledge of engineering devices) was written and illustrated by Badīʿ al-Zamān al-Jazarī in 603–604/1206, and it drew heavily on works from the Abbasid era. But, although it was subsequently copied and illustrated in the Mamluk period, I know of no evidence that any Mamluk sultan or governor actually commissioned any of these ingenious mechanical contrivances. Rather than serving as a construction manual, copying al-Jazarī’s treatise with its beautiful illustrations was an end in itself. Such manuscripts allowed its readers to take pleasure in dreaming about the marvels of the Abbasid technology, and those marvels always seemed to belong to the past. There was also an antiquarian flavor to the production of cookery books in the Mamluk period. One of the elaborate elite dishes of the period, Ibrāhīmiyya, actually took its name from Ibrāhīm b. al-Mahdī, the poet and son of the Abbasid Caliph al-Mahdī. An elaborate chicken dish, Maʿmūniyya, originated in the Abbasid period and took its name from Caliph al-Maʿmūn. Recipes were regularly copied from Abbasid cookbooks. According to Yungman, “the Mamluks saw themselves as the successors of former Muslim rulers, and as such they adopted the haute cuisine institutionalised under the Abbasid caliphate.”11 Turning now to quite a different aspect of late medieval retrospection, the Mamluk era saw a revival of the cult of the ẓarīf (dandy). This, too, partook of antiquarianism, since most of the important guides to the style and comportment of the dandy had been produced in the Abbasid period. Such guides included prescriptions on how to court girls, cultivate male friendships, dress elegantly, and perfume oneself. “In ʿAbbasid times it came to denote a comprehensive ideal of personal elegance and refinement, and implied such qualities

9 10 11

Kennedy, Court 154. Hill, Book of knowledge. Yungman, Medieval Middle Eastern court taste, 85–87, 95.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

303

as culture, urbanity, ethics, aesthetic sensibility, courtesy and amiability.”12 To a considerable extent ẓarf was a code of love, though the emphasis was usually on ʿUdhrite or chaste love of a rather literary kind. The Egyptian poet Bahāʾ alDīn al-Zuhayr (581–616/1186–1258) was accounted to be one of the ẓurafāʾ. In the fifteenth century the ẓurafāʾ made a cult of the comic and bawdy poetry of Ibn Sūdūn (ca. 810–68/1407–1464). In the same century, a few of the senior Mamluk amīrs presented themselves as ẓurafāʾ, including Jānibak al-Ashrafī and Taghrībirdī al-Maḥmūdī.13 Adab has a range of meanings; it sometimes refers quite narrowly to literary culture, but quite often its meaning also comprises education more broadly, as well as a mode of elegant comportment and speech. Although there is no direct evidence for an overlap between the ẓurafāʾ and a gay community in Cairo, the testimony of Shihāb al-Dīn Aḥmad al-Tīfāshī (580–651/1184–1253) suggests that there may have been an identifiable elite gay community. Al-Tīfāshī was a Tunisian who eventually migrated to Cairo. He wrote about precious stones and sex, and he was noted as an elegant dresser.14 His Nuzhat al-albāb fīmā la yūjad fī kitāb (Pleasure of the hearts regarding what one does not find in a book) included a section on how to recognise gay men. They will have elegant lodgings, furnished with bird cages, as well as a chessboard, volumes of poetry and erotic romances, collections of illustrated legends, and treatises on magic. They can be recognized by certain facial expressions, as well as their swaying way of walking.15 Moreover, it is noteworthy that Shams al-Dīn Muḥammad al-Tilimsānī (661–688/1263–1289), who was known as al-Shabb al-Ẓarīf (the young dandy) produced poetry that was strongly homosexual in content. From the mid-twelfth century onward there was a surge in the production of encyclopedias and anthologies. Though there was no medieval Arabic word for encyclopedia, it was nevertheless a typical literary product of the Mamluk age. (Al-Tīfāshī produced an enormous encyclopedia, of which only fragments and an abridgement survive.) The four-volume encyclopedia, Mabāhij al-fikar wa manāhij al-ʿibar (Delightful concepts and the paths to precepts) put together by the book dealer and stationer Jamāl al-Dīn al-Waṭwāṭ (632–718/1235–1318) is an early surviving example of an encyclopedia produced in the Mamluk era. It

12 13

14 15

Giffen, Ẓarf 821; cf. Giffen, Theory 14. On the Ayyubid and Mamluk cult of the ẓarīf, see Ghazi, Un groupe social; Vrolijk, Bringing a laugh 39–40. On the ẓurafāʾ in The thousand and one nights, see Bencheikh, La Volupté d’en mourir’. Rowson, Tīfāshī. Ahmad al-Tifâchî, Les Délices des coeurs 131–135.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

304

irwin

drew heavily on earlier Umayyad and Abbasid literature, and it in turn would be pillaged by al-Nuwayrī for his much grander compilation (on which see below). Two of the most famous and extensive enyclopedias were produced by Arab officials who worked for the government as chancery scribes. These were Ibn Faḍl Allāh al-ʿUmarī (700–749/1301–1349), the author of Masālik alabṣār fi mamālik al-amṣār (The routes of insight into the civilized realms), and Shihāb al-Dīn Aḥmad al-Qalqashandī (756–821/1355–1418), the author of Subḥ al-aʿshā fi kitābat al-inshāʾ (Dawn for the night-blind regarding chancery writing). Although these encyclopedias naturally concentrated on such matters as the drafting of official documents, calligraphy, protocol, and information about foreign regimes, a chancery scribe was also expected to have some acquaintance with the Arabic literary heritage, including, for example, poetry produced under the Abbasids and anecdotes concerning its production. Shihāb al-Dīn al-Nuwayrī (667–732/1279–1332) was also employed by the Mamluk regime, though he worked as a financial official in Egypt and Syria, and his encyclopaedia, Nihāyat al-arab fi funūn al-adab (The ultimate ambition in the art of erudition) owed a considerable debt to the culture of the Abbasid chancery. “The classical model of the Abbasid scribe and his encyclopedic erudition continued to be a leitmotif of descriptions of the scribal education from the Mamluk period and al-Nuwayrī was as guilty as many of his contemporaries in advancing a romanticized portrait of the chancery scribe.”16 Al-Nuwayrī’s debt to chancery culture notwithstanding, he drew on more literary sources and ranged more widely than the chancery encyclopedias of al-ʿUmarī and alQalqashandī. In Elias Muhanna’s introduction to a selected translation from alNuwayrī’s work, he has written eloquently as follows: “The world that appears in The Ultimate Ambition is not a plain reflection of Nuwayri’s quotidian reality but rather a medieval imaginary of warped dimensions, a dreamworld inhabited by real and fantastical creatures, living and long-dead monarchs, the sights and sounds of fourteenth-century Cairo alongside scenes from once-glorious Baghdad during its golden age.”17 In the introduction to the fourth book of the Nihāyat, which is devoted to plants, al-Nuwayrī, referring to the earlier botanical literature produced by Bedouin, Turkomans and Nabataeans, wrote what may be taken as a manifesto for his whole book: “My intention in presenting it has been to relate literary descriptions by the poets and literary epistles by the eloquent litterateurs, because that is what the attendant of literary gatherings is in need of and depends upon.”18 16 17 18

Muhanna, World 92. Muhanna, Introduction xii–xiii, in al-Nuwayri, Ultimate ambition. Al-Nuwayri, Ultimate ambition 181.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

305

As well as the selected translation of the Nihāyat, Muhanna has also produced a monograph on al-Nuwayrī’s encyclopedia, The world in a nook: AlNuwayri and the Islamic encyclopedic tradition. With respect to the compilation’s section on adab (in this context meaning literary culture), Muhanna makes the following observation: “a great deal of the older corpus of material anthologized in ninth- and tenth-century adab texts is the mainstay of alNuwayrī’s work. While there are several Mamluk-era texts among his sources, there is also an enormous amount of material that is centuries old, originating in the work of al-Jāḥiẓ, Ibn Qutayba, Ibn ʿAbd Rabbih, Ibn Waḥshiyya, al-Masʿūdī, al-Māwardī, Abū l-Faraj al-Iṣfahānī, and others.”19 Al-Nuwayrī preferred Abbasid poets to contemporary poets, and he has little to say about the high culture of his own time. Indeed, what is striking is the pastness of the adab that al-Nuwayrī celebrated. The quotations and anecdotes almost all hark back to Umayyad Damascus and Abbasid Baghdad and Samarra. What seems to be going on is a literary version of rescue archaeology where the aim was to revive the wit and wisdom of old Baghdad and elaborate on it in the Mamluk Sultanate. It is difficult to guess how realistic this mission was. Did al-Nuwayrī actually attend literary causeries in Cairo, Damascus, and Tripoli where the aim was to rekindle the wit, style, and erudition of the Abbasid cultural elite? Were there actually scholarly gatherings in Cairo, Damascus, and Tripoli where knowledge of the pearls of Abbasid high culture was a valuable asset, or was it all a fantasy based on al-Nuwayrī’s reading and his writing? This needs investigating. It is pleasant to think of causeries in Egyptian and Syrian cities whose participants might have traded quips and verses—somewhat in the manner of Robert Benchley, Dorothy Parker, and other members of the Algonquin Round Table as it flourished in New York from the 1920s to 1940s— but there seems to be little evidence for this. The sort of anecdotes about literary repartee and impromptu poetic inspiration that one finds, for example, in the works of Abū l-Faraj al-Iṣfahānī or al-Tanūkhī are largely absent in the Mamluk era. Perhaps it was more a matter of solitary scholars who read about Hārūn, Maʾmūn, Zubayda, Abū Nuwās, and Ibn al-Muʿtazz and then sighed for what was irretrievably lost. An intellectual salon of a kind did flourish under the patronage of Sultan Jaqmaq’s son, al-Nāṣīr Nāṣīr al-Dīn Muḥammad b. Jaqmaq (d. 848–849/1444), who held almost nightly intellectual soirees, which were sometimes attended by such scholarly eminences as Ibn Ḥajar and Ibn Taghrībirdī. Later, Sultan Qānṣawh al-Ghawrī presided over twice-weekly gatherings of scholars where

19

Muhanna, World 70.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

306

irwin

all manner of literary, religious, social, legendary, historical, and humorous matters were debated, but by then the discussions seem to have owed more to the tastes and preoccupations of Timurid international court culture than to an Abbasid model.20 There was an inevitable overlap between the encyclopedias with a strong adab content, such as those of al-Waṭwāṭ and al-Nuwayrī, and the literary anthologies that were also such a prominent feature of the Mamluk age. Most of these anthologies drew heavily from material transmitted from Abbasid times, though in many cases this was not a passive relaying of ancient lore. Rather, the compilers of the later anthologies sought to comment on, respond to, or outstrip the material that was being transmitted. Therefore, the relationship between the anthology and the material it anthologized might be intertextual. It was even possible that a particular poem or anecdote from Abbasid times might simply derive new significance from the material with which it was juxtaposed by the anthologist. Then again, the inclusion of a past poet in an anthology might confirm his canonical status. The poems of Ibn al-Muʿtazz (247–296/908) were particularly popular in the Mamluk period, as were the works of other Abbasid poets of the muḥdath (modern) school. The following are among the anthologies that were most widely read and whose materials were recycled by later writers (including the anonymous contributors to The thousand and one nights). The Dīwān al-ṣabāba (Diwan of ardent love) by Shihāb al-Dīn Aḥmad b. Abī Ḥajala (725–776/1325–1375) is devoted to the theme of chaste love, and this is illustrated by poems, anecdotes, and romantic narratives.21 Ibn Abī Ḥajala also put together Sukardān al-sultān, (Sugar-bowl of the Sultan) which (unusually) was written for a royal patron, the sultan al-Nāṣir Muḥammad b. Qalāwūn. The compilation includes humor, sermons, and literary anecdotes. Thamarāt al-awrāq (Fruits of the pages) by Ibn Ḥijja al-Ḥamawī (767–837/1366–1434) is a compilation of prose and poetry, which has been characterized as “an early-Abbasid style anthology,” consisting of an unstructured mixture of mostly Abbasid material mixed with Mamluk inshaʾ.22 Mustaṭraf fī kull fann mustaẓraf (The ultimate in all the branches of elegance) by Bahāʾ al-Dīn Muḥammad al-Ibshīhī (790–ca. 850/1388–ca. 1446) made use of adab material from earlier sources, especially al-Jāhiẓ in some sections, but he also drew on contemporary folk sources, and his anthology has a naïve and pietistic feel. It was widely read. On the other hand, Matālīʿ albudūr fī manāzil al-surūr (Risings of the full moons in the gardens of pleasure) 20 21 22

Flemming, Aus den Nachtgesprächen; Irwin, Political thinking. Giffen, Theory 38–40. Bauer, Mamluk literature 113–114 and n.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

307

by ʿAlāʾ al-Dīn ʿAlī al-Ghuzūlī (d. 815/1412) celebrated hedonism. Al-Ghuzūlī’s compilation has been well described by Everett Rowson as “a rather original compilation of poetry and prose anecdotes, organised according to the conceit of building a house and organizing a household; topics treated include neighbours, doors and fountains, but also wine, food and animals, and the work concludes with a chapter on the mansions of paradise” (manāzil al-surūr).23 AlGhuzūlī wrote to please those who, like himself, belonged to the leisure class, for, in the near absence of court patronage, the literati mostly wrote for each other. His chapters on the nadīm (cup companion) and the repertoire of the nadīm relied particularly heavily on Abbasid material. The Egyptian poet alNawājī (788–859/1386–1455) composed Ḥalbat al-kumayt (The racecourse of the bay), an anthology devoted to wine drinking. Though it had various original features, including the citation of medical sources, inevitably it also drew upon early Abbasid wine poetry. The taste for intertextual commentaries on past works of literature was widespread among the Mamluk sultanate’s literary elite. In this context, a further, somewhat curious work, which was neither an encyclopedia nor an anthology, though it partook of some qualities of both, requires mention. This was al-Ghayth al-musajjam fī sharḥ lāmiyat al-ʿajam, an extraordinarily discursive commentary on the twelfth-century poet and alchemist al-Ṭughrāʾī’s poem Lāmiyyat al-ʿajam (The verses rhyming in lām of the non-Arabs). This commentary was the masterpiece of Khalīl b. Aybak al-Ṣafadī (d. 764/1363), a calligrapher, poet, writer of stylish prose, and compiler of anthologies. In alṬughrāʾī’s lengthy qașīda he lamented the evils of his times and his sufferings in them, but the commentary on this poem served as a pretext for a display of al-Ṣafadī’s miscellaneous learning. So a 59-line poem received two volumes of dense exegesis. The writings of al-Jāḥiẓ, with their mingling of instruction and entertainment, furnished the model for al-Ṣafadī’s digressive masterpiece. In a study of the use of rhetorical devices and verses in one lengthy entry in alṢafadī’s biographical dictionary, Kitāb al-Wāfī bi-l-wafayāt (The abundant book on dates of death), Stephan Conermann makes the following observation: “AlṢafadī was a man with literary ambitions. He knew all the literary strategies and rhetorical tricks. He was very interested in Abbasid poetry and had an excellent knowledge of the sources.”24 “The poetry of lost love, of nostalgia for better days, of the pleasures of feasting and drinking wine can set up resonances across the ages […].”25 Few 23 24 25

Rowson, Ghuzūlī 254. Conermann, Tankiz 21. Kennedy, Court 114.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

308

irwin

Mamluk sultans showed any interest in poetry, though al-Ashraf Qāytbāy and Qānṣawh al-Ghawrī were exceptions in this respect. By contrast, many of the ʿulamāʾ did compose poems and they also produced works of literary prose. “There was […] a literary renaissance in the Ayyubid and early Mamluk period. Like all true renaissances, it had a backward-looking flavour and it harked back to the manners and literary productions of eighth- and ninth-century Iraq […] The poets of Cairo and Damascus studied the old Baghdadi prescriptions on how to court beautiful slave girls, cultivate male friendships, dress elegantly and perfume themselves.”26 Though literati of the Mamluk age paid the most attention to contemporary poets, they still studied the muḥdath poets of the Abbasid age and played allusive and intertextual games with what they read.27 Essentially the same point has been put forward by Everett Rowson: “For Mamluk writers, one is tempted to say, intertextuality was what literature is all about; and the more of a past one has to deal with, the more one can glory in ringing changes on, and playing with that past, to the ongoing enrichment of the Arabic literary tradition.”28 Though the word badīʿ originally denoted the “new” or original, writers had been working in the badīʿ manner since Abbasid times. The men (mostly men) who played such literary games were highly educated. Nevertheless their readiness on occasion to deploy dialect and colloquial forms was also a characteristic of the Mamluk age. As noted, the production of poetry in this period owed very little indeed to court patronage, and this is in striking contrast to what had been the case in Abbasid times. But hunting poetry (ṭardiyya), which had its heyday under the Abbasids, was now belatedly revived, and just possibly this reflected the interest and potential patronage of the hunting-mad Mamluks. Poetry featured prominently in the stories of The thousand and one nights. Josef Horovitz, who worked through the Macnaghten edition of the Nights (1839–1842), counted 1,420 pieces of poetry. What was quoted was mostly short anonymous extracts, and most verses that could be identified were dated as from Abbasid times or later. Ibn al-Muʿtazz was one of the few poets cited by name.29 In general, the poetry cited did little to advance the narratives, unless indeed the composition or recitation of poetry was precisely the theme of a story (and there are quite a few such stories). Elsewhere, poetry was commonly used to express heightened emotion, especially grief and longing. As we will see, quite a lot of the poetry seems to have been lifted from 26 27 28 29

Irwin, Mamluk literature 9. Bauer, Mamluk literature 113–114. Rowson, Alexandrian age 110. Horovitz, Poetische Zitate 375–378.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

309

muslim nostalgia

the fifteenth-century anthologies mentioned above, al-Ibshīhī, Ibn Ḥijja alḤamawī, al-Ghuzūlī, and al-Nawājī, either that or from the same, mostly Abbasid, sources that those anthologies drew upon. Geert van Gelder, having noted that the poetry recycled in the Nights is overwhelmingly classical Arabic poetry in the correct meter, makes the following important observation: “This is another proof of the fact that the Arabian Nights is not strictly a ‘popular’ collection […]. The Arabian Nights was never truly popular in the way that the epics of ʿAntar or the Bānī Hilāl were.”30 Though modern western readers usually skim the copious poetry found in the Nights, there is no reason to think that the premodern Arab audience for its stories did so. Earlier academic studies of the Nights have tended to exaggerate the importance of popular stories told on street corners and in markets and coffeehouses in shaping the origins and development of this story collection. (I certainly include myself among the guilty.) Yet, when one surveys the numerous stories that found their way into the Macnaghten edition of the Nights, their frequent debts to high culture and to such sources as Abūʾl-Faraj al-Iṣfahānī, al-Jāḥiẓ, and al-Tanūkhī becomes evident. Moreover, while many of the Nights stories about Hārūn al-Rashīd are effectively fairy stories, the stories about the caliphs al-Manṣūr and al-Maʾmūn more often derive from adab literature. A great deal of what is now accounted to be part of the Nights was added in the Ottoman period. But this discussion of the Abbasid content in the Nights will commence by examining those stories that are incontestably from the Mamluk era and, in the form that they have come down to us, cannot be dated later than the fifteenth century. These tales come from the oldest substantially surviving manuscript of the Nights, the one that formed the core of Antoine Galland’s translation in the early eighteenth century. One of these core stories, “The steward’s tale” (also titled “The reeve’s tale” in the Burton’s translation and in The Arabian nights encyclopedia) attracted the hostile attention of the great scholar and editor of the Nights, Muhsin Mahdi.31 This tale is part of the hunchback cycle of stories and is adapted from an event alleged to have happened in the early tenth century and then written about in the late tenth century in al-Tanūkhī’s Faraj baʿd al-shidda (Relief after grief). In al-Tanūkhī’s account, the alleged romance took place during the caliphate of al-Muqtadir, but in the Nights it has been shifted back to that of Hārūn alRashīd. It is the story of a shopkeeper who is visited in his shop by a beautiful woman, who is a handmaid of Hārūn’s wife Zubayda. The shopkeeper and the

30 31

Van Gelder, Poetry 13–14. Mahdi, From history to fiction 164–180; cf. Julia Bray, Caliph.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

310

irwin

handmaid fall in love and wish to marry. But the man has to be smuggled into the harem to be inspected by Zubayda before she may give consent to their marriage. Implausibly Zubayda has the youth stay ten days in the harem before arranging the wedding. But the youth eating zīrbājah (cumin ragout) without washing his hands leads to calamity, and his enraged bride then has his thumbs and big toes cut off. (The mutilation is thematically necessary in order to align it with the other tales of mutilation in the hunchback cycle.) Mahdi was apparently infuriated by the lack of historical accuracy and the implausibility of a Nights’ fantasy story: “It claims to be an account of historical events involving historical persons in a well-known time and place, and it does not even hint at anything that is impossible in itself.” It is judged to be “an outrageous distortion and corruption of history.”32 This seems an odd verdict on a story that was surely designed to entertain rather than dutifully chronicle the mores of a past century. (Incidentally, it is curious that so much fact and fiction has been handed down about the Abbasid harem and its inhabitants, and so little about the Mamluk harem.) The storytellers’ choice of the Abbasid era, and more specifically the caliphate of Hārūn al-Rashīd, as the favored setting for stories of romantic or improbable encounters did not originate in the Mamluk period. Kennedy, having noted that the earliest versions of the Nights date from the late middle ages, goes on to remark that “we can see the beginning of this mythologizing of the reign of Harun much earlier. Within a couple of generations of his death, writers such as Ibn Abi Tahir, the Grub Street hack of ninth-century Baghdad, were elaborating the stories of poets, singers, harems, fabulous wealth and wicked intrigues which were to form the basis of the legend.”33 The Hārūn of the Nights, who was so often sleepless and bored, was equipped by the storytellers with a supporting cast that included Zubayda, his queen, Jafʿar, the Barmakid vizier, Masrūr, the headsman, and Abū Nuwās, the famous poet.34 Many of the stories concerning Harun and his entourage, as well as other stories that were at least notionally set in the Abbasid period, were probably added to various recensions of the Nights during the Ottoman period. One of the stories that formed part of the Galland manuscript, “Nur al-Din ʿAli and his son Badr al-Din Hasan,” has been subjected to close historical and

32 33 34

Mahdi, From history to fiction 180. Kennedy, Court 51. From this point on the fictional Harun and other fictional characters will not be marked with diacritics, as they are not so marked in the Lyons’ translation of the Nights, nor in the Hadawi translation, nor in the Arabian Nights encyclopedia, to which repeated references will be made.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

311

linguistic analysis by Patrice Coussonet.35 It is a story inserted within the story of “The three apples,” and it is told by Jaʿfar to Harun to illustrate the power of fate or coincidence. Very briefly, it tells of two formerly friendly viziers who plan to cement their alliance by a wedding between their children, but who then fall out and thereupon the vizier Nur al-Din ʿAli leaves Egypt for Iraq. But eventually, thanks to the intervention of two jinn, Nur al-Din’s son Badr al-Din will be united in marriage with the daughter of the other vizier. Coussonet argued on the basis of the vocabulary and the familiarity with Mamluk institutions, including the chancellery and the barīd (postal express), that the story was composed in Egypt in the early fifteenth century. The encyclopedia of The Arabian nights adds the following: “The story contains many realistic references to contemporary society, suggesting that it does not belong to so-called popular folklore but was rather consciously conceived by an intellectually trained author with a solid knowledge of the chancellery, maybe a magistrate or secretary.”36 “The tale of the three apples,” which frames “Nur al-Din ʿAli and his son Badr al-Din Hasan,” is a detective story in which Harun in disguise, and accompanied by Jaʿfar and Masrur, sees a trunk brought out from the Tigris, and within the trunk is the mutilated body of a young woman.37 Jaʿfar is given three days to find the culprit, or else his life will be forfeit. (Here Harun features as a capricious tyrant on the pattern of Shahriyar.) Jaʿfar, though hesitant and timid, succeeds, and coincidentally the culprit turns out to be a member of his own household, and it is this coincidence that leads him to tell the tale of “Nur al-Din ʿAli and his son Badr al-Din Hasan.” Harun, in disguise and accompanied by Jaʿfar and Masrur, also features in “The porter and the three ladies of Baghdad,” one of the earliest stories in the Nights.38 The party, which begins with the three ladies and the porter having an erotic frolic, is interrupted by three one-eyed dervishes who seek to enter the house of the ladies. They are allowed in on the strict condition that they may not ask any questions about what they see or hear. A little later Harun and his companions seek to join them, and they are allowed to do so on the same condition. But when one of the ladies starts to savagely whip a pair of dogs and then embrace them in tears, the porter can contain his curiosity no longer and asks what has been going on. Whereupon black slaves armed with swords rush

35 36 37 38

Coussonet, Pour une lecture historique; cf. Arabian nights i, 129–172; Arabian nights encyclopedia i, 317–319. Arabian nights encyclopedia i, 318. Arabian Nights i, 126–129; Arabian nights encyclopedia i, 414. Arabian Nights i, 50–122; Arabian nights encyclopedia i, 324–326.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

312

irwin

in ready to kill all who have broken the interdiction. But then the dervishes and Jaʿfar tell stories in order that their lives may be spared and they may be allowed to leave. (The telling of stories under the threat of death of course echoes the overarching frame story of the Nights in which Scheherazade tells stories to Shahriyar in order to stay alive, as does the framing of stories within this story.) The following morning Harun, having resumed his caliphal status, summons all the participants at that strange soirée and obliges the ladies to tell stories that will explain their strange behavior. Harun also makes a grim appearance in the tragic love story of ʿAli b. Bakkar and Shams al-Nahar, as he brings to an end the romance of the two protagonists by having Shams al-Nahar imprisoned in the palace, which in turn leads ʿAli to die of grief. According to the Encyclopedia, the story probably originated in Baghdad, and the case for its origin in the Abbasid period has been made by Bencheikh.39 Harun makes a more benign appearance in disguise in the story of “Nur al-Din ʿAli and the damsel Anis al-Jalis.”40 In this story he dons a disguise and joins the lovers he has spotted partying in his Garden of Gladness. The (complicated) story ends with Harun making Nur al-Din ʿAli “king of Baghdad” and Nur al-Din ʿAli making a gift of Anis al-Jalis to the caliph (which seems a bizarre conclusion to the romance). Although “ʿAli the Cairene and the haunted house in Baghdad,” a story about wishes fulfilled by a jinn, is not included in the Galland manuscript, Coussonet dated it to the Mamluk period and, on the basis of mostly accurate historical and geographical references, specifically located its origin to Cairo at the beginning of the fifteenth century.41 However, there is no evidence of any familiarity with the real topography of Baghdad, nor any explicit reference to Harun or any other of the Abbasid caliphs, and Coussonet has argued that old Baghdad has here become a fantasy lieu de memoire: “Si cette dernière [Baghdad] est bien une ville réelle, nous verrons qu’elle se trouve dans espace mythique et imaginaire et qu’elle n’a en realité qu’une fonction symbolique.” Coussonet returns to this theme a little later. The Baghdad of the Abbasids represents the golden age of Islam, and its rulers are presented as just and pious in the idealizing collective memory. It is then a city of lost happiness.42 Coussonet’s presentation of the metaphorical location of Baghdad in this story is echoed in the Arabian

39 40 41 42

Arabian nights i, 650–693; Arabian nights encyclopedia i, 92–93; Bencheikh, La Volupté 276–283. Arabian nights i, 244–278; Arabian nights encyclopedia i, 316–317. Coussonet, Pensée mythique; Arabian nights ii, 255–273; Arabian nights encyclopedia i, 93– 94. Coussonet, Pensée mythique 34, 49.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

313

muslim nostalgia

nights encyclopedia: “Whereas Cairo in the story is described in a realistic way, Baghdad is depicted as a mythical city, being both the symbol of power and a place of nostalgic harmony.”43 The appearance in some of the Nights stories of Harun al-Rashid served a similar function to the words “Once upon a time” in English fairy tales. His presence signals that what is being related is fiction. To take one fantastic example among many, in the story of “ʿAbdallah b. Fadil and his brothers.” Harun is able to restore those brothers to human form because the jinn recognize that the caliph is more powerful than they are.44 But other Abbasid caliphs feature in stories, which may perhaps have been part of a recension of the Nights in the Mamluk period or, probably more often, they were added in the Ottoman period. But what is striking about these stories is that they frequently draw upon Abbasid anecdotes, and frequently those anecdotes seem to have been adapted from or paralleled in one or other of the Mamluk anthologies discussed above. Thus, “The story of ʿAbbas,” which is set during the caliphate of al-Manṣūr, is also found in al-Ibshīhī’s Musṭatraf.45 Part of the story of “Abu l-Hasan of Khurasan” has a close parallel with a story in Ghuzūlī’s Maṭālīʿ al-budūr. The story is set in the caliphate of Muʿtaḍid bi-llah, and it includes an account of the murder of the caliph.46 In “Abu Yusuf with Harun al-Rashid and Queen Zubayda,” Qadi Abu Yusuf’s witty declaration that the merits of fruits should not be judged in their absence is also found in al-Jāḥiẓ’s al-Bayān wa-l-tabyīn.47 In “Harun al-Rashid and Queen Zubayda in the bath,” Harun, on seeing Zubayda in the bath, composes the beginning of an erotic poem, which later Abū Nuwās (who, among others, in this story is gifted with clairvoyance) is able to complete the poem without being told what Harun’s verses were. A very similar tale was attached to Caliph al-Mahdī in Abū l-Faraj’s al-Iṣfahānī’s Kitāb al-Aghānī.48 The Nights story of “Harun al-Rashid and the slave-girl and the Imam Abu Yusuf” is also found in al-Nawājī’s Ḥalbat al-kumayt.49 In “Ibrahim al-Mahdi and the barber-surgeon,” Ibrahim, who was Harun’s half-brother, opposed Maʾmun’s succession to the caliphate and was consequently hunted down, but after he was discovered hiding in Baghdad, he was

43 44 45 46 47 48 49

Arabian nights encyclopedia i, 94. Arabian nights iii, 654–690: Arabian nights encyclopedia i, 63–65. Arabian nights encyclopedia i, 63. Arabian nights iii, 594–608; Arabian nights encyclopedia i, 68–69. Arabian nights ii, 193–194; Arabian nights encyclopedia i, 78. Arabian nights ii, 186–187; Arabian nights encyclopedia i, 203. Arabian nights ii, 6–8; Arabian nights encyclopedia i, 204.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

314

irwin

pardoned by the new caliph. This story is found in al-Iṣfahānī, al-Masʿūdī, alGhuzūlī, and Ibn Ḥijja al-Ḥamawī.50 It and many of the other Nights stories found in the various manuscripts suggest that the audience for the Nights had a taste for real history, or at least for what they took to be real history. Again, the story of “Ibrahim al-Mahdi and the merchant’s sister,” in which Ibrahim crashes a party and competes with a singing girl in the composition of verses, is found in various earlier literary and historical sources.51 A similar observation applies to “Ibrahim of Mosul and the devil,” a story about the devilish inspiration of music, which features in such early sources as al-Iṣfahānī and al-Masʿūdī, as well as the much later Ḥalbat al-kumayt of al-Nawājī.52 The story of “Ishaq of Mosul,” set during the caliphate of al-Maʾmun, has Ibrahim’s son, also a poet and musician, as the protagonist. His story features in Ibn ʿAbd Rabbih’s tenth-century adab anthology ʿIqd al-farīd. Although Ibn ʿAbd Rabbih was an Andalusian, his work drew almost entirely on eastern sources such as al-Jāḥiẓ. In the Mamluk era, al-Nawājī included the same story.53 The poetry-centered story of “Ishaq of Mosul and the merchant” is found in al-Iṣfahānī and al-Nawājī.54 Again, “Ishaq of Mosul and his mistress and the devil” is poetry centered. In this story, Ishaq of Mosul hired a blind man to sing for him and a slave girl whom he fancied, but it eventually turns out that the inspired blind singer is really the devil. AlNawājī also included this story in his anthology.55 “Jafʿar the Barmakid and the old Badawi” features Harun, Abu Nuwas, and Jafʿar, and it is a story about a great fart, which first appeared in al-Jāḥiẓ’s Kitāb al-Qawl fī al-bighāl (Book of mules).56 In “Jaʿfar b. Yahya and ʿAbd al-Malik b. Salih the Abbasid,” Jaʿfar promptly secures favors for ʿAbd al-Malik from Harun and is praised by Harun for his promptness. Ibn Ḥijja al-Ḥamawī, al-Nawājī, and al-Ibshīhī all have this story.57 In “The lovers of al-Madina,” a tale told by Ibrahim al-Mosuli serves as a vehicle for lots of poetry, and this story is also found in al-Nawājī.58 The confused tale of “The mad lover,” featuring the famous grammarian al-Mubarrad, Caliph al-Mutawakkil, and a lunatic, appears in al-Masʿūdī’s Murūj al-dhahab and in Ibn Ḥijja al-Ḥamawī’s Thamarāt al-awrāq

50 51 52 53 54 55 56 57 58

Arabian nights i, 891–898; Arabian nights encyclopedia i, 229. Arabian nights ii, 110–113; Arabian nights encyclopedia i, 229. Arabian nights encyclopedia i, 230. Arabian nights i, 903–908; Arabian nights encyclopedia i, 232. Arabian nights ii, 227–231; Arabian nights encyclopedia i, 233. Arabian nights ii, 794–796; Arabian nights encyclopedia i, 233–234. Arabian nights ii, 203–204; Arabian nights encyclopedia i, 236. Arabian Nights ii, 200–201; Arabian nights encyclopedia i, 236. Arabian nights ii, 796–799; Arabian nights encyclopedia i, 274–275.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

315

muslim nostalgia

among other sources.59 In “Al-Mutawakkil and his concubine Mahbuba,” the caliph and the concubine quarrel but then are reconciled after having the same dream. This story is also found in pseudo-Jāḥiẓ, al-Masʿūdī, and al-Ibshīhī.60 The above has been a short listing of some of the stories in the Nights that were at least notionally set in the Abbasid era (and, of course, there are other historical or fanciful Nights stories set in pre-Islamic, Umayyad, and Buyid times). When Sir Richard Burton set about on his translation of the Nights in the 1880s, he chose to draw heavily on expert folklorists, such as William Alexander Clouston and William Forsell Kirby, for his introductory matter and annotations. Burton’s overemphasis on the folkloric and popular nature of the Nights stories was, in part at least, due to the fact that he was pretty ignorant about classical Arabic literature. Subsequently, Duncan Black Macdonald was to take the Nights as a valuable source on contemporary Muslim folk beliefs. “The shell that separates the Oriental from the Unseen is still very thin, and the charm or amulet of the magician may break it. The world of the Arabian Nights is still his world […].”61 The anonymity of the compilers or authors of the stories also encouraged a populist interpretation of the story collection. Yet a more careful survey of the stories that found their way into the Nights suggests that a surprisingly large proportion derived from literary sources and, in some cases at least, they had formed part of the storytelling repertoires of cup companions, dandies, ʿulamāʾ, and other scholars—cultured folk who cherished the literary legacy of the Abbasids. The debt of the Nights to Abbasid belles lettres was large indeed. In addition to stories that explicitly refer to Abbasid personalities, there are many other tales that derive from other early works composed in classical Arabic. Thus, for example, among the barber’s tales of his six brothers, which are found in the hunchback cycle of stories and therefore in the Galland manuscript, the barber’s tale about his first brother draws on al-Iṣfahānī’s Kitāb al-Aghānī. The story of the second brother echoes a story in Ibn ʿAbd Rabbih’s ʿIqd al-farīd, and the fifth brother’s story derives from Kalīla wa-Dimna by Ibn al-Muqaffaʿ. As for the barber’s tale about himself, variant versions of this story are found in al-Masʿūdī’s Murūj al-dhahab, Ibn Ḥijja al-Ḥamawī’s Thamarāt alawrāq, and elsewhere.62 Some stories included in the Nights, such as that of the downfall of the Barmakid clan, seem to be nothing more than the retelling of historical events.

59 60 61 62

Arabian nights ii, 233–235; Arabian nights encyclopedia i, 277–278. Arabian nights ii, 120–122; Arabian nights encyclopedia i, 312. Macdonald, Religious attitude 126. Arabian nights i, 217–241; Arabian nights encyclopedia i, 115–121.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

316

irwin

Nevertheless, it should also be noted that the appearance of tales concerning the Abbasid caliphs and their entourages in late medieval anthologies and in the Nights stories was not necessarily a simple matter of the compilers of anthologies and the anonymous contributors to the great story collection reworking, elaborating, or distorting the original historical facts, for, as Tayeb El-Hibri has argued the extant ʿAbbasid historical narratives were not intended originally to tell facts, but rather to provide commentary on a certain political, religious, social, or cultural issue that may have derived from a real and controversial historical episode. Narrators writing before and during the era of al-Ṭabarī crafted the literary form of qișșa or khabar (narrative report), often with the intention of discussing the controversial results of a political, social or moral point.63 Again: “the entertaining tales of The Thousand and One Nights represents a spin-off from an earlier form of idealization whose roots can still be found in the historical sources.”64 More specifically, El-Hibri suggests that a modern reader might be distracted by the image of the joyful caliph scouting the streets of Baghdad disguised, with his vizier Jaʿfar, but although this image provides the setting for many stories describing the masquerading and trifling excursions of Harun under the cover of dark, one could say that this style of concealed travel by al-Rashid probably originated as an imitation of the orthodox image of the caliph ʿUmar i, touring the streets of Medina with his servant Aslam, seeking to check up on the welfare of his subjects.65 As that supreme expert on nostalgia, Marcel Proust, cautioned, “remembrance of things past is not necessarily remembrance of things as they were.” The conclusion of Kennedy’s The court of the caliphs runs as follows: The cultural legacy of the Abbasid court was immensely influential. The poets they patronized are still read and acknowledged as among the greatest in the Arabic language, and the translations they sponsored

63 64 65

El-Hibri, Reinterpreting Islamic historiography 13. El-Hibri, Reinterpreting Islamic historiography 21. El-Hibri, Reinterpreting Islamic historiography 25.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

317

formed the basis of higher learning not only in the Islamic world but in the medieval West as well. The Abbasids defined the style and performance of Muslim monarchy: they showed how a caliph and vizier should behave, how to decorate a palace, how to compose a formal proclamation; the flowing elegant script was itself a product of the Abbasid chancery.66 An examination of the culture of Mamluk Egypt and Syria only serves to underline this manifesto for the pervasiveness of the Abbasid legacy. Poets from the Abbasid period continued to be highly esteemed and imitated by poets in the Mamluk era. Scholars and scribes adopted and adapted tropes from the Abbasid rhetorical arsenal. Treatises on ingenious mechanical contrivances and ancient military tactics were looked on as treasures from a lumber room to be marveled at. Abbasid fashions and modes of comportment were taken as constituting an ideal of life. Anonymous storytellers worked on their audiences’ desire to relive what once had been—if only in their imagination.

Bibliography Sources The Arabian nights: Tales of 1001 nights, trans. M.C. Lyons, 3 vols., London 2008. al-Nuwayri, The ultimate ambition in the arts of erudition: A compendium of knowledge from the classical world, E. Muhanna (ed. and trans.), New York 2016. Ahmad al-Tifâchî, Les Délices des coeurs, trans. R. Khawam, Paris 1981.

Studies Amitai, R., The fall and rise of the ʿAbbasid Caliphate, in jaos 116 (1996), 487–494. Ayalon, D., Studies on the transfer of the caliphate from Bagdad to Cairo, in Arabica 7 (1960), 41–59. Bannister, M., The Abbasid Caliphate of Cairo, 1260–1517: Out of the shadows, Edinburgh 2021. Bauer, T., Mamluk literature: Misunderstandings and new approaches, in msr 9 (2005), pt. 2, 105–132. Bencheikh, J.E., La Volupté d‘en mourir’, in J.E. Bencheikh, C. Bremond, and A. Miquel (eds.) Mille et un contes de la nuit, Paris 1991, 276–283. Bray, J., A caliph and his public relations, in Middle Eastern literatures 7 (2004), 159–170. Conermann, S., Tankiz ibn ʿAbd Allāh al-Ḥusamī al-Nāșirī (d. 740/1340) as seen by his contemporary al-Șafadī (d. 764/1363), in msr 12 (2008), pt. 2, 1–24. 66

Kennedy, Court 296.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

318

irwin

Coussonet, P., Pour une lecture historique des mille et une nuits: Essai d’analyse du conte de deux viziers égyptiens, in Revue de l’Institut des Belles Lettres Arabes 48 (1985), 85–115. Coussonet, P., Pensée mythique, idéologie et aspirations sociales dans un conte des Mille et Une Nuits, Cairo 1989. Flemming, B., Aus den Nachtgesprächen Sultan Gauris, in Folia Rara: Festschrift für Wolfgang Voigt, Herbert Franke, Walther Heissig, and Wolfgang Treue (eds.), Wiesbaden 1967, 22–28. Garcin, G.-C., Histoire, opposition politique et piétisme dans le Husn al-Muhādarat de Suyūṭī, in ai 7 (1969), 33–89. Gelder, van G., Poetry and the Arabian nights, in U. Marzolph and R. van Leeuwen (eds.), The Arabian nights: An encyclopedia, i, Santa Barbara, CA 2004, 13–17. Ghazi, F.M., Un groupe social: Les raffinés (Zurafāʾ), in si 11 (1959), 39–71. Giffen, L.A., Theory of profane love among the Arabs: Development of the genre, London 1972. Giffen, Ẓarf, in J.S. Meisami and P. Starkey (eds.), Encyclopedia of Arabic literature, 2 vols., ii, London 1998, 821–822. Hassan, M., Longing for the lost caliphate: A transregional history, Princeton 2017. El-Hibri, T., Reinterpreting Islamic historiography: Hārūn al-Rashīd and the narrative of the Abbāsid Caliphate, Cambridge 1999. El-Hibri, T., The Abbasid Caliphate: A history, Cambridge 2021. Hill, D., The book of knowledge of ingenious devices by the Banū Mūsā ibn Shākir, Dordrecht 1979. Holt, P.M., Some observations on the Abbasid Caliphate of Cairo, in bsoas 47 (1984), 501–507. Horovitz, J., Poetische Zitate in Tausend und eine Nacht, in Festschrift Eduard Sachau zum Siebzigsten Geburtstage, Gotthold Weil (ed.), Berlin 1915, 375–378. Irwin, R., Mamluk literature, in msr 7 (2003), pt. 1, 1–29. Irwin, R., The political thinking of the “virtuous ruler” Qansuh al-Ghawri, in msr 12 (2008), pt. 1, 37–50. Kennedy, H., The armies of the caliphs: Military and society in the early Islamic state, London 2001. Kennedy, The court of the caliphs: The rise and fall of Islam’s greatest dynasty, London 2004. Little, D.P., Religion under the Mamluks, in mw 73 (1983), 165–181. Macdonald, D.B., Religious attitude and life in Islam, Chicago 1909. Mahdi, M., Appendix 3. From history to fiction: The tale told by the king’s steward, in The thousand and one nights (Alf layla wa layla) from the earliest known sources, M. Mahdi (ed.), iii, Leiden 1984–1994, 164–180. Marzolph, M., and R. van Leeuwen (eds.), The Arabian nights: An encyclopedia, 2 vols., Santa Barbara, CA 2004. = Arabian nights encyclopedia - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

muslim nostalgia

319

Muhanna, E., The world in a book: Al-Nuwayri and the Islamic encyclopedic tradition, Princeton 2018. Rowson, E.K., Ghuzūlī, in J. Meisami and P. Starkey (eds.), Encyclopedia of Arabic literature, 2 vols., i, London 1998, 254. Rowson, E.K., Tīfāshī, in J. Meisami and P. Starkey (eds.), Encyclopedia of Arabic literature, 2 vols., ii, London 1998, 772. Rowson, E.K., An Alexandrian age in fourteenth-century Damascus: Twin commentaries on two celebrated Arabic epistles, in msr 7 (2003), pt. 1, 97–110. al-Sarraf, S., Mamluk furūsīyah literature and its antecedents, in msr 8 (2004), pt. 1, 141–200. al-Sarraf, S., Furūsiyya literature of the Mamluk period, in D. Alexander (ed.), Furusiyya, vol. 1: The horse in the art of the Near East, Riyadh 1996, 118–135. Stewart, S., On longing: Narratives of the miniature, the gigantic, the souvenir, the collection, Durham, NC 1993. Thorau, P., The lion of Egypt: Sultan Baybars i and the Near East in the thirteenth century, trans. P.M. Holt, London 1987. Vrolijk, A., Bringing a laugh to a scowling face, Cairo 1998. Young, G.M., Victorian England: Portrait of an age, 2nd ed., London, 1953. Yungman, I., Medieval Middle Eastern court taste: The Mamluk case, in Y. Ben-Bassat (ed.), Developing perspectives in Mamluk history; Essays in honor of Amalia Levanoni, Leiden 2016, 76–96.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 17

The al-Mustanṣiriyya madrasa in Baghdad and Its founder, al-Mustanṣir Carole Hillenbrand

1

Introduction

figure 17.1

Al-Mustanṣiriyya madrasa: Courtyard from the south after schmid, die madrasa des kalifen al-mustansir in baghdad

The architectural features of the beautiful thirteenth-century al-Mustanṣiriyya madrasa in Baghdad have been analyzed by a number of distinguished scholars in the twentieth century, including a painstaking and detailed book written by Schmid in 1980.1 This chapter will focus on the religious, political, and educational aspects of the role and influence of the Mustanṣiriyya.

1 Schmid, Madrasa.

© Carole Hillenbrand, 2022 | doi:10.1163/9789004525245_018

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

2

321

Baghdad before the Beginning of the Reign of Caliph al-Mustanṣir (624–640/1242–1226)

Initially, under the early Abbasids, Baghdad had been the spiritual, and for a while the political, center of the Sunni Muslim world. In the second half of the sixth/twelfth century, Iraq and the venerated center of the Abbasid caliphate, Baghdad, fell under the sway of a Seljuq Turkish ruler, Ṭughril ii (r. 573–590/1177–1194). But he and the Turkish sultans who ruled after him acknowledged the Abbasid caliph as the spiritual leader of the Muslim world, and they sought his public recognition of their role as the military rulers. After Seljuq power collapsed in Iraq, the Abbasid Caliph al-Nāṣir (d. 622/1225) tried to make a final attempt to restore the Abbasid Caliphate to its former glory. During his 45-year rule, he showed himself to be a skillful and wise caliph. However, his successors proved to be politically weak and insignificant, and the Sunni Muslim caliphate in the east, in the face of dangerous external enemies, was destined to disintegrate rapidly as a political power.2 After only nine months and 24 days after his accession, al-Ẓāhir, al-Nāṣir’s son, died, and he was succeeded as caliph by his own son, al-Mustanṣir.3 The present paper can serve as a coda to the deftly drawn background to the later Abbasid Caliphate in Hugh Kennedy’s masterly overview of that institution throughout Islamic history.

3

An Introduction to the Reign of al-Mustanṣir

It is important to remember that the years of this caliph’s rule suffered from an uneasy military and political atmosphere because of two factors: the lurking proximity of Jalāl al-Dīn Khwārazmshāh and his campaigns in Iraq, and the ever-increasing threat of the Mongols. The primary sources for many areas of al-Mustanṣir’s life and activities within Baghdad are disappointingly sparse in detail. It is recorded that he was the eldest son of al-Ẓāhir, and his mother was a Turkish slave. He was proclaimed caliph on 13 Rajab 623/11 July 1226.4 He is often mentioned as a petty territorial ruler who received visits from neighboring potentates; for example,

2 Spuler, Disintegration 162. 3 For the death of al-Ẓāhir, see Ibn al-Athīr, al-Kāmil xii, 456. For Hugh Kennedy’s assessment of the later Abbasid caliphate, see Kennedy, Caliphate 175–214. 4 Ibn al-Athīr, al-Kāmil xii, 295.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

322

hillenbrand

Ibn Wāṣil (d. 697/1298) records in some detail that in 633/1235 two nearby potentates, Badr al-Dīn Luʾluʾ of Mosul and al-Malik al-Nāṣir Dāʾūd of Damascus, visited al-Mustanṣir.5 What is very clear in the sources, however, is that al-Mustanṣir was widely admired for his most significant role as the Sunni caliph of the Muslim world. He was praised for his generous almsgiving, especially in the last year of his reign when a plague hit Baghdad. Ibn Wāṣil eulogizes al-Mustanṣir’s character and his manner of governing the caliphate: “His way of life, may God have mercy on him, was the best conduct in justice, goodness towards (his) subjects and affection and sympathy for them. In all of that he was taking the path of his father, the Imam al-Ẓāhir bi-amr Allāh. So, he followed his path in faith in the madhhab of the Sunna and the Community, and having an aversion for the madhhab of the Rāfiḍites.6 The lands flourished greatly in the days of alMustanṣir bi-llāh.”7

4

What Was the Motivation for Building madrasas?

Building madrasas in the eastern Islamic world dates to the fourth/tenth and fifth/eleventh centuries. At that time, the word madrasa came to designate schools of higher education.8 As early as the end of the fourth/tenth century, because of the religious threat of the controversial group known as the Qarmathians, the Shāfiʿites of Khurasan had adopted the title madrasa to denote their schools, and they built four big madrasas in Nishapur. By the fifth/eleventh century Sunni Muslim rulers and religious lawyers realized the crucial necessity to strengthen Sunni Islam, to crush what they viewed as the frenetic propaganda of the Shiʿites—the Twelvers and the Ismaʿilis— and to remove any possible heterodox tendencies which may have been present in higher education until then. In the second half of the eleventh century the famous Seljuq Persian vizier, Niẓām al-Mulk, chose to establish three Shāfiʿite madrasas in Baghdad, including the celebrated madrasa in Baghdad known as the Niẓāmiyya, which opened in 459/1067. The enthusiasm and energy of Niẓām al-Mulk marked the beginning of a new period of brilliance for the

5 Ibn Wāṣil, Mufarrij v, 100–102. 6 Watt, Rāfiḍites; Halm, Schia 49. Halm writes that Rāfiḍites was the derogatory name for all the Shiʿites except the Zaydis. 7 Ibn Wāṣil, Mufarrij v, 316. 8 Massignon, Medresehs 77.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

323

madrasa. During his travels in the sixth/twelfth century, Ibn Jubayr mentions some thirty madrasas in Baghdad, in the eastern part of the city.9 What were the gifts that a religious scholar should possess? The celebrated legal and medical scholar of Baghdad, ʿAbd al-Laṭīf (d. 629/1231), gave the following advice to students on this matter: I commend you not to learn your sciences from books unaided, even though you may trust your ability to understand. Resort to professors for each science you seek to acquire; and should your professor be limited in his knowledge take all that he can offer, until you find another more accomplished than he. You must venerate and respect him […]. He who has not endured the stress of study will not taste the joy of knowledge.10 Such an environment for a madrasa was clearly important for the would-be scholar wishing to master the religious sciences in early thirteenth-century Baghdad.

5

The Reign of al-Mustanṣir (623–640/1226–1242)

In 625/1227, the 37th and penultimate Abbasid caliph, al-Mustanṣir, founded a magnificent madrasa named after him (see fig. 17.2); it took only six years to build. It stood right on the bank of the Tigris, by a bridge high enough for ships to sail under. The wide building includes a central courtyard. The solemn festivities of its inauguration in 630/1233 set the seal on its pre-eminent status among the madrasas of the Muslim world. Its endowments were bigger than those of other madrasas; they were valued at nearly two million gold dīnār (see fig. 17.3).11 Up until that time almost all such institutions were built for only one of the Sunni Muslim legal madhhabs; this one, however, was devoted to all four Sunni madhhabs: Ḥanbalī, Ḥanafī, Mālikī, and Shāfiʿī. The caliph invested great wealth in his new foundation. He wanted to supplant and eclipse the celebrated

9 10 11

Ibn Jubayr, Travels 238. Hourani, History 165. Al-Ḥawādith 48. This work was attributed by some scholars to an Iraqi scholar, Ibn alFuwaṭī (d. 718/1318), but its true authorship remains to be ascertained; Rosenthal, Ibn al-Fuwaṭī 769.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

324

hillenbrand

figure 17.2

Al-Mustanṣiriyya madrasa: Façade of the prayer hall after schmid, die madrasa des kalifen al-mustansir in baghdad

nearby Niẓāmiyya. The building was unprecedentedly large; its measurements were an oblong of circa 105×44×49 meters on two floors.12 The professor of the caliphal palace, Muḥammad b. al-ʿAlqamī, was in charge of building the caliph’s madrasa. The income the caliph bequeathed ensured that all members of the madrasa would be cared for appropriately; they numbered 400 persons. Around 300 of them actually lived in the Mustanṣiriyya madrasa itself. It was for that reason that this madrasa, as well as containing teaching rooms and a famous library, also had residential rooms for the large staff and students, as well as a kitchen, bathroom, and its own hospital.13 All in all, it is likely that this was the largest and most ambitious of all medieval madrasas. A parallel with the residential colleges of Oxford and Cambridge springs to mind. The opening of the madrasa was on the fifth day of the month of Rajab in 631/1233. The caliph often visited the madrasa and listened from a special seat to the lectures and scholarly discussions. The method of teaching was by lectures and learning by heart.14 Occasionally, he organized official receptions in the madrasa for heads of state and other eminent visitors.

12 13 14

Hillenbrand, Madrasa 1,148. Schmid, Madrasa 1. Petersen, Madrasa 1,123–1,134.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

figure 17.3

6

325

Al-Mustanṣiriyya madrasa: Axonometric view after hillenbrand, islamic architecture

The Establishment of the Mustanṣiriyya

Ibn Wāṣil writes the following description of how and where al-Mustanṣir organized the placing, structure, and arrangement of his new madrasa in Baghdad: He built on the bank of the Tigris on the eastern side near the house of the caliphate a madrasa called the Mustanṣiriyya. A madrasa lovelier than that had not (ever) been built on the face of the earth nor (had there been) a greater endowment (waqf ). He installed four teachers from the four madhhabs. Each of them had a lofty chair with a back to lean on. He established a splendid library in the madrasa where there were very many valuable books about current branches of knowledge, such as fiqh, literature and science.15 He arranged them so that the fuqahāʾ could read and copy them, and he put paper and pens there for those who wanted to copy.

15

Al-Ḥawādith 54; Hillenbrand, al-Mustanṣir.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

326

hillenbrand

He founded a hospital for the madrasa for those who were ill. In it were all sorts of medicines, aromatic roots, and drinks. He put there those doctors who could heal the fuqahāʾ.16 Bar Hebraeus, the Syriac Orthodox Eastern Christian chronicler (d. 685/1286), also writes a very approving and detailed description of how al-Mustanṣir established the Mustanṣiriyya: He built that House of Instruction which is called by his name, and the like of which exists nowhere in the world. And he appointed therein four doctors of their Four Faiths, and three hundred chief instructors, and every instructor in the Law was provided with adequate rations from day to day. And he built a bath for their private use, wherein no stranger might enter, and he appointed for them also a physician who would administer medicine unto them when they were sick.17 These many functions go far to explain the ambitious layout of the building with its stunning array of vaulting, while its rich terracotta ornament underscores its importance (see fig. 17.4). The celebrated Arab traveler Ibn Baṭṭūṭa (d. 771/1369) also gives a description of the teaching in the Mustanṣiriyya: All four schools are included in it, each school having an īwān18 with its own mosque and lecture room. The teacher takes his place under a small wooden canopy, on a chair covered with rags: he sits (on this) in a grave and quiet attitude, wearing robes of black and his (black) turban, and with two assistants on his right and left, who repeat everything that he dictates. This system is followed in every formal lecture of all four sections. Inside this college there is a bathhouse for the students and a chamber for ablutions.19 As well as the rooms allocated to the four Sunni madhhabs, the Mustanṣiriyya also included two other rooms, a Dār al-ḥadīth (for instruction on the sayings of the Prophet Muḥammad) and a Dār al-Qurʾān (for teaching the seven ways of reading the Quran). Six professors were appointed. 16 17 18 19

Ibn Wāṣil, Mufarrij v, 316. Bar Hebraeus, Chronography 390. An īwān is a vaulted or flat-roofed hall, open at one end; Hillenbrand, Islamic architecture 598. Ibn Baṭṭūṭa, Travels ii, 332.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

327

figure 17.4 Al-Mustansiriyya madrasa: Corridor (īwān) after hillenbrand, islamic architecture

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

328

hillenbrand

How were the rooms allotted? According to the Tarʾīkh Baghdād, the quarter to the right of the qibla was given to the Shāfiʿīs, the quarter to the left to the Ḥanafīs, that to the right of the entrance to the Ḥanbalīs, and that to the left to the Mālikīs.20 Everyday needs in the madrasa were generously met. Olive oil was used for lighting the college and lamps were provided for the students. In the storerooms of the madrasa were food, drinks, and medicines. The caliph went almost every day to inspect the building. He arranged for the laying out of a private garden, in which there was an observatory overlooking the madrasa. There he used to sit near a window and watch what was going on and even listen to the lectures and discussions with the students.21

7

The Form and Importance of Monumental Islamic Inscriptions

There is no doubt that medieval Islamic culture greatly valued monumental inscriptions. They were unusual in several respects in comparison with inscriptions found within other Mediterranean and West Asian cultures. Indeed, Muslim monumental inscriptions served a number of different purposes. Their locations were often grandiose, for they occupied lofty positions on the walls of mosques and madrasas. Their content recorded the titles and achievements of the founder of the building and historical information, including its date. But above all, the inscriptions were written for the glory of God and to stress the importance of Muslims’ relationship with Him. The most usual format of such inscriptions favored a special order: first, the bismillāh, then a verb to indicate what had been ordered done, followed by the object of the verb, then the name of the person who had ordered the inscription, and lastly the date of the construction of the building.22 Medieval Muslim architectural inscriptions were usually long and confined to bands. The position of inscriptions was carefully planned, with obvious propagandistic aims. A section, a word, or most frequently a cluster of laudatory titles and grandiose attributes belonging to the caliph were often situated in a special place in the inscription, such as around an arch or over a door.23

20 21 22 23

Creswell, Muslim architecture ii, 124–125; Massignon, Medresehs 80–82. Le Strange, Baghdad 267–268. Blair, Islamic inscriptions 1. Hillenbrand, Islamic architecture 222.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

figure 17.5

8

329

Al-Mustansiriyya madrasa: The wall facing the Tigris from the northwest after schmid, die madrasa des kalifen al-mustansir in baghdad

The Mustanṣiriyya Inscriptions

A key factor in the position chosen for the inscriptions on the Mustanṣiriyya was the aim of propagating their message as widely as possible. It is important to note that the caliph could have chosen to put up religious buildings, and especially the Mustanṣiriyya, inside the city of Baghdad itself. But he decided to build his most famous monument right by the river. The Mustanṣiriyya is situated on the eastern side of the Tigris on the downstream side of the Shuhadāʾ bridge, right on the riverbank. This was the way to make the building constantly visible to more people, both inhabitants of Baghdad on both sides of the river as well as the sailors and passengers in the ships moving toward the dock. Commerce between the Persian Gulf and India and China had flourished greatly under the Abbasids, as long as Baghdad remained the metropolis of the Middle East.24 Baghdad and Basra’s geographical position made them the main western termini for maritime trade between Iraq and China in Abbasid times between the seventh and tenth centuries. As al-Ṭabarī (d. 310 /923) records:

24

Hourani G., Arab seafaring 53.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

330

hillenbrand

Upon founding the city, al-Manṣūr would have declared: ‘Here is the Tigris with nothing between us and China, and on it arrives everything that the sea can bring.’25 Similar enthusiasm for the riverine position of Baghdad is provided by alYaʿqūbī (d. 284/897) when he writes that Baghdad “is an island between the Tigris and the Euphrates […] and a waterfront for the whole world.”26 The caliph’s decision to place his grand madrasa on the riverbank was a way to make the building constantly visible to more people, both inhabitants of Baghdad on both sides of the river, as well as the sailors and passengers in the ships regularly moving from distant lands toward the dock. The location of this madrasa made the most of Baghdad’s status as an inland port open to oceangoing vessels and opening onto vast horizons.

9

Translations of Some Selected Arabic Inscriptions on the Mustanṣiriyya

1) A long inscription band on the west side of the building stretches along the dock. It is written in Ayyubid cursive script in large letters.27

figure 17.6

Al-Mustanṣiriyya madrasa: Riparian inscription in Ayyubid-style cursive writing as transcribed by Niebuhr, with emendations by Herzfeld after sarre and herzfeld, archäologische reise i, 43

There was ordered to be built this noble madrasa for the students of knowledge, called the great madrasa, by the radiant ruler, the auspicious one among created beings, the shining white way in the eye of God,28 and 25 26 27 28

Al-Ṭabarī, History xxviii, 238; George, Direct sea trade. Hourani, Arab seafaring 64. Sarre and Herzfeld, Archäologische Reise i, 42–43. Caliph al-Mustaʾṣim’s general, Afshīn, dreamed of restoring the Persian empire and the “white religion”; Goldziher, Muslim studies i, 140. The color white in Islam symbolizes purity and peace.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

331

the one chosen by Him as the caliph on earth, the caliph Abū Jaʿfar alManṣūr al-Mustanṣir bi-llāh, the Prince of the Believers. May God give him a radiant rule for the benefit of the Muslims and strengthen his kingdom, in which He grants him a long life! That happened in the year 630 (1232–1233). This splendid inscription extended the full length of the river façade to the north of the bridge.29 It can be regarded as a crowning glory of the Mustanṣiriyya. It reflects very eloquently, and for the last time, the majesty and power of an Abbasid caliph who wished to establish a teaching institution for all Sunni Muslims and to propagate his message throughout all the lands of Islam, reclaiming in that process some of the status the office of caliph had long lost. Van Berchem points out that most of this inscription has now disappeared and has been replaced by a modern one with similar but very ugly lettering. This bears the date 1282 (1865–1866) and can be seen in the customs shed at the Tigris dock.30 2) Part of an inscription on the wall overlooking the Tigris.

figure 17.7

Al-Mustanṣiriyya madrasa: Inscription on the wall facing the Tigris répertoire 11/2, 42

In the name of God, the Merciful, the Compassionate. This was completed by order of the Commander of the Faithful and Caliph of the Master of the Worlds, who has covered the lands with his goodness and his justice and immersed the worshippers of God in his piety and his merit, Abū

29 30

Bell, Amurath 191. Sarre and Herzfeld, Archäologische Reise i, 43.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

332

hillenbrand

Jaʿfar al-Manṣūr al-Mustanṣir bi-llāh. May God crown his noble orders with success and ease and his armies with support and victory […] That (was completed) in the year 630 (1232–1233).31 3) Part of the inscription on the easternmost portal of the Mustanṣiriyya, over the door and under the high, no longer accessible, arch of the entrance.32

figure 17.8

Al-Mustansiriyya madrasa: The southwestern exterior wall with the inscription as restored in 1865 schmid, die madrasa des kalifen al-mustansir in baghdad

It was ordered that a madrasa should be built for lawyers of the four madhhabs by our Lord and Master, the Imām of the Muslims and the Caliph of the Lord of the World, Abū Jaʿfar al-Manṣūr, al-Mustanṣir billāh, the Commander of the Faithful. May God, the Exalted, firmly establish the educational institutions of religion throughout the duration of his rule. May He animate the hearts of men of knowledge by the multiplying of His proofs of grace and His help. It was completed in the year 630. God’s blessing on our Master Muḥammad, the Prophet, and his family.

10

Merits of al-Mustanṣir

Especially noteworthy are the panegyric words of the Syriac Christian, Bar Hebraeus (d. 685/1286): Mustanṣir ruled for seventeen years. This man began (to rule) in an open fashion, he rode and he went in and came out, and hid himself from no

31 32

Répertoire 11/2, 42; Massignon, Medresehs 81. Sarre and Herzfeld, Archäologische Reise ii, 164.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

333

man. And there was exhibited by him a justice greater than those of his father, and he began to distribute alms generously and he restored many buildings.33 Al-Mustanṣir also erected buildings in addition to the Mustanṣiriyya. They included the khān of Sābūs near Wāsiṭ.34 He also restored the Jāmiʿ al-Qaṣr in Baghdad, and he installed in it benches on which students could sit and have discussions after performing the prayers. It is recorded that during his reign he also dug many canals to irrigate the lands. Al-Mustanṣir was especially generous in his almsgiving, especially when the plague struck Baghdad in the last year of his life.

11

Death of al-Mustanṣir (640/1242)

Under the year 640/1242 Ibn Wāṣil provides some precise and informative details about the death of al-Mustanṣir, his genealogy, and the way he died: The caliph al-Mustanṣir bi-llāh, Abu Jaʿfar al-Mustanṣir b. al-Zāhir bi-amr Allāh Naṣr Aḥmad b. al-Nāṣir li-Dīn Allāh Abū l -ʿAbbās Aḥmad died […] in the early morning of Friday, with ten days remaining of Jumāda ii of this year, meaning the year 640. We have mentioned that he took over the caliphate eleven nights before the end of Rajab in the year 623/1226. The length of his caliphate was one month short of seventeen years. The reason for his death was according to what Wajīh al-Dīn b. Suwayd al-Tikrītī told me, (and he was familiar with what happened), that he was stabbed with a poisoned knife. And that had been mentioned before. And if this was true, the statement that agrees that every sixth one of the Banū ʿAbbās is deposed or killed, is not wrong.35 Ibn Taghrībirdī gives a detailed description of al-Mustanṣir’s appearance; he had red hair and pale skin, and he was short and corpulent.36

33 34 35 36

Bar Hebraeus, Chronography 390. Ibn al-Ṭiqṭaqa, Fakhrī 567–568. Ibn Wāṣil, Mufarrij v, 317. Ibn Taghrībirdī, al-Nujūm vi, 345–346.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

334

hillenbrand

figure 17.9

12

Al-Mustanṣiriyya madrasa: The portal inscription after schmid, die madrasa des kalifen al-mustansir in baghdad

Influence of the Mustanṣiriyya

This building had grandiose aims. It was established as a teaching institution for all Sunni Muslims; it reached out to students from many areas of the Muslim world. Its universality was paramount. The inscriptions carved in the Mustanṣiriyya praise Caliph al-Mustanṣir’s power and majesty. It was usual for a caliph to finance a new mosque or a mausoleum. This exceptional building was a madrasa, and thanks to its lavish layout and organization it had much greater prestige. Its library became celebrated. The caliph would welcome and entertain heads of state there. While the Seljuq vizier, Niẓām al-Mulk, in the eleventh century, had already chosen Baghdad as the location for his famous madrasa, the Niẓāmiyya, this other very notable building was established in a period of greater stability for the Abbasid caliphate in the early seventh/thirteenth century. The Mustanṣiriyya, seen retrospectively, appears as the last moment of Abbasid caliphal fame and grandeur before the horrendous events of the coming of the Mongols, the sacking of Baghdad, and the murder of al-Mustanṣir’s son, al-Mustaʿṣim, in 656/1258. The subsequent revival of the Abbasid Caliphate located in Cairo never reached the heights of prestige enjoyed in the heyday of Abbasid power in Baghdad. In its own historical context, the Mustanṣiriyya was the first madrasa built by a caliph. It proclaimed its presence by its sheer size and monumentality. It was visible to visitors arriving by ship on the river. Its location alone could be manipulated for propagandistic purposes. It may well be that by establishing the Mustanṣiriyya, described by Ḥamdallāh Mustawfī as “the most beautiful building in Baghdad,”37 al-Mustanṣir, who was such an unusual and still little37

Ḥamdallāh Mustawfī, Nuzhat 142.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

335

studied Abbasid caliph, was proclaiming (far too late in the day) the paramount need for unity in the Muslim world. The modern Arab scholar Najī Marouf provides many examples of how the practices of the Mustanṣiriyya were copied in later times. Discussing the events of 631/1233, he points out the following: “When al-Mustanṣir opened the abovementioned madrasa to teach fiqh to the four madhhabs, and to teach tafsīr and the knowledge of the Quran and ḥadīth and medicine and Arabic and religious exercise, by (doing that) Baghdad was ahead of all the Arab and Islamic countries and it introduced them to the practice of building madrasas for the four madhhabs.” He then goes on to mention the example of Egypt: “Ten years passed until the first madrasa for four madhhabs was opened in Egypt in the year 641/1243. It is the al-Ṣalāḥiyya madrasa which al-Malik al-Ṣāliḥ Najm al-Dīn Ayyūb established in Cairo.”38 Marouf then highlights the somewhat different example of a madrasa in Mecca, built in 814/1411 by al-Malik al-Manṣūr: “He decided that there should be in it four teachers—at that time there were four qāḍīs of Mecca and sixty lawyers—twenty for Shāfiʿism, twenty for Ḥanafism, ten for Mālikism and ten for Ḥanbalism.”39 Equality of importance in the four Sunni madhhabs is obviously not maintained here. In the two lengthy volumes of his work, Marouf meticulously provides short biographies, and occasionally longer ones, of many scholars who taught in the Mustanṣiriyya. Their biographies are recorded in separate chapters devoted to each of the four Sunni madhhabs. He also records the names and details of the medical doctors who taught there and he explains where they were housed in the building.40

13

Later History of the Mustanṣiriyya in Baghdad

The city of Baghdad was greatly damaged by the Mongol conquest and the Abbasid dynasty of Baghdad was brought to an end with Hülegü’s murder of the last caliph, the son of al-Mustanṣir, al-Mustaʿṣim, in 656/1258. The Mustanṣiriyya survived the city’s terrible destruction, but its library was robbed, and it was closed as a madrasa. Only a century later was it re-opened, and it remained a madrasa until the ninth/fifteenth century, but it suffered much

38 39 40

Marouf, Taʾrīkh ʿulamāʾ i, 37–38. Marouf, Taʾrīkh ʿulamāʾ i, 41. Marouf, Taʾrīkh ʿulamāʾ i, 391.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

336

figure 17.10

hillenbrand

Al-Mustanṣiriyya madrasa: Part of the portal inscription after sarre and herzfeld, archäologische reise ii, 164

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

337

thereafter.41 In the sixteenth century it was a lair for brigands before becoming a customshouse under Turkish domination. When the famous traveler to the Arab world, Carsten Niebuhr (d. 1815), saw the Mustanṣiriyya, he copied some of its inscriptions,42 but he found that the building was occupied by businessmen. In 1865 the Turkish Sultan ʿAbd al-ʿAzīz renewed the southern parts of the river wall, against which a quay had been built and was shattered. The northern parts of the river wall were hidden behind a coffeehouse, which stood on the bank of the Tigris. Inside, the building itself was disfigured by makeshift installations, and it served as the main customs office for Baghdad. The intrepid British traveler, art historian, and administrator, Gertrude Bell, writing in the early years of the twentieth century, was still very impressed by the buildings she saw in Baghdad. She declared that: “the mastery of structural problems shown by the architects of Islam in the thirteenth century is nothing short of amazing.”43 On April 1, 1945, the Iraqi Directorate of Antiquities management took over the Mustanṣiriyya and began to restore it. In 1960 the former madrasa was given a new role as the Museum for Islamic Culture and Art. The building tasks were completed in 1962. In March 2021 Iraq’s Ministry of Culture, Tourism and Antiquities announced that a project had begun to relaunch “the historic Mustanṣiriyya.”44 This announcement presages yet another reworking of this spectacular example of late Abbasid architecture, and bodes ill for the still further refurbished image it will present to modern viewers of this storied building. Every attempt to restore it in the past 80 years has taken it further away from its original state, since the aim of each successive restoration has not been to conserve what was still there but to make it look spick and span. So, one can only be grateful for the photographs taken by Herzfeld in 1909 of the monument in its original though ruined state, priceless documents of its vanished grandeur.43

41 42 43 44

Al-Janābī, Studies 73. Niebuhr, Carsten Niebuhrs Rejsebekrivelse ii, 295. Bell, Amurath 192. For the 2021 announcement, see Iraq al-monitor: The pulse of the Middle East, March 11, 2021.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

338

hillenbrand

Bibliography Sources Bar Hebraeus, The chronography of Gregory Abuʾl-Faraj, 1225–1286, trans. E.A.W. Budge, Oxford 1976. al-Fakhri, Al-Fakhri on the systems of government and the Moslem dynasties, trans. C.E.J. Whitting, London 1947. Ḥamdallāh Mustawfī, Nuzhat al-qūlūb, partial trans. G. Le Strange, Leiden and London 1919. al-Ḥawādith al-jāmiʿa wa-l tajārūb al-nāfiʿa fī l-maʾa al-sābiʿa, ed. M. Jawād, Baghdad 1932. Ibn al-Athīr, al-Kāmil fī l-taʾrīkh, ed. C.J. Tornberg, vol. 12, Beirut 1853. Ibn Baṭṭūṭa, The travels of Ibn Baṭṭūṭa a.d.1325–1354, trans. H.A.R. Gibb, vol. 2, New Delhi 1993. Ibn Jubayr, The travels of Ibn Jubayr, trans. R.J.C. Broadhurst, London 1952. Ibn Taghrībirdī, al-Nujūm al-ẓāhira fī mulk Miṣr wa-l-Qāhira, ed. W. Popper, vol. 6, Cairo 1929–1956. Ibn al-Ṭiqṭaqa, Fakhrī, trans. E. Amar, Paris 1910. Ibn Wāṣil, Mufarrij al-kurūb fī akhbār Banī Ayyūb, ed. H. Rabie, vol. 5, Cairo 1395/1975. Sibṭ b. al-Jawzī, Mirʾat al-zamān, vol. 8, in 2 parts, Hyderabad 1951–1952. Textes et traductions d’auteurs orientaux, vol. 1: Yaʿḳūbī: Les pays, trans. G. Wiet, Cairo 1937. al-Ṭabarī, The history of al-Ṭabarī, vol. 28: Abbasid authority affirmed, trans. J.D. McAuliffe, Albany 1995. al-Yaʿqūbī, Kitāb al-Buldān, ed. M.J. de Goeje, vol. 7, Leiden 1892.

Studies Bell, G., Amurath to Amurath, New York 1911. Blair, S., Islamic inscriptions, Edinburgh 1998. Creswell, K.A.C., The Muslim architecture of Egypt: Ayyubids and early Bahrite Mamluks a.d.1171–1326, 2 vols., New York 1978. George, A., Direct sea trade between early Islamic Iraq and Tang China: From the exchange of goods to the transmission of Ideas, in jras 25 (2015), 579–624. Goldziher, I., Muslim studies: Muhammedanische Studien, ed. S.M. Stern, vol. 1, London 1961. Halm, H., Die Schia, Darmstadt 1988. Hillenbrand, C., al-Mustanṣir, in ei2, vii, 727–729. Hillenbrand, R., Islamic architecture, Edinburgh 1994. Hillenbrand, R., Islamic monumental inscriptions: Location, content, legibility and aesthetics, in L. Korn. and A. Heidenreich (eds.), Beiträge zur Islamischen Kunst und Archäologie, iii, Wiesbaden 2021, 13–38. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the al-mustanṣiriyya madrasa in baghdad

339

Hillenbrand, R., Madrasa, in ei2, v, 1,136–1,154. Hourani, A., A history of the Arab peoples, London 1991. Hourani, G., Arab seafaring in the Indian Ocean in ancient and early medieval times, Princeton 1951. al-Janābī, T., Studies in mediaeval architecture, Baghdad 1982. Le Strange, G., Baghdad during the Abbasid Caliphate, London, Dublin, and New York 1972. Kennedy, H., The caliphate, St. Ives 2016. Makdisi, G., The rise of colleges: Institutions of learning in Islam and the west, Edinburgh 1981. Marouf, N., Tarikh ʿulamāʾ al-mustanṣiriyya, vol. 1, Baghdad 1384/1965. Massignon, L., Les Medresehs de Baghdad, in bifao 7 (1907), 77–86. Niebuhr, C., Carsten Niebuhrs Rejsebekrivelse fra Arabien 09 andre omkingliggende lande, vol. 2, Copenhagen 2002–2003. Petersen, J., Madrasa, in ei2, v, 1,123–1,134. Répertoire Chronologique d’Épigraphie Arabe, eds. Ét. Combe, J. Sauvaget, and G. Wiet, vol. 12, Cairo 1942. Rosenthal, F., Ibn al-Fuwaṭī, in ei2, iii, 769–770. Sarre, F., and E. Herzfeld, Archäologische Reise im Euphrat und Tigris-Gebiet, vol. 2, Berlin 1920. Schmid, H., Die Madrasa des Kalifen al-Mustansir in Baghdad: Eine baugeschichtliche Untersuchung der ersten universalen Rechtshochschule des Islam, Mainz am Rhein 1980. Sourdel-Thomine, J., L’art de Baghdad, in Arabica 9 (1962), 449–465. Spuler, B., The disintegration of the caliphate in the east, in P.M. Holt., A.K.S. Lambton, and B. Lewis (eds.), The central Islamic lands (chi), i, London and New York 1970, 143–174. Watt, W.M., The Rāfiḍites: A preliminary study, in Oriens 16 (1963), 110–121.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 18

Hārūn al-Rashīd in Premodern Arabic Literary Imaginary: Ideology of Monogamy, Harem Politics, and Court Intrigues Wen-chin Ouyang

1

Hārūn al-Rashīd the Icon

It is not an exaggeration to say Hārūn al-Rashīd (r. 170–193/786–809) is even now a global celebrity. He made it into John Canning’s (b. 1920) list of 100 great kings, queens, and rulers of the world (1968), which includes only five names from the Islamic Middle East: “Mohammed,” “Abd Al-Rahman,” “HarunAl-Rashid,” “Saladin,” and “Mohammed ii” (the Conqueror). Fame has its trappings. He is at once glamorous and dark, famous and notorious. He is the inspiration behind André Clot’s history of the early Abbasid rule, Harun al-Rashid and the world of The thousand and one nights (1986), because the complexity of his character and rule can inform historical narratives that bring out all the contradictions of an era. This is how Clot’s history is advertised: A symbol of the fabled Orient, Harun al Rashid, the caliph portrayed in The Thousand and One Nights, where we see him living grandly in his palace in Baghdad, surrounded by his wives, his concubines, musicians, and learned men, is not merely a figure of legend. He was the son of a Yemenite slave who cleared his path to power, very probably by poisoning the reigning caliph, her older son. Harun reigned for a quarter-century, and was the most famous caliph of the Abbasid dynasty. Through Arab chronicles, the author corrects our vision of Harun the Good, and gives a remarkable account of his development as a ruler. Though in Western countries he is remembered for the presents he sent to Charlemagne— notably the famous elephant, Abul Abbas—he was first and foremost a successful soldier who made war on the Byzantines. His empire was shaken by religious and social insurrections, and he did not shrink from annihilating the Barmecides, a powerful family whose wealth and influence he finally found unbearable. As a patron of pets and intellectuals, Harun contributed greatly to the cultural supremacy of Baghdad,

© Wen-chin Ouyang, 2022 | doi:10.1163/9789004525245_019

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

341

whose merchants and navigators spread the name of the caliph throughout the world.1 He is at the same time the epitome of tyranny in Milton Klonsky’s The fabulous ego: Absolute power in history (1974), the psychoanalysis of whom can help us to understand the human penchant for tyranny and how to temper our ego so as to transcend it. Since his rise to iconic fame in the second/eighth century, Hārūn al-Rashīd and his portrayals in both Arabic and orientalist historical and literary writings have understandably been subject to continuous scrutiny, each in turn producing a new icon based on but departing from previous ones. In recent Arabic scholarship on Hārūn al-Rashīd, there are many attempts to rehabilitate his image and rescue his reputation from tyranny and, more importantly, from the association between the type of tyranny he exemplifies and political and moral corruption. If Aḥmad Amīn (1886–1954) presents Hārūn al-Rashīd in all his multifaceted splendor in Hārūn al-Rashīd (n.d.),2 and Muḥammad ʿAbd al-Raḥmān Yūnus daringly relates political tyranny to sexual debauchery in the stories structured around him in al-Istibdād al-sulṭawī wa-l-fasād al-jinsī fī alf layla wa-layla (2007), scholars intolerant of ambiguity, against the grain of medieval Arabic-Islamic culture,3 feel they must rise to his defense and restore his reputation as not only a just ruler but also a pious Muslim who observes the strict rules of Islam. Shawqī Abū Khalīl in Hārūn al-Rashīd: Amīr al-khulafāʾ waajall mulūk al-dunyā (1996), Aḥmad al-Qaṭṭān and Muḥammad Ṭāhir al-Zayn in Hārūn al-Rashīd: al-Khalīfa al-maẓlūm (2001), and Ḥasan ʿAbd al-Ghaffār in Hārūn al-Rashīd: al-Khalīfa al-muftarā ʿalayhi (2009), just to name three examples, remove any mention of his abuse of power, concubines, and drinking parties from his life and rule, harem, and court, and offer us a biography of a devout Muslim who ruled justly, wisely, and magnanimously. He loved his first wife, Zubayda, and married only three more women, all mahāʾir, whom he wed properly as evidenced by the dowry he paid for them. Anecdotes about his conduct as ruler at court in the premodern adab tradition and the stories about his nocturnal adventures in The thousand and one nights are expunged from these accounts. Such a cleansing operation, however, tells us more about Hārūn al-Rashīd’s powerful presence in the Arabic cultural and literary imaginary. The ways rep1 Clot, Harun al-Rashid (e-book, accessed August 31, 2021). 2 Aḥmad Amīn, Hārūn al-Rashīd via Hindawi.org: https://www.hindawi.org/books/51719608/ (accessed August 31, 2021). 3 See Bauer, Culture of ambiguity.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

342

ouyang

resentations of his court and harem are tinkered with, edited, and transformed give us a sense of the importance of the symbolic value of Hārūn al-Rashīd, the icon, for generations of Arabs and Muslims who lay a claim to him as the symbol of their community and history. This is evident even in the more liberal representations of his character and rule. Two television adaptations suffice as examples: the prize winning 1997 Egyptian production written by ʿAbd alSalām Amīn and directed by Aḥmad Tawfīq,4 and the 2018 Syrian production funded by the Emirates written by ʿUthmān Juḥā and directed by ʿAbd al-Bārī Abū l-Khayr.5 These two series produced for the Ramadan season are by definition pious. However, they do not shy away from the complexity of his political career and social life but rather bring them together to make statements about the style and quality of his conduct and rule. Nūr al-Sharīf (1946–2015), who is himself an iconic Egyptian actor, portrays Hārūn al-Rashīd as a Sufi who, albeit fully cognizant of his responsibility as the leader of the Muslim community, thirsts for a peaceful life spent away from court intrigues and harem politics and instead in remembrance of God and the companionship of a pious and wise woman. Another iconic Hārūn al-Rashīd, that of the Syrian actor Quṣay Khūlī, on the other hand, is a skillful politician who manages the competing claims made on him by the members of both his court and harem successfully and imposes a delicate balance among the various players in the games of power unfolding before him. He is able to ensure stability, peace, and justice in his domain. In the two types of the contemporary rehabilitation of Hārūn al-Rashīd, one idealizing him as a devout Muslim and another romanticizing him as a Machiavellian but just commander of the faithful, each refashioning entails selecting material from the extensive body of writing on and around this global celebrity and channeling it into a narrative discourse out of which the desired ideal character emerges. Regardless, the selection process always and inevitably involves making a decision about what to do with The thousand and one nights and, to a lesser extent, the adab anecdotes about his indulgence in poetry, music, and, of course, sex. Even a small detail can change the intended message. The choice of Abū l-ʿAtāhiya (130–210 or 211/748–825 or 826) or Abū Nuwās (139–ca. 199/756–ca. 814) as a member of his entourage for his nocturnal tours of Baghdad is of key significance. In the 1997 Nūr Al-Sharīf version, choosing Abū l-ʿAtāhiya to be his mentor and companion follows the Sufi path of the 4 Available online at: https://www.youtube.com/watch?v=IcmGKTXdbXM&list=PL9KkecclNU BRfyIfbHycHhs_YOAbhqCZz. 5 Available online at: https://www.youtube.com/watch?v=qI5a‑4golhQ&list=PLDsEOSGNngQ 0zx0mpMqN30RF0hsnJBF5E.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

343

Egyptian tv Hārūn al-Rashīd. The choice of Abū Nuwās in Hanan al-Shaykh’s 2011 “reimagining” of One thousand and one nights signals a subversive agenda that seeks to turn the world of patriarchal authority upside down. By inserting Abū Nuwās, here as a homosexual, libertine, wine-imbibing, ritual clown, into the Hārūn al-Rashīd cycles in the Nights, al-Shaykh takes the side of “joie de vivre” in opposition to the austere religious life required for all Muslims by the fundamentalists, and at the same time it creates lines of escape for the women characters from the perceived total tyranny of patriarchy. Here, she overlaps fiction with history, Shahriyār with Hārūn al-Rashīd, and the fictional Hārūn al-Rashīd with his historical counterpart. Al-Shaykh’s One thousand and one nights: A new re-imagining is not the first work by an Arab to interrogate patriarchal tyranny in such a way. As early as 1984, the Egyptian television series that rewrote the Nights for that year’s Ramadan season—hailed as the most successful adaptation of the Nights in Arabic, not just Egyptian television, and written by Aḥmad Bahjat, directed by ʿAbd al-ʿAzīz al-Sukkarī and Saʿīd al-Rashīdī, and starring the two icons of Egyptian cinema, Najlāʾ Fatḥī and Ḥusayn Fahmī—has Shahrazād time travel from modern Cairo to Shāhriyār’s kingdom in time immemorial and treat and heal him from his narcissism and tyranny. Calling him a dictator (diktātūr), Shahrazād instills in Shāhriyār what we would call today gender equality, as well as democracy. In the retelling of the Hārūn al-Rashīd stories, Fatḥī and Ḥusayn, who play Shahrazād and Shāhriyār, also impersonate, for example, Zubayda and Hārūn al-Rashīd, thus overlapping the patterns of Shāhriyār’s power with those of Hārūn al-Rashīd. Of course, they also play merchants, working-class characters, and slaves—and here Najlāʾ Fatḥī embodies the erudite concubine figure in the Nights and premodern Arabic writings— and together, they take us on a journey of educating passion into love, tyranny into justice, and dictatorship into democracy. In this story of the education of the king and kingship, power and sex are inseparable, as Yūnus reminds us in alIstibdād al-sulṭawī wa-l-fasād al-jinsī. Is it only premodern Arabic storytelling, in the Nights and adab, that make such a link, leaving it to us today to transform the stories of love and adventure into modern tales of subversion? I return to the imaginings of the character, entourage, and rule of Hārūn al-Rashīd in classical Arabic storytelling and adab works and explore the role this body of stories and anecdotes plays in interrogating tyranny and, more importantly, delineating lines of escape. I am interested in the iconic, not the historical, Hārūn al-Rashīd, making my starting point the overlap of three patterns of political authority, here, the “kingship” of Shāhriyār, Hārūn al-Rashīd, and the King of China or Sultan of Egypt or Basra, and the intersection between court intrigues and harem politics. I am most particularly intrigued by the inter-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

344

ouyang

vention of what I would call the “ideology of monogamy” in the ways the Nights stories weave together historical information and adab anecdotes, culled from a culture of practice so alien to it, into fictional tales of wonder and marvel. The Nights’ strategic fabrication extends our temporal and spatial purview from here and now to there and forever, and from the Nights to a multigeneric body of writing Ibn Khaldūn calls adab. At the same time, it focuses our view on “kingship,” its style and quality, its scope and reach, and its influence and effect, and on the “kinship” into which this “kingship” is integrated. Polity, society, and family are collapsed into one—structure of power centered on gender. Similarly, the elite, merchants, and laborers are brought into the very structure of power. Expressions of power, including political authority, are underpinned by gender relations, which are in turn guided by either love or lust. The Nights, like the contemporary reimaginings of Hārūn al-Rashīd, playfully but judiciously selects material from adab and reconfigures this material for its own purposes, imitating and mocking adab all in one breath.

2

The thousand and one nights and adab

The Nights’ relation to adab is known. Even though it is not possible (as yet) to map in detail the influence of adab on the Nights, or vice versa, we do know that upon the translation of the Persian Hazār afsān into Arabic, eloquent men of letters reworked them, refining and embellishing them, and produced similar types of storybooks: “tanāwalahu l-fuṣaḥāʾ wa-l-bulaghāʾ, fahadhdhabūhu wa-namaqqūhu wa-ṣannafū fī maʿnāh mā yushbihuhu.”6 As early as the fourth/tenth century, Ibn al-Nadīm already mentioned the ways in which the author of Kitāb al-Wuzarāʾ, Abū ʿAbdallāh Muḥammad b. ʿAbdūs al-Jahshiyārī, who was a government bureaucrat educated in adab, compiled a collection of Arabic, Byzantine, and Persian stories (asmār al-ʿarab wa-l-ʿajam wa-lrūm) from oral and written sources, from storytellers and books, and organized them into “nights” as well but kept each “night” a separate, independent, full story. Ibn al-Nadīm saw this book in, give or take, 50 folios, which contain the 480 nights or stories al-Jahshiyārī collected before his death.7 Perhaps it is beside the point to map this mutual influence by tracing the movement of stories to and from the Nights and adab, for adab itself defies any content-driven definition. Rather, it might be more productive to pursue a line of inquiry delineated by Ibn Khaldūn. Adab is one of the four sciences of the 6 Ibn al-Nadīm, al-Fihrist 363. 7 Ibn al-Nadīm, al-Fihrist 363–364.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

345

Arabic language (ʿulūm al-lisān al-ʿarabī), preceded in importance by grammar (ʿilm al-naḥw), lexicography (ʿilm al-lugha), and style (ʿilm al-bayān). Ibn Khaldūn defines adab as an area of intellectual activity, ʿilm, not interested in any subject (mawḍūʿ); rather, its objective is the effective, or even aesthetic, use of language in poetry or prose.8 Accordingly, adab means poetry and prose distinguished not just by their form but also by their “play” with language. The Nights does “play with language,” but its play is of a different order; it transforms the “word play” of adab, as in the deployment of al-badīʿ devices, such as ṭibāq and jinās, into “play with narration,” setting genres such as “epic” and “romance” against each other dialogically,9 for example, and following the principles of doubling (izdiwāj), opposition (muqābala), paradox (aḍdād), and derivation (ishtiqāq) in the generation of motifs and their organization as tableaux in the arabesque tapestry of the stories.10 In its playful reconfiguration of the anecdotes (akhbār) about Hārūn al-Rashīd, as I will show in the following, the Nights tells not only a fantastic tale of his rule but also articulates a fantasy of an ideal community, here, as I will demonstrate, framed by an ideology of monogamy. I will focus on one of the many Nights variants so as to be able to highlight in detail how the composition of each story-text devises a subversive narrative strategy. The Hārūn al-Rashīd stories in the fifteenth-century Galland-Mahdi text11 will serve as the locus of my inquiry, through which I will bring into focus the other relevant Nights tales as well as the adab material underpinning and framing them. I will then extend our vision out to the fantasy of an ideal community inherent in the Nights’ Hārūn al-Rashīd tales. I do not look at the stories of his sexual orgies with his slave girls in his harem that appear in later nineteenth-century texts12 but at two stories that could have easily been lifted out of Kitāb al-Aghānī, which Abū l-Faraj al-Iṣfahānī (284–356/897–967) compiled around the 100 “ṣawt” selected at the behest of and for Hārūn al-Rashīd himself.13 Even though Ibn Khaldūn does not include it among his four pillars of adab—and these are Ibn Qutayba’s Adab al-kātib, al-Mubarrad’s al-Kāmil, alJāḥiẓ’s al-Bayān wa-l-tabyīn, and Abū ʿAli al-Qālī’s Āmālī—he considers it the most comprehensive and in fact the epitome of adab work.14

8 9 10 11 12 13 14

Ibn Khaldūn, Dīwān al-mubtadaʾ wa-l-khabar i (Muqaddima), 753–764. Ouyang, Romancing the epic; and Ouyang, Epical turn. Ouyang, Trickster jester. Mahdi (ed.), Kitāb Alf layla wa-layla. For a list of these stories, see Hârûn al-Rashîd in Marzolph and van Leeuwen, Arabian nights ii, 585–586. Al-Iṣfahānī, Kitāb al-Aghānī i, 2, 8. Ibn Khaldūn, Dīwān al-mubtadaʾ wa-l-khabar i, 763–764.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

346

ouyang

This resonates with the stories of “martyrs of love” brought into one compilation, and genre, by al-Sarrāj (417 or 419–500/1026 or 1028–1106) under the title Maṣāriʿ al-ʿushshāq. “Anīs al-Jalīs and Nūr al-Dīn b. Khāqān” smacks of a tale alSarrāj relates about Hārūn al-Rashīd and the slave girl (unnamed) of a drummer (zalzal) by the name of Manṣūr who refused to be sold even to the caliph after her master’s death. In the end, Hārūn al-Rashīd bought her, freed her, and gave her an income as a reward for her devotion to her original master.15 The pairing of Anīs al-Jalīs with Nūr al-Dīn, like that of Nūr al-Dīn b. Bakkār with Shams al-Nahār in another Hārūn al-Rashīd tale, echoes famous love-mad couples epitomized by Majnūn Laylā: Qays and Laylā, Qays and Lubnā and Jamīl and Buthayna, to name but three, who choose to devote themselves exclusively to each other against the lifestyle led by Hārūn al-Rashīd and his slave girls. In Arabic, the phrase “Hārūn al-Rashīd wa-jawārīhi” in today’s use conjures up the world evoked by Kitāb al-Aghānī and at the same time pokes fun and frowns upon the excess associated with his lifestyle and of course polygamy, and more importantly, it denounces sexual excess as an expression of male power, and of tyranny. Analyses of the Nights’ subversive discourse tend to focus on the frame tale, and on how Shahrazād forestalls death and heals Shāhriyār and puts a stop to his killing spree. Shahrazād is often related to the figure of the “erudite concubine” in classical Arabic writings and enframed Nights’ tales who, benefiting from the wisdom she derives from her extensive learning, successfully restores Shāhriyār back to sanity and justice from madness and despotism. The gendered discourse in the frame tale is carried over into the enframed tales, and the Nights is often hailed as the Arabic female tradition that gives voice to women and subverts the Arabic male tradition in which Shahrazād is versed. The Nights, as such is a satire of patriarchy and male authority in Arabic-Islamic culture. I need not rehearse this interpretation. Rather, I push this discussion further and show the ways in which the Nights not only subverts patriarchy but also constructs an ideal community in the interstices of the dialogue between the Nights and adab, in this instance, in the “play” with the stories of “martyrs of love,” bringing their inherent “ideology of monogamy” into our reading of “harem politics,” here, overlapping with “court intrigues,” and into our reading of adab in general, such as Kitāb al-Aghānī and Maṣāriʿ al-ʿushshāq.

15

Al-Sarrāj, Maṣāriʿ al-ʿushshāq i, 34–35.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

3

347

Hārūn al-Rashīd in the Nights

Hārūn al-Rashīd’s power irradiates from the harem but reaches times immemorial and the farthest corners of the universe. “The porter and the three ladies of Baghdad,” “ʿAlī b. Bakkār and Shams al-Nahār,” and “Anīs al-Jalīs and Nūr alDīn” begin in the harem, respectively, of the three ladies and his own, then moves out to his court toward the end of “The three ladies,” and in “The three apples,” which enframes “The two viziers.” These Hārūn al-Rashīd tales are sandwiched between “The merchant and the genie” and “The fisherman and the genie,” two tales inspired by the biblical Solomon legends, and the mythical “Jullanār,” which ends the Galland-Mahdi text, and “The hunchback,” set in the reign of al-Muntaṣir (r. 247–248/861–862), inserted between “The three apples” and “Nūr al-Dīn and Shams al-Nahār.” These stories cover a vast geographical area—with its center in Baghdad—from Cairo to Damascus, Basra, and Kashgar in China in the Baghdad cycles, and to Persia, Iran, Samarkand, India, and Indochina in the frame tale. His mobile cabinet is an entourage of three: himself, his vizier Jaʿfar (alBarmakī), and his bodyguard (al-sayyāf ) Masrūr. Together, they disguise themselves as merchants and wander the streets of Baghdad at night. The purpose of their nocturnal journeys is not clear. Hārūn al-Rashīd is possibly touring his capital to make sure its inhabitants are doing well, or seeking distraction from the pressures of court. The stories of injustice that unfold before Hārūn alRashīd are variations on the same theme: unrequited or betrayed love leads to tragedy or crime. He then intervenes to right wrongs, punishing the wrongdoers and rewarding the do-gooders, and bestow happiness on all. Justice follows a simple formula. In “The porter and the three ladies of Baghdad,” upon hearing the three qalandars’ and ladies’ love-gone-awry stories, he marries them to each other and takes as his prize the lady of the house. In “The three apples” he does not solve the murder. Rather, he orders Jaʿfar to do so and threatens to hang him within three days if the culprits are not found. The details of the murder are revealed from the perspective of hindsight. This crime-of-passion story comprises a series of coincidences. It begins with a merchant buying three apples from the orchards of Hārūn al-Rashīd in Basra for his ill wife. Seeing one of the apples in the hands of a black slave in the street, and that one of the three apples is missing from the three he bought for his wife, he gets jealous, kills his wife, and throws her body into the Tigris. The resolution unfolds through a similar series of coincidences. Jaʿfar sees a Basra apple in the hand of his young daughter, inquires about it, and finds out that one of his black slaves snatched the apple from the young son of the murderer, who in turn stole it from the three apples his father bought for his mother. In response to Hārūn al-Rashīd’s

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

348

ouyang

amazement, Jaʿfar tells him the story of “The two viziers, Nūr al-Dīn ʿAlī al-Miṣrī and Badr al-Dīn al-Baṣrī,” where the enduring love between the daughter and son of the estranged brothers, the two viziers, guides the family reunion that takes place at the end of the story. The key ingredient and the mover-and-shaker of the plot, just like the apple in “The three apples,” is a pomegranate dish. This story “is the wonder of wonders,”16 Hārūn al-Rashīd exclaims upon hearing it, and in a moment of ecstasy, he “freed the slave and gave the young man one of his choice concubines, settled him on a sufficient income, and made him one of his companions to the end of his days.”17 The Nights presents a caricature of his rule. He is a benevolent tyrant. He gives orders and threatens his officers with death if they do not deliver, be it the criminal or truth. At the same time, like ʿUmar b. al-Khaṭṭāb, he lives among his people, not in an ivory tower. At night, after a day at court, he often forsakes the pleasures of his harem and goes out into the street to inspect the affairs of his dominion in person, mingles with those he rules, and listens to what they have to say before he dispenses justice. Justice is primarily righting wrongs and compensating for them. Sufficient income, his company, and happily ever after are an added bonus. The stories told about and in the harem are brought to light and linked to other stories told at court. The three qalandars tell their stories in the harem at night in “The porter and three ladies of Baghdad,” but the three ladies tell theirs at court the next morning. Hārūn al-Rashīd holds court more like a patron of an adab séance in Kitāb al-Aghānī than a commander of the faithful leading a prayer, going on pilgrimage, conducting a military campaign, managing the complex political and financial relations with his allies and foes, and giving audience at court. Even the texture of the story resembles Kitāb al-Aghānī’s prose, even as it does away with the anecdotal structure of adab (organized around independent units of an anecdote framed by a chain of transmission) for the sake of narrative coherence. The events in the love story of a merchant’s son, ʿAlī b. Bakkār, and one of Hārūn’s favorite concubines, Shams al-Nahār, are structured around love poems, or songs, uttered and at times exchanged between the two lovers. There are no two identical love stories in the Nights, even though they are derived from the very same paradox of love, loyalty, and betrayal, from which further similar but different stories are generated and juxtaposed. These love stories are additionally structured around the concept of friendship-companionship-brotherhood (ṣadāqa-ṣaḥāba-ukhuwwa) central to Arabic-Islamic

16 17

Haddawy (trans.), Arabian nights 206. Haddawy (trans.), Arabian nights 206.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

349

imaginings of political community from Ibn al-Muqaffaʿ to Ibn Ḥazm and alTawḥīdī, to name but three who have written on the subject. Friendship, just like erotic love, is a part of love, and both are at the heart of family and community,18 with husband and wife and brotherhood serving as the kernels of the family and the intimate connection between family and community— brotherhood encompassing both friendship and companionship, the two pillars of any harmonious community. The ideal man-woman love overlaps with the ideal man-man or woman-woman friendship, which often slips into brotherhood and sisterhood, and both serve as the exemplars of human relations at the core of an ideal community.

4

Adab in The thousand and one nights

We already catch a glimpse of how friendship works as the communal glue in the stories of the three ladies of Baghdad. Their world and life fall apart thanks to the betrayal of their flesh-and-blood sisters. The dysfunctional sisterhood is matched by the improper erotic love in the stories of the three qalandars. In these stories, the key ingredients of community as prescribed by al-Tawḥīdī, “al-ṣadāqa wa-l-ʿishra wa-l-muʾākhāt wa-l-ulfa,” are missing: “al-riʿāya wa-l-ḥifāẓ wa-l-wafāʾ wa-l-musāʿada wa-l-naṣīḥa wa-l-badhl wa-l-muʾāsāt wa-l-jūd wa-ltakarrum.”19 The required exemplification of friendship (ṣadāqa) and brotherhood (muʾākhāt), living together (ʿishra) and intimacy (ulfa) through care (riʿāya), preservation (ḥifāẓ), support (musāʿada), counsel (naṣīḥa), spending (badhl), consolation (muʾāsāt), magnanimity ( jūd) and generosity (takarrum), are disrupted. These stories present the consequences of this disruption. We see how the overlap between love and friendship plays out in “ʿAlī b. Bakkār and Shams al-Nahār” and “Anīs al-Jalīs and Nūr al-Dīn,” where the outside world intrudes on his private home. These two stories demonstrate how community coheres around love and loyalty. ʿAlī b. Bakkār has a love affair, albeit unrequired, with one of Hārūn alRashīd’s favorite concubines, Shams al-Nahār, right under his nose, in the harem of his new palace. The descendant of a Persian royal family meets Shams al-Nahār in the shop of a perfumer (al-ʿAṭṭār) and they, with the help of the perfume merchant, and later a jeweler, together with his friends, her maids, even robbers, police officers, and palace guards, manage to meet in person twice:

18 19

See al-Shaar, Ethics; and Ouyang, Ṣadāqat al-rijāl wa-l-ukhuwwa. Al-Tawḥīdī, al-Ṣadaqa wa-l-ṣadīq 1–2.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

350

ouyang

once in her private chambers in the harem, and another time in the home of the jeweler. But they are interrupted on both occasions, by the arrival of Hārūn alRashīd and the intrusion of a band of robbers respectively. Fear hovers in the air and everyone helps keep the secret of their love affair, here conducted through the exchange of news, messages, and poems delivered by their go-betweens, and they escape being caught when they meet. This story reads like a modern thriller written in high classical Arabic literary style. The two protagonists, with the aid of their sidekicks, hide in chests, concealed alcoves, balconies, and invisible street corners; they run away along darkened garden paths and alleyways, escape on boats, and meet in strange places, all the while looking out for palace guards and eunuchs, malicious spies, and whistleblowers. Such suspenseful escapades are punctuated by lush descriptions of: the beauty of the protagonists; the rooms in which they do meet, the curtains, the rugs, the seats and cushions; the feasts, and what and how food and wine are served; the people in waiting, their appearance, dress and posture; and the song and dance performed, and the poems that mark the highs and lows of the story. The protagonists die, and on the same day, too, as one would find in the stories of Maṣāriʿ al-ʿushshāq, martyrs of their unrequited love. Their friends even manage to bury them next to each other. Hārūn al-Rashīd does hear murmurs of the affair from a maid who is angry because of a beating she received on the orders of Shams al-Nahār, but he chooses not to investigate. He takes an active role in “The story of Anīs al-Jalīs and Nūr al-Dīn b. Khaqān” when the two lovers squat in his palace when they flee from their hometown, Basra, to Baghdad. They even make themselves at home and stage song and dance parties. When Hārūn al-Rashīd hears noise from afar and goes to investigate, disguised as a fisherman, he finds out about the injustice behind the lovers’ exile. Nūr al-Dīn is the son of the good vizier, Fadl al-Dīn b. Khāqān, who at the request of his good king, Muḥammad b. Sulaymān al-Zaynabī, purchases a slave girl, Anīs al-Jalīs. Despite his warnings, Anīs al-Jalīs meets Nūr al-Dīn, Fadl al-Dīn’s philandering son, and they fall in love. In order to keep himself and his family from the evil vizier, al-Muʿīn b. Sāwī, Fadl al-Dīn orders Nūr al-Dīn to marry Anīs al-Jalīs, and more importantly never to take another wife or concubine or resell Anīs al-Jalīs. He soon dies. Nūr al-Dīn squanders all his inheritance. His friends turn away from him, even those who received gifts and help from him. He contemplates selling Anīs alJalīs at the slave market and has an altercation with al-Muʿīn b. Sāwī, whereby the latter figures out the secret of Anīs al-Jalīs. When al-Muʿīn b. Sāwī comes after him, Nūr al-Dīn runs away, taking with him Anīs al-Jalīs, with whom he cannot be parted. Hearing this, Hārūn al-Rashīd sends Nūr al-Dīn home with a letter to the king of Basra for him to remove al-Muʿīn b. Sāwī as vizier and

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

351

appoint Nūr al-Dīn in his stead. He even sends Jaʿfar to make sure all is in order. Indeed, al-Muʿīn b. Sāwī plots to have Nūr al-Dīn killed, but Jaʿfar arrives in time to save the day. He takes Nūr al-Dīn back to Baghdad. Hārūn al-Rashīd frees Anīs al-Jalīs and gifts her to him, appointing them both his boon companions. Hārūn al-Rashīd is a benevolent tyrant at his best, who brings to life the Nights’ description of the king of Basra: “bountiful as the sea, so that even the proud were glad to be his servants and even the days and nights awaited his command, for he was a man who rejoiced in sharing his wealth with those who served”;20 and of Shāhriyār, “a towering knight and a daring champion, invincible, energetic, and implacable. His power reached the remotest corners of the land and its people, so that the country was loyal to him, and his subjects obeyed him.”21 Unlike Shāhriyār, however, Hārūn al-Rashīd and the king of Basra overlook or forgive certain trespasses. We have already witnessed the magnanimity of the former. Al-Zaynabī, too, overlooks the Khāqān family’s violation of his trust: they absconded with 1,000 gold dīnār set aside as the price for Anīs al-Jalīs, together with the slave girl herself. However, their power and grace are not unconditional. Their effect and reach are premised on fulfilling the conditions set up in the “ideology of monogamy” expressed through the juxtaposition between two types of love: exclusive devotion of the lovers to each other or vice versa. Nūr al-Dīn b. Khāqān’s only virtue is his devotion to Anīs al-Jalīs. Here, “monogamy” is an episteme, and is coeval with the loyalty and solidarity associated with brotherhood and friendship in the Nights and adab, and its opposite, the unrestrained pursuit of sex, with the total disregard for the code of conduct inherent in the ideology of monogamy that must underpin the imagined ideal community. The code of conduct matters more than the king, for in the Nights, the king loses his kingdom when he or his queen violates this code. In the frame tale, Shāhriyār’s queen breaks the code and puts the kingdom at risk. In “The fisherman and the genie,” the frame tale, about a Hārūn-al-Rashīd-like king, who goes out to inspect an out of ordinary occurrence in his kingdom, then rights the wrong, rewarding the good and punishing the evil, and the enframed tale, of a king whose queen behaves like Shāhriyār’s queen and loses his kingdom, offer us the consequences of following or violating this code of conduct. Similarly, “Jullanār,” the last story in the Galland-Mahdi text, is the story of educating desire into love. The childless king Shahrimān gives up all his concubines and

20 21

Haddawy (trans.), Arabian nights 344. Haddawy (trans.), Arabian nights 3.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

352

ouyang

marries Jullānār. A beautiful son is soon born. Their son, Badr Jāsim, however, must go through a similar experience of learning how to love from lessons on, let us say, lust, or the pursuit of desire for its own sake. He has to get lost, wander around the world, experience desire, and learn to love properly before he can be king. This code of conduct is applicable to the officers of the court, especially the viziers, and the members of the community, merchants and laborers alike.22 The viziers, from Shahrazād’s father in the frame tale to “The fisherman and the genie” and Jaʿfar in the Hārūn al-Rashīd stories are mirrors of their kings. The character and conduct of good viziers are accentuated in their juxtaposition with the evil viziers in “King Yūnān and the Sage Dūbān,” told within “The fisherman and the genie,” “The two viziers,” and “Anīs al-Jalīs.” The fictional viziers overlap with fictionalized historical figures, Jaʿfar b. Yaḥya al-Barmakī (d. ca. 190/803) and the two Khāqāns: al-Fatḥ b. Khāqān (d. 247/861), who served as the governor of Egypt and Syrian provinces; and Abū l-Ḥasan ʿUbayd Allāh b. Yaḥyā b. Khāqān (d. 263/877), who served as the head of the Dīwān al-Kharāj and director of the tribunal of maẓālim (grievances). These good viziers, characterized, to use al-Tawḥīdī’s criteria, by the care (riʿāya), preservation (ḥifāẓ), support (musāʿada), counsel (naṣīḥa), spending (badhl), consolation (muʾāsāt), magnanimity ( jūd), and generosity (takarrum) they show their king, family and community, are juxtaposed to evil viziers, exemplified by al-Muʿīn b. Sāwī in “Anīs al-Jalīs,” who initiate and carry out conspiracies to remove their kings or coviziers and rob their subjects of their women and worldly possessions. Total tyranny is their objective. The juxtaposition of “Nūr al-Dīn and Shams al-Nahār” and “Anīs al-Jalīs and Nūr al-Dīn,” in such close sequence, plays out the doubling (izdiwāj) and opposition (muqābala) structured around two concepts, each containing two contradictory meanings (aḍdād): love in its platonic and lustful forms; and friendship in its good and evil manifestations. Nūr al-Dīn and Shams al-Nahār, despite their infringement on Hārūn al-Rashīd’s rights—but Hārūn al-Rashīd is the icon of the magnanimous king—evade detection and capture thanks to their friends and allies. Nūr al-Dīn’s friends, the perfumer and jeweler, and Shams al-Nahār’s maid and housekeeper, serve as their go-between messengers, organize their meetings, hide them and aid their escape, and help them cover their tracks. Anīs al-Jalīs and Nūr al-Dīn b. Khāqān are not as fortunate. The former has only Nūr al-Dīn as her friend and ally, and the latter loses all his friends as soon as he goes bankrupt. His generosity toward his friends is not returned.

22

See Ouyang, Utopian fantasy.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

353

Friends turn into debtors and come after him. A friend even advises him to sell Anīs al-Jalīs, to break the promise he made to his father. Luckily, his temper gets the better of him, and at the very last minute, in the middle of the slave market, he decides against handing Anīs al-Jalīs over to al-Muʿīn b. Sāwī, and they run off to Baghdad. They are saved by their love and, of course, the good caliph.

5

“Ideology of Monogamy” and Tyranny

The good caliph, interestingly, does not have to abide by the love paradigm, not even in the Nights, but he must embody friendship. This separation between love and friendship in the conduct of the good king opens up lines of escapes from any type of total control. I now come to the paradoxical relationship between tyranny and the “ideology of monogamy.” If the “ideology of monogamy” breaks down tyranny’s lines of defense, tyranny also renders the “ideology of monogamy” vulnerable to its excess. Neither is total. This is because tyranny is conceptualized in the Nights and adab as the manifestation of unruly desire, and monogamous love as its foil. In the grey area between, let us say, love and sex, or the interstices between love and friendship, there is space for fitna. Fitna, which means distraction or diversion in both erotic and political senses, brings together “harem politics” and “court intrigues” and at the same time separates them. In “The three apples” the death resulting from the missing apple in one household causes conundrum at his court, from suspicion of infidelity to murder and robbery, or fitna, as Hārūn al-Rashīd calls the series of events in the story and in its enframed story, “The two viziers,” and in “Anīs al-Jalīs.” This fitna also diverts his attention from court, as we see in “Anīs al-Jalīs,” or his harem, as in “Nūr al-Dīn and Shams al-Nahār.” The slippages between “court intrigues” and “harem politics,” public and private, community and individual, and power and love, open up lines of escape from tyranny in both ideologies and practices of power. Even Hārūn al-Rashīd cannot exercise total tyranny. “The mock caliph,” one of the most popular Hārūn al-Rashīd stories in the Nights not included the Galland-Mahdi text, tells the story of a wealthy jeweler who gets entangled in the caliph’s harem intrigues. He falls in love with Jaʿfar’s sister, Dunyā, but Zubayda causes a rift when she calls him away from their rendezvous, and as a result Dunyā has him beaten up and banishes him from her affections. He sells all his possessions, hires a boat, and plays the caliph presiding over a feast and a party of song and dance every night until the caliph chances upon the party. Hārūn al-Rashīd comes to his rescue again, of course,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

354

ouyang

and reunites the couple. The point of the story is less about the caliph’s power and more about the cracks in the structure of his power. Contemporary interrogations of power, especially within the Arabic-Islamic context, have placed the burden of tyranny on patriarchy and its symbol, Shāhriyār and Hārūn al-Rashīd. In Hanan al-Shaykh’s “reimagining,” for example, the subversion of patriarchy is addressed to Hārūn al-Rashīd, who is now the flesh and blood incarnation of Shārayār, and the defiance is expressed through the three ladies of Baghdad refusing the caliph’s match-making and marriage proposal. Such subversion engages keenly with the relationship between politics and erotics in the Nights, but it misses the ways in which the Nights interrogates power and its structures and shows us glimpses of lines of escape. One exception is Saʿdallāh Wannūs’s (1941–1997) 1977 play, al-Malik huwa al-malik (The king is the king). Inspired by “The mock caliph,” this play shows that power resides more in how we perceive it than its actual pervasiveness. In a plot that resembles Alexandre Dumas’s Man in the iron mask (1840s) and Mark Twain’s The prince and the pauper (1881), Wannūs replaces the king and his entourage, the Hārūn al-Rashīd threesome, with a different cast, a merchant and his sidekicks, and shows us that power resides in symbols, here the crown, robe, and staff, and it is through these symbols that power is exercised. The Nights does something similar, and more, through its iconization and interrogation of the iconizations of Hārūn al-Rashīd. It tells us repeatedly but variously that no tyranny is total.

Bibliography Sources Haddawy, H. (trans.), The Arabian nights (based on the text by Muhsin Mahdi), New York 1990. Ibn Khaldūn, Dīwān al-mubtadaʾ wa-l-khabar fī tārikh al-ʿarab wa-l-barbar, ed. K. Shiḥāda, vol. 1: Muqaddima, Beirut 2001. Ibn al-Nadīm, al-Fihrist, ed. R. al-Mazandarānī, Beirut 1988. al-Iṣfahānī, Kitāb al-Aghānī, 25 vols., Beirut 1992. al-Khayr, ʿA.B.A. (dir.), Hārūn al-Rashīd (2018), written by ʿU. Juḥā, available online: https://www.youtube.com/watch?v=qI5a-4golhQ&list=PLDsEOSGNngQ0zx0mpMqN 30RF0hsnJBF5E. Mahdi, M. (ed.), Kitāb alf layla wa-layla min uṣūlihi al-ʿarabiyya l-ūlā, Leiden 1984. al-Sarrāj, Maṣāriʿ al-ʿushshāq, 2 vols., Beirut n.d. al-Shaykh, H., One thousand and one nights: A new re-imagining, London 2011. Tawfīq, A. (dir.), Hārūn al-Rashīd (1997), written by ʿAbd al-Salām Amīn, available online: - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

hārūn al-rashīd in premodern arabic literary imaginary

355

https://www.youtube.com/watch?v=IcmGKTXdbXM&list=pl9KkecclNUBRfyIfbHycH hs_YOAbhqCZz. al-Tawḥīdī, al-Ṣadaqa wa-l-ṣadīq, ed. ʿA. Mutawallī, Cairo? 1972.

Studies ʿAbd al-Ghaffār, ḥ., Hārūn al-Rashīd: al-Khalīfa al-muftarā ʿalayhi, Giza 2009. Abū Khalīl, S., Hārūn al-Rashīd: Amīr al-kulafāʾ wa-ajall mulūk al-dunyā, Beirut 1996. Amīn, A., Hārūn al-Rashīd, n.d, accessed via Hindawi.org: https://www.hindawi.org/​ books/51719608/ (last accessed August 31, 2021). Bauer, T., A culture of ambiguity: An alternative history of Islam, trans. H. Bietersfeldt and T. Tunstall, New York 2021. Canning, J., 100 great kings, queens, and rulers of the world, New York 1968. Clot, A., Harun al-Rashid and the world of The thousand and one nights, trans. J. Howe, [Publisher] 1986, (e-book accessed August 31, 2021). Klonsky, M., The fabulous ego: Absolute power in history, New York 1974. Marzolph, U., and R. van Leeuwen (eds.), The Arabian nights: An encyclopedia, 2 vols, Santa Barbara, CA 2004. al-Qaṭṭān, A., and M.Ṭ. al-Zayn, Hārūn al-Rashīd: al-Khalīfa al-maẓlūm, Alexandria 2001. Yūnus, M.ʿA.R., al-Istibdād al-sulṭawī wa-l-fasād al-jinsī fī alf layla wa-layla, Beirut 2007. Ouyang, W., Romancing the epic: ʿUmar al-Nuʿmān as narrative of empowerment, in Arabic and Middle Eastern literatures 3 (2000), 5–18. Ouyang, W., The epical turn of romance: Love in the narrative of ʿUmar al-Nuʿmān, in Oriente Moderno 19 (2002), 485–504. Ouyang, W., Utopian fantasy or dystopian nightmare: Trajectories of desire in classical Arabic and Chinese fiction, in A. Chraïbi, F. Bauden, and A. Ghersetti (eds.), Le repertoire narrative arabe medieval: Transmission et ouverture, Geneva 2008, 323–351. Ouyang, W., Ṣadāqat al-rijāl wa-l-ukhuwwa fī l-thaqātayn al-ṣīniyya wa-l-ʿarabiyya, in Alif: A journal of comparative poetics 36 (2015), 145–172. Ouyang, W., Trickster jester: On humour, word play and laughter in the Arabian nights, in R. Bendix and D. Noyes (eds.), Festschrift zum 65. Geburtstag von Ulrich Marzolph, i, Dortmund 2018, 32–58. al-Shaar, N., Ethics in Islam: Friendship in the political thought of al-Tawḥīdī and his contemporaries, New York 2014.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 19

The Representation of the Barmakids in Bodleian Manuscript Ouseley 217 and Other Monographs Arezou Azad and Pejman Firoozbakhsh

1

Introduction

In his 1990 article, ‘The Barmakid revolution in Islamic government’, Hugh Kennedy discussed the Barmakid legacy in Abbasid administration. His work on the Barmakids and related subjects made important contributions to the now-commonplace nuance to the “Islamic rupture” narrative with studies of, and appreciation for, the hybridity and coexistence of pre-Islamic and Islamic practice in the first centuries of caliphal rule. When studying the Barmakids, Kennedy and other scholars of Islamic history have focused crucially on medieval Arabic source texts, leaving Persian and later Arabic monographs largely untouched. In this chapter, we hope to complement these earlier studies by introducing some of the fascinating material that comes out of Persian monographs, about the Barmakids in particular, that were written from the time of the Samanids to the Muẓaffarids of Iran. Our chapter will show the original and important contribution that the Persian material makes to our understanding of the narratives on the Barmakids across traditions and periods. 1.1 Organization, Approach, and Main Points of the Chapter At the core of our study lies Oxford’s Bodleian Library manuscript Ouseley 217, consisting of 116 folios (232 recto and verso sides), known as Akhbār-i Barmakiyān.* We use it as a reference point from which we undertake a diachronic study of the representation of the Barmakids in Persian and Arabic texts pro-

* Arezou Azad wishes to thank the Bodleian Library’s Bahari Fellowship, which enabled her to study this manuscript in depth from September 2019 to February 2020. Both authors thank the Arts and Humanities Research Council and European Research Council (grant agreement no. 851607) for funding the research that resulted in this chapter. An annotated translation of the full manuscript of Ouseley 217 is also currently in process. Dr Azad also thanks the participants at the School of Abbasid Studies and the Bodleian Art of the Persian Book Conference, both in July 2021, for their insightful comments that informed the contents of this

© Arezou Azad and Pejman Firoozbakhsh, 2022 | doi:10.1163/9789004525245_020 -

978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

357

duced between the tenth and eighteenth centuries ce throughout the Islamicate world in cities far apart from one another, including Baghdad, Cairo, Istanbul, Shiraz and Delhi. Our approach is fundamentally historiographical, that is, rather than using texts as factual sources for history, we analyze the way in which they represented history, in this case, of the Barmakids, across time and space. Specifically, we look at the transmission history of texts, as well as editorial amendments, omissions, and additions to narratives, and try to understand the reasons for these authorial interventions. We trace how the early authors depicted the Barmakids primarily as historical agents whose powers were the highest and lowest during the time of the Abbasid court of Hārūn al-Rashīd (r. 170–193/786– 809). We then mark out how the narrative emphasis tilts in later centuries (and until this day) toward portraying the Barmakids as embodiments, or apotheoses, of the principles of good rulership, social justice, and ethical behavior. In this light, the shadow of the Barmakids extends well beyond the relatively brief period of the Abbasid “golden age” in the eighth and ninth centuries ce. 1.2 Background: Who Were the Barmakids? No article on the Barmakids would be complete without an overview of the key dramatis personae who arguably, according to Islamic historiography, belonged to the best-known family of the newly conquered lands, having risen the ranks up to the highest echelons of early Islamic state structures. Their ancestral home was the remote but prosperous province of Balkh, in today’s northern Afghanistan. Their ancestors wielded much power in Balkh before the Muslim conquests. Their namesake, the “barmaks” (given as a personal name in the Islamic sources, although in its original use was a title coming from the Sanskrit pramukha, meaning “keeper” or administrator1) were the keepers of Balkh’s famous Naw Bahār monastery and temple which was probably commissioned by a ruler of the Kushan Empire (first–fourth century ce). Naw Bahār was an important stopping place on Buddhist pilgrimage routes, on account of its immense wealth, high levels of scholasticism, and Buddhist relics held there.2 The Naw Bahār institution was financed and maintained through the revenues chapter. Most of the writing of the chapter and its historiographical and historical elements, as well as information on Ouseley 217, was done by Arezou Azad, while Pejman Firoozbakhsh led the research on parallel manuscripts and monographs and their transmission. 1 De la Vaissière, É., De Bactres à Balkh 529–530. 2 The Buddhist temple was remarked upon by the Buddhist pilgrim and envoy Xuanzang, who had been sent on a mission by the Tang emperor and visited the most important Buddhist sites. He qualified the Naw Bahār of Balkh as one of the richest and most bejeweled he had seen on his extensive journey, which had taken him from Chang’an in China, through the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

358

azad and firoozbakhsh

obtained from the vast agricultural lands it owned, which may have covered much of the Balkh oasis extending over an area of 5,000 km2 from the foothills of the Hindukush to the south through the old city of Balkh and up to the Oxus River in the north that forms the border between Afghanistan and Uzbekistan today.3 The barmaks administered the revenues from the estate, acting almost as de facto rulers of this prosperous and flourishing oasis. The Barmakids entered the caliphal court in the last decades of Umayyad rule and, in the next generations, rose on the wave of the Abbasid revolution to political power and influence far away from their home in eastern Khurasan. The “barmak” appears in the Islamic sources as one of the local aristocrats who initially revolted in 86/705 against the Umayyad capture of Balkh by the general Qutayba b. Muslim (d. 96/715) in 90/708–709. The revolt failed, and this “barmak” was brought, together with other Khurasani hostages and at least one of his sons, to the court of Hishām (r. 105–125/724–743) in Syria. He converted to Islam as a mawlā (client), as did his son, a youth now called Khālid b. Barmak, who became a close friend of Hishām’s son Maslama. The first historical Barmakid is Khālid (90–165/709–781/2), who appears in al-Ṭabarī’s annals as an active member in the Abbasid daʿwa (mission) by 124/742 and in the Abbasid revolution which came to an end in 137/749–750. This, together with Khālid’s success as an administrator, and his family’s close relationship with the Abbasid family from the outset, led to the Barmakids’ preeminent role from the start of the new government. His son Yaḥyā b. Khālid b. Barmak (115/733 or 119/737–190/805) became one of the most powerful and influential men of his time. This was due to Yaḥyā’s upbringing in the court beside his father and, especially, to his role as the foster father and tutor of Hārūn, the future Caliph al-Rashīd. Their wives nursed each other’s sons, creating a bond practically equivalent, in Arab custom, to kinship. Yaḥyā oversaw Karakorum and Himalayan ranges, to India, Afghanistan, and Central Asia in 633ce, ca. 70 years before the settlement of the Muslims. See Beal (trans.), Si-yu-ki, i, 46; Fuchs (trans.), Huei-ch’aos Pilgerreise 426–469. 3 Ibn al-Faqīh’s ninth-century geographical account, especially in one important manuscript held in Mashhad, Iran, and published in facsimile edition by Fuat Sezgin, gives the dimensions (which are repeated in other contemporary geographies, too), but with the added information that many slaves tilled the land here. Ibn al-Faqīh, Kitāb al-Buldān 322–324; Sezgin (trans.), Collections 321–324; Azad Herzig, and Mīr Anṣārī, Faḍāʾil-i Balkh 251, 257. A Buddhist monastery managing massive estates might surprise those who are not familiar with the history of Buddhist institutions, but actually, this was also common in Tibetan Buddhist practice. The Tibetan term chos gzhis (monastic estate) refers to an estate or collection of villages that pay taxes to the monastery. On the relationship between monasteries and the state in eighth-century Tibet, see Dargyay, Sangha and state 122. On the early modern chos gzhis (monastic estate), see Travers, A compiled list, 5–6.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

359

the representation of the barmakids

Hārūn’s education and administered his princely affairs. Yaḥyā’s sons Faḍl (148– 193/765–808) and Jaʿfar (150–187/767–803) were appointed to high offices. Like their father, they were both considered generous and able administrators. But in 187/803, al-Rashīd had Jaʿfar executed, and his corpse divided and exposed in various parts of Baghdad for a year. Yaḥyā and Faḍl were incarcerated, and it was in prison that they died. The family history, as we have just described it, is well-known to Islamic historians.4 But perhaps the most remarkable thing about the Barmakids’ rise and removal from power by Caliph Hārūn al-Rashīd was that their story has been told and retold so many times, over the centuries and until this day. It is to this aspect that we now turn.

2

Textual Transmissions in Arabic and Persian of Texts on the Barmakids

2.1

The Long Story of Ouseley 217 (Akhbār-i Barmakiyān): Makers, Copiers, and Exporters from the Delhi Sultanate to the Victorians Our story starts with Ouseley 217. This manuscript has eluded modern scholarly interest since it entered the Bodleian Library in the eighteenth century ce and its catalogue in the 1850s.5 This chapter offers, therefore, the first study of the manuscript. On the first folio of Ouseley 217, the author introduces himself as the well-known chronicler Ḍiyāʾ al-Dīn Baranī (d. ca. 758/1357), and he explains that he produced this work for his patron, the Delhi Sultan Fīrūzshāh Tughluq (r. 752–790/1351–1388). The work was completed in the early years of Fīrūzshāh’s reign in the 750s/1350s. The author adds that he translated the book from Arabic to Persian (“Kitāb-i akhbār-i barmakiyān ki banda Ḍiyāʾ-i Baranī az ʿibārāt-i ʿarabī bi-pārsī tarjuma karda ast”).6 Baranī, before he was banished from the court of Fīrūzshāh, also wrote the better-known Tārīkh-i Fīrūzshāhī.7 But Ouseley 217 is not an autograph copy: it was made ca. 300 years after Baranī had produced it, and thus, for a much later royal patron. We know

4 This is a summary of accounts is found in: Bouvat, Les Barmécides; Sourdel, Le vizirat ʿabbāside; Barthold and Sourdel, al-Barāmika; Abbas, Barmakids; van Bladel, Barmakids; van Bladel, Bactrian background; de la Vaissière, De Bactres 531; Meisami, Masʿūdī on love; Sajjādī, Tārīkhi Barmakiyān 51–187. 5 Sachau and Ethé, Catalogue i, 162 (entry no. 308.) 6 Ouseley 217, fol. 1v. 7 Hardy, Baranī; Baranī, Tārīḵh.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

360

azad and firoozbakhsh

this because, although the manuscript lacks fine illuminations that one would expect from a royal commission, it has a wonderfully executed frontispiece, with striking lapis lazuli and gold paint work (see fig. 19.1), including golden wavy dividers between the 15 lines on the opening folios, while the nastaʿlīq script is penned by a well-trained scribe. There is no colophon, nor incipit, in which the scribe may have indicated the name of the specific royal personage for whom he had executed the copy. Inside the flyleaf, though, there are plenty of seals of which the Bodleian’s cataloguer Hermann Ethé deciphered one that is dated 1124/1712.8 The earliest seal is dated 1121/1709, which gives us a terminus post quem for the copy. This, together with the particular nastaʿlīq style that is typical of the seventeenth century, allows us to narrow down the dating of the copy to the latter half of the seventeenth century. This is a period that followed the Mughal “golden age” (of which Emperor Akbar, r. 1556–1605, was the prime representative). The manuscript was made in a period of Mughal downturn, and this also explains the lack of fine illuminations in it. Ouseley 217 was acquired by the diplomat and linguist of Georgian and Victorian England Sir Gore Ouseley (1770–1844), who traveled with his brother Sir William (1767–1842) to the east. It came to the Bodleian’s cataloguers and archivists along with many other items in Sir Gore’s collections, including a highly prized Shāhnāma. What could possibly have interested the Ouseleys and the Bodleian Library in Akhbār-i Barmakiyān? The manuscript had no stunning imagery, and its production postdated the eighth-century events it described by nearly a millennium. Akhbār-i Barmakiyān was also not one of the famous Persian universal histories. Was it a random coincidence that it found itself among an illustrious group of Persian manuscripts? Before answering the question, it is worth exploring why we ask it in the first place. At the root of our judgement (i.e., that Ouseley 217 was a “lesser” text) lies what has become something of a truism in the field of Islamic history, but one that is increasingly being proven false: namely, that Persian texts were secondary to Arabic ones as sources for studying early Islamic history. This modern scholarly bias is particularly pronounced toward Persian texts that are said to derive from “Arabic originals.” There was, in fact, a trend in Arabic-to-Persian translations in the tenth–twelfth centuries ce, and more often than not, these purported Arabic originals have not survived. This has led some scholars to suggest that medieval Persian authors fabricated the existence of such Arabic originals, as a way to add authenticity and credibility to their accounts.9

8 9

Sachau and Ethé, Catalogue i, 162 (entry no. 308.) See, for example, Ahmed Asif, Book of conquest 20.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

361

the representation of the barmakids

figure 19.1

Frontispiece on fol. 1v ouseley 217, bodleian library persian manuscript

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

362

azad and firoozbakhsh

Setting aside the question of whether purported or lost Arabic originals were real or not, and if we assume that Arabic source texts did form the basis, then we need to appreciate that such Persian texts were never mere derivatives, and they were neither direct nor exact translations of the Arabic. The editors of the recently published edition and translation of Faḍāʾil-i Balkh found this to be the case in the text that was written in the thirteenth century ce. The Persian text was based on a purportedly lost twelfth-century Arabic original and sometimes included new material in, and usually embellishments to, the ḥadīth-related anecdotes. This suggests to the editors that some material was either lost in the Arabic historiography and preserved in Persian, while other Persian material was new.10 Why would Arabic texts lose material? It seems highly plausible that the early Arabic-writing authors were deliberately shedding material they encountered in their sources, simply because the information did not fit within the bounds of the genre in which they were writing, notably tārīkh (chronicle), rijāl (biographies), and naṣīhat al-mulūk (mirrors for princes). Conversely, Persian texts did not (need to) follow the Arabic genre strictures and, instead, delivered a mix and compilation of excerpts from Arabic genre works. The result was a hybrid Persian text, of which the organizing principle was often a locality or region, or in this case, a family.11 The aim and audience of these Persian hybrid texts is very different to those of the Arabic works, and what might have been considered irrelevant (e.g., “human interest”) in the more ḥadīth-based or administration-based texts ended up finding a home in the Persian versions.12 If we return to the question of what might have interested European collectors and archivists in Ouseley 217, the answer crystallizes when consulting the most important and comprehensive modern study of the Barmakids. Although Lucien Bouvat’s Les Barmécides d’après les historiens arabes et persans is more than a century old, no other single scholar has captured the full extent of the known sources on the Barmakids. In what is essentially a bio-bibliographical

10 11

12

Azad, Herzig, and Mīr Anṣārī, Faḍāʾil-i Balkh 10–17. Storey’s bibliography of Persian texts and Bregel’s continuation list local histories that were written between the eleventh and the nineteenth centuries ce for Qum, Isfahan, Nāʿīn, Kāshān, Yazd, Fārs, Shabānkāra, Khurasan, Herat, Kirmān, the Caspian provinces (Ṭabaristān, Rūyān, Ṭabaristān-Rūyān-Māzandarān, Gīlān, Gīlān-Daylamistān), Sīstān, Khūzistān, the Bakhtiyārīs, Azerbaijan, Bukhārā, Badakhshān, Nīshāpūr, Khīwa, Marw, Samarqand, Ferghāna, and Kashgar. Storey, Persian literature i, pt. 1, 348–393; Storey and Bregel, Persidskaia literature ii, 1,008–1,208. A number of these themes are touched upon in the literature on Persian local histories, such as those enshrined in a special issue on the topic edited by Charles Melville in Iranian studies; see also in Daniel, Historiography 331–348.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

363

study, Bouvat surveys all the major Arabic and Persian texts that are either entirely dedicated to the Barmakids or devote significant sections to them.13 He goes further to offer a separate bibliography on European-language sources that dealt with the Barmakids: these are romance novels, plays, novellas, and so on. It seems there was an early modern European obsession with the Barmakids.14 The story of Jaʿfar and the caliph’s half-sister ʿAbbāsa—which made its rounds in Islamic historiography, and which may have come to Europe via links with the Ottoman, Safavid, and Mughal empires—seems to have captured the imagination of the post-Enlightenment European elite in particular:15 versions and spin-offs of the story were circulating in French, Spanish, and German. The Barmakids were powerful, rich, glamorous, and clever and became embroiled in intrigue and illicit love affairs. What was there not to love about them? Moreover, the popularity of the Barmakids came on the heels of the immense draw in Europe of the Arabian nights (in which the Barmakids feature as well, of course) after the stories were translated into French by Antoine Galland in 1730. The wider European fascination with “the Orient” in the arts, which had already found its expression in Shakespeare’s plays, operas of Puccini and Mozart, paintings, and more, was also part of this backdrop.16 With this cultural context in mind, we can understand the Ouseleys’ interest in a Persian history of the Barmakids. Almost tangentially, we find a growing interest in postclassical Arabic literature in the stories about the Barmakids. The most extensive and popular seems to have been Iʿlām al-nās bi-mā waqaʾa li-ʿl-barāmika maʾa bani ʾl-Abbās (Informing people concerning what happened to the Barmakids together with the Abbasids) written by the virtually unknown al-Itlīdī (d. ca. 1100/1689).17 It is possible that by the time al-Itlīdī was writing, he was using cross-translated works from Persian Mughal manuscripts, and these were then recycled into the Arabian nights (and vice versa).18

13 14 15 16

17

18

Bouvat, Les Barmécides 3–43. Bouvat, Les Barmécides 127–131. Bouvat, Barmécides, 20–21; Sadan, Death of a princess; Hámori, ʿAbbāsa bt. al-Mahdī; Hámori, Going down in style. For a fine study of the use of the Persian Turandot story by the opera composer Giacomo Puccini (1858–1924), see Mogtader and Schoeler, Turandot; and the review of the book by Paul. W.A. Mozart’s opera, Die Entführung aus dem Serail, was based loosely on Turkish music and inspired by contemporary interest in the “exotic” culture of the Ottomans. We thank Julia Bray for pointing us to multiple editions produced since the mid-nineteenth century under various titles that seem to emanate from this text by al-Itlidī. Personal email communication, November 2020. Marzolph, Leeuwen, and Wassouf credit Iʿlām al-nās for stories that made it into later ver-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

364

azad and firoozbakhsh

An unpublished postclassical Arabic collection of anecdotes on the Barmakids is Aḥsan al-masālik fi akhbār al-barāmik (The best paths for [knowing] the history of the Barmakids) compiled by Yūsuf b. Muḥammad al-Mīlawī, known as Ibn al-Wakīl (fl. 1131/1719).19 In his preface, he (wrongly) claims that no one before him had attempted to compile the notices or anecdotes relating to the Barmakids into one book.20 He attributes his sources mostly to wellknown authors, the latest of whom was al-Suyūṭī (d. 911/1505).21 It has survived in two manuscripts: Or. 4642 at the British Museum, copied before 1196/1782; and Arabe 2107 at the Bibliothèque nationale de France, copied in 1119/1707– 1708 or 1219/1804–1805.22 2.2 The Stemma of Baranī’s Book on the Barmakids Ouseley 217 is not the only Persian manuscript on the Barmakids. Bouvat, though not aware of Ouseley 217, knew about a lithograph of Baranī’s text that had been published in Bombay in the nineteenth century.23 The Iranian scholar Garakānī mentioned the same lithograph in his introduction to an edition of another work.24 And more recently, Sajjādī used the lithograph, together with two other manuscripts, for his edition of Baranī’s work.25 Our research enabled us to extend the list to seven manuscripts (in addition to the lithograph of Baranī’s text). The seven manuscripts have considerable differences, such as varying numbers of anecdotes, dissimilar prefaces, and divergent phrasings. Based on these differences, the manuscripts can be grouped into two versions. Version 1 is a text with around 69 anecdotes, and version 2 includes at least ten more stories than version 1 (see stemmatic diagram and table in appendix 19.2). Ouseley 217 belongs to version 1, together with two more manuscript: Ms. no. 1961 of the Library of India Office, copied in 1097/1686,26 and Ms. 3072/67 in Lahore’s Shir-

19 20 21 22 23 24 25 26

sions of the Arabian Nights, totally leaving out the directly Barmakid stories. Marzolph, Leeuwen, and Wassouf, Arabian nights 606. For further information about him, see Rosenthal, From Arabic books and manuscripts 452–454. Or. 4642, fol. 1v–2r. Rieu, Supplement 828–829. Misread as 1019 ah perhaps first by Baron de Slane (Catalogue des manuscrits arabes 374) and elsewhere since then. For example, in Bouvat, Les Barmécides 15, n. 2. Bouvat, Les Barmecides 10, n. 1. Garakānī, Tārīkh-i Barāmika 4–5, 14–19, 43–48, 80–84, etc. (introduction). Sajjādī, Tārīkh-i Barmakiyān 31–41. Ethé, Catalogue, 223–224.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

365

ani Collection at Punjab University, copied in 1266/1849–1850.27 The latter was consulted by Sajjādī for his edition of the history of the Barmakids. The remaining four manuscripts belong to version 2: 1) Ms. no. 282 of the National Library of Russia in St. Petersburg, which is attributed to the fifteenth century, making it the oldest of the seven.28 It was consulted by Sajjādī for his edition; 2) Or. 151 of the British Museum, copied in the seventeenth century; and 3) Ms. 14244/1 held at Mortazavi Library in Mashhad, copied in the eighteenth century. Neither of the last two manuscripts has been used in an edition. 4) We recently discovered a fourth manuscript, which distinguishes itself significantly for including finely executed illuminated paintings showing the Barmakids in scenes known from Baranī’s book (see fig. 19.2). This manuscript, now dispersed, was a real show piece that seems to have been commissioned by the Mughal Emperor Akbar (r. 1556–1605). We were able to identify it as resembling other manuscripts in version 2 based on the few lines of text that have survived on these painted sheets.29 Finally, also in version 2 is a lithograph, entitled Ikrām al-nās fī tārīkh-i āli barmak dar ahd-i banī ʿabbās, which was published in Bombay by Mīrzā Muḥammad Malik al-Kuttāb in 1889. The editor did not state which manuscript(s) he used for the lithograph copy. 2.3 Other Persian Monographs on the Barmakids that Have Survived Besides Baranī’s work, there are four Persian monographs on the Barmakids that have survived. Where it is mentioned, each monograph calls itself a translation from an Arabic “original,” and some specifically state that their translation is new, and/or the “definitive” or “superior” translation. Like Baranī in

27 28 29

Sajjādī, Tārīkh-i Barmakiyān 41. Sajjādī, Tārīkh-i Barmakiyān 41. Illustrated leaves from this manuscript were sold by Sotheby’s London rooms on July 1, 1969, lots 83–98. Two others were in the Warren Hastings Album (subsequently Phillipps ms. 14170) and sold on November 26, 1968, lots 376 and 377. Two illustrated leaves were sold in Sotheby’s New York rooms on April 15–16, 1985, lot 445, and March 21–22, 1990, lot 8, the latter formerly in the collection of Ed. Binney, 3rd. Another was sold at Christie’s London, October 14, 2003, lot 146. Some sold for £ 15,000 and £26,000. Owned and catalogued by: San Diego Museum of Art’s Edwin Binney Collection; Harvard University Library; and the Aga Khan Museum (formerly Prince Sadruddin Aga Khan collection) in Toronto (akm 126, 127). Publications in Falk, Smart, and Skelton, Indian painting, no. 7; Auer, Pen 169; Aga Khan Trust for Culture, Architecture 302; Welch and Welch, Arts of the Islamic book, no. 53, 155–157; and Canby, Princes, poets and paladins, nos. 87–88, 119–121.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

366

figure 19.2

azad and firoozbakhsh

Ḍiyāʾ al-Dīn Baranī, Akhbār-i Barmakiyān aga khan museum manuscript folio akm 126

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

367

Ouseley 217, who states that he had other translations available to him but had decided to make his own (presumably based on one or more Arabic texts he had on hand), and that he would leave it to his readers “to judge whose is best,”30 the authors of the other four surviving monographs wrote their own translations, too. Only one of the surviving monographs has received some western scholarly attention, and it goes back to the nineteenth century. It is ʿAbd al-Jalīl Yazdī’s Aḥwāl-i āl-i Barmak, begun in 762/1360 and dedicated to the Muẓaffarid ruler in Shiraz, Jalāl al-Dīn Abu l-Fawāris Shāh Shujā (d. 789/1384). This text was studied by French scholar Charles Schefer. Schefer included significant chunks of it based on his readings of manuscripts 1342 and 1351 at the Bibliothèque nationale de France in his Chrestomatie persane.31 Yazdī’s work cannot be called a recension of any versions of Baranī’s translation described above; it is a different book altogether. The interval between the two translations is very short, which probably means they were executed independently of one another. Yazdī’s chapter organization and content is different to Baranī’s, and there are anecdotes and transmitters mentioned in it that do not appear in Baranī’s book, and vice versa. Yazdī’s text is written in difficult literary prose with full poems, while Baranī’s is less eloquent. Yazdī quotes the Arabic transmitter ʿUmar al-Kirmānī (on him, see below) once in his work,32 while Baranī never mentions this Kirmānī. The second and third alternative texts are Akhbār-i Barmakiyān, written by Afḍal Kirmānī, probably Afḍal al-Dīn Abū Ḥāmid Kirmānī, a well-known Seljuq historian (d. ca. 615/1217–1218),33 and an Aḥwāl-i Barmakiyān produced by an anonymous author and translator in 597/1201. Both the latter works were edited by the Iranian scholar Muḥaddith in 2011–2012.34 We have identified a fourth text, called Tārīkh-i Barāmika, written by an unknown author, which was edited by the Iranian scholar Garakānī (based on his privately owned manuscript). We were able to identify Garakānī’s lost copy, as well as a later copy of it in Ms no. 9541 held in Tehran at the Library, Museum and Documentation Centre of the Islamic Consultative Assembly. Garakānī contended that the text he was editing was a lost translation from Arabic into Persian by a certain Muḥammad b. Ḥusayn b. ʿUmar Hirawī written 30 31

32 33 34

Ouseley 217, fol. 3v. Blochet, Catalogue i, 367–368. Schefer, Chrestomatie ii, 3,013. This two-volume compilation also included the first two sections of the local history of the Barmakid home of Balkh, the Faḍāʾil-i Balkh. Schefer, Chrestomatie i, 66–103. Supplément Persan 1342, fol. 7v. Bastani Parizi and Negahban, Afḍal al-Dīn Kirmānī. Muḥaddith (ed.), Aḥwāl-i wa Akhbār-i Barmakiyān.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

368

azad and firoozbakhsh

in the Samanid era.35 The text exhibits archaisms,36 but Garakānī’s specifically Samanid claim cannot be substantiated at the current stage of research. The monograph itself contains 23 anecdotes, all of which are attributed to a certain Abū al-Qāsim b. Ghassān (on him, see below). 2.4 Persian Monographs on the Barmakids that Have Not Survived There are, based on references in later sources, three more Persian monographs on the Barmakids that predate the Delhi sultans by centuries and have not survived. Baranī mentions two that would be near-contemporary with the Arabic classical texts.37 One is the abovementioned Samanid text by Hirawī, which Garakānī claims his manuscript to be. In Baranī’s words: “Chunīn gūyad Abū Muḥammad ʿAbdullāh b. Muḥammad ki mutarjim awwal-i akhbār-i barāmika būd, baʿd az Muḥammad b. Ḥusayn b. ʿUmar Hirawī.”38 The second of Baranī’s sources is a certain Abū Muḥammad ʿAbdullāh b. Muḥammad l’bry (i.e., Lābarī?), who Baranī says was the “first translator” (mutarjim) from Arabic. Baranī also tells us he was commissioned to produce this work by Sultan Maḥmūd of Ghazna (r. 388–421/998–1030): “Wa īn ʿAbdullāh-i mutarjimī gūyad ki akhbār-i karam wa sakhāwat wa dastgīrī-yi barāmika chandān-ast ki dar dafātir nagunjad, ammā ān-chi rawāyāt-i mashhūr wa maʿrūf būda, yakī az fuḍalā-yi Baghdād bi ʿibārat-i ʿArabī jamʿ karda-ast, wa man [Lābarī] ān rā bi ḥukm-i farmān-i aʿlāʾ bi-pārsī tarjuma karda-am.”39 The third lost Persian text is mentioned as a collected work by an anonymous author who also wrote the well-known Mujmal al-tawārīkh wa al-qiṣaṣ (ca. 520/1126–1127). The anonymous author notes: “[…] akhbār-i Barāmika bisyār ast az ʿahd-i Barmak tā ākhir-i dawlat-i īshān. Wa man ān rā kitābī mufrad sākhta-am wa tartībī nahāda rūzgār-i dawlat-i īshān rā wa ānchi karda-and dar ḥaq-i mardum wa rūzgār-i miḥnat wa sabab-i ān wa ānchi bar sar-i īshān āmad.”40 2.5 “Original” Arabic Monographs on the Barmakids None of the medieval “original” Arabic texts seems to have survived, and therefore, we cannot check how closely the purported translations are to the origin35 36 37 38 39

40

Garakānī, Tārīkh-i Barāmika 255–263 (introduction). Discussed in Garakānī, Tārīkh-i Barāmika 263–270 (introduction). Sajjādī, Tārīkh-i Barmakiyān 311, 337, 422. Sajjādī, Tārīkh-i Barmakiyān 337; citing Bombay lithograph, entitled Ikrām al-nās. Sajjādī, Tārīkh-i Barmakiyān 337; citing Bombay lithograph, entitled Ikrām al-nās. It is clear from the context, that the high order ( farmān-i aʿlaʾ) was given by Maḥmūd of Ghazna. Anonymous, Mujmal al-tawārīkh wa al-qiṣaṣ 508.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

369

als. Often these texts are only mentioned in passing by the Persian authors, and without reference even to authors’ names or titles. The one known exception is the Arabic transmitter (rawī) of a narrative of significant length on the Barmakids called ʿUmar b. Azraq (var. ʿAbd al-Razzāq) al-Kirmānī (fl. ca. 185/800). ʿUmar b. Azraq al-Kirmānī was a contemporary of the Barmakids, and his surviving excerpts focus on the early family history of Barmak and Khālid. Edmund Bosworth, and Ihsan Abbas before him, identified al-Kirmānī as a rāwī who was cited by al-Ṭabarī (d. 310/923) for his accounts of the Barmakids, and who, in turn, collected the accounts of ʿAbbās b. Bazīʿ, Bashshār b. Burd, Ayyūb b. Hārūn b. Sulaymān b. ʿAlī, Bakhtishū and his son Jibrail, Jaʿfar b. Muḥammad b. Ḥakīm al-Kūfī, Ibn Shāhik al-Sindī, Muḥammad b. al-Faḍl b. Sufyān, Mūsā b. Yaḥyā, Yaʿqūb b. Isḥāq, Ibrāhīm b. al-Mahdī, Zubayr b. Bakkār, and Jaʿfar b. al-Ḥasan al-Lahbī.41 However, now, through our study of the Persian monographs, we know of a second Arabic rāwī and compiler of a book on the Barmakids. His name is (Abū) al-Qāsim Muḥammad b. Ghassān al-Ṭāʾifī. He probably lived some decades after ʿUmar b. Azraq al-Kirmānī, and his account covers the full period of the Barmakid family history, all the way up to their time in the Abbasid court, their golden age, and downfall, as well as the aftermath. In Baranī’s translation, al-Ṭāʾifī is identified as either “the original compiler” or “the Arabic author” of this book. We only know about al-Ṭāʾifī what he writes himself, namely, that he transmitted from his father who had lived during the time of the Barmakids.42 Al-Ṭāʾifī’s name seems not to appear in the Arabic sources, other than on one occasion in what looks like a fourteenth-century Arabic manuscript called Akhbār al-barāmika wa dhikru ayyāmihim wa iḥsānihim.43 This manuscript refers to the rāwī as “al-Qāsim b. Ghassān,”44 much like in Garakānī’s manuscript (instead of Abū al-Qāsim b. Ghassān, as it appears in the other texts).45 2.6 The Barmakids in Works that Are Not Monographs The best-known historians to mention the Barmakids in significant excerpts are those who wrote in the Arabic tārīkh, tadhkīra, and rijāl traditions, such as Ibn Qutayba (d. 276/889), Ibn Wāḍiḥ al-Yaʿqūbī (d. 284/897–898), al-Ṭabarī

41 42 43

44 45

Bosworth, Abū Ḥafṣ ʿUmar al-Kirmānī; ʿAbbās, Shadharāt 11–17. Ouseley 217, fol. 12r, 13v. Ms. 4828 in the Fatih collection, Süleymaniye Library, fol. 24r. We thank John Nawas for drawing our attention to the publication of this manuscript as Akhbār al-Barāmika, edited by Jalīl al-ʿAṭiyya. Ms. 4828, fol. 24v. Garakānī decided to “correct” the manuscript rendering of the name to “Abu ʾl-Qāsim,”

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

370

azad and firoozbakhsh

(d. 310/923), al-Masʿūdī (d. 346/957), al-Jahshiyārī (d. 331/943), Abū l-Faraj alIṣfahānī (d. ca. 362/973), and Ibn Khallikān (d. 681/1282).46 The Barmakids feature in a variety of genres in Persian medieval texts as well, including in mirrors for princes (naṣīḥat al-mulūk), such as, Siyar almulūk by Niẓām al-Mulk (d. 485/1092) and Naṣīḥat al-mulūk by Muḥammad al-Ghazālī (d. 505/1111). Two other genres in Persian writing where the Barmakids appear frequently is tārīkh, such as Tārīkh-i Bayhaqī (470/1077), Zayn al-akhbār of Gardīzī (written ca. 442–443/1050–1051), Rawḍatu uli -l-albāb fī maʿrifat al-tawārīkh wa l-ansāb of Banākatī (d. 730/1330), and Tajārib al-Salaf by Hindūshāh Nakhjawānī (d. 730/1330), as well as in works of poetry and ḥikāyāt, or the collections thereof, most notably Jawāmiʿ al-Ḥikāyāt by Awfī (d. after 640/1242).

3

What the Texts Tell Us about the Barmakids

Having considered the transmission routes of the accounts on the Barmakids, we now turn to the content. We limit our discussion here to version 1. Baranī’s account in Ouseley 217 (and other manuscripts in version 1) includes around 70 stories: two are about Barmak, one about Khālid, 21 about Yaḥyā, 16 about Faḍl, 20 about Jaʿfar, and around ten about the Barmakids in general. The stories are sequenced, broadly in chronological order, starting with Barmak and ending with Jaʿfar and the few surviving Barmakids who purportedly lived in poverty after the Barmakid disgrace. Baranī, in his anecdotes, mentions many political and administrative court officials known from the historical records of the Abbasid court. To mention only a few star characters in the stories: al-Faḍl b. Sahl (d. 202/ 818), al-Ḥasan b. Sahl (d. 236/850–851), al-Faḍl b. Rabīʿ (d. 207/822– 823 or 208/823–824), ʿAlī b. ʿĪsā b. Māhān (d. 196/812), and ʿAbdallāh b. Mālik, the provincial governor of Khurasan under the Abbasid Caliph al-Mansur (r. 754–775).47 Other courtiers include musicians and poets, also known from the historical record, such as, Muslim b. al-Walīd al-Anṣārī (d. 208/823), Isḥāq b. Ibrāhīm al-Mawṣilī (155–235/772–850), Yaḥyā b. Maʿīn, Abū Zakariyyāʾ (158–

46 47

but the manuscript actually uses “al-Qāsim b. Ghassān.” See Garakānī, Tārīkh-i Barāmika 2, 9, 14, 19, 26–28, 32, ff. See Bouvat, Les Barmecides 5–9 for details. For further information about these officials, see Sourdel, al-Faḍl b. Sahl b. Zadhānfarūkh; Sourdel, al-Ḥasan b. Sahl; Sourdel, al-Faḍl b. al-Rabīʿ; Sourdel, Ibn Māhān; Kennedy, Early Abbasid caliphate 80–81.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

371

233/775–847), Abū Yūsuf b. Ibrāhīm al-Anṣārī al-Kūfī (d. 182/798–799), Yaḥyā b. Muʿādh al-Rāzī (d. 258/872), and Abū ʿAmr Kulthūm b. ʿAmr al-ʿAttābī (d. ca. 220/835).48 The aim of the book is not to give a terse historical account but to build the historical characters into stories that have a narrative arc: the message being cautionary and aimed at rulers as much as subjects. The main message is that good rule must come, above all, with generosity and grace. Storytelling is, therefore, a narrative device in which fiction and nonfiction come together. Baranī explains it best when, citing a Maʿāthir Maḥmūdī by Imām Qaffāl, explaining why Sultan Maḥmūd of Ghazna (r. 388–421/998–1030) was interested in the stories of the Barmakids. Baranī writes: “Maḥmūd loved the stories of the prophets and the first saints, but in the next tier of importance to him were the stories of the generous people [emphasis added].”49 Then, Baranī inserts himself, stating that the reason why he wrote his book on the Barmakids was to warn rulers to “heed the lesson not to kill the generous ones.” The not-soveiled reference was to the Barmakids, who were, in Baranī’s worldview, the apotheosis of generosity.50 While, above all, Baranī’s is a book of advice literature, he also was set on entertaining his readers and listeners. His stories are often embellished with fantastical tales about exotic lands, kings, miracles, and travels. The narrative arc reaches its climax in the cruel acts of Caliph Hārūn al-Rashīd, who toppled his trusted Barmakid advisors—so the narrative goes—turning against them in a cowardly act of envy and spite. Another laudable trait highlighted frequently in the text is the Barmakids’ exceptional hospitality, including as hosts of some of the best parties in eighth-century Baghdad.51 But all this goodness became the reason for their downfall: Hārūn al-Rashīd and his close circle, envious of Barmakid fame, orchestrated their downfall. In an interesting twist, the postscript in the text is that Hārūn felt deep remorse for these acts, but that his pride kept him from doing the right thing, which would have been to reinstate them in their state functions.52 Some of the stories (ḥikāyāt or akhbār) in the text are well-known, such as that of the nominal marriage and its illicit consummation by the caliph’s half-

48 49 50 51 52

Kratschkowsky, Muslim b. al-Walīd; Wright, Isḥāq b. Ibrāhīm al-Mawṣilī; Leemhuis, Yaḥyā b. Maʿīn; Schacht, Abū Yūsuf; Forūzandi, Siyrī dar zindigī; Blachère, al-ʿAttābī. Ouseley 217, fol. 2v. Ouseley 217, fol. 3r–v. E.g., see Ousely 217, fols. 33r–234v. On Hārūn’s purported remorse, see Ouseley 217, fols. 25v, 27v.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

372

azad and firoozbakhsh

sister ʿAbbāsa and the Barmakid minister Jaʿfar. This story is often depicted as setting in motion the beginning of the Barmakid decline (and by extension that of the Abbasid realm).53 Other stories are less known but have parallels in further surviving sources. Baranī injects humor in some stories and embellishes well-known events with direct speech and details that could not possibly have happened in that exact way. Some are clearly fictitious but, at the same time, give a wonderful and entertaining glimpse into how courts must have functioned—if not in the Abbasid times, then surely in the Delhi Sultanate or Mughal court in which these manuscripts were copied. We get a sense of how people vied for power, and how the general populace attained access to the court in ways that surely resonated with premodern audiences, just as it did in the nineteenth century. These accounts, and an analysis of what in them is original versus later redaction, could provide interesting research material for court historians. If Baranī’s account, and those in the other monographs on the Barmakids, belong to a hybrid genre of historical fiction, what is their value to the historian? Can historians treat them as serious historical sources? We would argue that, while the stories in these texts should not be accepted wholesale, they still offer details on the administrative culture and the way premodern administrators lived that are not mere coincidences or fabrications but represent a worldview that existed either at the time of writing or at the time about which the authors were writing. And beyond this, the texts, being numerous as they are and consistently copied, are a testament to the lasting Barmakid legacy throughout the Islamicate realm.

4

Case Study: Story of the King of Ṭabaristān, the Ring, and the Fish

We will now test our hypotheses of transmission, shedding, and authorial intervention discussed above through one of Baranī’s most charming anecdotes. The main gist of the story, which is embedded in the very first ḥikāyat of Ouseley 217, is this:54 Barmak arrived in the court of the Umayyad Caliph ʿAbd al-Malik b. Marwān (r. 65–86/685–705) in Syria. In his attempt to win the caliph’s ear, Barmak woos him with exciting stories about his journey from his homeland in Balkh to Syria. One is a story about a ruler from Ṭabaristān (in today’s Māzandarān

53 54

Meisami, Masʿūdī on love; Hámori, ʿAbbāsa bt. al-Mahdī. Ouseley 217 in fols. 4r–5v.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

373

province, Iran), which the Barmak would have passed on his way. The king (malik) and Barmak were at the Caspian shores when the king removed his stone-studded ring from his finger and offered it to Barmak as a present. When Barmak refused to accept such a generous gift, the king threw his ring into the sea. This saddened Barmak, who asked the king why he had done such a thing. At this point, the king ordered one of his retinue to bring a chest. From it, the king brought out a magical fish. The king lobbed the fish into the sea while Barmak and the king’s entourage watched in amazement, for the fish re-emerged from the water, with the ruler’s ring around its head!55 The story is also mentioned by Ibn Isfandiyār (fl. early thirteenth century ce) in his Tārīkh-i Ṭabaristān, who in turn, cites a Tārīkh-i Barmakiyān as his source on the malik, whom Ibn Isfandiyār refers to by the name of Māhiya-sar (Fishhead). Ibn Isfandiyār inserts himself here, adding that he will not expand on the story because of its fictitious nature. Ibn Isfandiyār further states that there are other fantastical sources that he has read in a book by a certain Yazdādī, but that these are beyond belief, which is why he has not “translated” them in his book.56 Baranī, like Ibn Isfandiyār, acknowledges that the fish story is not believable but makes it clear that his purpose is not to convince us, his readers, of its veracity. He, rather, marvels at the Barmakid capacity to convince the caliph of the veracity of such a story. He attributes it to the Barmakids’ exceptional tact, intelligence, and mild manner. Baranī also takes from this story the advice, which he passes on to his readers, that “wise people have said that stories that are not believable should not be told in the presence of kings.”57 Thus, the message is: do not try this at home; only a Barmakid can get away with telling a powerful ruler fantastical stories without reprimand or punishment! From the story we learn three important things: first, that Ibn Isfandiyār’s source (Tārikh-i Barāmika) was written in Arabic; second, that a similar sequence of stories exists in Ibn Isfandiyār’s source and Baranī’s work. A third point is that Baranī and Ibn Isfandiyār may have used a common source on the history of the Barmakids, but while the former used the story to support a point

55 56

57

Ouseley 217, fol. 6v–6r. Ibn Isfandiyār, Tārīkh-i Ṭabaristān 85. On Ibn Isfandiyar’s sources, see Melville, Ebn Esfandiyar. Iqbāl, the editor of Tārīkh-i Ṭabaristān, also thinks the story is fabricated on account of the fact that the Ṭabaristān ruler lived at the time of the Prophet and not at the time of ʿAbd al-Malik (Ibn Isfandiyār, Tārīkh-i Ṭabaristān 85, n. 3). Niẓam al-Mulk Ṭūsī also relates part of the story in his Siyar al-Mulūk 210–216. Ouseley 217, fol. 7r.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

374

azad and firoozbakhsh

of advice, the latter had no use for it in a tārīkh genre piece, thus considering it unworthy of translation. Ibn Isfandiyār shed the material; Baranī kept and embellished it.58

5

Conclusion

We have drawn two important conclusions from our study. First, is that the Persian histories of the Barmakids are at least as important as the Arabic ones. They draw on contemporary or near-contemporary sources, just like the Arabic ones do. We have seen that the earliest Persian monographs are an earliest extant text written in the twelfth which was preceded by a lost tenth century text that was written when Abū ʿAlī Balʿamī (d. ca. 363/974) was translating al-Ṭabarī’s History. Al-Ṭabarī, in turn, was using al-Kirmānī (who is also cited by Yazdī, the author of a fourteenth-century work on the Barmakids). That there was a second rāwī called al-Ṭāʾifī is plausible, too; and this transmitter has survived mainly in the Persian sources, although we have also found him mentioned in one Arabic source so far. Thus, no history of the Barmakids is complete without the Persian texts of tārīkh, advice literature, Barmakid family histories, or collections of stories (ḥikāyāt). We hope that further scholars will follow suit to conduct more detailed studies of the contents of the Persian and later Arabic texts on the Barmakids to capture the full extent of Barmakid accounts. In this regard, we encourage scholars working in both western institutions and Persian-speaking and Arabic-speaking academic circles to collaborate and take note of each other’s findings so that a more comprehensive and balanced picture can be drawn up of the Barmakids. The second important conclusion is that the Barmakid legacy in Islamic historiography is long and deep, and extends beyond the divides of language, ethnicity, tribe, region, or a particular dynasty. Once ensconced in the caliphal court in the eighth century ce, the Barmakids were part of an imperial project of Islamization. Their appeal to authors lay in their perfection of the universally accepted qualities of generosity, good thoughts, and good deeds. Whoever judged them for being inferior on account of their origins was of a small mind. Thus, Baranī delivers this story attributed to Isḥāq b. Ibrāhīm al-Mawṣilī

58

The story is also omitted from the Akhbār-i Barmakiyān, while it is included in Aḥwāl-i Barmakiyān, both edited by Muḥaddith, Aḥwāl-i wa Akhbār-i Barmakiyān 143–144.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

375

(d. 235/850), who was “one of the most eloquent speakers and exceptional singers of his time”.59 The singer describes his encounter with ʿAbdallāh b. Mālik al-Khuzāʿī,60 a well-known senior official at the Abbasid caliphal court. Al-Khuzāʿī was drunk on evening wine and ordered, somewhat brazenly, for Isḥāq to “[p]ick up an instrument and sing a song, so that I may become even jollier than I am already!” Isḥāq was insulted by this rude manner and sang without his usual joy and passion. Al-Khuzāʿī noted to his boon companions that, “When Isḥāq sings for the Barmakids it warms the heart and is beautiful; why is he not doing the same now?” Al-Khuzāʿī laments sarcastically that, “[t]hese Barmakids were originally infidels (gabr). It was only the commander of the faithful who elevated Khālid; whereas, we Arabs have great genealogies (nasab). What do the Barmakids have that is greater or better than us?” Isḥāq was deeply perturbed by al-Khuzāʿī’s utterances, feeling distraught and irritated by them, “so much that I felt like each of my hair follicles was on fire!” Isḥāq replied: “You are my superior and should not speak in this way. You shouldn’t compare your generosity or charity with that of the Barmakids, for no one can reach the hights of their deeds. No one today—neither amongst the Arabs nor the people of ʿAjam—can reach their levels of generosity and beneficence!”61 Isḥāq’s emotional outburst reveals what must have lain at the root of the Barmakid appeal: they represented all that was good in Islam, to Arab and nonArab Muslims alike, to speakers of Persian, Arabic, and other languages. It was the Barmakids’ universality in a multi-ethnic and multilingual environment of the Islamicate world that must have accounted for the retelling of their stories over the past millennium in all parts of the Islamic world. In Mughal India, where Muslim rulers were grappling with majority Hindu and other Indian religionists, this message would have rung true. In these narratives, the Barmakids were everyone’s Muslims. They were not Arabicized, as some scholars have contended, but a democratizing and diversifying force of Islamization.

59

60

61

Isḥāq (155–235/772–850) son of Ibrāhīm al-Mawṣilī (d. 188/804), illustrious court musician of the early Abbasid period, known for his scholarly prowess and musical expertise. Fück, Isḥāq b. Ibrāhīm al-Mawṣilī; minor alterations in Wright, Isḥāq b. Ibrāhīm al-Mawṣilī; on the accounts about the Barmakids attributed to him, notably in Kitāb al-Aghānī, see Bouvat, Les Barmécides 8, 10, 52, 64–66. ʿAbdallāh b. Mālik al-Khuzāʿī belonged to the Khuzāʿa, as ancient Arabian tribe. Some Khuzāʿīs who went to Khurasan “were among the ʿAbbāsid agents who paved the way for the new dynasty.” Kister, Khuzāʿa. Ouseley 217, fols. 49r–50r.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

376

azad and firoozbakhsh

table 19.1 Stemmatic Diagram of Baranī’s Akhbār-i Barmakiyān

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Muḥammad b. Ḥusayn b. ʿUmar Hirawī Abū Muḥammad ʿAbdullāh b. Muḥammad LʾBRY/al-ʾBRY (?) Attributed to Hirawī (see above) by the editor

Unknown

Unspecified Anon. author of Mujmal altawārīkh wa al-qiṣaṣ

[Tārīkh-i Barāmika]

Unknown

Persian translator/ compiler

Name

Before 1126

tenth or eleventh century

Presumably at the time of Samanids Between 998 to 1030

Date (ce)

Unknown

None

Commissioned by Maḥmūd of Ghazni

Unknown

Dedication/ commission

Lost (reported by Baranī)

Lost (reported by Baranī)

Manuscripts (copy dates in ce)

Abū al-Qāsim b. Ms 5138, Library, Museum and Ghassān (for all Documentation Centre of Islamic anecdotes) Consultative Assembly. Copied sometime between 1617 to 1640. Ms 9541, Library, Museum and Documentation Centre of Islamic Consultative Assembly. Nineteenth-century copy. Unknown Lost (reported in Mujmal altawārīkh wa al-qiṣaṣ)

Unknown

Unknown

Transmitter

table 19.2 Table of Premodern Persian Monographs on the Barmakids (Lost and Surviving Manuscripts and Published Editions)

N/A

Garakānī, Tārīkh-i Barāmika.

N/A

N/A

Unknown

23

Unknown

Unknown

Edition (see No. of bibliography for anecdotes full reference)

the representation of the barmakids

377

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Translated by Late twelfth or Afḍal Kirmānī early thirteenth (probably Afḍal century al-Dīn Abū Ḥāmid Kirmānī, d. ca. 1218)

[Akhbār-i Barmakiyān]

Akhbār-i Ḍiyā al-Dīn Baranī Early 1350s Bar(d. ca. 1359) makiyān/ Aḥwāl-i Āl-i Barmak

1201

Anonymous

[Aḥwāl-i Barmakiyān]

Date (ce)

Persian translator/ compiler

Name

Transmitter

Manuscripts (copy dates in ce)

A few anecdotes Ms 8959, Library, Museum and by Abū al-Qāsim Documentation Centre of Islamic b. Ghassān Consultative Assembly. Copied in August 1612. ṣadr-i cālī … A few anecdotes Ms 8959, Library, Museum and nasīb wa ḥasīb- by Abū al-Qāsim Documentation Centre of Islamic i Khurāsān wa b. Ghassān Consultative Assembly. Copied in Kirmān Abū August 1612. Bakr bin cAlī Ms 4936, Malek National Library and Museum Institution. Copied in January 1895. Fīrūzshāh Abū al-Qāsim Version 1: Tughluq (r. 1351– Muḥammad [b. Ms Ouseley 217, Bodleian Library 1387) Ghassān] Ṭāyifī in Oxford. Copied in seventeenth (named as “ori- century. ginal compiler, Ms 1961 at the Library of India compiler of the Office. Copied in 1686. Arabic book”) Ms 3072/67, Shirani Collection, Punjab University Library in Lahore. Copied in 1850.

None

Dedication/ commission

table 19.2 Table of Premodern Persian Monographs on the Barmakids (Lost and Surviving Manuscripts and Published Editions) (cont.)

Partially in Sajjādī. Tārīkh.

Muḥaddith, Aḥwāl

Muḥaddith, Aḥwāl

71

73

54

Edition (see No. of bibliography for anecdotes full reference)

378 azad and firoozbakhsh

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Persian translator/ compiler

[Aḥwāl-i Āl- ʿAbd al-Jalīl i Barmak] b. Yaḥyā b. ʿAbd al-Jalīl b. Muḥammad b. ʿAbd al-Bāqī alYazdī

Name

1360

Date (ce)

Jalāl al-Dīn Abu ʿl-Fawāris Shāh Shujāʿ (d. 1384)

Dedication/ commission

Manuscripts (copy dates in ce)

Version 2: Ms 282, National Library of Russia in Saint Petersburg. Probably fifteenth-century copy. Ms Or. 151, British Museum. Copied in seventeenth century. Ms 14244/1, the Mortazavi Library in Mashhad. Eighteenth-century copy. Scattered folios from a lost Ms. Copied ca. 1595–1600. One by ʿUmar b. Supplément Persan 1342 in bnf. ʿAbd al-Razzāq Copied ca. 1400. Kirmānī Supplément Persan 1351 in bnf. Ten by Abū Copied in 1519. al-Qāsim b. Ms 522, Maktabat Shaikh al-Islām Ghassān ʿĀrif Ḥikmat. eighteenth-century copy.

Transmitter

table 19.2 Table of Premodern Persian Monographs on the Barmakids (Lost and Surviving Manuscripts and Published Editions) (cont.)

Partially in: 84 Charles Schefer. Paris 1885.

Sajjādī, Tārīkh. Only Ms 282 and lithograph copy of Malik al-Kuttāb.

Malik al-Kuttāb, 82 Ikrām al-nās (lithograph copy).

Edition (see No. of bibliography for anecdotes full reference)

the representation of the barmakids

379

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

380

azad and firoozbakhsh

Personal Note A final personal word from us authors: Hugh has been working with us on the Invisible East program in Oxford since 2019, and specifically with Arezou since she was a doctoral student. We have both benefitted immensely from Hugh’s breadth of knowledge on Abbasid history, his infectious good humor, and his always constructive comments. We can probably speak for the whole team when we say that Hugh has been a colleague and a friend. We are honored to be able to present a chapter in this wonderful collection celebrating his immensely influential work.

Bibliography Sources ʿAbbās, I., Shadharāt min kutubi mafqūda fī tārīkh, Beirut 1408/1988. Anonymous, Mujmal al-tawārīkh wa al-qiṣaṣ, ed A. Naḥwī, Tehran 1399/2020. Anonymous, Akhbār al-barāmika, ed. J. al-ʿAṭiyya, Beirut 2006. Azad, A., E. Herzig, and A. Mir Ansari, Faḍāʾil-i Balkh: Or the merits of Balkh, annotated translation with commentary and introduction of the oldest surviving history of Balkh in Afghanistan (Gibb Memorial Trust Series), London 2020. Baranī, Ḍ., Akhbār-i Barmakiyān, Bodleian Library manuscript Ouseley 217. Baranī, Ḍ., Tārīkh-i Fīrūzshāhī, ed. S.A. Khan, Calcutta 1862. Bosworth, C.E., Ḥudūd al-ʿālam. “The regions of the world”; a Persian geography, 372a.h.–982a.d., trans. V. Minorsky, London 1970 (rev. ed.). Garakānī, ʿA., Tārīkh-i Barāmika, az behtarīn munshaʾāt-i qarn-i chahārum wa panjum, bā muqaddami-yi mufaṣṣal-i tārīkhī wa adabī wa intiqādī wa ḥawāshī wa taʿlīqāt, Tehran 1313/1934. Ibn al-Faqīh, Kitāb al-Buldān, ed. M.J. de Goeje, Leiden 1967. Ibn Isfandiyār, Tārīkh-i Ṭabaristān, ed. ʿA. Iqbāl, Tehran 1320/1941–1942. Malik al-Kuttāb, M.M., Ikrām al-nās fī tārikh-i āl-i barmak dar ahd-i banī ʿabbās, Lithograph copy, Bombay 1889. Muḥaddith, M.H. (ed.), Aḥwāl-i wa Akhbār-i Barmakiyān, Du risāli az du niwīsandi-yi nāshinākhti (muḥtamilan dar qarn-i shishum), Tehran 1390/2011. Niẓām al-Mulk Ṭūsī, Siyar al-mulūk, ed. M. ʿĀbidī, Tehran 1398/2019. Sajjādī, Ṣ., Tārīkh-i Barmakiyān, 3rd ed., Tehran 1398/2019–2020.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

381

Studies Abbas, I., Barmakids, in EIr, iii, 806–809. Aga Khan Trust for Culture, Architecture in Islamic arts: Treasures of the Aga Khan Museum, Geneva 2011. Ahmed Asif, M., A Book of conquest: The Chachnama and Muslim origins in Central Asia, Cambridge, MA 2016. Auer, B., The pen, the sword, and the vizier, in Mirror of Persian kings: The origins of Perso-Islamic courts and empires in India, Cambridge 2021, 161–187. Barthold, W., and D. Sourdel, al-Barāmika, in ei2, i, 1,033–1,036. Bastani Parizi, M.E., and F. Negahban, Afḍal al-Dīn Kirmānī, in Encyclopaedia Islamica, Leiden, consulted online (last accessed September 14, 2021). Beal, S. (trans.), Si-yu-ki: Buddhist records of the western world, 2 vols., London 1906. Blachère, R., al-ʿAttābī, in ei2, i, 751. van Bladel, K., Barmakids, in ei3, consulted online (last accessed September 4, 2021). van Bladel, K., The Bactrian background of the Barmakids, in A. Akasoy, C. Burnett, and R. Yoeli-Tlalim (eds.), Islam and Tibet: Interactions along the musk routes, Farnham and Burlington 2011, 43–88. Blochet, E., Catalogue des manuscrits persans de la Biblitothèque nationale, 3 vols., Paris 1905–1928. Bosworth, C.E., Abū Ḥafṣ ʾUmar al-Kirmānī and the rise of the Barmakids, in bsoas 57 (1994), 268–282. Bouvat, L., Les Barmécides d’après les historiens arabes et persans, Paris 1912. Canby, Sh.R., Princes, poets and paladins: Islamic and Indian paintings from the collection of Prince and Princess Sadruddin Aga Khan, London 1998. Daniel, E., Historiography—iii. Early Islamic period, EIr, xii, 330–349. Dargyay, E.K., Sangha and state in imperial Tibet, in E. Steinkellner (ed.), Tibetan history and language: Studies dedicated to Uray Géza on his seventieth birthday, Wien: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien, Vienna 1991, 111–128. Ethé, H., Catalogue of Persian manuscripts in the India office library, London 1903. Falk, T., E.S. Smart, and R. Skelton, Indian painting: Mughal and Rajput and a sultanate manuscript, London 1978. Forūzandi, M., Siyrī dar zindigī wa andīshihā-yi Yaḥyā bin Muʿādh Rāzi, in Faṣlnāmi-yi dānishkadi-yi adabiyāt wa ʿulūm-i insānī (dānishgāh-i Shahr-i Kurd) 4 (1388/ 2009– 2010), 145–165. Fuchs, W. (trans.), Huei-ch’aos Pilgerreise durch Nordwest-Indien und Zentral-Asien um 726, Sitzungsberichte der Preussischen Akademie der Wissenschaften, phil.-hist. Klasse 22, Berlin 1938, 426–469. Fück, J.W., Isḥāq b. Ibrāhīm al-Mawṣilī, in ei2, iv, 110–111. Hámori, A.P., ʿAbbāsa bt. al-Mahdī, in ei3, consulted online (last accessed September 6, 2021). - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

382

azad and firoozbakhsh

Hámori, A.P., Going down in style: The pseudo-Ibn Qutayba’s story of the fall of the Barmakīs, in Princeton papers in Near Eastern studies, 3, Princeton, NJ 1994, 89–125. Hardy, P., Baranī, Żīāʾ-al-Dīn, in EIr, iii, 753–754. Kennedy, H., The Barmakid revolution in Islamic government, in Pembroke papers 1 (1990), 89–98. Kennedy, H., The early Abbasid Caliphate: A political history, London 1981. Kister, M.J., Khuzāʿa, in ei2, consulted online (last accessed April 27, 2022). Kratschkowsky, I., Muslim b. al-Walīd, in ei2, consulted online (last accessed April 27, 2022). Leemhuis, F., Yaḥyā b. Maʿīn, in ei2, consulted online (last accessed April 27, 2022). Lewin, B., al-Aṣmaʿī, in ei2, consulted online (last accessed April 27, 2022). Marzolph, U., R. van Leeuwen, and H. Wassouf, Arabian nights: An encyclopaedia, Santa Barbara, CA 2004. Meisami, J.S., Masʿūdī on love and the fall of the Barmakids, in The Journal of the Royal Asiatic society of Great Britain and Ireland 2 (1989), 252–277. Melville, C. (ed.), Iranian studies 33 (2000). Melville, C., Ebn Esfandīār, in EIr, vii, 20–23. Mogtader, Y., and G. Schoeler, Turandot. Die persische Märchenerzählung. Edition, Übersetzung, Kommentar, Wiesbaden 2017. Paul, L., Review: Youssef Mogtader und Gregor Schoeler, Turandot. Die persische Märchenerzählung. Edition, Übersetzung, Kommentar, Wiesbaden: Reichert Verlag, 2017, 134+57 S., isbn 978-3-95490-283-5, in Der Islam 98 (2021), 277–281. Rieu, Ch., Supplement to the catalogue of the Arabic manuscripts in the British Library, London 1894. Rosenthal, F., From Arabic books and manuscripts, in jaos 83 (1963), 452–457. Sachau, E., and H. Ethé, Catalogue of the Persian, Turkish, Hindūstānī and Pushtū manuscripts in the Bodleian Library, 2 vols., Oxford 1889–1900. Sadan, J., Death of a princess, in Stefan Leder (ed.), Story-telling in the framework of non-fictional Arabic literature, Wiesbaden 1998, 130–157. Schacht, J., Abū Yūsuf, in ei2, i, 164–165. Schefer, C., Chrestomathie persane à l’usage des élève de l’É cole spéciale des langues orientales vivantes, 2 vols., Paris 1883–1885. Sezgin, F., (trans.), Collections of geographical works by Ibn al-Faqīh, Ibn Faḍlān, Abū Dulaf al-Khazrajī, facs., Frankfurt am Main 1987. Slane, W.M., Catalogue des manuscrits arabes, Paris 1883–1895. Sourdel, D., al-Faḍl b. al-Rabīʿ, in ei2, ii, 730–731. Sourdel, D., al-Faḍl b. Sahl b. Zadhānfarūkh, in ei2, ii, 731–732. Sourdel, D., al-Ḥasan b. Sahl, in ei2, iii, 243–244. Sourdel, D., Ibn Māhān, ʿAlī b. ʿĪsā b. Māhān, in ei2, iii, 859. Sourdel, D., Le vizirat ʿabbāside de 749 à 936 (132 à 324 de l’hégire), 2 vols., Damascus, 1959–1960. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the representation of the barmakids

383

Storey, C.A., and Y. Bregel, Persidskaia literatura: Bio-bibliograficheskii obzor, 3 vols., Moscow 1972. Storey, C.A., Persian Literature: A bio-bibliographical survey, 5 vols., London, 1927–1952. Travers, A., A compiled list of Tibetan districts (rdzong) and government estates of the Ganden Phodrang territory (1830–1959), in Revue d’Etudes Tibétaines 56 (2020), 5– 47. de la Vaissière, É., De Bactres à Balkh, par le Nowbahar, in Journale Asiatique 298 (2010), 517–533. Welch, A., and S.C. Welch, Arts of the Islamic book: The collection of Prince Sadruddin Aga Khan, Ithaca, NY 1982. Wright, O., Isḥāq b. Ibrāhīm al-Mawṣilī, in ei3, consulted online (last accessed September 7, 2021).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 20

Eutychius of Alexandria Vindicated: Muslim Sources and Christian Arabic Historiography in the Early Islamic Empire Robert Hoyland

1

Introduction

The Melkite clergymen Eutychius (Ar. Saʿīd) son of Baṭrīq (d. 328/940) and Agapius (Ar. Maḥbūb) son of Constantine (d. ca. 333/945) composed our earliest extant Christian Arabic chronicles. The former was patriarch of Alexandria in Egypt, and the latter was bishop of Manbij in northern Syria, and both were actively working on their historical compositions in the 930s. The two works have a superficial similarity: they are both described by their authors as a “chronicle” (kitāb al-taʾrīkh),1 they are of about the same length, beginning with Creation and continue up until the time of their respective authors, and in each case about two-thirds is devoted to the pre-Islamic period. They also both follow the same basic outline, which by their day was standard fare for Christian world chronicles: the Biblical world (Israel, Mesopotamia, and Egypt), ancient Persia culminating in the conquests of Alexander the Great, the Roman Empire and the emergence of Christianity, Christian Rome and its dealings with Sasanian Persia, and the rise of the Islamic empire. In very broad terms one might say that they have a common aim, namely to present a Melkite perspective on world history in the new lingua franca of Arabic and to show that the Melkite church, as an Arabic-speaking and monotheist institution, was an integral part of the caliphate.2 In addition, they both updated their work to take account of an Islamic worldview. Most obviously, this meant narrating the emergence of Islam and

1 Eutychius, Kitāb al-Taʾrīkh/Annales i, 3, which he qualifies with the phrase “compiled on the basis of investigation and verification” (al-majmūʿ ʿalā l-taḥqīq wa-l-taṣdīq); it is also given the title Naẓm al-jawhar (String of pearls) in some manuscripts. Agapius i, 9, later adding (i, 16) that the equivalent word in Greek is chronikon. 2 Griffith, Apologetics and historiography 85–89.

© Robert Hoyland, 2022 | doi:10.1163/9789004525245_021

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

385

the evolution of Muslim rule using Muslim as well as Christian sources, but it also meant adjusting the stories of those prophets and kings that were part of a shared historical tradition. For example, both authors identify Jethro, the father-in-law of Moses, and Enoch with the Quranic prophets Shuʿayb and Idrīs, respectively.3 And both authors incorporate some details from the Muslim biography of Alexander the Great into their portrayal of his exploits.4 In this way, Agapius and Eutychius contributed to a rapprochement between Christian and Muslim accounts of world history. Their success in this respect can be gauged by the positive endorsement they both received from their younger Muslim contemporary, ʿAlī al-Masʿūdī (d. 344/956): “The best Melkite compositions that I have seen on the history of the kings and prophets, of the nations and countries, and the like, are the books of Maḥbūb b. Constantine of Manbij and of Saʿīd b. Baṭrīq.”5 Nevertheless, these two chronicles are very different when one delves into their details. Both have their own particular polemical agenda: Agapius seeks to demonstrate the superiority of the Greek translation of the Old Testament over the Hebrew and Syriac versions and to polemicize against Jewish tampering with Scripture,6 whereas Eutychius wishes to champion the Melkite understanding of the Christian faith against that of all rival versions.7 Agapius is happy to talk about the Greco-Roman pagan past, but Eutychius sticks close to the Bible and church history. They also each have access to sources not used by the other, and even where they do have sources in common, they usually make different use of them, shaping them according to their own needs and wants.8 For example, Agapius uses a universal Syriac chronicle of the Umayyad period and the historical work of the Melkite scholar Qusṭā b. Lūqā (d. ca. 300/912) to write about the classical world, and also the chronicle of the astrologer-scholar Theophilus of Edessa for Byzantine and Muslim affairs during the seventh and

3 Eutychius, Kitāb al-Taʾrīkh/Annales i, 29; and Agapius i, 120 (Jethro-Moses). Eutychius, Kitāb al-Taʾrīkh/Annales i, 11; and Agapius i, 35 (Enoch-Idris). For further discussion of this question with regard to Eutychius, see Simonsohn, Biblical narrative. 4 See below for examples and references. 5 Al-Masʿūdī, Tanbīh 154. 6 Wasserstein and Wasserstein, Legend of the Septuagint 144–152; and Conterno, Found in translation. 7 Ebeid, Melkite-Chalcedonian reading. 8 Compare, for example, the difference in their use of the Letter of Aristeas (Wasserstein and Wasserstein, Legend of the Septuagint 142–144 and 144–1452), the Romance of Alexander (Voigt, Recherches 73–75 and 80–83) and the Legend of Constantine (Stutz, Constantinus Arabicus 77–113 and 114–145).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

386

hoyland

eighth centuries.9 For his part, Eutychius expands his Biblical section with material from the Cave of treasures and his section on the clash between Christian Rome and Sasanian Iran with narratives drawn from the Persian Book of lords tradition.10 In addition, Eutychius makes more substantial use of Muslim sources than Agapius, and it is the aim of this paper to examine this particular issue. I intend this not as a prelude to an assessment of the reliability or worth of Eutychius’s chronicle for historical reconstruction, but rather as an attempt to see what his selection of sources and the way in which he deployed them can tell us about the work itself and what its author wanted to achieve. In what tradition of history writing did Eutychius seek to align his work, and what is the significance of his decision to compose it in Arabic? In short, I am interested in Eutychius’s chronicle as a cultural product and in what it can tell us about the world in which Eutychius lived and the intellectual milieu in which he operated.

2

The Date of Eutychius’s Chronicle

First, it is necessary to dispel the notion of Michel Breydy that there is an earlier, shorter Alexandrian recension of Eutychius’s chronicle and a later, longer Antiochene recension. In 1975 Breydy discovered an Arabic manuscript in the library of St. Catherine’s monastery on Mount Sinai (Sinai Ar. 582, henceforth S) that contained a text consisting of numerous discrete episodes from history recounted in a somewhat disjointed fashion, with little connection made between the episodes. Breydy observed that much of this material appeared, in expanded form, in the chronicle of Eutychius as published by Louis Cheikho in 1909 (henceforth C). There were two obvious ways to make sense of this discovery: either C was an expansion of S, or S was an abridgement of C. For various reasons Breydy opted for the first explanation.11 Indeed, he went further and argued that S was an autograph copy of Eutychius’s chronicle and that C was an expansion of it by the eleventh-century physician Yaḥyā al-Anṭākī. Thus, postulated Breydy, there were two distinct recensions: an Alexandrian one and an Antiochene one. This hypothesis became the prevailing view, and it has tended to put readers off using the text, faced as they are with a choice between a disjointed collection of source extracts or what Breydy has told them is a late reworking of it. There were a few dissenting voices, including Hugh Kennedy, 9 10 11

Hoyland, Agapius; and Hoyland, Theophilus. See below for references to Eutychius’s use of these two sources. Set out in Breydy, Etudes.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

387

who recognized the value of Eutychius’s chronicle for writing the history of the Melkite church,12 but it is the recent and brilliant study of Maria Conterno that demonstrates the inadequacies of Breydy’s theory. She shows that the second explanation fits the evidence much better (i.e., that S is an abridgement of C). As she puts it: “The Sinai manuscript is not the archetype nor a full copy of the Annals [i.e., the chronicle of Eutychius], but an informal copy made for personal use by someone who abbreviated/excerpted the text [of Eutychius], deliberately leaving out or summarising entire portions of it.”13 There is one possible alternative explanation. Conterno points out that S “is a paper codex of very small size, literally pocket size (16.5 × 14 cm), that looks more like a notebook than like a book.”14 The handwriting is not easy to date, especially as it is a private hand, not a public one. Breydy maintained that it was very ancient, from around the tenth century, whereas Conterno favors the twelfth–thirteenth century, alleging that it resembles that of Fatimid and Ayyubid documents. It is a fairly unremarkable angular script that could fall anywhere in the range assigned by these two scholars. So it could be that a much later person copied out chunks of C, as Conterno proposes, or, if we wanted to preserve some of Breydy’s theory, we could argue that S is an earlier source book, which Eutychius either made or used, containing excerpts from various texts found in the library at St. Catherine’s or at some monastery in Palestine. In any case, we can take it that C does largely represent Eutychius’s own work, which therefore dates to the early fourth/tenth century and not to the fifth/eleventh. This is supported by another important finding of Conterno, namely that in the 20 or so manuscript copies of C the text is fairly stable, with just occasional omissions and additions (most notably, polemics against Jacobites and Nestorians),15 and so it is also much more cohesive and coherent than Breydy maintained. Here Conterno is talking about the subject matter, but in regard to wording, it is true that there are numerous minor variations in the surviving manuscripts, which reflects the fact that it was a very popular work, copied a great many times.

12 13 14 15

Kennedy, Melkite church 329–330, and see ns. 1, 7–9. and 11. See also Griffith, Apologetics and historiography 78–80; Ebied, Saʿīd b. Baṭrīq; Stutz, Constantinus Arabicus 134. Conterno, Recensions 402. Conterno, Recensions 402. See especially the table at Conterno, Recensions 398, which illustrates that all but five of the many passages not in S (identified by Breydy as later interpolations) are present in all the other manuscripts, and four of these are present in most manuscripts.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

388 3

hoyland

The Nature and Content of Eutychius’s Chronicle

The work of Conterno allows us to restore not only the earlier date to Eutychius’s chronicle but also to give a better assessment of its quality. In order to explain how S, very much a cut-and-paste job, could be Eutychius’s work in its final form, Breydy had to say that its author was “no real historian,”16 just extrapolating material from various sources with little input of his own. Yet this is at odds with al-Masʿūdī’s positive assessment cited above and with the popularity of the text in subsequent centuries. Other scholars have been disappointed by it because it seemed to contain little new factual data but many fantastical yarns.17 However, it is important to judge the work for what it is, not for what it is not, and it is clearly intended as a readable manual of world history, a summary of what is known, not new research. Eutychius states this himself in his introduction: Anyone who wants to discourse knowledgeably about any one of the sciences must be conversant with the foundation of that particular science. That being the case, it is a matter of necessity that I lay a foundation that may serve as a reference and a reliable guide for the inquirer. People have adopted many contradictory positions regarding history. What in my opinion is the right way forward, after lengthy investigation and much exertion, is to bring together what is in the Torah and other trustworthy compositions. I shall, then, set out (the material thus assembled) in a comprehensive and clear way, and I shall ensure it is in the form of succinct and concise narratives, so that this book of mine will be a sufficient resource, removing the need to refer to any other text to ascertain some point of history.18 The seventeenth-century editors of Eutychius’s chronicle, John Selden and Edward Pococke, called it Annales, and that is the title by which most scholars refer to it. However, this is rather unsatisfactory, as the text does not proceed in an annalistic fashion. Occasionally Eutychius provides calculations of years between major junctures in history, such as the Creation, the Flood, the birth of Abraham, the reign of David, the Babylonian exile, the death of Alexander,

16 17 18

Griffith, Apologetics and historiography 78, who gives a good summary of Breydy’s theory and its problems. For an interesting response to such a characterization, see Carrasco, Un pasaje. See also Mayerson, Procopius or Eutychius. Eutychius, Kitāb al-Taʾrīkh/Annales i, 5.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

389

the destruction of Jerusalem by Titus, the Prophet Muḥammad’s emigration to Medina (hijra), and, rather hubristically, his own birth, but he does not order his work into yearly entries. Rather, his system for structuring his narrative is by significant persons. For the Biblical period this means patriarchs, prophets, judges, and kings, but from the time of Cyrus the Great onward the ordering is purely by ruler, whether of the Persians, Romans, or Muslims. Within each entry events are assigned to a particular year of that ruler’s reign. It is only with Eutychius’s own time that we begin to see the provision of more absolute dates, given according to the hijrī calendar. Often these entries will be quite brief, but where Eutychius has found an anecdote to his liking, they can be very long. This not only concerns famous men, like Alexander the Great and Constantine i, but also lesser known figures. For example, Eutychius recounts at great length how Emperor Theodosius i had once been a poor man in Constantinople, and he and his friend Theophilus chopped wood together and fed the poor with the proceeds from its sale, and yet they became, respectively, emperor (379–395) and patriarch of Alexandria (384–412).19 Perhaps more surprisingly, Eutychius also includes long anecdotes about some of the Sasanian emperors (see below), but this may just reflect that they featured in the source that he used for Persian history and appealed to his love of a good story. Eutychius’s chronicle thus possesses that same marriage of chronology and narrative that is present in the historical works of his Muslim contemporaries, such as Aḥmad al-Yaʿqūbī (d. ca. 297/910), Muḥammad al-Ṭabarī (d. 310/923) and ʿAlī al-Masʿūdī. In the late Roman world, the genres of history (subdivided into church and secular/classicizing) and chronicle (most famously represented by Eusebius’s chronological canons) were relatively distinct.20 Obviously histories had some truck with chronology, just as chronicles would often include some longer narratives, but the divisions between them were reasonably clear, reinforced by the fact that the former usually dealt with a limited section of time and the latter commonly went back to Creation/Adam or at least to some significant moment in the distant past. In the Islamic sphere, there were works that primarily focused on chronology and/or the calculation of time, such as those by al-Layth b. Saʿd (d. 175/791),21 Ḥamza al-Iṣfahānī (d. ca. 350s/960s) and Muḥammad al-Bīrūnī (d. 439/1048), and some that gave a lengthy narrative account of a single topic, such as that of Sayf b. ʿUmar 19 20 21

Eutychius, Kitāb al-Taʾrīkh/Annales i, 141–149. For a useful introduction to late Roman/Byzantine and early Islamic chronography, see Foot and Robinson (eds.), Oxford History chs. 10 and 12. Zychowicz-Coghill, First Arabic annals, gives a plausible reconstruction of Layth’s text.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

390

hoyland

(d. 180/796) on the reconquest of Arabia (al-ridda) by Caliph Abū Bakr and that of Naṣr b. Muzāḥim (d. 212/827) on the battle of Siffin (Waqʿat Ṣiffīn) between the caliphs ʿAlī and Muʿāwiya. However, much more popular were those works that blended these two approaches. Our first extant Muslim example is the chronicle of Khalīfa b. Khayyāṭ (d. 230/845), which incorporates narratives of varying length into an annalistic scheme.22 In a loose way, Muslim authors of such works are likely to have been inspired by Greek and Syriac works that employed the same format, such as the Greek chronicle of Theophanes the Confessor (d. 202/818), which was itself influenced by and a continuation of the chronicle of George Synkellos, who compiled much of his material while a monk in Palestine in the mid-eighth century.23 Among George’s sources was John Malalas of Antioch (d. ca. 570), who produced one of the earliest examples of this genre, which tend to be referred to as world chronicles or sometimes as universal histories.24 Although the combination of chronology and narrative is the hallmark of all these aforementioned works, they still exhibit a fair degree of diversity when one compares them with one another. This diversity is evident in two principal respects. The first is the exact form of chronological ordering, whether by years (Theophanes, Ibn Khayyāṭ, al-Ṭabarī) or by reigns (al-Yaʿqūbī, Eutychius, alMasʿūdī). The second is the geographical and ethnographic extent, that is, how many regions and peoples are included. In the Islamic case, the minimum coverage will be the Israelites, Arabians, and Persians, as they were perceived to be the main wellsprings of Islamic civilization; in the maximal works, coverage is expanded to include the Greeks, Indians, Chinese, Franks, and so on. The more restricted view tends to go with religiously conservative authors, Islamic scholars like al-Ṭabarī, and the maximal view with those open to foreign wisdom, courtiers, and nonreligious scholars like al-Yaʿqūbī and al-Masʿūdī. Eutychius belongs at the religiously conservative end, and he only has Biblical peoples, Persians, Ptolemies (who presumably get in thanks to Eutychius’s Egyptian bias), Romans, and Muslims.25

22 23 24 25

Wurzel, Khalifa ibn Khayyat. Note that his chronicle begins with Muḥammad, so one might call it an Islamic chronicle, but it was not an idea that caught on. Hoyland, Arabic, Syriac and Greek historiography. Jeffreys designates it “the earliest extant example of a Byzantine world chronicle,” Beginning of Byzantine chronography 497. For these and other aspects of the world chronicle in Christian and Islamic culture, see Foot and Robinson (eds.), Oxford History ch. 21 (by A. Marsham).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

391

eutychius of alexandria vindicated

4

The Muslim Sources of Eutychius’s Chronicle

4.1 The Ancient World Eutychius’s section on the Biblical world largely consists of abridged and paraphrased passages from the Bible, supplemented with material from Christian apocryphal texts, in particular the sixth-century Cave of treasures.26 It is difficult to be sure whether Eutychius made use of Muslim sources in this section, but a number of close correspondences and common perspectives “support the notion of a cultural world shared by the Melkite patriarch and his Muslim contemporaries.”27 Muslims accepted the overall structure and narrative of the Bible, and Melkite Christians accepted some subtle modifications introduced by Muslims, such as the enhanced status of Ishmael and the identification of some of the prophets with characters in the Quran. The next section concerns the dynasties of the Achaemenids, Ptolemies, and (pagan) Romans. The first two items are simply lists of rulers with little detail beyond the length of their reigns. More attention is given to the pagan Roman rulers, but principally regarding their attitude toward early Christians— whether tolerant or hostile. The basic chronological framework relies upon an early sixth-century Alexandrian chronography that is influenced by the chronological works of the astronomer-mathematician Claudius Ptolemy (d. ca. 170) and the early Christian chronographer Julius Africanus (d. ca. 240).28 There would seem to be no input from Muslim sources, which are ignorant of the Achaemenids and have minimal interest in the Ptolemies and pagan Roman rulers. Between the Achaemenids and Ptolemies, Eutychius inserts a biography of Alexander the Great, who was a very popular subject for Muslim histories, in part for his mention in the Quran (as Dhū l-Qarnayn) and in part for his fame as a world conqueror. Eutychius’s depiction of him is substantially derived from the Greek tradition, whether directly from the Greek Alexander romance or from other Greek texts dependent upon it. An example of the latter is Eutychius’s account of the invasion of Egypt by the Persian Emperor Ochus and the flight of the pharaoh to Macedonia, seat of King Philip, father of Alexander, which is close to that of the chronicle of John Malalas.29 But there are 26 27 28 29

Götze, Die Nachwirkung der Schatzhöhle 155–169. Simonsohn, Biblical narrative 48. See also Monferrer-Sala, Cast out Hagar; and MonferrerSala, Gn 6, 1–4. Burgess, Date, purpose and historical context; Frick, Chronica minora xc–clxiv. Voigt, Recherches 73–75, though the pharaoh is called Nectanebo in the Greek tradition but Shanaq by Eutychius, which is close to the Sarnaqos of the Syriac Alexander Legend (75).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

392

hoyland

additions that are not present in any of the Greek versions and that plausibly hail from a Muslim source. For example, regarding the gifts exchanged between Alexander and the Persian Emperor Darius iii, which are meant to convey subtle messages, Eutychius has the same additions as the Muslim tradition, namely sesame grains from Darius, intended to signify the numerousness of his troops, and mustard seeds from Alexander, indicating his army’s pungent vivacity.30 And the vague hints about Alexander’s poisoning in the Greek biography are greatly elaborated upon in both Eutychius and the Muslim tradition.31 Then there is the consolation letter from Alexander to his mother and the sayings of the philosophers at the graveside of Alexander, neither of which are found in the ancient Greek versions but which Eutychius recounts at length and which were very popular in Muslim circles.32 4.2 Sasanian Emperors Already in the nineteenth century the Orientalist Theodor Nöldeke had noticed that Eutychius drew upon Muslim sources for his entries on the Sasanian emperors. His observation came amid his lament for the loss of the History of the Persian kings by ʿAbdallāh b. al-Muqaffaʿ (d. ca. 140/757), a high-ranking bureaucrat in the Islamic imperial government and of Persian ancestry, for it was, says Nöldeke, the most pre-eminent of the translations of the Book of lords, the putative royal history of the Persian realm.33 Our chances of finding Ibn alMuqaffaʿ’s original are very low, he predicts, but many individual passages from it can be assembled from those who depend upon it. As an example, he refers to the account of the Persian Emperor Peroz i (459–484) and his fight against Akhshunwār, king of the Hephthalites, presented by Eutychius and the Muslim scholar ʿAbdallāh b. Qutayba (d. 275/889). The two narratives are so close, he maintains, that they must rely upon the same source, and “this can only be the work of Ibn al-Muqaffaʿ.” On the basis of this correspondence, he argues, “one would be inclined to attribute Eutychius’ other reports about the Sasanids to Ibn al-Muqaffaʿ.”34 30 31

32

33 34

Eutychius, Kitāb al-Taʾrīkh/Annales i, 78; al-Ṭabarī, Taʾrīkh i, 695; Voigt, Recherches 11–12. Eutychius, Kitāb al-Taʾrīkh/Annales i, 81; though the poisoning of Alexander in Muslim sources appears seldom and does not resemble Eutychius’s account, which may depend rather upon a Greek or Syriac text, Voigt, Recherches 240–247. Eutychius, Kitāb al-Taʾrīkh/Annales i, 81–84; al-Yaʿqūbī, al-Taʾrīkh i, 162–163; Voigt, Recherches 247–257. A Syriac version of the graveside sayings of the philosophers is very close to that found in Eutychius, but as the editor notes, the transmission could have gone from Syriac to Arabic or vice versa; Brock, Laments. For discussion of its possible nature and contents and its transmission into Arabic, see Hämeen-Anttila, Khwadāynāmag; and Hoyland, History of the kings. Nöldeke, Geschichte xx–xi. For a partial translation of the Peroz-Akhshunwār narrative and some discussion, see Hoyland, History of the kings 135–141. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

393

There is no unambiguous evidence that the two authors draw upon Ibn al-Muqaffaʿ—Ibn Qutayba just tells us that he found his narratives on the Sasanian dynasty in “the biographical histories of the Persians” (siyar al-ʿajam)— but it does seem sure that their accounts are related. As well as the aforementioned Peroz-Akhshunwār episode, there is also the reconnaissance mission of the Persian Emperor Sapor ii (309–379) in the camp of his Roman counterpart. The following passage is identical in Eutychius and Ibn Qutayba except for the minor differences that I note in parentheses: I want to go into the territory of the Romans (iq/E: secretly/alone) so that I can learn about (the extent of their strength and readiness/the condition of their kings and armies) and the roads of their lands. When I have achieved this, I will return to my (kingdom/country) and then I will (proceed/march) to them (with troops/on the basis of this knowledge). His men warned him (against imperiling himself/and made known their fears for him). He, however, did not accept (iq: their advice and their objection). He journeyed in disguise until he reached (their/Roman) territory. He (remained among them/continued traversing it). While thus engaged, news reached him that a son of (Caesar/Maximian) was holding a banquet and that (he/his father) had ordered that (E: the lowly people and) the poor might gather and be given food (E: after the nobles had eaten). Sapor set off (in the guise/in the form) of a beggar so that he might be present at the gathering (iq: and attend the meal). (iq: While he was sitting at the table) there was brought to (Caesar/Maximian) one of the vessels of Sapor, on which was engraved a bust of Sapor. The servants began serving wine in it (E: to the king and the grandees around him). When the vessel reached one of (the grandees/sages), who was knowledgeable in (E: astrology and) physiognomy, he inspected the image, having previously looked at the face of Sapor (E: while he was sitting in the assembly). (iq: He grasped the vessel and said: I see something amazing, and Caesar said: What is that?). He said: “I see among those seated (the owner of this image/a man like the image that is on this cup).”35 The Roman emperor interrogated Sapor, who at first denied his identity, but the physiognomer was so insistent that eventually Sapor confessed. He was locked up—in a cow-shaped statue according to Eutychius—and taken along with the 35

Eutychius, Kitāb al-Taʾrīkh/Annales i, 119; Ibn Qutayba, al-Maʿārif 657. Eutychius is wrong in making Maximian (286–305) the contemporary of Sapor ii; this would have been Julian the Apostate (361–363).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

394

hoyland

Roman army as they launched an assault upon the Persian Empire. Eventually, however, he was able to escape and join his people in time to repel the Romans from the walls of the city of Gondeshapur. In this latter portion of the Sapor narrative Ibn Qutayba and Eutychius diverge somewhat, and we encounter this same pattern of close correspondence and divergence in their entries for other Sasanian emperors: Ardashir i and the capture of Hatra, Yazdgird i and the supernatural horse, Bahram Gor’s exploits in India, and Kavad i’s dealings with the ruler of the Turks and with the heresiarch Mazdak. It is difficult to be sure what accounts for this pattern. The frequent divergences would seem to exclude the idea that Eutychius is directly reliant on Ibn Qutayba. In other genres of historiography one would not hesitate to postulate a common source, but the Persian Book of lords tradition enjoyed enormous popularity and circulated in many subtly differing versions. Indeed, Eutychius implies that he used more than one source,36 so it may be that the divergence in Ibn Qutayba and Eutychius is to be explained by the two authors combining multiple sources rather than both reworking a single common source.37 In any case, Eutychius did have access to other material, as is clear from his account of the role played in the accession of Khusrau ii (591– 628) by his relatives Bindoi and Bistam, which is unknown to Ibn Qutayba, and from his description of Khusrau’s marriage to Maria and conversion to Christianity, which is likely from a Byzantine source.38 4.3 Conquests and Caliphs As one would expect, Eutychius makes much more substantial use of Muslim sources for the Islamic period. Breydy claims to have identified one of these sources, namely the history of the conquest of Egypt by the Egyptian scholar ʿUthmān b. Ṣāliḥ (d. 219/834), maintaining that there was “full overlap” (pleine coïncidence) between his account and that of Eutychius. Though the two accounts are certainly close, Breydy’s statement requires some qualification. First, it is important to bear in mind that the work of ʿUthmān is not extant, but survives only in quotations from it by later authors, chiefly the notable lawyer and historian Ibn ʿAbd al-Ḥakam (d. 257/871), though the latter does appear

36 37

38

Eutychius, Kitāb al-Taʾrīkh/Annales i, 178: “some of the Persians say […].” This is argued by Maristo, Bahrām Čūbīn 190–200, with respect to the common features in the Muslim narratives about Bahram Chobin. He perhaps does not take sufficient account of the fact that the transmission of historical accounts was not bound by the same strict conventions as for legal and scientific texts, and that free rewording was acceptable, but his point is still a good one. Eutychius, Kitāb al-Taʾrīkh/Annales i, 213–215.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

395

to have been careful with his citations.39 Second, in some places only the gist of the story is the same, with little or no correspondence in wording. Here, for example, is the opening part of the narratives of Eutychius and ʿUthmān on the conquest of Egypt as presented by Breydy:40 Eutychius: He (ʿUmar) wrote to ʿAmr b. al-ʿĀṣ to march to Egypt, and so he set off from Palestine. Then ʿUmar b. al-Khaṭṭāb had second thoughts about that and he wrote to him (ʿAmr): If my letter overtakes you when you have already entered Egypt then march on in God’s name, but if it overtakes you when you have not yet entered (Egypt) then return. ʿUthmān: When ʿUmar b. al-Khaṭṭāb came to Jābiya, ʿAmr came before him and got him on his own, saying to him: “O commander of the faithful, permit me to march to Egypt,” and he urged him on until ʿUmar inclined to that and entrusted to him four thousand men […] Go ahead and I will consult God about your going. My letter will come to you quickly God willing, and if my letter that reaches you orders you to withdraw from Egypt before entering it, or any part of its territory, then withdraw. But if you have entered it before my letter comes to you, then go ahead. Even when their accounts are very close in wording, minor variations will occur. Here, for instance, is the account from both authors about the attempt of the Egyptian leader known as the Muqawqis41 to initiate a dialogue with the Muslim general ʿAmr b. al-ʿĀṣ, the words common to both accounts are marked in bold:42

39 40

41

42

For the way in which Ibn ʿAbd al-Ḥakam used ʿUthmān’s work, see Zychowicz-Coghill, Conquests of Egypt 185–193. Breydy, La conquête arabe 384. Breydy quotes from S; for C, see Eutychius, Kitāb alTaʾrīkh/Annales ii, 21. Note that this same incident of ʿUmar’s letter to ʿAmr occurs earlier (Eutychius, Kitāb al-Taʾrīkh/Annales ii, 19), perhaps indicating that Eutychius has another source for the conquest of Egypt. Modern scholars debate whether the Chalcedonian Patriarch Cyrus or the anti-Chalcedonian Patriarch Benjamin is intended, but in the Islamic tradition he is removed somewhat from the realities of the early seventh century and has become a partially fictionalized character. See Zychowicz-Coghill, Conquests of Egypt 57–67. Breydy, La conquête arabe 387–388; cf. Eutychius, Kitāb al-Taʾrīkh/Annales ii, 22. Note that historians from the ninth century and later divide the Egyptian people into Romans, meaning Greek-speaking Chalcedonians, and Copts, meaning Coptic-speaking anti-Chalcedonians, the former portrayed as anti-Muslim and the latter as pro-Muslim, but the seventh-century situation was not so clear cut. See Zychowicz-Coghill, Conquests of Egypt 53–78.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

396

hoyland

The Romans went out, with the Muqawqis and a group of Coptic notables, from the southern gate of the citadel, while others stayed to fight against the Arabs. They embarked on the boats, came onto the island, which is today a place of artisans, and they cut the bridge, and this was when the Nile was in flood. Then the Muqawqis sent word to ʿAmr b. alʿĀṣ and said: “You came into our country and insisted on fighting us. For too long a time you have now been in our territory. [ʿUthmān: You are a small group and the Romans have drawn up close to you, have mobilized against you and have much equipment and weaponry]. You are surrounded by the Nile and you are prisoners at our mercy. Send us therefore one of your men so that we can hear his words and maybe reach an agreement between you and us that pleases both sides, and so put an end to this war.” This phenomenon of numerous slightly differing versions of the same incident is extremely common in Arabic literature, and it tends to obfuscate the question of the original authorship of any one account. Eutychius may have used ʿUthmān’s historical work directly, but it is also possible that he relied on a dependent of ʿUthmān or else on a source shared by him and ʿUthmān. This problem is even more marked with Eutychius’s account of the early Muslim-Byzantine confrontation at the village of Dāthin, near Gaza.43 This account has distinct clusters of variant versions; one of these clusters, which ultimately goes back to a certain Jaʿfar b. ʿAbdallāh (d. ca. 125/743), enjoyed great popularity, in part because it was allegedly the first encounter between the Muslims and the Byzantines, and perhaps in part because it contained a parley between the commander of the Byzantine forces in Gaza and the Muslim general ʿAmr b. al-ʿĀs. Often this conversation contains a discussion about why the Muslims have come, ʿAmr explaining that they want a greater share of the fertile lands possessed by the Romans, but sometimes, as in Eutychius, it is just an occasion for ʿAmr to put before the commander the three standard options offered to all non-Muslims: conversion, poll-tax, or war. In either case no agreement is reached, and so ʿAmr prepares to leave. In the reworking of Jaʿfar’s account by the historian Muḥammad al-Wāqidī (d. 207/823), the Byzantine commander is said to suspect from ʿAmr’s nonchalant and authoritative manner of speaking that he is actually a leader and not just a messenger, and so he orders his men to assassinate him before he exits the city. ʿAmr is saved by 43

Eutychius, Kitāb al-Taʾrīkh/Annales ii, 10–11. The toponym Dāthin comes from our earliest extant account of the battle (al-Balādhurī, Futūḥ 109), but Eutychius has Tādūn, and Ibn ʿAsākir either Thādan (Taʾrīkh ii, 82) or Bādhan (Taʾrīkh xlvi, 155).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

397

a Greek-speaking servant of his who overhears and warns him, allowing him to slip away safely back to the Muslim camp. This version is closest to the account of Eutychius, but there is very little exact correspondence in wording, except that both authors have ʿAmr say upon his escape: “I will never return to such a man as this” (lā aʿūdu li-mithli hādha abadan).44 Either Eutychius is relying upon a source that has not survived or he is giving a very free rendering of the account of al-Wāqidī. Besides one or more texts on the Muslim conquests, Eutychius seems to have employed some sort of work on the caliphs. For example, he is able to name the chief of police and the chamberlain of every caliph, and this information corresponds well to the lists of these office holders found in Muslim sources.45 He also offers details about the physical traits of caliphs, which can be found in one or other Muslim source, even the more recondite ones. For example, he agrees with al-Yaʿqūbī that ʿUmar i was bald and ambidextrous, that ʿUthmān had a bushy beard and a gold brace on his teeth, and that Muʿāwiya I had bulging eyes and a big arse,46 but we have to go to later sources to find agreement with Eutychius that ʿAlī had an extensive beard and a large belly.47 Surprisingly, Eutychius says nothing at all about the emergence of Islam except that the emigration (hijra) of the Prophet Muḥammad to Medina occurred in the first year of Heraclius’s reign (which is incorrect).48 His only other mention of Muḥammad concerns his death, which he places in the 16th year of Heraclius’s reign (also incorrect) and the eleventh year of the Hijra (which is correct), at the age of 63, leaving no children but Fāṭima. Christians living in the Byzantine Empire mostly wrote polemical accounts of Muḥammad and the rise of Islam, whereas those under Muslim rule tended to be more neutral, if not actually positive, but Eutychius’s silence is unusual. Perhaps it is intended, along with the indication of Muhammad’s lack of a male heir, as a subtle polemic, the most that a prominent Melkite cleric writing in Arabic in the Islamic empire dared to do for fear of incurring censure. It is possible that Eutychius obtained all his Islamic material from one Muslim world chronicle that has not come down to us, but the diversity of 44 45 46 47 48

Eutychius, Kitāb al-Taʾrīkh/Annales ii, 11; Ibn ʿAsākir, Taʾrīkh xlvi, 156. E.g., Eutychius’s list of chamberlains conforms well to that given by Ibn Ḥabīb, al-Muḥabbar 258–260 (asmāʾ al-ḥujjāb). Eutychius, Kitāb al-Taʾrīkh/Annales ii, 27 (ʿUmar), 33 (ʿUthmān), 38 (Muʿāwiya) = alYaʿqūbī, al-Taʾrīkh ii, 185, 205, 283. Eutychius, Kitāb al-Taʾrīkh/Annales ii, 33 = al-Nuwayrī, Nihāyat xx, 1. Eutychius, Kitāb al-Taʾrīkh/Annales ii, 1. The Hijra occurs in the 12th year of Heraclius’s reign. Eutychius adds that Muḥammad “made the pulpit” (ʿamila al-minbar) in the eighth year, which is unclear.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

398

hoyland

his accounts suggest that he had a number of Muslim sources at his disposal and that he selected and fashioned narratives from them according to his own designs.49 Moreover, he would on occasion combine them with material of non-Muslim origin. For example, the above quotation about the Muqawqis is prefaced by an introduction to this character that seems to be of Byzantine provenance, informing us that he was a Jacobite who hated the Romans but hid his doctrinal allegiance lest the Romans kill him, and he deceived the Romans into surrendering the citadel to the Muslims. In addition, Eutychius uses nonMuslim sources to document church history, which is still an important part of his narrative even in the Islamic period, especially the actions of patriarchs and the problems of heresy, and to record the reigns of Byzantine emperors, even if his notices on them are mostly quite brief. His last substantial entry for a Byzantine emperor concerns Heraclius, about whom he has much unique information, in particular his campaign against Khusrau, his dealings with Manṣūr b. Sarjūn,50 financial governor of Damascus, his massacre of Jews in Palestine, his involvement with monotheletism (what Eutychius calls the Maronite heresy), and some unusual items, such as the decision of the general Māhān to become a monk on Mount Sinai after losing the battle of Yarmūk,51 and the employment of columns bearing the busts of Emperor Heraclius and Caliph ʿUmar as boundary markers in the vicinity of Qinnasrīn.52 In sum, Eutychius would seem to be more active in selecting and shaping his material than is usually assumed. I might add a final indication of this, namely his predilection for certain topics, which are present across his work. The most obvious example is his interest in buildings; for the Islamic period this mainly concerns churches, but also mosques and some public structures. For example, he speaks of construction, whether new build, extensions, or refurbishment, in connection with: the Church of the Nativity in Bethlehem; the congregational

49

50

51

52

Griffith thinks that Eutychius “consulted different documentary sources” based on his frequent use of the phrase “in another copy,” Eutychius of Alexandria 158; this is possible, though the phrase may come from the scribe copying Eutychius’s chronicle on the basis of multiple manuscripts. Eutychius, Kitāb al-Taʾrīkh/Annales ii, 5, 7, 13–15. He is depicted as a traitor, like the Muqawqis, facilitating the surrender of the Romans to the Muslims. Note that Eutychius (Kitāb al-Taʾrīkh/Annales ii, 69) says that Patriarch Elias iii of Jerusalem (ca. 265–294/879– 907) was a descendant of this Manṣūr, which would mean an impressive survival of this family’s elite status. Eutychius, Kitāb al-Taʾrīkh/Annales ii. 15. S has Fāhān; this must be the Bāhān/Baanes of Arabic/Greek sources (the letters m, s, and b are easily confused in a carelessly written Arabic manuscript); see Nichanian, Le maître. Eutychius, Kitāb al-Taʾrīkh/Annales ii, 19–20.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

399

mosques of Fusṭāṭ, Jerusalem, and Damascus; churches at Helwan, Qaṣr Shāma, Abū Qīr, and Mount Muqaṭṭam; the Church of the Apostles at the monastery of al-Quṣayr; and the Church of the Resurrection in Jerusalem. And he notes that the great church of Alexandria, which he asserts to be originally the temple of Saturn commissioned by Cleopatra, caught fire in 300/912, and that Muslims attacked Christian holy sites in Ramla, Ascalon, Caesarea, and Damascus in 311 and 312 (923–924). Concerning public structures, he speaks of Cleopatra’s wall53 to keep out the Romans, and he pays a lot of attention to Nilometers, whether the establishment of new ones or the repair of old ones. The majority of these notices about construction pertain to Egypt, and this brings me to another of Eutychius’s predilections, namely, for his homeland. This is particularly the case for his own lifetime, when there is an increasing focus on Egypt, reflecting an emerging sense of Egyptian identity during that period.54

5

Conclusion

As I noted above, there are a number of similarities between the chronicles of Eutychius and Agapius in their general layout and subject matter, but in the particulars of their content there are many differences. Eutychius seems much more at home with Muslim sources, deploying them to a much greater extent than Agapius. This may be chiefly a consequence of what was available to him. In Manbij in northern Syria, Agapius was near to Edessa, the heart of Syriac Christianity, and this probably gave him better access than Eutychius to Syriac materials, whether in the original or in Arabic translation.55 At least in part, though, the difference between them reflects the character of these two authors and of their cultural milieu. It is unlikely that Eutychius could find no information in Alexandria about the Greek pagan past, but he does not touch on this topic at all, whereas Agapius devotes much space to it.56 The same goes for the Byzantine emperors of the Islamic period. Eutychius has much to say about Heraclius, but thereafter he generally only reports their accession to the throne and the length of their reign, whereas Agapius still conveys much 53 54 55

56

Attributed in Muslim sources to a certain Dalūka, who reigned after the pharaoh defeated by Moses; see Sijpesteijn, Building an Egyptian identity. Sijpesteijn, Building an Egyptian identity. An example of this may be his account of how the Amazon warriors would procreate once a year and raise only the resulting female progeny, killing the boys, which I previously asserted to be of unknown provenance (Hoyland, Agapius 20–21), but which I noticed recently is close to the account of Bardaisan of Edessa (Book of the laws, 48–49). Hoyland, Agapius.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

400

hoyland

information about their actions. Here we can tell that Eutychius’s reticence is not due to lack of sources, for where there is a question of religious controversy he is roused to comment, writing at length on the correspondence between Emperor Constans ii (20–48/641–668) and Pope John iv on monotheletism and the flirtation of Emperor Theophilus (214–227/829–842) with iconoclasm.57 About both historians the question has often been posed why they wrote in Arabic.58 By this time, however, it would be surprising if they had not written in Arabic. Melkite Christians residing within the Islamic empire had begun writing in Arabic in the second half of the eighth century; through the ninth century they gradually translated from Greek into Arabic much of their religio-cultural heritage.59 In the same century an enormous amount of Greek scientific literature was also translated into Arabic,60 and by the early tenth century Arabic had become the intellectual medium of much of the Islamic world so that it was no longer only translations that we see in Arabic but original compositions, too, such as the great historical works of al-Ṭabarī and al-Masʿūdī. It was, therefore, only to be expected that Eutychius and Agapius would write in Arabic. The only exceptions to this trend were those communities and polities with a very distinct identity that was bound up with their own language, who might therefore choose to write in that language as well as, or even instead of, Arabic, such as Armenians, Georgians, Syriac Christians, and, in a different way, certain Persian dynasties.61 There is no sure indication that Eutychius and Agapius used any non-Muslim texts in their original language, and so it is likely that much of their source materials circulated in Arabic translation already by their time. We know that Agapius made use of the historical work of Qusṭā b. Lūqā, who was a major figure of the Greco-Arabic translation movement, translating numerous scientific texts from Greek and Syriac into Arabic, and plausibly doing the same for history.62 This suggests that more historical materials pertaining to nonMuslim affairs had been translated into Arabic already in the ninth century than has usually been supposed, and this muddies somewhat the long-held 57 58 59 60 61

62

Eutychius, Kitāb al-Taʾrīkh/Annales ii, 28–32, 63–64. On the latter topic, see Griffith, Eutychius of Alexandria. Most recently, see Conterno, Christian Arabic historiography, which provides a thoughtful discussion of this topic. Treiger, Christian Graeco-Arabica. Gutas, Greek thought. Kennedy, Survival of Iranianness. For thoughts on the identity formation of Christian communities in the Islamic period, especially Syriac Christians, see Ter Haar Romeny (ed.), Religious origins of nations. Hoyland, Agapius.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

401

notion that there was little interest in translating historical, as opposed to scientific or philosophical, texts into Arabic. It is of course true, though, that we are not talking here about the grand narrative historians like Thucydides and Tacitus, but rather shorter themed works, focused on heroic characters, such as Alexander the Great, Constantine i, Julian the Apostate, and celebrated holy men,63 as well as on specific topics and episodes from Biblical and classical history,64 and we are also not talking about close translations but rather creative adaptations.65 Recent studies of such narratives have revealed that these texts circulated across Muslim and Christian communities very easily, facilitated by their rendering into the lingua franca of Arabic.66 Eutychius and Agapius were, therefore, not so different from their Muslim contemporaries in the field of history. Certainly, they were all beneficiaries of the translation of historical materials into Arabic; the difference lay rather in the choices they made in the selection and combination of these materials.

Personal Note I dedicate this chapter affectionately to Hugh, whom I have known since my graduate days and from whom I have learned much over the years.

Bibliography Sources Agapius = Vasiliev, A. (ed.), Kitab al-ʿUnvan: Histoire universelle écrite par Agapius (Mahboub) de Menbidj, 2 parts, in Patrologia orientalis 5 (1910) 505–691; 7 (1911), 459–591; 8 (1912), 399–547; 11 (1915), 9–144.

63

64

65

66

Voigt, Recherches (Alexander in Arabic); Stutz, Constantinus Arabicus; Tayyara, Tabalwur. The Life of Mar Saba, used by Eutychiuis, seems to have existed in Arabic already in the eighth century, see Van Esbroeck, Incidence. Creative adaptation in Arabic of Biblical stories effectively begins with the Quran and continues in commentaries thereon. For Greco-Roman material in Arabic, see Hoyland, Agapius. Note the remark of Voigt that Agapius “creates a hybrid narrative” (Recherches 63), combining allusions to the Greek Alexander romance and elements from Arabic historiography; the same could be said of Eutychius. Note the observation of Stutz that in Christian and Muslim retellings of the history of Constantine we see no obvious divisions along confessional lines (“keine eindeutig konfessionellen Trennungslinien”; Constantinus Arabicus 317).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

402

hoyland

al-Balādhurī, Futūḥ al-buldān, ed. M.J. de Goeje, Leiden 1866. Bardaisan of Edessa, The book of the laws of countries, ed. and trans. H.J.W. Drivers, Piscataway, NJ 2007. Eutychius, Kitāb al-Taʾrīkh/Annales, ed. L. Cheikho, 2 parts in 1 volume, Beirut 1906– 1909. [The earlier edition of J. Selden and E. Pococke was published at Oxford in 1658–1659. For an edition of Seldon, see M. Breydy, Das Annalenwerk des Eutychius von Alexandrien, Louvain 1985.] Ibn ʿAsākir, Taʾrīkh Dimashq, ed. ʿA. al-ʿAmrawī, 80 vols., Beirut 1995. Ibn Ḥabīb, al-Muḥabbar, ed. I. Lichtenstädter, Hyderabad 1942. Ibn Qutayba, al-Maʿārif, ed. Th. ʿUkāsha, Cairo 1969. Ibn Qutayba, ʿUyūn al-akhbār, 4 vols., Beirut 1997. al-Masʿūdī, Kitāb al-Tanbīh wa-l-ishrāf, ed. M.J. de Goeje, Leiden 1894. al-Nuwayri, Nihāyat al-arab fī funūn al-adab, 33 vols., Cairo 2002. al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, ed. M.J. de Goeje, 16 vols., Leiden 1879–1901. al-Yaʿqūbī, al-Taʾrīkh. ed. M.T. Houtsma, 2 parts in 1 vol., Leiden 1883.

Studies Breydy, M., La conquête arabe de l’Egypte, in Parole de l’Orient 8 (1977–1978), 379–396. Breydy, M., Etudes sur Saʿīd ibn Baṭrīq et ses sources, Louvain 1983. Brock, S., Laments of the philosophers over Alexander in Syriac, in jss 15 (1970), 205– 218. Burgess, R.W., The date, purpose and historical context of the original Greek and the Latin translation of the so-called “Excerpta Latina Barbari,” in Traditio 68 (2013), 1– 56. Carrasco, C.M., Un pasaje controvertido en los Annales de Eutiquio de Alejandría. El ataque judío a la ciudad de Tiro, in Collectanea Christiana Orientalia 11 (2014), 53– 73. Conterno, M., Christian Arabic historiography at the crossroads between the Byzantine, the Syriac and the Islamic traditions, in A.M. Butts and R.D. Young (eds.), Syriac Christian culture: Beginnings to renaissance, Washington, DC 2020, 212–225. Conterno, M., Found in translation: Agapius, the Septuagint and the “falsified” Torah of the Jews, in M. Conterno and M. Mazzola (eds.), Intercultural exchange in late antique historiography, Leuven 2020, 143–168. Conterno, M., The recensions of Eutychius of Alexandria’s Annals: ms Sinai 582 reconsidered, in Adamantius 25 (2019), 383–404. Ebeid, B., Melkite-Chalcedonian reading of history: The case of Eutychius of Alexandria and his Annals, in Aram 31 (2019), 51–83. Ebeid, B., Saʿīd b. Baṭrīq the theologian: New considerations on his historical work, “The annals,” in Parole de l’Orient 42 (2016), 165–190. Foot, S., and C. Robinson (eds.), The Oxford history of historical writing, vol. 2: 400–1400, Oxford 2015. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

eutychius of alexandria vindicated

403

Frick, C., Chronica minora, vol. 1, Leipzig 1892. Götze, A., Die Nachwirkung der Schatzhöhle, in Zeitschrift fūr Semitistik 3 (1924), 153– 177. Griffith, S.H., Apologetics and historiography in the Annals of Eutychius of Alexandria: Christian self-definition in the world of Islam, in R. Ebied and H. Teule (eds.), Studies on the Christian Arabic heritage, Louvain 2004, 65–89. Griffith, S.H., Eutychius of Alexandria on the emperor Theophilus and iconoclasm in Byzantium, in Byzantion 52 (1982), 154–190. Gutas, D., Greek thought Arabic culture: The Graeco-Arabic translation movement in Baghdad and early Abbasid society, London 1998. Hämeen-Anttila, J., Khwadāynāmag: The Middle Persian book of kings, Leiden 2018. Hoyland, R.G., Agapius of Manbiǧ, Qusṭā ibn Lūqā and the Greco-Roman past: The beginnings of Christian Arabic and Muslim historiography, in Quaderni di studi arabi 16 (2021), 7–41. Hoyland, R.G., Arabic, Syriac and Greek historiography in the first Abbasid century: An inquiry into inter-cultural traffic, in Aram 3 (1991), 211–233. Hoyland, R.G., The “History of the kings of the Persians” in three Arabic chronicles, Liverpool 2018. Hoyland, R.G., Theophilus of Edessa’s chronicle, Liverpool 2011. Jeffreys, E., The beginning of Byzantine chronography, in G. Marasco (ed.), Greek and Roman historiography in late antiquity, Leiden 2003, 497–527. Kennedy, H., The Melkite church from the Islamic conquest to the Crusades: Continuity and adaptation in the Byzantine legacy, in The 17th international Byzantine congress: The major papers, New Rochelle, NY 1986, 325–343; reprinted in The Byzantine and early Islamic Near East, Burlington 2006, no. 6. Kennedy, H., Survival of Iranianness, in V. Curtis and S. Stewart (eds.), The rise of Islam, vol. 4 of The idea of Iran, London 2009, 13–29. Maristo, J., Bahrām Čūbīn in early Arabic and Persian historiography: Why so many stories?, PhD diss., Helsinki 2020. Mayerson, P., Procopius or Eutychius on the construction of the monastery at Mount Sinai: Which is the more reliable source, in basor 230 (1978), 33–38. Monferrer-Sala, J.P., Cast out Hagar and Ismael her son from me: Text and intertext in Eutychius of Alexandria’s Annals, in Hugoye: Journal of Syriac studies 14 (2011), 225– 247. Monferrer-Sala, J.P., Gn 6, 1–4 a la luz de un fragmento exegético contenido en el Kitāb al-taʾrīj al-maŷmūʿ ʿalā l-taḥqīq wa-l-taṣdīq de Eutiquio de Alejandría, in meah: Sección árabe-islam 49 (2000), 117–130. Nichanian, M., Le maître des milices d’Orient, Vahan, in B. der Mugrdechian (ed.), Between Paris and Fresno: Armenian studies in honor of Dickran Kouymjian, Costa Mesa, CA 2008, 321–337.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

404

hoyland

Nöldeke, T., Geschichte der Perser und Araber zur Zeit der Sasaniden, Leiden 1879. Sijpesteijn, P., Building an Egyptian identity, in A. Ahmed, B. Sadeghi, and M. Bonner (eds.), The Islamic scholarly tradition: Studies in history, law, and thought in honor of Michael Cook, Leiden 2011, 85–105. Simonsohn, U., The biblical narrative in the Annales of Saʿīd ibn Baṭrīq and the question of medieval Byzantine-Orthodox identity, in Islam and Christian-Muslim relations 22 (2011), 37–55. Stutz, J., Constantinus Arabicus: Die arabische Geschichtsschreibung und das christliche Rom, Piscataway, NJ 2017. Tayyara, A., Tabalwur qiṣṣat al-qayṣar Yūliyānūs fī l-fikr al-islāmī al-taʾrīḫī, in al-Majmaʿ: Studies in Arabic language, literature and thought 6 (2012), 111–140. Ter Haar Romeny, B. (ed.), Religious origins of nations? The Christian communities of the Middle East, Leiden 2010. Treiger, A., Christian Graeco-Arabica: Prolegomena to a history of the Arabic translations of the Greek church fathers, in Intellectual history of the Islamicate world 3 (2015), 188–227. Van Esbroeck, M., Incidence des versions arabes chrétiennes pour la restitution des textes perdus, in Institut de recherche et d’histoire des textes (1989), 133–143. Voigt, C., Recherches sur la tradition arabe du Roman d’Alexandre, Wiesbaden 2016. Wasserstein, A., and D.J. Wasserstein, The legend of the Septuagint from classical antiquity to today, Cambridge 2006. Wurzel, C., Khalifa ibn Khayyat’s history on the Umayyad dynasty (660–750), Liverpool 2015. Zychowicz-Coghill, E., Conquests of Egypt: Making history in ʿAbbasid Egypt, PhD diss., Oxford 2017. Zychowicz-Coghill, E., The first Arabic annals: Fragments of Umayyad history, Berlin 2021.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 21

Bureaucrats on the Move: Messengers in Fourth/Tenth-Century Iraq Maaike van Berkel, Nadia Maria El Cheikh and Letizia Osti

1

Introduction

At the end of 323/935, a year into the caliphate of al-Rāḍī, the vizier Ibn Muqla found himself forced to negotiate with the governor al-Ḥasan b. Abī l-Hayjāʾ b. Ḥamdān, the future Hamdanid ruler Nāṣir al-Dawla.1 He wrote to him, offering amnesty. The letter was delivered by an unspecified messenger: “Ibn Ḥamdān kissed the letter but said to the messenger: ‘I have no dealings with this person—meaning Ibn Muqla—nor can I accept his guarantee: he has no honour, good faith, or loyalty. I won’t hear anything he says unless ʿAlī b. ʿĪsā acts as mediator and vouches for him.’”2 What was the bearer of the letter supposed to do with this answer? Who could he turn to? This story of the negotiations between Ibn Ḥamdān and Ibn Muqla illustrates the agency required of a messenger: he needs both initiative and diplomacy to decide his next steps in this negotiation process. How and to whom should he deliver the oral message from Ibn Ḥamdān? Is he supposed to return to the current vizier, Ibn Muqla, who had sent him, or is it better to approach the former vizier and Ibn Muqla’s main antagonist, ʿAlī b. ʿĪsā, directly, suggesting that he act as mediator?3 This narrative also tells us something about the limitations of the messenger’s authority: if the receiver does not trust the sender, the messenger is unable to negotiate a deal or act on his own authority. Clearly, Ibn Ḥamdān does not

1 This paper was written collectively and is the latest result of many years of joint work and friendship between us and with Hugh Kennedy. For formal purposes, Maaike van Berkel is responsible for the introduction and the sections on al-Ḥasan b. Hārūn; Nadia Maria El Cheikh is responsible for the section on Muslim envoys; Letizia Osti is responsible for the sections on al-Kūfī and the conclusion. 2 Miskawayh, Tajārib i, 326–327. Throughout the chapter we rely on Margoliouth’s translation of the text, with some adjustments in wording and transliteration. 3 On the long career of ʿAlī b. ʿĪsā (d. 334/946), see Bowen, Life and times; van Berkel, ʿAlī b. ʿĪsā. Ibn Ḥamdān was granted his wish: ʿAlī, who had been exiled to al-Ṣāfiya (near Dayr Qunnā), at the request of Ibn Muqla was enlisted to help with the negotiations (see Bowen, Life and times 347–350).

© M. van Berkel, N.M. El Cheikh and L. Osti, 2022 | doi:10.1163/9789004525245_022 -

978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

406

van berkel, el cheikh and osti

trust the sender of the letter, namely, Ibn Muqla. He needs somebody else, a person of higher authority than the messenger, to guarantee Ibn Muqla’s good faith. Nevertheless, Ibn Ḥamdān treats Ibn Muqla’s letter with the utmost respect, kissing it upon delivery and thus implying that he recognizes the office of the vizier as a representative of the caliph, even though he may not respect the individual who currently holds this office. The mission of this messenger is a mix of two functions: formally, he facilitates internal communications between an acknowledged ruler and his subject; in practice, however, he acts as a diplomat (i.e., an emissary) to the sovereign ruler of another territory. While superficially the two functions are similar, the latter requires considerably more independence and initiative if the messenger wants to accomplish and survive his mission. Another element of note is the reason for the negotiations: Ibn Muqla had written to Ibn Ḥamdān “inviting him to become again a loyal subject and offering him amnesty”: like many regional leaders of the time, Ibn Ḥamdān no longer sent regular payments to Baghdad, leaving the central government in dire straits and unable to keep the bureaucratic machine running smoothly. It may be presumed, then, that the bearer of the vizier’s message must have some knowledge of the financial implications of Ibn Ḥamdān’s loyalty. In other words, besides diplomatic skills and initiative, our messenger may well have needed some financial competence. Modern scholarship has examined many aspects of these issues, but diplomacy, on the one hand, and administrative communication—which of course includes sending a delegation from the periphery to the center—on the other, have mostly been treated as separate topics.4 In this chapter, we will explore the intersection of these two topics, looking at the role, influence, and agency of messengers in the early fourth/tenth century. This period is particularly interesting because it sees an acceleration in the political fragmentation of the caliphate and, consequently, of infrastructures such as the postal system, which in previous centuries had been a reliable means of internal communication.5 In this context, we take two general observations as our starting point: – In this period, the center of the empire becomes less prominent. The gradual loss of Baghdad’s attraction as the capital of the caliphate is accelerated, and all provinces become de facto independent. As a consequence, a weakened

4 See, for example, El Cheikh, Byzantium; Tillier, Représenter la province; van Berkel, Communication between centre and periphery; van Berkel, Political communication. 5 Silverstein, Postal systems 53–121.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

407

bureaucrats on the move

central government must negotiate power with regional leaders and, increasingly, regional leaders negotiate—or clash—among themselves with no intervention from the central government. – As centralized administrative structures and procedures become less solid, the bureaucratic machinery is increasingly dependent on informal bonds and personal relations of trust. Our working hypothesis is that the fourth/tenth century sees the emergence of a type of bureaucrat6 who acts as internal messenger and becomes a crucial mediator between a formally recognized center and its periphery. As the messenger acquires importance, the career of a successful bureaucrat becomes less statically tied to administrative bureaus, absorbing a more dynamic and informal role. The messenger will no longer be a nameless rasūl as in the episode above but will become an active participant in the story, whose career can be followed and who will have his own agency in the course of events. Narrative and normative literature provide plenty of examples of messengers, extrapolating prescriptions on the qualities and ideal behavior of an envoy. In general, prescriptive texts are more interested in foreign diplomatic missions than in—at least formally—internal communication, while more descriptive texts concentrate on the moral qualities and oratory skills of the messengers rather than their practical skills. In what follows, we will illustrate the contents of such normative literature on diplomats (i.e., envoys to independent rulers) and see which aspects it addresses; we will then contrast these ideal qualifications with two practical examples of internal diplomacy taken from chronicles on the fourth/tenth century. This will highlight specific additional characteristics that this type of internal diplomacy requires.

2

Muslim Envoys: Qualities, Attributes, and Skills

From the dawn of Islam, we read about embassies and envoys. Early reports testify that the Prophet Muhammad sent envoys to Arab tribes, to the patriarch of the Copts in Egypt, to the Byzantine emperor, the Persian king, and the negus of Abyssinia.7 In his Ṭabaqāt, Ibn Saʿd mentions that the Prophet, having returned from Hudaybiyya in Dhū l-Ḥijja 6/628, sent messages to the kings, 6 Throughout this paper, depending on the context we use different translations of the term kātib and its plural kuttāb (secretary, clerk, bureaucrat) to highlight the different ranks this term can describe. 7 Hamidullah, Six originaux des lettres 149–172; Serjeant, Early Arabic prose 141–142.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

408

van berkel, el cheikh and osti

inviting them to join the community of believers.8 These envoys and the messages they carried had the mission of inviting these rulers to embrace Islam. The role of the envoys was an important one and, with the establishment of the Islamic state, would increase in magnitude as multiple embassies both inside the Islamic caliphate and as part of international diplomacy were undertaken in order to deliver official letters, carry gifts, sign peace treaties, exchange captives, debate, etc. The Mirrors of Princes are treatises that aim to advise rulers about the most appropriate comportment to follow on both the ethical and practical levels. This genre belongs to adab, a concept central to Abbasid cultural practice, which in addition to denoting “a corpus of varied literary knowledge,” referred to “a constellation of courtly manners and tastes to be conditioned and exhibited.”9 In time, descriptions of the qualities envoys should possess were elaborated, as well as principles for the envoys to follow with respect to their conduct. One particular text within this genre is Ibn al-Farrāʾ’s (fl. early fifth/eleventh century) Kitāb Rusul al-mulūk wa man yaṣluḥ lil-risāla wal-safāra (Messengers of the kings and those who act appropriately as messengers and ambassadors). This fourth/tenth-century treatise is almost unique in its focus on the conduct of Muslim envoys and draws on Quranic examples to define the moral and solemn qualities of the ideal envoy. Typical of the Mirror of Princes and of adab, broadly, Kitāb Rusul al-mulūk aims to instruct about past and best practices in the choice and dispatch of envoys, their profile and qualities, their duties and conduct, and the correct way to deliver a message.10 As he states in his introduction, Ibn al-Farrāʾ aims to expose the significance of the role of the messenger and who would be best suited to undertake such missions. His treatise discloses the attributes of the ideal messenger and highlights stories of those messengers who were praised or blamed by the ancient kings and wise men. Like any authentic adab text, the treatise comprises passages of prose and poetry with regards to this topic.11 The Muslim envoy of Caliph al-Muʿtaṣim (218–227/833–842) to the Byzantine emperor assigns the function of messenger third place in the caliphate, after the military and the judiciary: “The caliphs have servants who perform different types of service […] they work in their special field and nothing else is required of them. Some are trained for conquest and so they take up weapons and lead armies; others are 8 9 10 11

Ibn Saʿd, Ṭabaqāt i, 258–259. Ali, Arabic literary salons 33–35. Vaiou (trans.), Diplomacy 1 and 16. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 21–22.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

409

prepared for judiciary service […] and others like me, are fit to be sent by caliphs to kings and to carry their messages.”12 Kitāb Rusul al-mulūk starts with Quranic injunctions and citations from verses in chapters 2 (al-Baqara), 4 (al-Nisāʾ), 5 (al-Māʾida), and others. Quotations refer to the Prophet Muhammad, the ideal messenger. This ethical framing underlies the virtues expected of envoys as well as the principles of conduct for messengers/prophets.13 This highly important role requires specific qualities, attributes, and skills. According to Ibn al-Farrāʾ, the envoy should be a man: of sound judgment, eloquent, shrewd, and resourceful […] who seizes every opportunity […] One whose tongue is sharp. […] One who understands the subtleties of conduct […] when wishing to dispute something on your behalf, he should surpass the most famous debaters […] with a knowledge of law, traditions, regulations, and biographies so that he may follow the example of those who preceded him […] he should know the rules of taxation and account-keeping […] Let him be from noble and respectable families. […] When you find these qualities in one, make him part of your entourage (biṭāna) and disclose to him all your affairs, large and small and consult with him.14 The messenger should have a loud and clear voice, should be well-known, handsome, dignified, firm, courageous, and eloquent. He should represent his sovereign faithfully.15 He needs to be patient and in full control of his temper, as a disrespectful ruler may insult him. If the messenger is overcome by anger and rage, his determination and resolution will be undermined and he will therefore fail to carry out his mission successfully.16 The messenger should not be inferior to his sender “in wisdom and reason, appearance and nobility.” If anything, the messenger should exceed his sender in these matters in order to reflect positively on him.17 The fear of betrayal is central. The messenger should be “guarded and honest without excessive anguish and modesty so that he may not be motivated, by what is shown to him as a great favour, to sell his religion and betray his

12 13 14 15 16 17

Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 65; translation in Vaiou (trans.), Diplomacy 86–87. I have depended on Vaiou’s translation with some adjustments in wording. Vaiou (trans.), Diplomacy 28–29. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 33–34; Vaiou (trans.), Diplomacy 66. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 35–37; Vaiou (trans.), Diplomacy 67. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 40; Vaiou (trans.), Diplomacy 71. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 44; Vaiou (trans.), Diplomacy 73.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

410

van berkel, el cheikh and osti

sender or sell his trust.”18 It is crucial that the messenger not exceed the purpose of his mission and neither misjudge nor misunderstand the one who has charged him with this task. The messenger should thus “deliver the letter in accordance with its content.”19 The envoy of Caliph al-Muʿtaṣim stressed the fundamental importance of trust in his reply to the Byzantine emperor: “If they did not have trust in me and were not sure of whether I was honest and truthful in bringing and delivering a message, they would not have thought me fit to be sent to you.”20 The importance of this element will be seen below in one of our examples. Ibn al-Farrāʾ delineates the physical characteristics of the ideal messenger: “He should have a perfect figure and a well-built body, so that he should not look short or small […] the way one looks makes an impression before one speaks and the body hides the mind […]. The kings […] make that a compulsory quality for their messengers.”21 Messengers with remarkable moral, physical, verbal, and intellectual qualities were selected. The way they dressed, the gifts they brought, and their general attire had to impress. One of the goals of embassies was to project power and prestige through the demeanor and attire of the envoy as well as through the grandiose gifts he carried. This was the case of the envoy of Caliph al-Muʿtaṣim to the Byzantine emperor, who is said to have beheld his “dignified (hayba) appearance” and “the large quantity of his embellishments, and the baggage and equipment he brought with him.”22 Given that the most important task of the messenger is to deliver a verbal or written message, eloquence is central to this role. In his adab anthology, al-ʿIqd al-farīd, Ibn ʿAbd Rabbih (d. 328/940) talks about “the magic of eloquent speech that affects the soul with its subtlety and agrees with it in finesse. Fine speech captivates hearts, for it can sometimes win over someone burning with rage.”23 He defines eloquence as “the choice of words and making others understand well” or “fewness of words and conciseness of right ideas.” The eloquent man is thus described as someone “whose words are few and who achieves decisive ideas and meanings.”24 Quoting from Kitāb Akhlāq al-mulūk, Ibn al-Farrāʾ confirms that “the messenger should possess the right instinct and temperament, verbal competence, and an ability to enunciate […]; He should be cognizant

18 19 20 21 22 23 24

Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 44; Vaiou (trans.), Diplomacy 73. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 32. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 65; Vaiou (trans.), Diplomacy 87. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 47; Vaiou (trans.), Diplomacy 75. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 64–65; Vaiou (trans.), Diplomacy 86–87. Ibn ʿAbd Rabbih, al-ʿIqd al-farid ii, 122; Boullata (trans.), Unique necklace ii, 1. Ibn ʿAbd Rabbih, al-ʿIqd al-farid ii, 261–262; Boullata (trans.), Unique necklace ii, 102.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

411

of the articulation of sounds […]; He should convey the words and meanings of the king […] and safeguard what he has been asked to deliver.”25 In addition to trustworthiness, eloquence, and good looks, one additional attribute of the ideal messenger involves him advising the ruler. Caliph al-Manṣūr (136– 158/754–775) is said to have been inspired by the advice of an envoy of Emperor Constantine v (741–775ce) in deciding how to lay out Baghdad. Asked for his evaluation about the new capital, the Byzantine ambassador said: Everything I saw is great and noble except for three things […]; The soul is green and it [Baghdad] does not have greenery; water is life and you do not have life [i.e., water]; your enemy lives with you, meaning his subjects in the markets. The markets were in proximity to the palace. […] When the messenger left, al-Manṣūr investigated the matter and realized that the messenger was correct in his advice. He consequently built al-ʿAbbāsiyya, diverted water from the Karkhāyā river and ordered that the markets be moved to al-Karkh.26 Here, a foreign messenger is ascribed great influence, successfully interfering in effecting the urban planning of the new Abbasid capital. Writing in the fifth/eleventh century, the vizier Niẓām al-Mulk added another function to the Muslim envoy: “It should also be realized that when kings send ambassadors to one another their purpose is not merely the message of the letter which they communicate openly, but secretly they have a hundred other points and objectives […],” gathering information about roads and passes, the size of the army, the court protocols, the king’s hobbies, and the company he keeps, as well as his “personal qualities and manners and his designs and intentions.”27 Like other adab texts, this treatise focuses on a performance based on a scrupulous sense of propriety, the elegant, decorous, and urbane being merely the outer signs of inner virtue and harmony. Adab here showed itself in the context of embassies and in the qualities, demeanor, and eloquence of the envoy. The envoy needs to master his words, his gestures, his facial expressions, his feelings, and his temper. He is expected to follow codes of behavior and ethical norms.

25 26 27

Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 50; Vaiou (trans.), Diplomacy 77. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk 75–76; Vaiou (trans.), Diplomacy 92. Al-Khaṭīb alBaghdādī includes this episode in Taʾrīkh Baghdād i, 80. Nizam al-Mulk, Book of government 95.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

412 3

van berkel, el cheikh and osti

From Theory to Practice

By and large, then, advice literature, when laying down norms and ideals for messengers, refers to diplomatic missions and not to negotiations between actors who, at least in theory, acknowledge the same caliphal authority. To be sure, we do have other sources: historical narratives provide many examples of this second type of messenger, describing their role and assessing their behavior. However, these narratives do not describe a new profession with a specific code of conduct. In fact, it may be argued that internal messengers were not a professional category at all, given that most of them had official positions in the administration, army, or court—they were, for example, judges, princes from the caliphal household, or army leaders. Still, there is no doubt that, formal category or not, they did exist, and they mostly came from among the ranks of the kuttāb, the bureaucrats. In fact, if we go back to normative literature, most of the ideal qualities of diplomatic envoys align with the ideal qualifications of kuttāb laid down in administrative literature (adab al-kātib texts), which reached its peak in the same period, the late third/ninth and early fourth/tenth century. In their manuals authors from this period, such as Qudāma b. Jaʿfar, al-Ṣūlī, and ʿAbdallāh al-Baghdādī, emphasize very similar qualities and skills for the kuttāb as those described by Ibn al-Farrāʾ in his Kitāb Rusul al-mulūk. They mention, for example, that the kuttāb need to be eloquent, loyal, and well-mannered. Speaking of the ideal envoy, Ibn al-Farrāʾ states that he should have knowledge of law, traditions, regulations, and biographies, as well as taxation and account-keeping. This is the kind of knowledge that is specific to the kuttāb. Another important ideal qualification of both the kātib and the envoy is their trustworthiness, which Ibn al-Farrāʾ’s text and the adab al-kātib manuals describe in similar terms. A conspicuous difference between the two, however, is that while kuttāb are mostly praised for their eloquence in writing, it is the (oral) debating skills of envoys that are stressed. The emphasis on writing and good penmanship in normative literature for kuttāb might also explain why we find so few ideal descriptions of internal messengers in this literature: the bureaucratic ideal was one of a written culture; from this perspective, the internal messenger is the informal replacement of this formal written bureaucracy. He is the oil that made the machinery run, but he does not represent the bureaucratic ideal as presented in the administrative literature.28

28

Van Berkel et al., Crisis and continuity 99–103.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

413

On a more practical level, it should not surprise us to find many kuttāb among the internal messengers of the caliphate. They were the ones with sound understanding and expert knowledge on the financial matters that often had to be negotiated between Baghdad and regional powers during these missions. Indeed, while the skills of the epistolary, and not the financial bureaucrat, epitomize the ideal image in the administrative literature, in everyday life the financial kuttāb were in great demand, and precisely these financially skilled bureaucrats are prominent among the internal messengers.29 This focus on financial expertise might be another reason for the lack of descriptions of the internal messenger in the administrative literature. To be sure, this might have been different for diplomats who negotiated with foreign rulers and who were not, in the first instance, required to deal with detailed internal financial matters but rather with borders, exchanges of prisoners, or commercial exchanges. Still, as mentioned, Ibn al-Farrāʾ lists knowledge of taxation and account-keeping among the requirements for diplomats. Two of these kuttāb who operated as internal messengers, al-Ḥasan b. Hārūn and al-Kūfī, form the core of our next investigation into the role and position of the internal messenger as gleaned from the historical narratives. They were both classically trained kuttāb whose circumstances diverged from those of earlier generations in that, rather than rising through the ranks of a chancery, they traveled between cities and served different masters. In other words, their careers reflect the political instability of their time, as will be seen presently.

4

Al-Ḥasan b. Hārūn

The career of our first case study, Abū ʿAlī al-Ḥasan b. Hārūn al-Hamadhānī (fl. first half of the fourth/tenth century), strikingly illustrates the changing political and administrative structures of the period: over the course of 21 years, he appears in the sources sporadically, every time in a different office, in a different place, and at the service of a different authority. He is first mentioned in Miskawayh’s chronicle for the year 314/926–927. At this time, he is serving Yūsuf b. Abī l-Sāj, the leader of Armenia and Azerbaijan. Ibn Abī l-Sāj, who is advancing on Kufa, sends al-Ḥasan ahead to Wāsiṭ to negotiate with the vizier of the caliph al-Muqtadir, al-Khaṣībī. Miskawayh describes al-Ḥasan as “a clerk of Yūsuf b. Abī l-Sāj, who used to serve him in private business under the auspices of a certain person named Abū ʿAbdallah Muḥammad b. Khalaf alNirmānī.”30 29 30

Van Berkel et al., Crisis and continuity 103–107. Miskawayh, Tajārib i, 148.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

414

van berkel, el cheikh and osti

Viziers, 313–334/925–945 With dates of accession Al-Muqtadir (295–320/908–932) – Aḥmad al-Khaṣībī 313/925 – ʿAlī b. ʿĪsā 315/927 – Ibn Muqla 316/928 – Sulaymān b. al-Ḥasan 318/930 – ʿUbaydallāh al-Kalwadhānī 319/931 – Al-Ḥusayn b. al-Qāsim 319/931 – Al-Faḍl b. Jaʿfar b. al-Furāt 320/932 Al-Rāḍī (322–329/934–940) – ʿAbd al-Raḥmān b. ʿĪsā 322/934 – Ibn Muqla 322/934 – ʿAbd al-Raḥmān b. Īsā 324/936 – Abū Jaʿfar al-Karkhī 324/936 – Sulaymān b. al-Ḥasan 324/936 Al-Muttaqī (329–333/940–944) – Aḥmad b. Maymūn 329/941 – Abū ʿAbdallāh al-Barīdī 329/941 – Al-Qarārīṭī 329/941 – Abū Jaʿfar al-Karkhī 329/941 – Abū ʿAbdallāh al-Barīdī 330/941 – Al-Qarārīṭī 330/941 – Abū ʿAbdallāh al-Barīdī 330/941 – Al-Qarārīṭī 330/942 – Aḥmad b. ʿAbdallāh al-Iṣfahānī 331/943 – ʿAlī b. Muḥammad b. Muqla 331/943 figure 21.1

Viziers appointed between 313/925 and 334/945

In the negotiations with al-Khaṣībī, which will be looked at closer later, al-Ḥasan proves his value to Ibn Abī l-Sāj, and the following year, 315/927–8, after al-Ḥasan conspires against his direct superior Muḥammad b. Khalaf alNirmānī, Ibn Abī l-Sāj dismisses al-Ḥasan’s superior and appoints him in his place. According to al-Ṣūlī, al-Ḥasan subsequently takes part in the battle of Kufa against the Qarāmiṭa, a Shiʿite rebel movement operating from Baḥrayn. During this battle, al-Ḥasan is said to have handled the sword himself, even

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

415

getting wounded in the heat of the battle.31 The combination of military and bureaucratic roles in al-Ḥasan is noteworthy in a time when the people of the pen and the people of the sword were generally presented as distinct categories, each with their own role and identity within society. Al-Ḥasan’s associations with military leaders continues in the later stages of his career. We next hear of him in 320/932, when he is appointed secretary of ʿAlī b. Yalbaq, chamberlain (ḥājib) to the caliph al-Qāhir and son of the army leader Yalbaq.32 This employment does not last long: a few months later, following an unsuccessful plot to unseat al-Qāhir, ʿAlī b. Yalbaq is arrested together with his co-conspirators. Crucially, the arrest is carried out by the Sājī troops, the former army of Yūsuf b. Abī l-Sāj, with whom al-Ḥasan had fought against the Qarāmiṭa a few years earlier.33 Despite al-Ḥasan’s previous association with the Sājīs, the arrest of his master is a sign for him that the tables are turning and that he should go into hiding. The following year, however, al-Ḥasan is collaborating again with both the Sājī and the Ḥujarī troops in a plot against al-Qāhir, who is eventually deposed. In 323/934–935 he flees from Jabal, where he had apparently held the office of governor, to avoid arrest.34 After his arrival in Baghdad, he manages to negotiate a safe conduct by paying a fine to the vizier Ibn Muqla.35 From this moment onward, al-Ḥasan b. Hārūn becomes more prominent as a representative of the central administration in Baghdad. The next step in al-Ḥasan b. Hārūn’s career is in 324/935–936, when he turns up at the court of al-Rāḍī in the company of the former viziers al-Khaṣībī and Sulaymān b. al-Ḥasan, both returning to Baghdad after the fall of Ibn Muqla. All three are treated with honor by the new vizier, ʿAbd al-Raḥmān b. ʿĪsā. In referring to al-Ḥasan together with these two senior Baghdadi kuttāb, Miskawayh seems to imply that al-Ḥasan is becoming prominent in the capital.36 In the same year, he leaves Baghdad again when he is appointed head of the administration of the land tax in Isfahan and Fars, while Badr al-Kharshanī is appointed head of public security.37 Apparently the two had been closely related

31 32 33 34 35 36 37

Al-Ṣūlī, Akhbār 268. Miskawayh, Tajārib i, 242–243. See van Berkel et al., Crisis and continuity 124; Kennedy, Armies of the caliphs 162. See also al-Ṣūlī, Akhbār 268. Al-Ṣūlī, Histoire i, 125, n. 2. Al-Ṣūlī, Akhbār 70. Miskawayh, Tajārib i, 337. Miskawayh, Tajārib i, 338.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

416

van berkel, el cheikh and osti

Amīr al-umarāʾ, 324–334/936–945 Ibn Rāʾiq (324–326/936–938) Bajkam (326–329/938–941) Kurankij (329/941) Ibn Rāʾiq (329–330/941–942) Ibn Ḥamdān, Nāṣir al-Dawla (330–331/942–943) Tūzūn (331–334/943–945) figure 21.2

The first five amīr al-umarāʾ, 324–334/936–945

for some time, because Badr al-Kharshanī is said to have lived in al-Ḥasan’s house in Baghdad.38 Once again, al-Ḥasan’s position does not last long: later in the chronicle of the same year and without further comment, Miskawayh mentions that Ibn Rāʾiq, once appointed to the new office of amīr al-umarāʾ, proceeded to arrest al-Ḥasan together with the Sājī troops with whom he had collaborated a few years earlier and who were now stationed in Wāsiṭ. The following year Ibn Rāʾiq puts all the Sājīs to death, but for unspecified reasons, he spares al-Ḥasan’s life.39 The appointment of Ibn Rāʾiq as the first amīr al-umarāʾ has highly symbolic value in Miskawayh’s chronicle: it annihilates the power of the vizier and represents the definitive end of the central military and administrative authority over the provinces.40 The disbandment of the Sājī troops, which had been loyal to the caliph al-Rāḍī, is narrated in this context. That al-Ḥasan was spared the fate of the Sājīs is a testimony to his political savvy and versatility. When we hear about him again, in fact, he is still occupying key positions: from 328/939 to 330/941–942 and then once more in 331/943, he is mentioned as governor of Kufa,41 and slightly later Miskawayh lists him among the officers (quwwād, once again a military role) who accompanied the Hamdanid general Sayf al-Dawla in his flight from Baghdad on the approach of his rival Tūzūn.42 The following year, al-Ḥasan is mentioned as fleeing with the caliph al-Muttaqī from Mosul when Tūzūn advances on that city, and later on he acts as messenger for al-

38 39 40 41 42

Al-Ṣūlī, Akhbār 82. Miskawayh, Tajārib i, 351. In particular, see Miskawayh, Tajārib i, 352. Miskawayh, Tajārib ii, 26; al-Ṣūlī, Akhbār 139 and 240. Miskawayh, Tajārib ii, 26 and 44.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

417

bureaucrats on the move

Muttaqī in the negotiations with Tūzūn for the former’s return to Baghdad.43 His role as messenger on this occasion will be discussed in more detail below. Two years later, we find al-Ḥasan b. Hārūn in yet another influential position: he is appointed to personally attend to the first Buyid ruler of Iraq, Muʿizz al-Dawla, and cater to him, his staff, and his household.44 In 335/946–947 he is said to have gone into retirement after being attacked by his former superior Abū Jaʿfar al-Ṣaymarī.45 Finally, Miskawayh mentions him one last time, in 339/950–951, as one of the potential candidates for the personal secretaryship of Muʿizz al-Dawla. He does not get the position, and we never hear of al-Ḥasan b. Hārūn after this. Miskawayh does not explicitly comment on al-Ḥasan b. Hārūn’s character and career. He refers to him on several occasions in his Tajārib, but it is hard to infer al-Ḥasan’s personal characteristics or merits from these (often short) descriptions. Al-Ṣūlī, however, is explicit about the qualities of al-Ḥasan b. Hārūn, whom he clearly admires. He devotes a long paragraph to his qualities, referring to him as one of the main kuttāb of his age, with a good reputation and a glorious career.46 In this same section, al-Ṣūlī also provides us with a short biography of al-Ḥasan, describing him as a poet, and highlights the qualities Caliph al-Rāḍī saw in him. In many ways, al-Ṣūlī presents al-Ḥasan b. Hārūn as a model messenger and ideal secretary and servant. Interestingly, al-Ṣūlī not only praises him as man of the pen but also sees him as courageous and a good horseman (ḥasan al-furūsiyya), qualities generally attributed to the people of the sword. These qualifications are in line with al-Ḥasan b. Hārūn’s exceptional career in which he occupied both secretarial and military positions.47 It is not possible to establish whether al-Ḥasan is representative of his contemporaries or an outlier—especially the military side, which does not seem to be usual among his colleagues. For the purposes of this paper, we concentrate on elements that both Miskawayh and al-Ṣūlī emphasize: his constant moving up and down between center and periphery, and his frequent role as a mediator. As such, his position clearly represents the informal bonds and personal relations of his age. These latter aspects are evident in another kātib of this period, Abū ʿAbdallāh Aḥmad b. ʿAlī al-Kūfī.

43 44 45 46 47

Miskawayh, Tajārib ii, 67; al-Ṣūlī, Akhbār 267–269. Miskawayh, Tajārib ii, 91. Miskawayh, Tajārib ii, 106. Al-Ṣūlī, Akhbār 267–269. Al-Ṣūlī, Akhbār 267–269.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

418 5

van berkel, el cheikh and osti

Al-Kūfī

Our second case study, Abū ʿAbdallāh Aḥmad b. ʿAlī al-Kūfī (d. after 331/943) is a contemporary of al-Ḥasan b. Hārūn and had a similarly complicated career. We can follow him in the sources only for the span of ten years, but within this time we observe him, like al-Ḥasan, moving between different roles, places, and employers. Like al-Ḥasan, his trajectory illustrates the opportunities a skillful person could create for himself in such unstable times, rising through the ranks and acquiring control of key functions. However, whereas, as we have seen, al-Ḥasan occasionally assumed military roles, al-Kūfī’s activity is strictly connected with financial matters. His first recorded post is under Isḥāq b. Ismaʿīl al-Nawbakhṭī, tax farmer for Wāsiṭ and the irrigation of the Euphrates, who was buried alive by Caliph alQāhir in 322/933. Al-Kūfī was arrested together with his employer a few months earlier, but he did not suffer the latter’s fate; on the contrary, he is mentioned in the same year in Baghdad as secretary to the sometimes vizier al-Qarārīṭī. In 323/934, he placed himself under the protection of the vizier Ibn Muqla and was appointed secretary to the latter’s son, Abū l-Ḥusayn. However, according to Miskawayh, he was only waiting for a chance to get away, and this came when Abū ʿAbdallāh al-Barīdī (d. 332/944), who controlled Ahwāz, sent a dispatch “stating that there was no chance of any money being transmitted to the capital at that time,” claiming that he had spent everything to pay the army. Al-Kūfī then suggested that he be sent to Ahwāz to negotiate. As soon as he got there, al-Kūfī wrote to Abū l-Ḥusayn b. Muqla that he was unable to complete his mission, and he remained with al-Barīdī, for reasons which will be looked into below, until the general Ibn Rāʾiq took control in the capital as the first to hold the title of amīr al-umarāʾ in 324/936. At this point, al-Kūfī persuaded al-Barīdī that he would be able to sow discord between Ibn Rāʾiq and his secretary, al-Ḥūsayn b. ʿAlī al-Nawbakhṭī, who was al-Barīdī’s old enemy. Indeed, in 325/937 al-Kūfī was dispatched to Baghdad as an emissary of al-Barīdī, and not long afterwards Ibn Rāʾiq appointed him as his secretary, replacing al-Nawbakhṭī. When, in 326/938, Ibn Rāʾiq was forced to flee Baghdad upon the arrival of the new amīr al-umarāʾ Bajkam, al-Kūfī was arrested and fined. However, he must have managed to work his way into the administration of the new amīr, as in 329/940 he entered the service of Ibn Shirzād, secretary to Bajkam. When the latter had Ibn Shirzād arrested, al-Kūfī replaced him, retaking charge of finances (al-naẓar fī l-amwāl).48 For the next two years

48

Miskawayh, Tajārib i, 415.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

419

we see him moving between Baghdad and Wāsiṭ to conduct Bajkam’s business in the capital and bring him the sums he gathered. After the death of Caliph al-Rāḍī, he steered the decision on who should succeed as new caliph toward Bajkam’s choice. After the death of Bajkam in the same year, al-Kūfī became the deputy of the new vizier, Aḥmad b. Maymūn. However, when Abū ʿAbdallāh al-Barīdī marched on Baghdad, he went into hiding, fearing repercussions for having served al-Barīdī’s rival Ibn Rāʾiq. Still in the same year, when Ibn Rāʾiq returned to Baghdad, al-Kūfī reentered his service, taking charge of finances once again. In 331/943 we find him in the same position, this time serving under the Hamdanid ruler Nāṣir al-Dawla, who in turn sent him to his brother Sayf al-Dawla in Wāsiṭ to deliver payment for the troops.49 The last mention of al-Kūfī shows him fleeing Baghdad with Sayf al-Dawla, possibly in the same group as al-Ḥasan b. Hārūn, on the arrival of the new amīr Tūzūn.50 The sources do not seem to have a good opinion of al-Kūfī. While it is customary for the person in charge of finances to have his predecessor tortured in the hope of extracting money, Miskawayh mentions a few cases where alKūfī treated his prisoners with particular cruelty. Al-Ṣūlī, besides recording personal interactions with al-Kūfī, who had been in charge of paying him on several occasions, mentions that, when Ibn Rāʾiq was preparing to flee the capital on the arrival of Bajkam in 326/938, the population cried at his passage “this is God’s punishment for having al-Kūfī as secretary and giving him absolute power! (li-stiktābika al-Kūfī wa-taslīṭika iyyāhu ʿalā l-nās).”51 On the other hand, only a few pages later he claims to have always sung the praises of alKūfī to Caliph al-Rāḍī, “assuring him of his competence and his loyalty” and listing the professional and personal qualities he had been able to observe and experience, having long been a neighbor and friend of his.52 And after all, more than one general had deemed al-Kūfī loyal enough to entrust him with the delivery of large sums of money. In particular, Nāṣir al-Dawla had him deliver 400,000 dīnār to his brother in Wāsiṭ for the payment of his army. When alKūfī found the troops in revolt, al-Ṣūlī says that he brought the money back to Baghdad.53 Al-Kūfī’s ambiguous reputation and unstable career, constantly on the brink of disgrace, reflect the political instability of the period. Like al-Ḥasan b. Hārūn,

49 50 51 52 53

Al-Ṣūlī, Akhbār 241. Miskawayh, Tajārib ii, 45. Al-Ṣūlī, Akhbār 106. Al-Ṣūlī, Akhbār 107. Al-Ṣūlī, Akhbār 241.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

420

van berkel, el cheikh and osti

he frequently changes role and employer and frequently travels outside the capital, not privately in periods of disgrace as ʿAlī b. ʿĪsā but in an official capacity as the emissary of an employer.

6

Four Missions

Al-Ḥasan b. Hārūn and al-Kūfī share three basic features: they are competent administrators, serve under many different masters, and are constantly on the move between Baghdad and the neighboring cities, not only following their employers but also going on missions on their behalf. The fact that al-Kūfī was unquestionably a Baghdadi, while al-Ḥasan’s training was presumably outside the capital—or at least his first recorded employment was—does not seem to play a part in their professional attributes. At the same time, the sources at our disposal evaluate them differently. This is especially clear in the accounts of their work as messengers. We have selected the accounts of four missions— two carried out by al-Ḥasan, two by al-Kūfī—which will now be looked at in detail. The missions differ in content, the way they are related, and the amount of detail given in the sources. Taken together, they provide information about the role and agency of internal messengers, which can be compared with the normative literature on envoys illustrated in the previous section. The four missions in chronological order: – in 314/926–7, al-Ḥasan b. Hārūn was sent by Ibn Abī l-Sāj to negotiate with the vizier al-Khaṣībī; – in 323/934, al-Kūfī was sent by the vizier’s son Abū l-Ḥusayn b. Muqla to negotiate with al-Barīdī; – in 324/935, al-Barīdī sent al-Kūfī as his representative to amīr al-umarāʾ Ibn Rāʾiq; – in 332/943–4 al-Ḥasan b. Hārūn negotiated with amīr al-umarāʾ Tūzūn in the name of the caliph al-Muttaqī.

7

Al-Ḥasan b. Hārūn for the Governor: Loyal and Firm Negotiator

As mentioned earlier, Miskawayh refers to al-Ḥasan b. Hārūn for the first time in the year 314/926–927 as a private clerk of Yūsuf b. Abī l-Sāj.54 At that time, Ibn Abī l-Sāj is on his way to Wāsiṭ with an army to support Iraq against attacks

54

Miskawayh, Tajārib i, 148.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

421

by the Qarāmiṭa. Authorities in Baghdad ask him to take up this task since the troops stationed in the capital were proving unable to counter the attacks. Ibn Abī l-Sāj sends al-Ḥasan b. Hārūn ahead of the troops to negotiate the conditions for the payment of the army with the vizier al-Khaṣībī. It is unclear whether these negotiations take place in Baghdad or Wāsiṭ—most probably in Baghdad, since we have no information about the vizier being in Wāsiṭ at the time. It is also uncertain whether al-Ḥasan directly negotiates with the vizier or through the vizier’s clerks, but the latter seems more probable: while describing the process of negotiation, Miskawayh states that “the clerks demanded that alḤasan b. Hārūn should accept a stipulation.”55 If we look at the description of the process and the outcome of the deal alḤasan b. Hārūn was able to broker for his master, it is clear that Miskawayh sees him as a firm negotiator who operates in his master’s interest. Al-Ḥasan asked for a substantial amount, 5,000 dīnār a month, for the payment of the troops of Ibn Abī l-Sāj, arguing that his master was not inferior to Aḥmad b. Ṣuʿlūk, ruler of Rayy, who was said to have received a similar amount for his troops during the reign of Ḥāmid b. al-ʿAbbās.56 This monthly stipend is received by assigning the revenues from the Eastern provinces to his master. Then Miskawayh mentions a salient detail of the negotiations: “The kuttāb demanded that al-Ḥasan b. Hārūn should accept a stipulation that the central government should send a paymaster (munfiq) to expend the revenues of those regions upon the troops and the retainers of Ibn Abī l-Sāj. Al-Ḥasan b. Hārūn accepted all their demands and signed the documents containing them except this of the paymaster.”57 This narrative demonstrates al-Ḥasan’s political capacity and loyalty towards his master. It shows that he is an astute negotiator and one who is also firmly committed to his master’s interest through the further stipulations of the deal. When the clerks demand that he should accept a paymaster from the central government to distribute the revenues from the Eastern provinces to the troops, al-Ḥasan refuses. He realizes that this is not the kind of intervention his master would appreciate: Yūsuf b. Abī l-Sāj would not want to be seen as someone “who could not be trusted to be honest with his own men.”58

55 56

57 58

Miskawayh, Tajārib i, 148. Aḥmad b. Ṣuʿlūk is elsewhere referred to as the brother of Ṣuʿlūk by Miskawayh. The precise details of the deal this person was able to negotiate are unclear. Miskwayh mentions a messenger of the brother of Ṣuʿlūk visiting Baghdad in 308/916–917, bringing money and gifts, and receiving a robe of honor. Miskawayh, Tajārib i, 75. Miskawayh, Tajārib i, 148. Miskawayh, Tajārib i, 148.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

422

van berkel, el cheikh and osti

In his description of the negotiation process, Miskawayh approves of alḤasan b. Hārūn’s mission and firm position. Our other source, al-Ṣūlī, does not describe these specific negotiations between al-Ḥasan and the clerks of alKhaṣībī, but he does mention another interesting detail related to this moment in time. While Yūsuf b. Abī l-Sāj is in Wāsiṭ waiting to start his attack on the Qarāmiṭa, al-Ḥasan is not his only contact or messenger in Baghdad: al-Ṣūlī himself exchanges letters with Yūsuf b. Abī l-Sāj at the same time. He also reveals that Ibn Abī l-Sāj has informants among the palace eunuchs.59 Whether any of these other contacts were able to positively influence al-Ḥasan b. Hārūn’s mission is unclear. Nevertheless, we may conclude that Ibn Abī l-Sāj had various networks/connections at his disposal, both formal (through his messenger al-Ḥasan b. Hārūn) and less formal (through the eunuchs of the palaces) to exchange information and negotiate his interests with the center. At the same time, this mission and its context show the growing influence and power of regional leaders such as Ibn Abī l-Sāj, who had to be called upon in situations of military crisis. Although at this point regional leaders such as Ibn Abī l-Sāj still operated at the request of the center, they would soon aim to take over fully.

8

Al-Kūfī: Mission as an Escape Route

Al-Kūfī’s first mission brings us a decade forward, under a new caliph and after eight changes of viziers with increasingly short terms of office. In 323/934, Abū ʿAlī Ibn Muqla was appointed vizier for the third time by the newly installed caliph, al-Rāḍī. Al-Kūfī, who had been working under the kātib al-Qarārīṭī, now in disgrace, put himself at the disposal of Abū l-Ḥūsayn, son and assistant of Ibn Muqla. When the latter left for Mosul, leaving his son as de facto vizier, al-Kūfī attended his majlis, professing “sincere attachment; he was however anxious to get away from him and out of his reach.”60 An opportunity arose when Abū ʿAbdallāh al-Barīdī, a tax farmer who had seized control of Ahwāz, sent a dispatch “stating that there was no chance of any money being transmitted to the capital at that time,” claiming that he had spent everything to pay the army. Al-Kūfī managed to persuade Abū l-Ḥusayn that he should be sent to Ahwāz to negotiate with al-Barīdī. Miskawayh describes the process as follows: Abū lḤusayn wrote to al-Barīdī “that he could not accept any excuse for the delay in

59 60

Al-Ṣūlī, Akhbār 27. On this episode, see Osti, History and memory 48–49, 121–123. Miskawayh, Tajārib i, 327.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

423

transmitting the money, and that Baridi had compelled him to send Abū ʿAbdallāh Aḥmad b. ʿAlī al-Kūfī to discuss the financial question with him, and exact payment. The letter was despatched and Aḥmad b. ʿAlī followed it to Ahwāz.”61 However, as soon as he was in Ahwāz al-Kūfī gave up the negotiations, writing to Abū l-Ḥusayn that he was unable to carry out his mission. Al-Kūfī had thus achieved his aim to get away from Ibn Muqla and his son. Miskawayh also provides the text of a letter by Ibn Muqla, where the vizier, who saw that negotiations have failed, bypasses al-Kūfī and addresses al-Barīdī directly: A plague on al-Kūfī who has thwarted me; I sent him to win you to my side and he has turned you against me; he has aroused your cupidity, and you in your greed have listened to him. By God, I will amputate his hands and feet. As for you, I have hopes that you will not persist in your ingratitude for my favours and benefits, and that reflection will lead you to a proper sense of your obligations; that you will give me cause for satisfaction with you, and give me your help in this difficult situation to the like of which no occupant of my post under any dynasty was ever reduced; that you will avert what threatens me by a remittance of money, and thereby save your two fortunes, one of which is in my hands and the other in your own.62 In other words, Ibn Muqla blames the messenger, and it seems that he has reason to do so: al-Kūfī, violating the firm rule outlined by Ibn al-Farrāʾ, has betrayed the trust of the person who sent him. It is not unreasonable for Miskawayh to deduce that al-Kūfī may have feared that al-Barīdī would be persuaded by the vizier. Indeed, al-Kūfī seems to have been uncertain of his reception at Ahwāz: according to Miskawayh, he “felt apprehensive of al-Barīdī and wished to get away from him, being afraid of his impulses.”63 Again, al-Kūfī’s strategy was to deflect the attention away from himself and onto a potential need for a mission that would send him away. This time, he suggested that he might be able to sow discord in the entourage of Ibn Rāʾiq, who by that time had taken control of Baghdad as amīr al-umarāʾ, and get himself appointed as his secretary, where he could better represent the interests of Abū ʿAbdallāh al-Barīdī. Once again, as will be seen below, this came to pass.

61 62 63

Miskawayh, Tajārib i, 327. Miskawayh, Tajārib i, 329. Miskawayh, Tajārib i, 327.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

424

van berkel, el cheikh and osti

We may already point out two crucial differences between the account of al-Ḥasan’s mission in 314/926–927 and al-Kūfī’s in 323/934. The first concerns content: while we know the gist of al-Ḥasan’s negotiating process, al-Kūfī does not seem to attempt negotiation at all—in fact, most of Miskawayh’s attention is on al-Kūfī’s personal situation. The second difference is stylistic: Miskawayh narrates al-Ḥasan’s actions from an external point of view, whereas he spells out his own interpretation of al-Kūfī’s motivations, providing two pieces of evidence from the protagonists: Ibn Muqla’s letter, and an account in al-Kūfī’s own voice of comments he made years later in private: I never had a more agreeable time in my life than that which I spent with al-Barīdī. I was in his house for about a year, without occupation, not included among his underlings, and not worried with having to attend to any business. He associated with me in the friendliest manner, and I obtained from him in gold, silver and articles sent me the value of thirtyfive thousand dinars. I did not leave Ahwaz till I had been made secretary to Ibn Rāʾiq. I had by then been secured against Ibn Muqla through his arrest; from him I had reason to fear, and praise be to God who did not let him leave the world until he was as utterly ruined as he had ruined it. May God send his son after him, for he is the worse of the two; both are equally shameless, hard and mean; only the father with all his faults occasionally showed mercy and generosity to his attendants and his household though not to strangers.64 Al-Kūfī’s return to Baghdad proceeds along similarly tortuous lines.65 After Ibn Rāʾiq’s appointment as amīr al-umarāʾ, al-Barīdī gained an ally in Muḥammad b. ʿAlī Ibn Muqātil (d. 350/961), a merchant of humble origins who had come up in the world thanks to his attachment to Ibn Rāʾiq’s secretary al-Nawbakhṭī but was now prepared to betray his patron. Against the advice of al-Nawbakhṭī, Ibn Muqātil persuaded Ibn Rāʾiq to give al-Barīdī the farming of Ahwāz in exchange for 360,000 dīnār, to be paid in monthly installments. He further persuaded the amīr to receive al-Kūfī as the representative of al-Barīdī as a sign of respect to the latter. Thus, al-Kūfī arrived back in Baghdad as the representative of al-Barīdī at the court of Ibn Rāʾiq. The alliance with Ibn Muqātil continued: during an illness of al-Nawbakhṭī, Ibn Muqātil persuaded Ibn Rāʾiq that his

64 65

Miskawayh, Tajārib i, 328. Miskawayh, Tajārib i, 361–364.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

425

secretary would not recover enough to be able to serve him again and that alKūfī should be appointed in his place: He said: “Abū ʿAbdallāh Aḥmad b. ʿAlī al-Kūfī here is the equal of alNawbakhtī …. Al-Kūfī is absolutely trustworthy and honest and is on an intimate footing with Abū ʿAbdallāh al-Barīdī. If you appoint al-Kūfī secretary, you will have a capable, honest, and painstaking minister, and in addition you will draw those people into your party and secure their adhesion. We shall make al-Barīdī recognize that we have granted his request and made him your secretary, appointing as his representative Abū ʿAbdallāh al-Kūfī.” Ibn Rāʾiq replied: “Ask God’s blessing on your scheme and carry it out. Only you are to be responsible if al-Kūfī plays me false or favours al-Barīdī against me in any matter.” Ibn Muqātil replied: “I am prepared to guarantee on behalf of al-Kūfī all that is stipulated by the amīr.”66 According to Miskawayh, Ibn Muqātil was rewarded for his service to al-Barīdī with 20,000 dīnār. In his new position, al-Kūfī was able to obtain the farming of two further key cities, Basra and Wāsiṭ, for al-Barīdī. Miskawayh relates al-Kūfī’s argument as direct speech addressed to Ibn Rāʾiq: “As for Wāsiṭ, I administer it myself, and not even an unmounted messenger of theirs comes there. I undertake that the revenue shall be paid in full. As for Basra, I have settled the amount at four million dirham, on condition that he provides trustworthy sureties there.” Apart from the first installment from Ahwāz, none of the farming money ever materialized, and later that year al-Barīdī and Ibn Rāʾiq entered open conflict. Al-Kūfī, on his part, remained with Ibn Rāʾiq until the latter’s deposition, after which, having paid a fine, he became secretary to the new amīr al-umarāʾ Bajkam. The two missions display different attitudes on the part of the messenger: in the first case, al-Kūfī does not fulfil his task—in fact, in Miskawayh’s account he does not even attempt to carry it out. In the second, whatever al-Kūfī’s motivations may have been for leaving Ahwāz, he remains loyal to al-Barīdī, at least for the time it takes him to ensure his goal of unseating al-Nawbakhṭī and obtaining Basra and Wāsiṭ. Al-Kūfī seems very skilled at balancing his function as an envoy with his personal advantage. At the same time, we see an illustration of two aspects that seem to gain ever more importance in this period: the exis-

66

Miskawayh, Tajārib i, 363.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

426

van berkel, el cheikh and osti

tence of parallel forms of communication, which are at times conflicting, and the importance of the messenger’s other skills: it is al-Kūfī’s expertise as a kātib and his financial knowledge in particular that makes it possible for him to gain the trust of his successive employers—whatever his popularity, his skills are needed. This is illustrated by the fact that Bajkam seems to have had no great qualms about employing him although he had remained loyal to the last to his rival Ibn Rāʾiq. Taken together, al-Kūfī’s two missions contain all the elements posited in our introduction: firstly, al-Kūfī starts off as a Baghdad clerk but quickly moves to serve, and carry messages for, military leaders outside the capital; secondly, acting as a messenger/envoy becomes one of the main functions of his role; thirdly, this function is dependent on his personal relations with the people he serves rather than their office or his own.

9

Al-Ḥasan b. Hārūn for the Caliph

Al-Ḥasan’s second mission brings us another decade forward, and one caliph and eleven viziers later. Al-Ḥasan b. Hārūn operates once again as a messenger, but this time on behalf of the caliph al-Muttaqī. Both Miskawayh and al-Ṣūlī describe al-Ḥasan b. Hārūn’s mission in 332/943–944, emphasizing the precariousness of the caliph’s situation. Miskawayh points out that al-Muttaqī, at this point residing with the Hamdanids, first in Mosul and later in Raqqa, and thus away from Baghdad, realizes that his hosts are weary of accommodating him and that they want to get rid of him.67 For this reason he sends al-Ḥasan b. Hārūn and a certain Abū ʿAbdallāh b. Mūsā al-Hāshimī to Baghdad to negotiate the caliph’s return to the capital with Tūzūn, the current amīr al-umarāʾ. Al-Ṣūlī only refers to al-Ḥasan b. Hārūn and does not name other members of the mission, but he does mention a letter from the caliph that al-Ḥasan b. Hārūn brought along for Tūzūn.68 According to Miskawayh, Tūzūn received the request and the two messengers with “delight and eagerness.”69 The two were able to make Tūzūn swear his loyalty to al-Muttaqī and the vizier Abū l-Ḥusayn, son of the previous vizier, Ibn Muqla. The oath seems to have been validated by a number of people. Miskawayh goes to some lengths to describe the context and the attestation of the oath:

67 68 69

Miskawayh, Tajārib ii, 67; Al-Ṣūlī, Akhbār 267–269. Al-Ṣūlī, Akhbār 267–269. Miskawayh, Tajārib ii, 67.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

427

These two obtained from him [Tūzūn] solemn oaths and assurances of fidelity to al-Muttaqī and to the vizier Abū l-Ḥusayn Ibn Muqla, and Tūzūn summoned the judges, witnesses, representatives of the families of ʿAbbās and Abū Ṭālib, and the chief secretaries in order to swear in their presence to al-Muttaqī. A copy of the oath was carefully made out, and all those present set their attestations that Tūzūn had sworn it.70 His emphasis on the attestation of the oath does make sense if we look at the continuation of these events. A copy of the oath was given to the messengers, probably to be presented to the caliph, though Miskawayh is not clear about the latter.71 Al-Ṣūlī talks about the same oath and notes that al-Ḥasan b. Hārūn returned to the caliph alone with a letter from Tūzūn and a personal report of the events. In general, the mission seems to have gone very smoothly. According to al-Ṣūlī, all parties were satisfied, and he presents the mission as an illustration of al-Ḥasan b. Hārūn’s competence. All’s well that ends well, one could say, were it not for a rather unhappy follow-up of these events for the caliph: when al-Muttaqī finally decides to go back to Baghdad, Tūzūn meets him halfway and arrests, deposes, and blinds him, thus breaking his oath of allegiance.72 Thus, while al-Ḥasan’s second mission initially seems successful, handled by an experienced and trustworthy messenger, the final outcome is less positive. However, our sources do not seem to blame al-Ḥasan b. Hārūn for it. The mission’s failure is caused by the precarious situation of the caliphate and, according to al-Ṣūlī, the caliph’s mistake of leaving the capital, the seat of his rule. In this respect, the situation is very different from the mission al-Ḥasan undertook 18 years earlier when the caliph al-Muqtadir was still firmly seated in his palace in Baghdad and al-Ḥasan b. Hārūn negotiated with the vizier on behalf of Yūsuf b. Abī l-Sāj. Between the first and second mission, al-Ḥasan b. Hārūn’s own position seems to have changed considerably. He rose from messenger of a provincial leader to main messenger for the caliph. At first sight, this seems a successful career for a plausibly loyal and competent servant. However, his career coincided with the loss of power and importance of the caliph, and the question is whether it was really such an improvement to be a negotiator in the caliph’s

70 71 72

Miskawayh, Tajārib ii, 67. Miskawayh, Tajārib ii, 67. On these events, see Osti, Reaching out and going out. See also Mottahedeh, Loyalty and leadership 46–47.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

428

van berkel, el cheikh and osti

name when, in the end, not even an oath could save him. On the other hand, his power as a negotiator may well have grown: with the caliph forced to move outside the capital, away from his administration, the role and agency of the messenger must have expanded.

10

Conclusion

As we have seen, the careers of al-Ḥasan and al-Kūfī have many common elements: they are traditionally trained bureaucrats who serve both in the caliphal administration as well as serving military commanders and regional leaders. They travel frequently, together with their employers but also alone as their emissaries. In these respects, their careers are consistent with the political instability of the period and the loss of importance of Baghdad as the center of power: in order to find work, a kātib might have to move away from Baghdad and from the court and serve in the administration of the new leaders. In this more fluid context, the finely honed procedures and etiquette of the palace chancery must become more flexible and informal.73 To be sure, personal networks and informal alliances had always been an integral part of the administrative machinery, but this was a more radical development: the formal and extensive administrative apparatus, with its emphasis on written communication, had gradually been replaced by a more flexible and manageable informal system. In other words, personal relationships took precedence over official roles. This fluidity of roles is reflected in a lack of clear norms and descriptions of the messenger’s tasks, ideal qualifications, and position. Although the vast majority of Abbasid bureaucrats remain anonymous in the sources, their position, tasks, ideal qualifications, and character as a professional group are set out in detail in the normative administrative literature. The same is true for the role and ideal qualifications of diplomats, as these were laid down in, for example, Ibn al-Farrāʾ’s Kitāb Rusul al-mulūk. Internal messengers, however, besides being anonymous individually, are also invisible as a group, and their task is undefined. On the basis of our two case studies, we contend that, in the course of the fourth/tenth century, messengers acquire more visibility in historical narratives: a task that in earlier times may have been carried out by a nameless rasūl is now performed by a recognizable individual who features as

73

See also van Berkel et al., Continuity and crisis.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

429

a main character in the story. On the other hand, our case studies confirm that being a messenger is not a profession in itself; if anything, it seems to be an occasional task connected with other professional activities. If we accept al-Ḥasan and al-Kūfī as representing a new type of messenger/envoy, it is clear that this figure must perform the tasks and possess the qualities of both the kātib—being able to negotiate complex financial deals— and the envoy—having the diplomatic skills to face a potentially hostile interlocutor. It is not surprising then that the refashioned envoy’s presence and visibility are comparable to those of high-ranking kuttāb of earlier times. In other words, the waning influence of a formal, centralized bureaucracy in Baghdad coincides with the emergence of a new figure, the kātib/rasūl. His role may not have codified norms, but we have seen that Miskawayh and al-Ṣūlī do comment on the skills of al-Ḥasan and al-Kūfī as messengers. From their critical descriptions, we can distill the dos and do nots, ideal skills, and required attitude of this emerging official, the internal messenger. First and foremost, the messenger needed the technical skills of a classically trained kātib: knowledge of state finances and taxation, proficiency in writing, and knowledge of chancery practices. Al-Ḥasan’s military experience notwithstanding, both he and al-Kūfī were trained as kuttāb and remained, throughout their careers, officials who first and foremost used the pen (and the tongue) to communicate their messages. Yet, because many of their activities took place outside the traditional administrative bureaus and they had to operate individually, representing the employer in the latter’s absence, other qualities, such as honesty, loyalty, independence, competence, and diplomacy, were highly appreciated as well. In addition, a strong personal network of powerful people was certainly helpful to the envoys to help steer between rivals and positions in this new and highly competitive political arena. These latter qualities are, not surprisingly, in line with the ideal qualifications of the external diplomat as described by Ibn al-Farrāʾ and other normative sources. This is not to say that anonymous messengers do not continue to exist, nor that kuttāb were the only type of recognizable messengers. Rather, we argue that isolating this kātib/rasūl as a new category with a specific role brings into question the general understanding of this period as one of victory of the sword over the pen. Of course, we owe much of the historical record to the kuttāb themselves, who are bound to emphasize the importance of their own class. However, Hugh Kennedy’s work on the Abbasid Caliphate has highlighted how, on the basis of personality-centered historical narratives, it is possible to reconstruct much more than the bare outlines of political and military events. In particular, his investigation of the decline of Abbasid political power in the fourth/tenth century identifies crucial socioeconomic factors that go well bey-

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

430

van berkel, el cheikh and osti

ond a binary interpretation of history as a battle between good and evil.74 Above all, Hugh has taught us that how later generations make sense of the past is as important as the past itself. Thus, it is no coincidence that these personality-centered chronicles are mainly told through accounts set at the Baghdad court in the first part of the fourth/tenth century, as we have shown in our collective study of the caliphate of al-Muqtadir, but gradually shift their focus outside the capital. The case of messengers has illustrated the changes connected to this shift as well as elements of continuity in the crisis of the ʿAbbasid political power.

Bibliography Sources Hamidullah, M., Six originaux des lettres diplomatiques du Prophète de l’Islam, Paris 1985. Ibn ʿAbd Rabbih, al-ʿIqd al-farīd, ed. Aḥmad Amin et al., 6 vols., Cairo 1940–1949. Ibn ʿAbd Rabbih, The unique necklace: al-ʿIqd al-farīd, trans. I.J. Boullata, 3 vols., Readings 2009. Ibn al-Farrāʾ, Kitāb Rusul al-mulūk wa man yaṣluḥ lil-risāla wa-l-safāra, ed. Ṣalāḥ al-Dīn al-Munajjid, Beirut 1972. Ibn Saʿd, al-Ṭabaqāt al-kubrā, 7 vols., Beirut n.d. al-Khaṭīb al-Baghdādī, Taʾrīkh Baghdād, ed. Muṣṭafā ʿAbd al-Qādir ʿAṭā, 21 vols., Beirut 1997. Miskawayh, Tajārib al-umam: The eclipse of the Abbasid Caliphate, ed. H.F. Amedroz, trans. D.S. Margoliouth, 6 vols., Oxford 1920–1921. Nizam al-Mulk, The book of government or rules for kings: The Siyar al-muluk or Siyasatnama, trans H. Darke, London 1960. al-Ṣūlī, Akhbār al-Rāḍī bi-llāh wa-l-Muttaqī lillāh, aw, Taʾrīkh al-dawla al-ʿabbāsiyya min sanat 322 ilā sanat 333 hijriyya min Kitāb al-Awrāq, ed. J. Heyworth Dunne, Beirut 1935. al-Ṣūlī, Akhbār al-Rāḍī bi-llāh wa-l-Muttaqī lillāh: Histoire de la dynastie Abbaside de 322 à 333/933 à 944, trans. M. Canard, 2 vols., Algiers 1946 and 1950. Vaiou, M. (trans), Diplomacy in the early Islamic world: A tenth century treatise on ArabByzantine relations: The book of messengers of kings (Kitāb Rusul al-mulūk) of Ibn al-Farrāʾ, London 2015.

74

See Kennedy, Decline and fall. See also Kennedy, Prophet and the age of caliphates.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

bureaucrats on the move

431

Studies Ali, S., Arabic literary salons in the Islamic Middle Ages: Poetry, public performance, and the presentation of the past, Notre Dame 2010. Berkel, M. van, ʿAlī b. ʿĪsā b. Dāʾūd b. al-Jarrāḥ, in ei3. Berkel, M. van, Communication between centre and periphery in the early-tenthcentury Abbasid Empire, in W. Pohl (ed.), Shadows of empire: Imperial peripheries in early medieval Eurasia, Leiden and Boston forthcoming. Berkel, M. van, Political communication or how to reach the unreachable, in S. Heidemann and K. Mewes (eds.), The reach of empire: The early Islamic empire at work, Berlin forthcoming. Berkel, M. van, N.M. El Cheikh, H. Kennedy, and L. Osti, Crisis and continuity at the Abbasid court: Formal and informal politics in the caliphate of al-Muqtadir (295– 320/908–32), Leiden and Boston 2013. Bowen, H., The life and times of ʿAlí Ibn ʿÍsà, the “good vizier,” Cambridge 1928. El Cheikh, N.M., Byzantium viewed by the Arabs, Cambridge, MA 2004. Kennedy, H.N., The decline and fall of the first Muslim empire, in Der Islam 81 (2004), 3–30. Kennedy, H.N., The armies of the caliphs. Military and society in the early Islamic State, London and New York, 2001. Kennedy, H.N., The prophet and the age of the caliphates: The Islamic Near East from the sixth to the eleventh century, London and New York 1986. Mottahedeh, R., Loyalty and leadership in an Early Islamic Society, Princeton 1980. Osti, L., Reaching out and going out: Movement and power in mid-4th/10th century Iraq, in S. Heidemann and K. Mewes (eds.), The reach of empire: The early Islamic empire at work, Berlin forthcoming. Osti, L., History and memory in the Abbasid Caliphate: Writing the past in medieval Arabic literature, London 2022. Serjeant, R.B., Early Arabic prose, in A.F.L. Beeston et al. (eds.), The Cambridge history of Arabic literature: Arabic literature to the end of the Umayyad period, Cambridge 1983, 141–142. Silverstein, A.J., Postal systems in the pre-modern Islamic world, Cambridge 2007. Tillier, M., Représenter la province auprès du pouvoir impérial: les délégations (wufūd) égyptiennes aux trois premiers siècles de l’Islam, in Arabica 67 (2020), 125–199.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 22

Al-Ṭabarī’s Unacknowledged Debt to Ibn Abī Ṭāhir Ṭayfūr Sarah Bowen Savant

The history of citation practices in Arabic is a story waiting to be told. In this story, the isnād, or transmission chain, will receive its due, but so, too, will a wide variety of other, until now comparatively neglected, topics. A major concern running through all work on citation must be the “reuse” of earlier texts.* The concept of text reuse arises out of computer science, specifically the field of information retrieval. It is useful as a concept because it points to diverse writerly practices involving one person’s appropriation of the words of others. Computer software can identify, record, and visually display these patterns. The term “reuse” is intended to be value neutral; it is not equivalent to “plagiarism,” which carries a sense of literary theft—though such theft is one form of reuse. The numerous forms of reuse encountered in the Arabic literary tradition include (roughly from shorter to longer) literary devices (similes, metaphors, jokes, aphorisms, and parables); documentary formulae; poetry; taught texts, such as those used in ḥadīth, law, and history; excerpts, commentaries, and other adaptations of earlier written works; anthologies and other collections of writings; and different versions of the same text. In each of these cases, as in many others, the reuse of earlier works was culturally mediated. Authors worked with, and occasionally against, the expectations of their intended audiences. Citation encoded not just a debt but also status, respect for a system of referencing as much as for any individual person, and obligations concerning how to reuse other people’s words and how to acknowledge literary debts.

* This study is part of a project that received funding from the European Research Council (erc) under the European Union’s Horizon 2020 research and innovation program (kitab, grant agreement no. #772989). I would like to thank Masoumeh Seydi, Ryan Muther, Peter Verkinderen, and Mathew Barber for their work on the dataset and data visualization; Hamid Reza Hakimi and Peter Verkinderen for work on the digital version of al-Ṭabarī’s text; and Abdul Rahman Azzam, Lorenz Nigst, Letizia Osti, Aslisho Qurboniev, Hanna Siurua, Gowaart Van Den Bossche, and Peter Verkinderen for reviewing this chapter.

© Sarah Bowen Savant, 2022 | doi:10.1163/9789004525245_023

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

433

The following case is interesting because it touches on questions about citation, reuse, and the credit one does or does not owe one’s forebears. It involves the great polymath of medieval Islam, Muḥammad b. Jarīr al-Ṭabarī (d. 310/923), and on the surface it might appear to be a case of what we today would consider plagiarism. Hugh Kennedy and I have discussed it at some length, as I have sought to understand al-Ṭabarī’s working habits. The depth of Professor Kennedy’s engagement with al-Ṭabarī is virtually unrivalled today, and it goes back to his early career, when he read the entirety of al-Ṭabarī’s Taʾrīkh at Cambridge. His interest and advice convinced me that there is still much work to be done on this well-studied author and that the case I describe here is important and of interest for historians.

1

A Case of Literary “Theft”?

I start with the ostensible victim of the crime, Ibn Abī Ṭāhir Ṭayfūr (d. 280/893), who was born in 204/819–820 into a Khurasanian family that made its way to Baghdad, where Ibn Abī Ṭāhir enjoyed a productive career as a prolific writer of dozens of books. Among them we find titles such as “The book about the conciliatory king and the supportive vizier” (Kitāb al-Malik al-muṣliḥ wa-l-wazīr al-muʿīn), “The compendium of poets and accounts about them” (Kitāb al-Jāmiʿ fī al-shuʿarāʾ wa-akhbārihim), “The book about the boasting match between the rose and the narcissus” (Kitāb Mufākharat al-ward wa-l-narjis), and “The book about Wahb’s apology for breaking wind” (Kitāb Iʿtidhār Wahb min ḥabqatihi). The list of Ibn Abī Ṭāhir’s books contains three now-lost works pertaining to sariqa, a term that generally refers to theft but that can in this context be fairly translated as “borrowing” or, perhaps better, “plagiarism.”1 The subject of sariqa has been treated by scholars of Arabic literature, who paid particular attention to the period when Ibn Abī Ṭāhir lived.2 Two of his books on sariqa deal with the “modernist” poet Abū Tammām (d. ca. 232/845) and his student and rival al-Buḥturī (d. 284/897). These are “The book on the borrowings/plagiarisms of Abū Tammām” and “The book on the borrowings/plagiarisms of

1 Ibn al-Nadīm (d. 385/995) provides the most extensive list of Ibn Abī Ṭāhir’s works. See OpenITI, Ibn al-Nadīm, Fihrist, search term: ‫اخبار بن أبي طاهر‬. See also Toorawa, Ibn Abī Ṭāhir Ṭayfūr 95–96 and 168, n. 64. All OpenITI texts cited in this chapter can be accessed through the Zenodo release 10.5281/zenodo.5118900. Details of the underlying print editions can be found in the bibliography that accompanies this chapter. 2 E.g., Heinrichs, Evaluation of sariqa 358. Heinrichs notes that “the material offered by the sariqāt works is rather inhomogenous.” See also Osti, Author as protagonist.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

434

savant

al-Buḥturī from Abū Tammām” (Kitāb Sariqāt al-Buḥturī min Abī Tammām). The third book, “The book about the borrowings/plagiarisms of the poets” (Kitāb Sariqāt al-shuʿarāʾ), treats the general topic of sariqa in poetry.3 As well as writing about others’ literary theft, Ibn Abī Ṭāhir was himself accused of sariqa.4 His personal legacy thus carries considerable memory of unpaid literary debts. Given this legacy, there is a certain irony in the fact that thousands of words from one of his books, The history of Baghdad: Accounts concerning its caliphs, its governors, and events in their time (Kitāb Taʾrīkh Baghdād fī akhbār al-khulafāʾ waʾl-umarāʾ wa-ayyāmihim),5 were reused without attribution, often verbatim, by Ibn Abī Ṭāhir’s younger contemporary, Muḥammad b. Jarīr al-Ṭabarī (d. 310/923), a scholar who is normally regarded as a careful citer of past authorities. Only volume six of Ibn Abī Ṭāhir’s original Kitāb Baghdād survives. This volume pertains to the reign of the Abbasid Caliph al-Maʾmūn (r. 198–218/813– 833). The period covered in the volume begins with al-Maʾmūn’s arrival in Iraq from Khurasan in the month of Ṣafar 204/August 819 and ends with his death. Al-Ṭabarī’s evident and extensive reuse of material found in this surviving section of the book in his own famous history, The history of prophets and kings (Taʾrīkh al-rusul wa-l-mulūk), has been mentioned repeatedly in scholarship, but to my knowledge it has been discussed in detail only by Hans Keller, in his 1908 edition, German translation, and discussion of Ibn Abī Ṭāhir’s book.6 Keller labeled al-Ṭabarī an “Abschreiber” and a “Plagiator.” However, a scathing critique of Keller’s argument by Henry Frederick Amedroz, a senior British scholar of his day, may have discouraged later scholars from paying attention to Keller’s evidence.7 They may also have assumed that Keller underestimated the intertextuality of the written tradition. Writing for the second edition of the Encyclopaedia of Islam more than half a century after Keller, Franz Rosenthal concluded rather blandly that “Ibn Abī Ṭāhir’s treatment agrees widely with that of the later Ṭabarī.”8 3 Toorawa, Ibn Abī Ṭāhir Ṭayfūr 96. 4 Ibn al-Nadīm quotes another Baghdadi, Jaʿfar b. Ḥamdān, as saying that Ibn Abī Ṭāhir was the most given to plagiarism (asraq al-nās), copying as little as a half or a third of a verse. OpenITI, Ibn al-Nadīm, Fihrist, search term: ‫أسرق الناس‬. 5 OpenITI, Yāqūt, Muʿjam al-udabāʾ, search term: ‫أحمد بن أبي طاهر‬. 6 Keller, Stellung der “Annalen.” 7 Amedroz, Review of part 2 of Kitāb Baġdâd. 8 Rosenthal, Ibn Abī Ṭāhir Ṭayfūr. In his translation of al-Ṭabarī’s corresponding volume on al-Maʾmūn’s caliphate, C.E. Bosworth refers his readers to Keller’s treatment: “The whole question of the relationship between Ibn Abī Ṭāhir and Ṭabarī has been discussed in a highly detailed and masterly fashion, so as to require no further discussion here, by Keller in the Introduction to his German translation of the Kitāb Baghdād, ii, pages xiii–xxvi.” Bosworth, Translator’s foreword 7.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

figure 22.1

435

The first alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād (on the left) and al-Ṭabarī’s Taʾrīkh (on the right). Blue and green highlighting marks text present in one work but not in the other. Unhighlighted text is found in both. Beige highlighting indicates text contained in both works but in different locations. Based on the kitab Diff Viewer a new application created by peter verkinderen

Indeed it does. In early 2021, the digital humanities project that I oversee, kitab, created evaluation datasets using the Open Islamicate Texts Initiative (OpenITI) corpus of digital Arabic texts to assess the precision and recall of passim, the main software we use to generate our data on text reuse. I checked and corrected the passim alignments generated by Ryan Muther between the Kitāb Baghdād and al-Ṭabarī’s Taʾrīkh. There were 33 such alignments, amounting to a total of more than 14,000 word tokens of shared text.9 The first alignment, in a passage describing al-Maʾmūn’s arrival in Baghdad, is illustrative: the two passages are noticeably similar, but al-Ṭabarī’s also features rearrangement and paraphrase (see fig. 22.1). Ibn Abī Ṭāhir first names three narrators, saying that they relied on history books (kutub al-taʾrīkh) as their sources. Al-Ṭabarī jumps straight into the story,

9 For the alignments, see 10.5281/zenodo.5118900.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

436

savant

under a heading translated in the English edition as “Al-Maʾmūn’s arrival in Iraq and what happened there.”10 But the subsequent narrative is essentially the same in both works. Both depict al-Maʾmūn’s stop in Nahrawān, southeast of Baghdad, on his way to the latter city. They use the same phrasing to say that the caliph arrived on a Saturday and remained in Nahrawān for eight days. In both, members of his family and notables come out to meet him. Al-Ṭabarī also mentions military leaders in this context. In addition, he provides details concerning the date of al-Maʾmūn’s entry into Baghdad and the green attire, lances, and banners of his retinue that are not found in Ibn Abī Ṭāhir’s corresponding text shown here. However, the latter does include these details, too, slightly earlier in the work, before the beginning of the alignment detected by passim. Such rearrangements can be harder to detect than overlaps occurring in the same order. In gathering the alignments, we treated Ibn Abī Ṭāhir’s Kitāb Baghdād as the base text and compared it to al-Ṭabarī’s Taʾrīkh. The numbering of alignments thus follows the order of the text in the Kitāb Baghdād. Al-Ṭabarī followed a slightly different organizing principle, adhering closer to chronological order than did Ibn Abī Ṭāhir, so the order of the aligned passages diverges in his work. The dataset indicates the locations of the passages in each book. In alignments seven and eight, for example, al-Ṭabarī changed the order of his predecessor’s text. These passages treat the rebellion of Naṣr b. Shabath and quote a letter from al-Maʾmūn. Ibn Abī Ṭāhir’s text leads with the letter and then describes exchanges between al-Maʾmūn and Jaʿfar b. Muḥammad al-ʿĀmirī, who communicates with Naṣr on the caliph’s behalf. Al-Ṭabarī reverses the order, foregrounding the courtiers but necessitating some backtracking: he introduces the letter with “Before this […]” (see figs. 22.2 and 22.3). The dataset identifies the overlapping passages by page numbers in the Leiden edition and the Bosworth translation. These numbers reveal the comprehensive extent of al-Ṭabarī’s reuse of Ibn Abī Ṭāhir’s text in his account of the period 204–218/818–834 in the Taʾrīkh. Topics that pass freely from Ibn Abī Ṭāhir’s book into that of al-Ṭabarī include the caliph’s entry into Baghdad (including al-Maʾmūn’s adoption of the black clothes of his Abbasid forefathers and his rejection of green as his symbolic color); much relating to the Ṭāhirids; al-Maʾmūn’s consummation of his marriage with Būrān, the daughter of al-Ḥasan b. Sahl; the intrigues of the Abbasid court and the postings of governors and military units; ʿAlid and other rebels

10

Alignment 1: OpenITI, Ibn Abī Ṭāhir, Baghdād, lines 57–79; OpenITI, al-Ṭabarī, Taʾrīkh, lines 108,276–108,298; al-Ṭabarī, History of al-Ṭabarī xxxii, 94.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

figure 22.2

437

The seventh alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād (on the left) and al-Ṭabarī’s Taʾrīkh (on the right), showing the reproduction of a letter from al-Maʾmūn to the rebel Naṣr b. Shabath

(including Naṣr b. Shabath); al-Maʾmūn’s conflicts with the Byzantines; anecdotes showcasing the caliph’s generosity, poetic talents, and discerning ear; attitudes toward Arabs and non-Arabs; and much else. The parallel texts contain much poetry and long letters, including a missive of about 2,400 words from Ṭāhir b. al-Ḥusayn to his son, ʿAbdallāh, advising him how to govern (Bosworth called this letter “an early example of the ‘Mirrors for Princes’ genre in Arabic”)11 and a letter sent by al-Maʾmūn to his Byzantine counterpart, Theophilus, summoning the latter to Islam.12 Ibn Abī Ṭāhir’s treatment of the so-called miḥna—in which al-Maʾmūn required judges and traditionists to 11 12

Alignment 4; Bosworth, Translator’s foreword 3. Alignment 26; OpenITI, Ibn Abī Ṭāhir, Baghdād, lines 3,658–3,680; OpenITI, al-Ṭabarī, Taʾrīkh, lines 109,398–109,421; al-Ṭabarī, History of al-Ṭabarī xxxii, 195–197.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

438

figure 22.3

savant

The eighth alignment between Ibn Abī Ṭāhir’s Kitāb Baghdād (on the left) and al-Ṭabarī’s Taʾrīkh (on the right), in a passage describing exchanges between al-Maʾmūn and Jaʿfar b. Muḥammad al-ʿĀmirī about the rebel Naṣr b. Shabath

endorse the theological position that the Quran was created—is also found in al-Ṭabarī’s work, though the precise extent of the reuse cannot be ascertained because there is a gap in Ibn Abī Ṭāhir’s surviving text.13 At no point does al-Ṭabarī give credit to Ibn Abī Ṭāhir as his source; the latter is not named anywhere in the reused sections. Historians working on the Kitāb Baghdād and this section of al-Ṭabarī’s work have remarked on the close relationship, and Hugh Kennedy, in particular, has pointed out the different style of writing used in the latter parts of al-Ṭabarī’s text.14 But the extent of the reuse has gone largely unnoticed, and historians have not considered what it suggests about al-Ṭabarī’s working habits more generally. Yet examining the cases of alignment between the two texts yields some interesting insights.

13

14

Al-Ṭabarī’s text follows Ibn Abī Ṭāhir’s but then breaks off suddenly. See the last alignments in the dataset, as well as Ibn Abī Ṭāhir Ṭayfūr, Taʾrīkh Baghdād 346; and al-Ṭabarī, History of al-Ṭabarī xxxii, 199–209 (pages 209–222 continue to treat the miḥna but are not found in the surviving portion of Ibn Abī Ṭāhir’s text). Kennedy, Caliphs and their chroniclers; see also Borrut, Entre mémoire et pouvoir esp. 103– 108. For a different perspective on this period and on reading al-Ṭabarī, see El-Hibri, Reinterpreting Islamic historiography 12–13.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

439

To begin with, cutting out Ibn Abī Ṭāhir involved some work on al-Ṭabarī’s part. Consider the following two examples of rewriting, both by chance involving an account of a death: 1) Alignment 5: The circumstances of Ṭāhir b. al-Ḥusayn’s death Ibn Abī Ṭāhir: “Yaḥyā b. al-Ḥasan told me (ḥaddathanī) that ʿAbd al-Khāliq reported from Abū Zayd Ḥammād b. al-Ḥasan, who said: ‘Kulthūm b. Thābit b. Abī Saʿd, whose patronymic was Abū Saʿda, told me: “I was in charge of the intelligence and postal service in Khurāsān, and the place where I usually sat each Friday was at the foot of the pulpit […]” ’ ” Al-Ṭabarī: “It is related that Kulthūm b. Thābit b. Abī Saʿd, whose patronymic was Abū Saʿda, said: ‘I was in charge of the intelligence and postal service in Khurāsān, and the place where I usually sat each Friday was at the foot of the pulpit […]’”15 2) Alignment 33: The cause of al-Maʾmūn’s fatal illness Ibn Abī Ṭāhir: “Saʾīd al-ʿAllāf, the Quran reciter, told me (ḥaddathanī): ‘AlMaʾmūn sent [for me] when he was in the Byzantine lands. I was then brought to him at Budandūn […]’” Al-Ṭabarī: “It is mentioned from (dhukira ʿan) Saʿīd al-ʿAllāf, the Quran reciter: ‘Al-Maʾmūn sent for me, at the time when he was in the Byzantine lands, having entered them from Tarsus on Wednesday, the sixteenth of Jumādā ii [July 9, 833]. I was brought to him at Budandūn […]’ ”16 Such changes and omissions are found throughout the 33 alignments that passim detected and I reviewed. Al-Ṭabarī’s omissions are telling: in all of these cases, he appears to have decided not to cite Ibn Abī Ṭāhir directly, even though the latter says that he received the reports through personal contact with the major historical actors involved in the events. Al-Ṭabarī thus left considerable cultural capital just sitting on the table. Indeed, in his entire work, al-Ṭabarī cites Ibn Abī Ṭāhir by name only once, in connection with a Ḥusaynid rebellion in the year 250/864 that would have been addressed (if it was at all) in the now-lost later parts of the Kitāb Baghdād. However, even here al-Ṭabarī does not refer to the book itself.17

15 16 17

Al-Ṭabarī, History of al-Ṭabarī xxxii, 132. Al-Ṭabarī, History of al-Ṭabarī xxxii, 224. Al-Ṭabarī writes: “Ibn Abī Ṭāhir mentioned that Ibn al-Ṣūfī al-Ṭālibī told him (ḥaddathahu) that he […]”; OpenITI, al-Ṭabarī, Taʾrīkh, line 115,986.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

440 2

savant

What This Means for the Taʾrīkh

This evidence of reuse has implications for how we read and interpret the Taʾrīkh because it seems possible—even likely—that al-Ṭabarī relied on Ibn Abī Ṭāhir much more extensively than is currently recognized (see figure 22.4 below). In Figure 22.4, the Taʾrīkh is broken down into chunks of 1,513 characters— the average number of characters in a chunk of 300 word tokens (all texts in the OpenITI corpus are divided into 300-token chunks). The location of each isnād that starts with the active-voice ḥaddathanī/ḥaddathanā (“he told me”/“he told us”) is marked with a blue dot; isnāds beginning with dhukira ʿan (“it is reported from”) are marked in red. Imagine the figure as a great scroll on which the text is written with 1,513 characters per line. The top of the scroll is the figure’s left edge, and its bottom is the right edge.18 The data in Figure 22.4 allow us to make the following observations: first, the ḥaddathanī/ḥaddathanā formula is the dominant marker of citation in most of the book. Masoumeh Seydi and I are working on a separate study of this citation pattern in al-Ṭabarī’s writings, and we argue that this formula points to a small number of persons from whom al-Ṭabarī took down well-organized notes that he kept for decades and used to create his Taʾrīkh as well as his Quran commentary, Jāmiʿ al-bayān ʿan taʾwīl āy al-Qurʾān, and his Tahdhīb al-āthār, an incomplete work on traditions. Second, al-Ṭabarī’s method of citation changes with the founding of Baghdad in 145/762. He largely stops quoting his informants directly with ḥaddathanī/ḥaddathanā and instead begins using predominantly the passive voice dhukira ʿan. His use of the passive voice eventually tapers off, right around the point at which Ibn Abī Ṭāhir’s book reportedly ended—namely, with the reign of alMuhtadī (r. 255–256/869–870).19

18

19

For the more technically minded: to produce this figure we defined Regular Expression patterns to find the transmissive terms ḥaddathanī/ḥaddathanā and dhukira ʿan. We then searched for the terms in the text of the Taʾrīkh to determine the character positions of those terms. We scaled the positions to the x- and y-axes of the graph, which represent the length of the text and the approximate length of the 300-word token chunks, respectively. To identify the beginnings of sections, we searched for the section headers (annotated through OpenITI mARkdown) and marked these with vertical lines. The script for doing this is available at 10.5281/zenodo.5118900. This portion of the Taʾrīkh runs from the end of volume 28 through the middle of volume 36 in the English translation published by the State University of New York Press; in the Brill edition, it corresponds to series 3, pt. 1, 271–1,839.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

figure 22.4

441

The passive voice as a potential indicator of text reuse in al-Ṭabarī’s Taʾrīkh. Blue dots mark occurrences of the phrases ḥaddathanā/ḥaddathanī as the first element of an isnād. Red dots indicate uses of the passive-voice phrase dhukira ʿan. Vertical lines identify section beginnings as well as the segment that corresponds to the extant volume of Ibn Abī Ṭāhir’s Kitāb Baghdād figure created by peter verkinderen and masoumeh seydi

We do not know precisely where Ibn Abī Ṭāhir’s book began, but it is likely that his account started with the founding of Baghdad (the book, after all, is known as the Kitāb Baghdād). Therefore, al-Ṭabarī’s use of the passive voice is concentrated precisely in the part of the narrative that overlaps with Ibn Abī Ṭāhir’s book. Without a more complete version of the Kitāb Baghdād, we cannot prove that al-Ṭabarī relied on Ibn Abī Ṭāhir’s book for the history of Baghdad from its founding to 256/870, but the conspicuous shift in the citation pattern, together with the evidence of al-Ṭabarī’s rewriting provided by our dataset, points to a significant, though unacknowledged, debt to Ibn Abī Ṭāhir.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

442

savant

This presumption finds further support when we search the Taʾrīkh using an expanded list of transmissive terms (that is, phrases that introduce a citation) that also includes phrases such as ruwiya ʿan.20 An initial search using a list prepared by R. Kevin Jaques produces interesting results. The first part of the Taʾrīkh, up to the founding of Baghdad, represents 74% of the entirety of the book, but it contains 96% of all citations. In this part of the work, a transmissive term occurs on average every 331 words. By contrast, in the second part, which probably corresponds to Ibn Abī Ṭāhir’s book in its original form, such a term occurs on average every 1,728 words. Though the search requires further refinement for greater accuracy, it is safe to say that al-Ṭabarī is far less transparent about his sources in the later parts of his book.

3

Why Did al-Ṭabarī Not Acknowledge His Debt to Ibn Abī Ṭāhir?

There is much work to be done on citation practices, using data such as the above. In particular, we need to combine such data with a fresh consideration of terminologies, periodizations, topics, and literary and historical contexts. We will then be able to better understand literary borrowings such as al-Ṭabarī’s. What might al-Ṭabarī have been thinking? First, it is worth remembering that the earliest surviving physical evidence of books dates to al-Ṭabarī’s lifetime and the period when he was amassing material for his works.21 These were early days, when it may have been clear how one should cite certain types of written materials, but expectations were perhaps fuzzier for other writings, including books about events close to one’s own time. Ibn Abī Ṭāhir’s book may have been viewed as a “practical data carrier” that did not require citation, much as many people today view Wikipedia.22 Both Figure 22.4 above and the statistics yielded by the search of wider terms support this possibility. For the years 204–218 (that is, the period represented by the alignment dataset), al-Ṭabarī mentions only one other source, Aḥmad b. Muḥammad b. Makhlad, 20 21

22

Specifically, we searched the text for transmissive terms that begin a paragraph, as a useful though not perfect proxy for the beginnings of isnāds. The earliest dated Arabic codex written on paper is from 866, and it contains Abū ʿUbayd’s (d. 238/838) Gharīb al-Ḥadīth. Gruendler, Rise of the Arabic book 15; see also Déroche, Manuscrits arabes. Gruendler, Rise of the Arabic book 23; I thank Letizia Osti for the analogy. An alternative expression, from an era less publicly grateful to public servants, is “You don’t thank the postman.” Ilkka Lindstedt has noted also al-Ṭabarī’s selective quotation of al-Madāʾinī (d. ca. 228/843). Al-Ṭabarī does not provide an isnād when reproducing al-Madāʾinī’s material pertaining to the Prophet. Lindstedt, Life and deeds 243.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

443

by name, using the formula ḥaddathanī. This person had direct contact with al-Ṭabarī and was acknowledged for that. A second possible explanation for al-Ṭabarī’s practices lies in the general expectations surrounding the genre of history writing. In a similar case, discussed by Letizia Osti, Abū Bakr Muḥammad b. Yaḥyā al-Ṣūlī (d. ca. 335/947) copied from an older contemporary’s book (or notebooks) material that went back to an early Abbasid poet. Ibn al-Nadīm, judging al-Ṣūlī ill for it, writes, “He copied it altogether and claimed it as his own.” Pondering the effect on al-Ṣūlī’s reputation, Osti asks: “Can an author who is known to have stolen another’s work be considered a good eyewitness?” She notes, however, that the question “does not seem to have been taken into consideration by any of the historians whose opinion we have reviewed.” It appears, then, that the rules governing reuse of poetry seem to have differed from those applied to historiographical material, which permitted “the appropriation and reworking of themes and forms.”23 The concept of sariqa may have been deployed in conversations about poetry, less often in those about prose. If this is the case, the only extraordinary feature of al-Ṭabarī’s borrowing from Ibn Abī Ṭāhir would be the scale of the reuse. Given this scale, however, a third explanation seems even more persuasive to me. This relates to al-Ṭabarī’s view of Ibn Abī Ṭāhir, including the latter’s origins and status (he began his career as a schoolteacher), his reputation in literary circles, and the methods he favored. As Shawkat Toorawa has noted, Ibn Abī Ṭāhir is one of “very few writers who does not appear to have been involved in the many sectarian, doctrinal, and theological issues of the day.”24 There is nonetheless ample evidence that his contemporaries and the generations that followed disparaged him. Their criticisms arose in literary circles, but they express sensitivities that might well have resonated with al-Ṭabarī. For example, al-Ṣūlī complained that Ibn Abī Ṭāhir got his facts and citations wrong when recounting an anecdote in which a caliph sought a courtier’s opinion on an appointment. Ibn Abī Ṭāhir named the caliph as al-Maʾmūn, but al-Ṣūlī claims that the incident in fact took place in the earlier context of the Umayyad Caliph Hishām b. ʿAbd al-Malik (r. 105–125/724–743). According to al-Ṣūlī, the error is characteristic of Ibn Abī Ṭāhir: I have reported [the story] according to the transmissions of reliable authorities and from several sources, but Ibn Abī Ṭāhir attributes it to

23 24

Osti, Author as protagonist 244–245. Toorawa, Notes toward a biography 131.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

444

savant

al-Maʾmūn and Aḥmad b. Yūsuf without naming an authoritative source. This is because he is someone who gets his knowledge from books [without the aid of a teacher; a ṣaḥafī], someone who does himself harm by speaking too much (ḥātib layl). He imposes as a condition the selection of good poetry for inclusion in his anthologies, but he actually includes bad poetry. And he claims to be picky and careful. Furthermore, he relates untruths and makes mistakes in his dating and in his attribution of poetry. […] I saw him in Basra in 277 [890 or 891]; Aḥmad b. ʿAlī al-Mādarāʾī had summoned him there. I took down in writing two or three of his [Ibn Abī Ṭāhir’s] lectures, but when I realized he was a ṣahafī, in whom I saw nothing I wanted, I left him. I am sorry that I have to speak ill of and belittle a littérateur (aḥad min ahl al-adab), but I have no choice but to speak the truth and state matters as they are.25 As Toorawa and Osti point out, the label ṣahāfī that al-Ṣūlī applies to Ibn Abī Ṭāhir is loaded with meaning. It denotes “someone who relied on books and on libraries for his knowledge rather than on memory and on oral and direct acquisition from others” (Toorawa);26 in other words, “someone who relied on books exclusively without the aid of a teacher” (Osti).27 The term ṣaḥafī derives from the word for a leaf or a page of a book and, by extension, a piece of writing, a letter, or a book. Other derivatives of the root carry negative associations: taṣḥīf signifies corrupt speech. One also finds ṣaḥafī used to mean “someone who errs while reading or writing.” Toorawa observes that “the connection between the meaning ‘someone who relies on books’ and the meaning ‘someone who errs when reading’ is, of course, significant in a context where the precise value of book and book-related knowledge has not yet been settled.”28 It is possible that al-Ṭabarī considered Ibn Abī Ṭāhir’s method of collecting information unreliable, or feared that others would see it so.29 If

25 26 27

28 29

Translated by Toorawa, Ibn Abī Ṭāhir Ṭayfūr 22, citing al-Ṣūlī, Awrāq 209–210. Toorawa, Ibn Abī Ṭāhir Ṭayfūr 22. Osti, History and memory 70–71. Osti notes that al-Ṣūlī also refers to Ibn Abī Ṭāhir as a ḥātib layl, meaning “somebody who mixes reliable and unreliable material as if he were gathering wood in the dark.” Al-Ṣūlī’s assessment of Ibn Abī Ṭāhir “was not concerned with the books themselves, but with how they were used and composed.” For the origins and meaning of ḥātib layl, see also Toorawa, Ibn Abī Ṭāhir Ṭayfūr 140, n. 25. Toorawa, Ibn Abī Ṭāhir Ṭayfūr 22–23. Ibn al-Nadīm, too, paints an unflattering picture of Ibn Abī Ṭāhir. As noted above, he quotes Jaʿfar b. Ḥamdān (author of a Kitāb al-Bāḥir). Jaʿfar reports that Ibn Abī Ṭāhir began his career as a schoolteacher but went on to establish himself in the booksellers’ market in the eastern district of Baghdad. According to him, Ibn Abī Ṭāhir was a most prolific

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

445

not compelled to cite Ibn Abī Ṭāhir, he might have preferred to avoid naming him.30 In sum, al-Ṭabarī’s decision not to cite Ibn Abī Ṭāhir may have had several related reasons. He worked at a time when the rules for citation were still taking shape, especially for the medium and the genre in which he worked. The person of Ibn Abī Ṭāhir did not inspire him to extend the rules to accommodate a fellow compiler. But al-Ṭabarī was also a writer of these rules through the example he set. His own Taʾrīkh was reused extensively in other works, often without explicit citation. In this way, arguably, his choices regarding whom to cite and whom not to cite shaped both the legacy of Ibn Abī Ṭāhir—who is invisible within al-Ṭabarī’s work—and his own legacy, insofar as his work was subsequently mined without citation, much like Wikipedia is today.

4

Concluding Remarks

Thanks to the OpenITI corpus, it is now relatively easy to search texts for the terms mentioned in this chapter and to see whether an author includes or omits the name of a predecessor. Such searches are useful and important. When we find long chunks of overlapping text, we should ask what an author is doing, how he cites past authorities, and why. A starting point of this investigation, I believe, should be the assumption that authors such as al-Ṭabarī worked in ways that largely made sense to their contemporaries. Cases of extensive reuse like al-Ṭabarī’s should thus prompt us to consider the ideas and expectations that drove this practice. Tracing these across the body of material made available by the digital age will enable a fresh picture of authorial practices to emerge.

30

compiler, but also most banal (ablad ʿilman) and prone to grammatical errors. Ibn alNadīm also quotes an apparently embittered al-Buḥtūrī: “I have never seen anyone […] whose speech was more corrupt, whose mind was duller, and whose language was more ungrammatical. […] No one plagiarized more than he did (asraq al-nās).” Ibn al-Nadīm, Fihrist, “Akhbār b. Abī Ṭāhir.” It is worth noting, however, that the story that, according to al-Ṣūlī, Ibn Abī Ṭāhir recorded erroneously is also found in al-Ṭabarī’s Taʾrīkh (although briefly). See alignment 19 in the dataset; al-Ṭabarī, History of al-Ṭabarī xxxii, 179–180 (year 212, “al-Maʾmūn appoints Ghassān b. ʿAbbād governor over Sind”).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

446

savant

Bibliography The book files (structurally annotated in OpenITI mARkdown) and other data cited in this chapter are available online in a Zenodo open data repository under the doi 10.5281/zenodo.5118900.

Sources (OpenITI Files) Ibn Abī Ṭāhir Ṭayfūr (d. 280/893), Kitāb Baghdād, https://github.com/OpenITI/0300AH /blob/master/data/0280IbnTayfur/0280IbnTayfur.Baghdad/0280IbnTayfur.Baghda d.Shamela0005880‑ara1.completed (last accessed September 2, 2021); based on Ibn Abī Ṭāhir Ṭayfūr, Kitāb Baghdād, ed. ʿI. al-ʿAṭṭār al-Ḥusaynī, Cairo 2002. Ibn al-Nadīm (d. 385/995), al-Fihrist, https://github.com/OpenITI/0400AH/blob/maste r/data/0385IbnNadim/0385IbnNadim.Fihrist/0385IbnNadim.Fihrist.Shia003355‑ar a1.mARkdown (last accessed September 2, 2021); based on Ibn al-Nadīm, Fihrist, ed. R. Tajaddud, [Tehran] [1971 or 1973]. al-Ṭabarī (d. 310/923), Taʾrīkh al-rusul waʾl-mulūk, https://raw.githubusercontent.com/​ OpenITI/0325AH/master/data/0310Tabari/0310Tabari.Tarikh/0310Tabari.Tarikh.Sh amela0009783BK1‑ara1.completed and .mARkdown (last accessed: 25 May 2022); both based on al-Ṭabarī, Taʾrīkh al-Ṭabarī (Taʾrīkh al-rusul waʾl-mulūk), ed. M. AbūʾlFaḍl Ibrāhīm, 10 vols. (vol. 11 excluded as contains additional texts), Cairo 1387/1967. Yāqūt al-Ḥamawī (d. 626/1229), Muʿjam al-udabāʾ (Irshād al-arīb ilā maʿrifat al-adīb), https://raw.githubusercontent.com/OpenITI/0650AH/master/data/0626YaqutHa mawi/0626YaqutHamawi.MucjamUdaba/0626YaqutHamawi.MucjamUdaba.Sham ela0009788‑ara1.mARkdown (last accessed September 2, 2021); based on Yāqūt alḤamawī, Muʿjam al-udabāʾ, ed. I. ʿAbbās, 7 vols., Beirut 1993.

Sources (in Print) Ibn Abī Ṭāhir Ṭayfūr, Kitâb Baġdâd, ed. and trans. [into German] H. Keller, Leipzig 1908. al-Ṣūlī, Kitāb al-Awrāq: Qism min akhbār al-shuʿarāʾ, ed. J. Heyworth-Dunne, Cairo 1934. al-Ṭabarī, The history of al-Ṭabarī, vol. 32: The reunification of the ʿAbbāsid caliphate (a.d.813–833/a.h.198–218), trans. C.E. Bosworth, Albany 1987. al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, ed. M.J. de Goeje et al., 15 vols. in 3 series, Leiden 1879–1901.

Studies Amedroz, H.F., Review of part 2 of Kitab Baġdād von Ahmad ibn Abi Ṭāhir Ṭaifūr (Sechster Band), ed. and trans. H. Keller, jras 40 (1908), 855–64. Borrut, A., Entre mémoire et pouvoir: L’espace syrien sous les derniers Omeyyades et les premiers Abbassides (v. 72–193/692–809), Leiden 2011.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

al-ṭabarī’s unacknowledged debt to ibn abī ṭāhir ṭayfūr

447

Bosworth, C.E., Translator’s foreword, in al-Ṭabarī, The history of al-Ṭabarī, xxxiii, Albany 1987, 1–8. Déroche, F., Les manuscrits arabes datés du iie/ixe siècle, in Revue des études islamiques 55–57 (1987–1989), 343–379. Heinrichs, W.P., Sariḳa, in ei2, http://dx.doi.org.iij.idm.oclc.org/10.1163/1573‑3912_islam​ _COM_1446 (last accessed July 27, 2021). Heinrichs, W., An evaluation of sariqa, in Quaderni di Studi Arabi 5–6 (1987–1988), 357– 368. El-Hibri, T., Reinterpreting Islamic historiography: Hārūn al-Rashīd and the narrative of the Abbasid caliphate, Cambridge 2008. Gruendler, B., The rise of the Arabic book, Cambridge, MA 2020. Keller, H., Die Stellung der “Annalen” Ṭabarîs zum Kitâb Baġdâd, in Ibn Abī Ṭāhir Ṭayfūr, Kitâb Baġdâd, ii, Leipzig 1908, xiii–xxvi. Kennedy, H., Caliphs and their chroniclers in the middle Abbasid period (third/ninth century), in C. Robinson (ed.), Texts, documents and artefacts: Studies in honour of D.S. Richards, Leiden 2002, 17–36. Lindstedt, I., The life and deeds of ʿAlī b. Muḥammad al-Madāʾinī, zgaiw 20–21 (2012– 2014), 235–270. Osti, L., The author as protagonist: Biographical markers in historical narratives, in jsai 45 (2018), 239–275. Osti, L., History and memory in the Abbasid caliphate: Writing the past in medieval Arabic literature, London 2021 (forthcoming). Rosenthal, F., Ibn Abī Ṭāhir Ṭayfūr, in ei2, http://dx.doi.org.iij.idm.oclc.org/10.1163/1573​ ‑3912_islam_SIM_3056 (last accessed July 27, 2021). Toorawa, S.M., Ibn Abī Ṭāhir Ṭayfūr and Arabic writerly culture: A ninth-century bookman in Baghdad, London 2005. Toorawa, S.M., Notes toward a biography of Ibn Abi Tahir Tayfur (820–893), in Social Sciences & Humanities and Law & Management Research Journal (University of Mauritius) 1 (1998), 121–140.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 23

Heraqleh: A New Interpretation Andrew Petersen

This paper was directly inspired by Hugh Kennedy, who first introduced me to Abbasid architecture in his lectures at St Andrews in 1983.* In the same year Kassem Toueir published an article in which he suggested that the early Islamic site of Heraqleh near Raqqa was built as a “victory monument to Hārūn alRashīd.”1 While this seemed to be an interesting explanation for this enigmatic structure, it was not entirely satisfactory because it did not seem to relate well to the architecture of the monument. My interest in the site was renewed about a year ago when I was asked to write an article about the archaeological evidence for the use of astronomy and astrology in the planning of early Islamic cities.2 While the role of astrologers in planning the Round City of al-Mansūr in Baghdad is well-known, the early Abbasid city itself has long disappeared so that there is no way of assessing how the actual layout of the city reflected astronomical or astrological principles. In other cases, where there are extant remains of early Islamic cities, there is little evidence for astronomical or astrological influence in their planning. The only exception to this is the role of the qibla in the orientation of buildings and streets, although in the majority of cases other factors, such as topography, wind direction, or existing roads, also seem to be major influences on urban design. The two sites for which an astrological or astronomical significance seem likely are the octagon at Qādisiyya near * Many people have generously helped with the research for this chapter, providing both advice and data. I am particularly indebted to Andrea Ricci, both for providing her own images as well as access to data from the Deutches Archaeologisches Institute. Louise Rayne provided invaluable advice on the irrigation systems of Raqqa as well as a 1967 corona image and access to her own photographs of the site. Dan Lawrence generously sent me copies of Tony Wilkinson’s photographs of Heraqleh taken in 1995. Dan Demeter provided me with recent photographs of the site, which include details not previously visible. Andrea Luigi Corsi provided valuable information and advice on the dating of the decorative stonework. The astronomer William Tobin provided a careful and objective appraisal of Heraqleh’s possible use as an observatory based on his own experience of astronomical instruments and the location in ninth-century Syria. Fabio di Silva not only gave advice on the astronomical alignment of the monument but also set it within the wider context of cultural astronomy. It should be stated that the conclusions along with any errors are my own responsibility. 1 Toueir, Heraqlah. 2 Petersen, Commentary on Ibrahim Allawi.

© Andrew Petersen, 2022 | doi:10.1163/9789004525245_024

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

449

heraqleh: a new interpretation

figure 23.1

Map of the Middle East showing the location of Raqqa and other sites mentioned in the text map by ifan edwards

Samarra and the monument at Heraqleh outside the early Islamic city of Raqqa. The octagon at Qādisiyya has been studied in detail by Alastair Northedge, who came to the conclusion that it was built by the Abbasid Caliph Hārūn al-Rashīd when he was looking for an alternative place to locate his capital outside Baghdad and before he established himself at Raqqa.3 The fact that the site at Qādisiyya was not finished, together with some details of its architecture, which clearly indicate a date in the early ninth century, confirm this hypothesis.4 The octagonal plan of the outer walls together with a rectangular central building clearly suggest a symbolic function. By analogy with Baghdad, which it was said to resemble, the plan of the Qādisiyya octagon was probably based on astrological calculations. The ground plan of the monument at Heraqleh, a square within a circle, also strongly suggests astrological symbolism and bears comparison with the octagon at Qādisiyya. According to Kassem Toueir and the majority of modern

3 Northedge, Historical topography 82–94. 4 Northedge and Kennet Archaeological atlas 138–139.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

450

figure 23.2

petersen

Map showing the location of Heraqleh in relation to Raqqa and other sites in the vicinity map by ifan edwards after original by louise rayne 2014

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

451

heraqleh: a new interpretation

scholars Heraqleh was also built by Hārūn al-Rashīd who resided at the nearby city of Raqqa for 12 years begining in 180/796. While it is easy to suggest an astrological or astronomical association for this building, many other questions remain unanswered, including its precise function, its relationship to other structures in the area, and the reason it was apparently unfinished. Although it may not be possible to fully answer all of these questions, it is hoped that this paper will provide a deeper understanding of this structure and its potential significance.

1

Discovery and Historiography

First of all, it should be made clear that there is no near contemporary text that discusses the structure currently identified as Heraqleh. Possibly the earliest surviving eye-witness record of the site is by the geographer Yāqūt al-Ḥamawī (574/75–626/1179–1229), who gives the following account of a visit: “I came across this amazing place during a pleasant journey to Raqqa and found a structure next to the Euphrates between Rāfiqa and Bālis, which is called Heraqlah after the Heraqlah, which is in the land of the Byzantines and the crumbling remains of the fort that has survived to the present day. The ruins comprise a large complex with wonderous buildings, and it is close to Ṣiffīn, which is on the west side of the site.”5 The first European description of Heraqleh appears in the accounts of the Euphrates Expedition, which visited in 1836 as part of the survey work for a fast route to India. The site is mentioned in the memoirs of William Ainsworth, one of the members of the expedition, and is described as follows: “In the woods on the opposite side rose a castellated building, called Arakla by the Arabs, and apparently ancient Zenodotium, on the prolongation of the causeway northwards.”6 The site is also shown as a circular structure on the map produced by the expedition. However, the first detailed description of the building appears in the writings of the German traveler Sachau, who visited the site in December 1879. As this is the earliest modern description, it is worth quoting in full. On Thursday 18 December I rode from Raqqa wnw partly in the floodplain and partly in the steppe to the ruins of Heraqleh, which I reached after 1 hour 48 minutes. It lies on the edge of the steppe and comprises 5 Yāqūt al-Ḥamawī, Kitāb Muʾjam al-Buldān iv, 121 (my translation). 6 Ainsworth, Personal narrative 276.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

452

petersen

figure 23.3

corona image from 1967 showing Heraqleh in relation to the Euphrates floodplain (darker shade) and an adjacent partially eroded circular structure to the south-east. image courtesy of louise rayne

a not very large square with extraordinarily high strong walls built from unhewn stones of milk white gypsum, and some vaults covered by a mixture pebble and earth mortar are still visible. The location of the entrance gate is not clear, but it was presumably on the northwest side. There is no evidence of a large town here, but I would like to point out that in a small circle around the square wall there are some walls and small hills in

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

453

heraqleh: a new interpretation

the ground (ruins), which indicate that in earlier times there were some houses around the outside of the castle. Therefore, I consider the ruins of Heraqleh to be a castle guarding the Euphrates crossing from Elḥammâm (Thapsacus?) to protect the east bank from the Bedouin. It reminded me of the Roman castles I saw in Syria, which have an equally solid design, and is, therefore, probably a building of the Roman-Greek period, not the Arab Middle Ages. There are no traces of ornament or inscriptions. It goes without saying that part of the ruin was blown away by the desert sand.7 Both these early descriptions agree that Heraqleh was some form of fortification and that it was of pre-Islamic origin; however, neither was able to attach the monument to any historical source. In November 1907 the site was visited by Ernst Herzfeld, who was the first to make detailed plans of both the central structure and the outer ring with its four gates. Although there are some inaccuracies in Herzfeld’s plans (e.g., the shape of the gateways and no depiction of the inner circle), his drawings are an invaluable record of the structure before it was damaged by modern development. Features of particular interest are the semicircular buttress towers on the outer face of the circular enclosure wall, which are no longer clearly, and the canal that cuts through the southern part of the enclosure wall.8 Although Herzfled’s partner on the Euphrates project, Friedrich Sarre, originally dated the building to the Parthian period, Herzfeld linked the building to a story related by Yāqūt al-Ḥamawī. Specifically, Herzfeld stated that “the true interpretation of the ruin only emerged afterwards, it is nothing more than a castle of Hārūn al-Rashīd. In Shawwal 190/august 806 Hārūn had conquered Herakleia on the Taurus. And Yāqūt relates that among the prisoners was the daughter of the Patriarch of Herakleia who Hārūn took to Raqqah. And he built a castle (ḥiṣn) on the way from Rāfiqah to Bālis on the Euphrates so the castle was called Hiraqlah.”9 This new dating was not initially accepted, and the famous British archaeologist Gertrude Bell argued that the form of vaulting used in the structure must have been built using centering (i.e., on a wooden frame) and this “would be enough to show that the building does not belong to the Mohammedan period.”10 However, by the 1960s the building was generally accepted as an Islamic monument, and K.A.C. Creswell suggested that the building should be 7 8 9 10

Sachau, Reise 244–245 (my translation from German original). Sarre and Herzfeld Archäologische Reise 161–163. Yāqūt al-Ḥamawī, Kitāb Muʾjam al-Buldān iv, 962. Bell, Amurath to Amurath 53.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

454

petersen

figure 23.4 Herzfeld’s drawing of Heraqleh before modern disturbances—note the canal that crosses the outer enclosure wall to the south and the semicircular buttress towers between the gates. Interestingly, there is no depiction of the inner ring around the central square building

one of the priorities for excavation by the Syrian Department of Antiquities. In response, a series of excavations were carried out at the site by Kassem Toueir starting in 1969, with further campaigns in 1976 and 1990. Based on his research, Kassem Toueir confirmed the Islamic dating of the structure but was skeptical about Yāqūt’s story that it was built as a palace for the daughter of the patriarch. In place of Yāqūt’s explanation, Toueir argued that Heraqleh was built by Hārūn al-Rashīd as a victory monument to celebrate his capture of the city of Herklea (modern Turkish Eregli) in Anatolia. Toueir based this interpretation of the building on a number of factors, starting with “the exaggerated hymns of praise for the victory of Hārūn against Byzantium” recorded in the accounts of Yāqūt and al-Ṭabarī. Other reasons for this attribution were the unfinished condition of the monument caused by the death of Hārūn al-Rashīd in 183/809 ce, Hārūn al-Rashīd’s imperial ambitions and the units of measurement used, which it is argued have a symbolic value (Zahlensymbolik), as do the geometrical forms used in the gates.11 The attribution of the building to the early Islamic period, and in particular the time of Hārūn al-Rashīd, is now generally accepted. However, Toueir’s interpretation of the building as a victory monument is less satisfactory. For example, the 1989 revised version of Early Muslim architecture by K.A.C. Creswell compares Heraqleh to the Dār al-Imāra at Merv and the palaces of Samarra. Although the idea of a victory monument is not rejected out of hand, Creswell states “its unfinished condition means that we have no way of deciding what sort of structure was planned above it.”12

11 12

Toueir, Das Hiraqla 141. Creswell, A short account 278.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

455

heraqleh: a new interpretation

2

Review of Archaeological Excavations

Although two major summaries of this work have appeared in international publications, no detailed record of the excavations has so far been published. The excavations concentrated on the central platform and the four different shaped gates in the outer enclosure (S-square, W-hexagonal, N-circular and E-circular). In particular, there are no section drawings or any indication of the nature of the material removed from the interior of the central structure. However, from the general accounts it appears that much of the central mound was filled with soil to create a terrace, while any fill removed from the vaults or other chambers within the structure was mostly filled with collapsed debris. In any case, there do not appear to have been many finds from the excavations that support the idea that much of the interior was filled in, and this also adds to the impression that the building was not used or reused as a place of occupation. Among the few finds reported from the excavation, there were some fragmentary pieces of stucco as well as glass vessel fragments and one small glass flask. Both the stucco fragments and the glass are consistent with an early Islamic (ninth century date). In addition to the portable finds, there were a few interesting examples of carved stones incorporated into the building, including two inscriptions, and two symbols carved in relief. The most interesting inscription is a short Arabic text written in Kufic script, part of which was inserted upside down. The inscription is located on the east side of the south tower and is 0.36 m long and 0.10m high and extends over three stones. On the first stone, the words māʿamala hāḍā can be deciphered, on the other two stones the name Huḍayl appears twice, and on both occasions it has been inserted upside down. Also on the southern face of the south tower, there is an inscription that denotes the letter ‘E’. Of more interest, there is a Greek cross carved in relief and incorporated into one of the walls in the southwest quarter of the central building. The position of the cross within the wall clearly indicates secondary reuse. Most of the carved stones seem to have been reused and are not in their original locations; however, there are two examples of carved decoration in the western gateway of the circular outer enclosure wall. The decoration on each of the stones is identical, although on one of the stones the upper part has been worn away. The decoration is carved in relief and consists of a central ring enclosed within two squares forming an eight-pointed star. This form of decoration has a long history, dating back to at least the fourth century ce in Roman mosaics in Italy, and continued to be used into the Abbasid period, where it appears in stuccos at Samarra. However, the specific form used here with intersecting outlines is more characteristic of Umayyad and pre-Islamic

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

456

petersen

figure 23.5

Google Earth Image of Heraqleh showing central building and concentric inner and outer enclosures. Note four different shaped gates in outer enclosure West (hexagonal), North (octagonal), East (circular) and South (square). google earth 7.3.2022. 35°57’24”n, 38°55’51.”e, elevation 250m [online] available through: http://earth.google.com/web/​ @35.95654874,38.93384356,249.39530517a,916.15033364d,35y,‑0h,0t,0r [accessed 19.07.22]

decoration in the region, perhaps suggesting a date before the time of Hārūn al-Rashīd.13 Taken together, the evidence from the carved and inscribed stones found at Heraqleh indicate that the structure dates to the Islamic era, both because of the incorporation of reused stones bearing Arabic inscriptions and a reused cross, which is unlikely to have been removed from its original context when the area was under Byzantine rule. The decoration on the other hand suggests a terminus anti-quem significantly before the ninth century.

13

Personal communication with Andrea Luigi Corsi (March, 27 2021).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

heraqleh: a new interpretation

3

457

The Canal

The circular outer enclosure of Heraqleh is cut by the line of a canal, which according to Herzfeld and others dates to the early ninth century. Kassem Toueir made a study of the course of the canal based on aerial photographs, showing that it ran for approximately 16km parallel to the Euphrates.14 The canal originated in an old bed of the Euphrates eight km to the west of Heraqleh and terminated in the marshes of the Balīkh valley to the east of Raqqa. In support of an early ninth-century date, Toueir cites Yāqūt, who records that a canal known as the Nahr al-Nīl was excavated on the orders of Caliph Hārūn alRashīd. In further support of his argument, he notes that the line of the canal serves the agricultural area to the north of Raqqa as well as the palaces built during the time of Hārūn al-Rashīd’s residence in Raqqa (180–192/797–808). However, recent archeological work carried out by the late Tony Wilkinson and Louise Rayne has called into question the ninth-century date for the canal.15 Their main argument is that the canal cannot be dated to the time of Hārūn al-Rashīd, as it clearly postdates the circular outer enclosure at Heraqleh. Louise Rayne also points out that there are a series of qanat holes to the east and north of Heraqleh that predate the canal. Within the site of Heraqleh itself, Wilkinson and Rayne identified traces of the qanat in the form of two wells ten m apart on either side of the eastern iwan of the central building and “a raised line between them suggested that an underground tunnel connected the wells which appears to be contemporary with the site.” According to this, the qanat should be dated to the Umayyad period, although it is not clear what this dating is based upon.16

4

Summary of Dating Evidence

Louise Rayne’s post-Abbasid dating of the canal rests on her acceptance of Heraqleh as an Abbasid monument built during the reign of Hārūn al-Rashīd. However, if Heraqleh itself is of an earlier date, this would mean that a ninthcentury Abbasid date for the canal would not be a problem. Unfortunately, most of the dating material for Heraqleh is relative, and until some form of absolute dating has been carried out on the structure (either osl 14 15 16

Toueir, Le Nahr el-Nil 217–227. Rayne, Water and territorial empires 305–307; Wilkinson and Rayne, Hydraulic systems 4–5. Kamash, Water supply 4.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

458

petersen

figure 23.6 Plan of the central building of Heraqleh after toueir 1983

or C14 accelerator dating would be suitable), any dating must be provisional. However, both the evidence from the decoration and the canal suggest that Heraqleh may predate Hārūn al-Rashīd.

5

A Victory Monument?

While the precise dating of Heraqleh may need some refinement, the evidence presented above shows that it is clearly a building from the early Islamic period. A bigger problem seems to be the unique form and size of the structure, which defies easy characterization. While Yāqūt’s story of a palace for the captured Byzantine princess seems too fanciful, Toueir’s identification of the building as a victory monument is also problematic. For example, there is no external evidence suggesting that Hārūn built such a monument and also nothing about the structure indicates a victory monument. Although victory monuments were known at the time and often figured in the conflicts between the Byzantines and the Sasanians, less is known about their use in the Islamic period. Certainly there were elaborate diplomatic exchanges between the Byzantines and the Abbasids, which involved overawing their rivals with the size of their capital cities and the luxuriousness of their palaces.17 There are, however, no structures that have been clearly identified as victory monuments dating from the early Islamic period (there are, however, a few examples from later periods, such as the now-destroyed Mamluk victory monuments at Haṭṭīn and ʿAyn Jalūt). The art historian Oleg Grabar claimed that the Dome of the Rock was built as a victory monument, with its symbolically significant location in Jerusalem and mosaic decoration including depic-

17

Canepa Two eyes of the earth 100–121.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

heraqleh: a new interpretation

459

tions of jewels crowns and other insignia that could be linked to Sassanian or Byzantine rulers.18 However, the main theme of the decoration appears to be a glorification of God and an assertion of the core beliefs of Islam (i.e., the word of the Quran and the role of Muḥammad as prophet). Even if one accepts Grabar’s interpretation of the Dome of the Rock as a form of victory monument, it can only be regarded as a general statement of the superiority of Islam over other religions and does not commemorate a particular event or location. In a perceptive article on the subject of Islamic victory monuments, Thomas Leisten suggested that permanent victory memorials were rare in the Islamic world.19 Instead, victories were recorded in poems and paintings rather than in the large grandiose structures that were characteristic of Roman and Byzantine times. For example, he cites the famous Abbasid-era poet al-Mutanabbī (303–354/915–965), who described a painting commemorating the victory of the Hamdanid ruler Sayf al-Dawla over the Byzantines at the fortress of alBarzūya north of Antioch.20 The painting has a depiction of the Byzantine emperor on his knees kissing a carpet on which the Hamdanid amīr was seated and in the background were groups of fighting animals representing the battle between Muslims and Byzantines. While the painting clearly showed the victory of Islam, the fact that it was painted on the side of a tent rather than on a permanent building shows the ephemeral nature of such victory commemorations. A more famous example of a Muslim victory painting is the mural on the west wall of the bathhouse at Khirbat al-Mafjar in Jordan. The painting shows six kings paying homage to the seated Muslim ruler or caliph and recalls Parthian or Sasanian reliefs commemorating victories. However, the location of the painting in a remote desert location, and the fact that there are paintings on other themes throughout the bathhouse, reduces its significance as a victory monument.

5

Alternative Hypotheses

If Heraqleh was not built as a victory monument, what was its intended function? Although Yāqūt’s story of a Byzantine princess seems unlikely, it is worth considering whether the structure would have been suitable as a palace or for-

18 19 20

Grabar, Umayyad dome of the Rock. Leisten, Mashhad Al-Nasr. Leisten, Mashhad Al-Nasr 23.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

460

petersen

figure 23.7

View of the central building or terrace at Heraqleh from the west, with the hexagonal west gate in the foreground image courtesy of cristina tonghini

figure 23.8

Heraqleh, central building from SW in 2010 image courtesy of andrea ricci

tress. It is easy to see how the height and size of the central structure led many of the early visitors to the site to regard it as some form of fortress. However, the circular enclosure wall with four gates and the lack of defensive features such as ditches or a single fortified entrance to the central building does not suggest a fortification. Creswell was clearly in favor of the idea that Heraqleh was intended to function as a palace. However, it should be pointed out that there were many Abbasid palaces at Raqqa, some with extravagant architecture, including features such as glass floors, but they were all based on rectangular orthogonal plans, and none resembled Heraqleh. Certainly the lack of structures on the top

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

461

heraqleh: a new interpretation

figure 23.9

Aerial view of Heraqleh circa 1985 image courtesy of deutsches archaeologische intstitut

figure 23.10 Stucco from excavations at Heraqleh with 80cm white and yellow scale divided into 10cm sections. image courtesy of the deutsches archaeologische institut

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

462

petersen

of the central building make it difficult to guess at the projected final form of the complex; however, it is worth looking again at the evidence that does survive to see if there are any clues about the intended function of the building. First of all, it should be noted that the building incorporates a range of geometrical forms, which, as Kassem Toueir recognized, must have had some symbolic value. The geometry runs through the structure at all levels, from the overall shape of the complex, two concentric circles around a central square, to the decorative panels in the gateways with circles inside eight-pointed stars. Even the metrology of the complex reflects harmonic geometric principles; thus, Toueir established that “all measurements in metres became round numbers without fractions when they were converted into dheraʿa […] we also discovered that all the measurement converted into dheraʿa are multiples of the number two. All of these phenomena cannot be regarded as coincidental.”21 The excavation also revealed a range of curiously shaped rooms within the central complex, including square, rectangular, and L-shaped rooms. These rooms were not connected by doorways, and despite the fact that they had fired brick floor surfaces, they were later filled in with river gravel and building debris. While Toueir recognized the potential symbolic value of the geometry and metrology of the building, he did not satisfactorily link this to the function of the building, only suggesting that it was an intrinsic part of imperial ideology. Given that the Abbasid caliphs employed both astrologers and astronomers who also dealt with geometry, it seems likely that the structure was conceived with a cosmic purpose. The problem for modern researchers is understanding how the architecture could have been used in a practical sense to relate to the cosmos. This “problem of methodology is at its most acute when we are dealing with a monument that is unique.”22 There are two possibilities that suggest themselves: as a location for religious rituals or as a base for measuring cosmic events. Neither of these suggested explanations are exclusive and do not preclude other functions, but they are simply another way of trying to understand this extraordinary structure.

21 22

Toueir, Das Hiraqla 141–142. Hoskin, History of astronomy 1.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

463

heraqleh: a new interpretation

figure 23.11

6

Greek cross reused within the central building at Heraqleh image courtesy of dan demeter

Heraqleh as a Religious Building

If this was a religious structure, it is unlikely to have related to the two religions of the book, Judaism and Islam, both of which have well-known, defined religious forms. On the other hand, the construction of a non-Muslim religious structure at the heart of the Abbasid Caliphate seems an anathema to the concept of the early Islamic state, yet it is known that other religions continued to be practiced and in some cases even continued to build new places of worship during this period.23 Within the region of northern Mesopotamia, the only semi-tolerated religious community of which we have knowledge from the Islamic sources are the Sabians from Ḥarrān. Although it seems that the inhabitants of Ḥarrān were not Sabians but another cult based around the movement the of stars and planets, they were given recognition as Sabians to make them more acceptable

23

Piccirillo, Umayyad churches.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

464

petersen

to Muslim authorities. Ḥarrān was first taken by the Arab Muslims in late 639 or 640ce, and significantly, the pagans (Sabians) negotiated a separate peace with the conquerors, distinct from that of their Christian neighbors. Under the Umayyads the Sabians continued their pagan practices and were not forced to convert until the early Abbasid period, during the reign of Caliph al-Maʾmūn. The caliph, who visited the city shortly before his death in 833ce, gave the pagans the choice of conversion to Islam or one of the sects legalized by the Quran as people of the book, of which the Sabians may have been one option. The beliefs and religious practices of the Sabians of Ḥarrān were recorded by a number of Muslim authors and have been discussed in detail by Bayard Dodge, and more recently by Tamara Green.24 The religion was based on the belief that the cosmos has a primal cause (too transcendent for humans to comprehend) and that the sun, planets, and stars were the home of intermediate deities. Within Ḥarrān there was a central temple surrounded by the temples of the lesser deities. Each of the planets was represented by a different shape, thus the shrines of Saturn and Mercury were hexagonal, the shrine of Jupiter triangular, Mars and the Sun were square, and the moon had a pentagonal shape. According to the tenth-century geographer al-Iṣṭakhrī, the place of religious worship was a lofty mound that served as a “high place” for ritualistic purposes and observing the rising and setting of heavenly bodies. In addition to the temples above ground, there were subterranean halls, which were used for initiation rites where priests spoke through tubes so as to make the idols talk. Unfortunately the temples and associated shrines at Ḥarrān were destroyed in the early eleventh century, but the nearby ritual site of Sumatar, which was investigated by Segal in the 1950s, still survives.25 The complex at Sumatar was built in the second century ce and was also based around planet worship and has the remains of a central mound, subterranean chambers, and chambers of different shapes—square, circular, and cylindrical on a circular base. While it is not possible to make direct comparisons between the temples at Ḥarrān and the monument at Heraqleh, there are certainly some similarities. For example, the square, circular, hexagonal, and octagonal gates at Heraqleh recall the different-shaped shrines for the planets at Ḥarrān. The central square building with the four subterranean vaults at Heraqleh also recalls the central mounds with subterranean chambers at both Ḥarrān and Sumatar. Interestingly, the holes in the vaults at Heraqleh also bring to mind the tubes used by the priests to communicate during initiation rites. The unfinished state

24 25

Dodge, Sabians of Harran; Green, City of the moon god 94–123. Segal, Pagan Syriac monuments.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

465

heraqleh: a new interpretation

figure 23.12

Interior of the western gate of the outer enclosure showing the location of carved stone wall decoration image courtesy cristina tonghini

of the monument might be explained by a changed attitude to the religion by the Abbasid state, perhaps following al-Maʾmūn’s visit to Ḥarrān when he threatened them with death if they did not convert to one of the accepted religions.26 Unfortunately, without corroborating written material it is impossible to prove any association between Heraqleh and the planet worshippers from Ḥarrān, but the proximity of the two sites and the fact that we know there was a community of Sabians in Raqqa during the ninth century suggests that this is a subject which could repay further investigation.

7

Heraqleh as an Observatory

As will be apparent from the discussion above, there was a close association between the Sabians and the twin sciences of astrology and astronomy. Certainly the Sabians were valued by the caliphs for their learning and ability in the

26

Dodge, Sabians 59.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

466

petersen

figure 23.13 Detail of the wall decoration in the western gate image courtesy of the deutsches archaeologische institut

cosmic sciences. One of the most famous was Thābit b. Qurra (210–288/826– 901), who was patronized by Caliph al-Muʿtaḍid and became a member of the group of astrologers and astronomers advising the caliph. Of more relevance is the case of the famous astronomer Muhammad al-Battānī (244–317/858–929), who was originally a Sabian from Ḥarrān, where his father Jābir b. Sinān alHarrānī had been a maker of astronomical instruments. From his name it is apparent that al-Battānī was a convert to Islam, at least nominally, although his great astronomical knowledge derived from his Sabian background and education. At some point before 264/877 he moved to Raqqa, where he established an observatory and made continuous astronomical observations over a period of more than 40 years, finishing in 305/918. He is also known to have visited Antioch in 288/901 and Baghdad in 317/929, where he represented a group from Raqqa claiming that they had been the victims of some injustice. His abilities and discoveries were famous and were included in Kitāb al-Fihrist (Book of the catalogue) by the tenth-century bibliographer Ibn al-Nadīm. The Fihrist states that “Nobody is known in Islam who reached similar perfection in observing the stars and scrutinizing their motions.”27 Al-Battānī’s work was well-known to Europeans, who had translated his works into Latin by the twelfth century, and his findings were later incorporated into the writings of Copernicus, Kepler, and Galileo. He wrote at least six books on astronomy, astrology, and mathematics, although his most important work is undoubtedly Zīj al-Ṣābiʾ, which is an astronomical handbook and set of tables based on the model of Ptolemy’s Almagest.28 The book is divided into 57 chapters, with the oldest surviving copy a twelfth-century manuscript held in the Escorial library in Madrid (Escorial

27 28

Ibn al-Nadīm, Fihrist x, xxxx. Van Dalen, Battani 101–103.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

467

heraqleh: a new interpretation

figure 23.14

Vault inside the central building showing the use of roughly square stone for the walls and fired brick for the vaults image courtesy of dan demeter

árabe 908).29 Among al-Battānī’s most important discoveries were: 1) the value of solar eccentricity, which was better than that achieved by Copernicus; 2) the obliquity of the ecliptic 23°35′ for the year 880; and 3) the geographical latitude of Raqqa. Al-Battānī argued that the observations of earlier astronomers were compromised by the fact that they did not make observations over extended periods of time (compared to his 41 years of data) and that the instruments they used were too small and thus unable to calculate precise astronomical measurements.30 In view of the fact that we know al-Battānī resided in Raqqa during the late ninth and early tenth century, could he have used Heraqleh as an observatory, and going further, was the monument built as an observatory? Certainly

29 30

Nallino, Al-Battani sive Albattani Opus Astronomicum. Sayili, Observatory in Islam 197–199.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

468

petersen

figure 23.15

One of a number of circular well-like holes giving access to the vaults below image courtesy of louise rayne

al-Battānī was wealthy enough to have constructed an observatory on this scale; however, in order to test this theory it is worth reviewing what we know of observatories in early Islam. Unfortunately, although we know of the existence of observatories in Baghdad, Damascus, and elsewhere, there are no physical remains, and the earliest observatory for which there is archaeological evidence is the Maragheh observatory in northeastern Iran, built in 1259.31 The Maragheh observatory has some similarities with Heraqleh, such as its elevated location (400 m × 150 m) outside the nearby city, its circular shape, and the presence of subterranean features, both for observations and deep inclines used for changing the angles of observational instruments. According to local tradition, there was a well that formed part of the observatory, from which daytime observations of the stars could be made.32

31 32

Vardjavand, Kavosh-e Rasadkhaneh; Saliba, Role of the Maragha observatory. Sayili, Observation well.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

heraqleh: a new interpretation

469

Although there is no description of al-Battānī’s observatory in Raqqa, there are some clues in his writings about the type of instruments used in his observations. The instruments mentioned include: 1) an astrolabe; 2) a gnomon divided into 12 parts but capable of division into smaller units; 3) horizontal and vertical clocks and an armillary sphere; 4) a parallactic ruler; and 5) a mural quadrant “the larger the dimension the more exact it becomes.” When discussing how to make parallax measurements, al-Battānī recommends the use of the large, or huge (ʿazīm), quadrant and the parallactic ruler. Also, a twelfth-century Arabic text on astronomy that seems to refer to al-Battānī states, “the owner or director (ṣāḥib) of the famous observatory after the time of al-Maʾmūn spent much money for the observatory.”33 Finally, it is worth reviewing the features of Heraqleh that might have been useful in an observatory. The most obvious feature is the outer circular enclosure with its division into quarters (marked by the four gates) and the 60 subdivisions (marked by the semicircular buttresses), which could easily have functioned as a solar calendar. The role of the inner circle is less clear, although it could also have functioned as a solar clock. The central square building or terrace would have served as a stable platform for astronomical instruments, while the rooms built into the terrace could have accommodated various instruments, such as wall quadrants. It is worth noting that the height of the central platform meant that the entire horizon was visible for 360° without the interference of natural features or buildings.34 This would be of great importance when observing the exact time of the rising and setting of celestial bodies. Also, the circular holes in the roofs of the four vaulted halls could have functioned as good places for viewing stars during the daytime.35 There is, however, one anomaly in the design of the structure that requires explanation. Although the four gates and iwans or vaulted halls are roughly oriented to the cardinal directions, they are not exactly aligned. This seems to be a strange anomaly, especially if this is considered to be an astronomically related building. One possibility is that the building might have been designed to face the qibla, which at this point is roughly south; however, it is not exact and given that determining the qibla was one of the primary functions of astronomical science, it seems unlikely to be the case. Another possibility is that the southern and western gates directly and very precisely face the setting of two of the

33 34

35

Sayili, Observatory in Islam 197. Raqqa Heraqlah, latitude 35.952945 N, longitude 38.881905 E, elevation 855ft above sea level (15 ft above ground). Information available at https://www.heywhatsthat.com/?view​ =BM9XMX22 (last accessed May 8, 2021). Sayili, Observation well.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

470

petersen

brightest stars in the night sky (at 900ce): Canopus in the south and Procyon in the west. This could be merely coincidental (except that to randomly hit two of the ten brightest stars in the sky is highly unlikely) were it not for the fact that both stars were not only well-known in Arabic astronomy (Canopus = Suhayl and Procyon = al-Shiʿrā al-Ghumayṣāʾ), but they were also linked together by a pre-Islamic legend.36 More research is clearly required on this point, but what is clear is that the Heraqleh’s unusual orientation cannot have been accidental.

8

Conclusion

Based on this review of Heraqleh, a few tentative conclusions can be drawn. First, the size of the site, its methods of construction, and its design clearly mark it out as a construction of unique significance. None of the current explanations for the function of the building are satisfactory. Although the building could have been either a fortress, a palace, or a victory monument, for reasons advanced in this paper, none of these seems likely. Instead, it is suggested that an explanation that takes account of the unusual geometric architecture, proportions, and building techniques is more likely to arrive at a realistic explanation. Both of the possible uses suggested in this paper are potentially controversial, yet the evidence of the structure itself indicates that the building served some specific and unusual function. Further research looking at the fabric of the structure to retrieve dating material is clearly required, and also further work on any astronomical alignments evident in the building.

Personal Note When I was an undergraduate at St Andrews I had the good fortune to have Hugh Kennedy as my tutor, which opened up a whole new dimension of Medieval History, both in terms of geography and ideology. When I asked whether it would be possible to study the Islamic world through the medium of archaeology as well as history, Hugh was very encouraging and was instrumental in getting me a place to study Islamic Art and Architecture at Oxford. Since then, Hugh has always been encouraging, especially in his enthusiasm for archaeology and in particular the Islamic archaeology of Iraq.

36

Fabio da Silva, personal communication (May 30, 2021).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

471

heraqleh: a new interpretation

Bibliography Sources Yāqūt al-Ḥamawī, Kitāb Muʿjam al-Buldān, ed. F. Wüstenfeld, vol. 4, Leipzig 1869.

Studies Ainsworth, W.F., A personal narrative of the Euphrates expedition, 2 vols., London 1888. Bell, G.L., Amurath to Amurath, London 1911. Canepa, M.P., The two eyes of the earth: Art and ritual kingship between Rome and Sasanian Iran, Berkley 2009. Creswell, K.A.C., A short account of early Muslim architecture, rev. ed. by J.W. Allan, Aldershot 1989. Dodge, B., The Sabians of Harran, in F. Sarruf and S. Tamim (eds.), American University in Beirut festival book ( festschrift), Beirut 1967, 59–86. Grabar, O., The Umayyad dome of the rock in Jeruslaem, in Ars Orientalis 3 (1953), 33– 62. Green, T., The city of the moon god, Leiden 1992. Hoskin, M., The history of astronomy: A very short introduction, Oxford 2003. Kamash, Z., Water supply and management in the Near East 63bc–ad636, DPhil thesis, University of Oxford 2006. Leisten, T., Mashhad Al-Nasr: Monuments of war and victory in medieval Islamic art, in Muqarnas 13 (1996), 7–26. Nallino, C.A., Al-Battani sive Albattani Opus Astronicum ad Fidem codicis escurialensis Arabice Editum, Rome 1903. Northedge, A., The historical topography of Samarra (Samarra studies 1), London 2008. Northedge, A., and D. Kennet, Archaeological atlas of Samarra (Samarra studies 2), London 2015. Petersen, A., Commentary on Ibrahim Allawi, “Some evolutionary and cosmological aspects of early Islamic town planning,” in Journal of Skyscape Archaeology 8 (2022), 1–18. Piccirillo, M., The Umayyad churches of Jordan, in Annual of the Department of antiquities of Jordan 28 (1984), 333–341. Rayne, L., Water and territorial empires, PhD diss., Durham University 2014. Sachau, E., Reise in Syrien und Mesopotamien, Berlin 1883. Saliba, G., The role of the Maragha observatory in the development of Islamic astronomy: A scientific revolution before the renaissance, in Revue de Synthèse 108 (1987), 361–373. Sarre, F., and E. Herzfeld, Archäologische Reise im Euphrat und Tigris (Gebeit 1), Berlin 1911. Sayili, A., The observation well in Ankara Üniversitesi Dil ve Tarih-Coğrafya Fakültesi Dergesi, vol. 2, Ankara 1953. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

472

petersen

Sayili, A., The observatory in Islam, Ankara 19882. Segal, J.B., Pagan Syriac monuments in the Vilayet of Urfa, in Anatolian studies (1953), 97–119. Toueir, K., Heraqlah: A unique victory monument to Harun al-Rashid, in World archaeology 14 (1983), 296–304. Toueir, K., Le Nahr el-Nil entre Raqqa et Heraqleh, in Bernard Geyer (ed.), Techniques et Practiques hydro-agricole traditionelles en domaine irrigué, i, Paris 1990, 217–227. Toueir, K. Das Hiraqla des Harun al-Rashid, in V. Daiber and A. Becker (eds.), Raqqa iii, Mainz am Rhein 2004, 137–142. van Dalen, B., Battānī: Abū ʿAbd Allāh Muḥammad ibn Jābir ibn Sinān al-Battānī alḤarrānī al-Ṣābiʾ, in T. Hockey et al. (eds.), The biographical encyclopedia of astronomers, New York 2007, 101–103. Vardjavand, P., Kavosh-e Rasadkhaneh-e Maragha, Tehran 1987. Wilkinson T.J., and L. Rayne, Hydraulic systems in the Middle East, in H. Selin (ed.) Encyclopaedia of the history of science, technology, and medicine in non-western cultures, Dordrecht 2014, https://doi.org/10.1007/978‑94‑007‑3934‑5_10228‑.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

part 4 Frontiers and the others

∵

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 24

The Interface between Byzantium and the Ilkhanids in Fourteenth-Century Book Painting Robert Hillenbrand

1

Introduction

The Ilkhanid realm had a physical border with Byzantium to the west and one with China to the east; so it is not surprising that these two traditions leave the most visible alien footprint in Ikhanid manuscript painting. But it also had an invisible though no less real border with western Europe as well, a border built on the lengthy exchange of official letters between the Mongol rulers and the powers of western Europe,1 on the envoys2 and missionaries despatched by the pope to those rulers,3 on the European craftsmen who worked, willingly or unwillingly, for the Mongol elite in distant Karakorum, on the trading links with the Italian city-states,4 on the regular flow of western ambassadors to Tabrīz5 and, more generally, on the Mongol tolerance of other faiths in their realm,6 which gave Persian artists the chance for insights into those faiths and how they were practiced, especially in the visual arts. This is a chapter of Ilkhanid art history that remains to be written. But the impact of western European art on Persian painting, though of secondary importance in this paper, will surface at intervals in the discussion that follows.7 Thus, the image of the Bier of Alexander in the great Mongol Shāhnāma8 is suffused with echoes of contemporary and earlier French and Italian art, from Jean Pucelle, as seen in the gestures of the mourners, to the Pisano brothers in the treatment of drapery and the reminiscence of a Pietà. Note, too, the French plait of Alexander’s distraught mother, Olympias, and the red-haired and red1 Mostaert and Cleaves, Trois documents mongols 434, 451, and 471; Boyle, Il-Khans of Persia 554–555; Boyle, Appendix 213–214; Richard, Les Mongols et l’Occident; Richard, Ambassade. 2 De Rachewiltz, Papal envoys, 84–124, 160–174, and 187–204. 3 Dawson, Mission. 4 Petech, Marchands; Prazniak, Siena. 5 Jahn, Täbris. 6 Spuler, Mongolen 165–209. 7 See Hillenbrand, Great Mongol Shahnama, ch. 9. 8 Grabar and Blair, Epic images 134.

© Robert Hillenbrand, 2022 | doi:10.1163/9789004525245_025

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

476

hillenbrand

bearded men. These are not Persians. But this is not the time to embark either on that topic or on the image of Alexander in Byzantine and Ilkhanid art,9 nor on the connections in the visual arts between eastern Christendom and the west,10 and indeed with China.11 Intercultural exchanges of this kind, especially in the transition from late antique to early Islamic times, have been a core feature of Hugh Kennedy’s scholarship throughout his career. The present paper deals with similar issues in a later period and in the context of quite another border. It explores the cognate topic of the role of Byzantine art as a source of ideas, motifs, techniques, and iconography for Persian painters in late Ilkhanid times.12 The very idea of a border carries with it some idea of difference, a binary division, a red line. It may be the place of a stand-off between two mutually hostile states, or a neutral no-man’s land, or a zone of friendly interaction. And one may also propose subsets of these general categories to accommodate the nuances of real-life situations. Borders may be linguistic, confessional, or racial, and in all these cases they may be blurred at the edges. But cultural borders are the most blurred of all, and this is especially true in the world of the visual arts, which cannot readily be policed even today by passports and frontier guards. After all, the agent of change may be a single person, like Guillaume de Boucher, a Parisian goldsmith who was minding his own business in Belgrade in 1242, trying to make a living in one of the remoter regions of Christian Europe, when a malign fate in the shape of a detachment of Mongol warriors spirited him away to outer Mongolia, where he was to spend the rest of his life making automata13 and other marvels for the Great Khans Güyük and Möngke.14 And the agent of change could just as well be not a man but a manuscript. Moreover, the movement of ideas, themes, motifs, techniques, styles, and objects from one culture to another is rarely a matter of simple transfer. Changes in context, function,

9

10 11 12

13 14

For Alexander in Byzantine art, see Galavaris, Alexander the great conqueror; and, in the great Mongol Shāhnāma, see Grabar and Blair, Epic images 112–135. Not surprisingly, similar scenes sometimes turn up in both traditions at about the same time, such as Alexander overseeing the building of the wall against Gog and Magog (Hellenic Institute, codex 5, f. 179v; and Great Mongol Shāhnāma, no. 37). Demus, Byzantine art and the west; Jolivet-Lévy, La peinture. Kouymjian, Chinese motifs; Kouymjian, Intrusion 121–129. For the multiple connections between Byzantine and Islamic art in earlier times, see Grabar, Islamic art and Byzantium; and Evans and Ratliff, Byzantium and Islam 4–11, 98–101, 136–143, and 244–258. Olschki, Guillaume Boucher 45–106. Olschki, Guillaume Boucher 1, 7–8, and 18; cf., for the presence of people from across the globe at the Mongol court in Karakorum, Allsen, Commodity 2.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

477

medium, meaning, or even in apparently “mere” supplementary details can be fraught with unexpected consequences, so much so that the original source is apt to become indistinct. This paper will try to test and to flesh out these generalizations in a single specific context, namely, the impact of Byzantine modes and iconography on the book painting of Ilkhanid Iran between 1300 and 1350. The discussion will move from specific themes to the close analysis of six paintings, which can serve as a paradigm of what was going on in this particular case of bordercrossing—which, by the way, was overwhelmingly from west to east.15

2

Historical Context

How does the historical context impact on the topics to be discussed? Two basic facts are particularly relevant here. The first concerns religious minorities. Before 1295, the date when the Mongol elite converted to Islam en masse,16 the Mongols—for a variety of reasons—having conquered Iran from 1220 onward, found it politic and convenient as non-Muslims to entrust much of the administration of their realm to other non-Muslims, principally Christians and Jews.17 They pursued this policy much more energetically than any of their predecessors in Iran. So these two minorities, and also the Buddhists, though to a much lesser extent, enjoyed an Indian summer of power and prestige, though it was destined to end badly for them.18 Thus even in the Bīrūnī image of the Baptism, to be discussed later, one may detect anti-Christian elements; and it is hard not to see a symbolic reference to this harsher attitude in the way that Buddhism is trashed under the cloak of an image depicting Abraham smashing the idols of the Sabaean astrolaters in the Bīrūnī manuscript of 1307. Note, too, how the carefully chosen images in that manuscript represent Judaism through the two greatest catastrophes to afflict the children of Israel: the loss of the Ark and the destruction of Solomon’s temple. The fact that these two images are on adjoining pages increases their sorrowful emotional impact.19 15 16 17

18 19

For the reverse process, see Hillenbrand, Lure of the exotic. Melville, Pādshāh-i Islām. For Saʿd al-Dawla, see Fischel, Jews 93, 98, and 103–105; Spuler, Mongolen 205; and Budge (trans.), Chronography 479. For Rashīd al-Dīn, see Morgan, Rashīd al-Dīn Tabib; and Krawulsky, Mongol Īlkhāns. For other Jewish statesmen, see Spuler, Mongolen 108, 207– 208, and 240. For the Jews, see Hillenbrand, Holy ark 50–51, 54–56, and 59; for the Buddhists, see Kadoi, Buddhism; Scarcia, Vihar; Curatola, Viar dragon; and Curatola, Grotte. Hillenbrand, Non-Islamic faiths 309–310.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

478

hillenbrand

As it happens, in the later thirteenth century the status of the Christians was further enhanced by the marriages of high-born Byzantine princesses to members of the Mongol elite.20 Maria Palaiologina, an illegitimate daughter of Emperor Michael viii Palaiologos, was dispatched from Constantinople to marry Hülegü, but he died before they even met, and so she was married to his son, the Ilkhan Abāqā. She became known as Despina Khātun, and for 15 years, until her husband’s death, she was an influential advocate of Christianity among the Mongol elite, many of whom were Nestorians.21 A convent in Constantinople—St. Mary of the Mongols—was named after her, and she ended her days there as a nun under the name Melania.22 She is depicted in the Deisis mosaic in the Kariye/Kahrie Cami.23 Her sister, Euphrosyne Palaiologina, was betrothed to Nogai Khān, the ruler of the Golden Horde. Later, two daughters of Andronicus ii were married to Toqto’a and to his successor Uzbek (1312–1341).24 Before then, Öljeitü, the son of Arghūn’s third wife, the Christian Uruk Khātūn, had been baptized under the name of Nicholas after Pope Nicholas iv;25 and Öljeitü himself married a sister of Andronicus ii.26 All these marriages continued a much older tradition of Muslim rulers taking Christian wives, recorded in both Umayyad and Abbasid times and also under the Seljuqs of Rum.27 The fact that these Byzantine consorts practiced their faith at court would have afforded opportunities for Muslim artists to have encountered Christian images. So for over half a century the Christian faith could be freely practiced in Mongol Iran. The second important factor in creating a context for this paper is religious figural art. The Mongols had become accustomed to the pervasive presence of Buddhist religious art, both in China and in Iran; Abāqā and Arghūn were both devout Buddhists,28 and Ghāzān and Öljeitü were both Buddhists in their youth.29 The Ilkhans also knew (though rather less well) that Christianity, too, made much use of religious images, both via the missionary Catholic friars30 20 21 22 23 24 25 26 27 28 29 30

Ryan, Christian wives, especially 416 on the favor shown to Christians by members of the Mongol elite. Spuler, Mongolen 60–61, with further commentary on the close diplomatic ties between Byzantium and Ilkhanid Iran under Abāqā. Ryder, Despoina; Müller-Wiener, Bildlexicon 204–205; Teteriatnikov, Place. Underwood, Kariye Djami i, 46–47; ii, pls. 36–37. Ryan, Christian wives 420. Roux, Histoire 408. Spuler, Mongolen 92. Vryonis, Decline 225. Spuler, Mongolen 149–153. Spuler, Mongolen 154 and 158 respectively. For the presence and impact of European works of art in the Mongol capital of Karakorum,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

479

and the Nestorian Christians of the eastern church.31 Despite the mass conversion to Islam engineered by Ghāzān, the Mongol elite had no preconceived prejudices against religious art in the Muslim sphere. It is therefore plausible that, with examples of Buddhist and Christian religious art conveniently at hand, the Mongols expected Islam to furnish something along the same lines. When they discovered that they were mistaken, they took swift steps to rectify the situation. Hence the sudden flowering—surely an artificial flowering—of Muslim religious images in the immediately following decades. It is thus no accident that in the first generation after the Mongols converted to Islam there survive—it is worth stressing that word because it implies that there were many more—no less than four cycles of images of the Prophet Muḥammad, though there had apparently been none produced within the Islamic world in the previous 700 years. So, out of the blue, figural religious images suddenly entered the eastern Muslim world, and there were no Muslim models, no established iconography of Muḥammad, to draw upon. The painters therefore had to devise a series of appropriate scenes at speed and from a standing start. No wonder they borrowed or adapted images and motifs from whatever source—usually a foreign one—came to hand. An outstanding example is the fragmentary Miʿrājnāma in Istanbul32 with its echoes of Byzantine33 and Sienese painting.34 And their lack of familiarity with the compositions or motifs that they borrowed practically guaranteed that they would misuse them. So it should come as no surprise that the Muslim versions of Byzantine themes—whether these were metropolitan, provincial, or from cognate traditions such as the Armenian or indeed Syriac one—were often arresting in their strangeness and unpredictability, and in their lack of a close fit. Before looking in detail at what forms these copies, borrowings, and adaptations took, it is appropriate to investigate how examples of Byzantine art might have reached Iran. Several means of transmission come to mind. One

31 32 33 34

whether these were booty, presents from Louis ix of France, or the precious illustrated manuscripts that Friar William of Rubruck had brought from the west, see Olschki, Guillaume Boucher 22–23 and 25–28. Note, too, the echo of a procession of Christian friars in the cortège of Rustam and Zavara in the Great Mongol Shāhnāma; Grabar and Blair, Epic images 104–105. Baumer, Church of the east 169–234; Baskhanov, Baskhanova, Petrov and Serikoff, Arts 104– 109; Olschki, Guillaume Boucher 18 and 43. Ettinghausen, Persian ascension miniatures; Blair, Ascending to heaven; Gruber, Ilkhanid book of ascension 24–31. For two twelfth-century images of enthroned Madonnas not hitherto noted as possible parallels, see Berenson, Due Dipinti, main plates. Smart, Dawn pls. 45, 47, 53, 104, 108, and 110.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

480

hillenbrand

is via Armenian art; after all, the Christians in the immediate ambit of Tabrīz were mainly Armenians,35 though Assyrian Christians also lived in the area.36 Both of these groups drew inspiration for their religious art from Byzantium, as did another neighboring region, Georgia.37 The retinue of Abāqā’s wife, the Byzantine Princess Maria Palaiologina, were permitted freedom of worship, and court artists would have seen Byzantine vestments, icons, manuscripts, and ritual objects at first hand, and might have seen wall decoration in Byzantine churches in the capital. Religious processions in the streets would have involved the display of Christian images or ritual objects.38 Gifts from Byzantine ambassadors stored in court treasuries could have been accessible to court artists. Finally, one must reckon with the possibility that the artists themselves might have seen east Christian art if they came from or had lived in areas where Christians were plentiful. The transmission of Byzantine ideas and motifs can be traced in two seminal illustrated texts: The chronology of ancient nations by al-Bīrūnī (1307) and the World history by Rashīd al-Dīn (1314). The range of borrowings is wide, so this paper will be limited to the evidence they provide on three core themes: the treatment of drapery, the use of haloes, and the reuse of Byzantine iconography.

3

Drapery

First, then, conventions for drapery. The complex drapery found in scenes from the Bīrūnī manuscript, dubbed Schnörkelfalten by German art historians, seems to derive, though at several removes, from a tradition found, for example, at Monreale Cathedral in Sicily, whose mosaics in a hybridized Byzantine style include a pair of angels visiting Lot (see fig. 24.1).39 This manner of depicting drapery as a chaotic, hectic mass of writhing folds, creating a tension wholly at variance with the stillness of the body they

35 36 37 38 39

For examples of their art at this time, see Mathews and Walker, Armenian gospels; and Soucek, Armenian and Islamic manuscript painting. Baumer, Church of the east 99–104; for a much fuller and up-to-date account, see Harrak, Monastère. Mepisashvili and Tsintsadze, Arts 167–172, 183–187, 199–200, 204, 207–209, 211–212, 229– 233, and 282–290. Such as the crozier and thurible in the funerary procession of Rustam and Zavara in the Great Mongol Shāhnāma; Grabar and Blair, Epic images 105. Demus, Mosaics of Norman Sicily pl. 104a.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

figure 24.1

481

Palermo, Monreale Cathedral: Mosaic of angels visiting Lot, late twelfth century

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

482

hillenbrand

figure 24.2 Al-Bīrūnī, Chronology of ancient nations, 1307: Muḥammad appoints ʿAlī as his successor

cover,40 had a wide currency in western Europe and Byzantium from the later twelfth century onward.41 So the Bīrūnī manuscript represents a late echo of this passing fashion. It is essentially a replacement for pattern, in that the swirling folds offer a focus of visual interest, just as patterned or flowered robes with an even surface do; but the Islamic delight in sheer pattern trumped that manner.42 So its presence in the Bīrūnī manuscript runs against the grain of Islamic practice, and it is not strange that their example was not followed. Folio 162a of the Chronology meticulously renders the gatherings of Muḥammad’s robe at the wrist and elbow, as well as the compressed ruffling of the sleeves and, this time in a very unByzantine way, a patterning of extremely delicate spiked escutcheons in the body of the robe (see fig. 24.2).43 So in some sense this is a fusion of the Byzantine and the Islamic attitude to drapery. In the Bīrūnī Annunciation scene, Gabriel illustrates another drapery

40 41 42 43

Demus, Mosaics of Norman Sicily 421–422. Demus, Mosaics of Norman Sicily 430; cf. more generally Demus, Byzantine art and the west 179–204. James, Masterpiece 198. Hillenbrand, Images 141.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

483

fourteenth-century book painting

figure 24.3

Hosios Lukas, mosaic of the Anastasis, first half of the eleventh century

quotation, namely the trailing end—of a sash, for example—that defies gravity by projecting at right angles from the body, as if it had a life of its own. Here again there are Byzantine parallels, as in the mosaics of the Anastasis at Hosios Lukas (see fig. 24.3) and Chios,44 though cognate forms are also known in religious images executed in various media in sites that date from the tenth to twelfth-centuries along the Silk Road in western China.45

4

Haloes

The second major theme for discussion here is the halo. The earliest surviving examples of Islamic book painting in both the Arab and Persian worlds frequently use haloes indiscriminately: in the Vienna Galen manuscript even birds have haloes.46 In the Rashīd al-Dīn manuscripts, which so often draw on Byzantine models, they occur only once, in a very specific borrowing from the

44 45 46

Demus, Byzantine mosaic decoration pl. 13a–b. Whitfield and Farrer, Caves 35, 50–51, 66–69, 71–72, 82, 84–85, 96–97, 102–104, and 132. Ettinghausen, Arab painting 91.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

484

hillenbrand

Byzantine Octateuchs, namely the crossing of the Red Sea.47 And then, while the scene itself plainly draws on Byzantine prototypes,48 the very particular borrowing of the halo form is not necessarily Byzantine at all but is more likely to come from western Europe.49 The Bīrūnī manuscript often depicts haloes, but in two distinctively different ways. The first follows the standard model of a golden roundel that encircles the entire head, including any headgear that is worn. The second is more unusual and is confined to the work of the artist whose work begins on f.129b. Here the halo is sited further down so that much less of it is visible, and it may also bear decorative red curlicues around the rim, as in the case of ʿAlī’s halo on f. 162a (see fig. 24.2).50 So far, so good. But Ahrimān the spirit of evil, the first man and the first woman both caught out in sin, Musaylima the false prophet, Ẓaḥḥāk who fed for centuries on the brains of babies, Nebuchadnezzar the destroyer of Solomon’s Temple, and the Christian bishops opposing Muḥammad—all of them either wicked or at least open to judgment—are shown with haloes,51 as are onlookers galore,52 while time and again the figure of central interest, such as the mother of Caesar, al-Hallāj, Farīdūn, Peroz, Eli and Mary, is pointedly denied one.53 Sometimes this practice leads to absurd conclusions, as when John the Baptist has a halo and Jesus does not (see fig. 24.4).54 These inconsistencies suggest a lack of awareness of the role of the halo, and it is therefore not surprising that after the fall of the Ilkhanids the halo largely drops out of use (the Great Mongol Shahnama and the Miʿrajnama are the major exceptions), perhaps a signal that Persian painters turned their back on Byzantine and indeed Christian art as a source.

47 48

49 50

51 52 53 54

Rice, Illustrations 60–61; Allen, Byzantine sources 125; Hillenbrand, Symbolism 37 and 45. For the Octateuch parallel (Istanbul, tks Library, Codex 8, f.197b), see Rice, Art of Byzantium 43 (in color); for further examples, see Lowden, Octateuchs figs. 133–134 and 136– 141. For an extensive discussion of the issue, see Cabrol and Leclercq, Dictionnaire xii, s.v. Nimbe, cols. 1272–1312. Hillenbrand, Images 141. It is noteworthy, given the fervent Shiʿism on display in this image, that all the other Rāshidūn also have haloes, although these are plain, so that ʿAlī stands out. Soucek, Illustrated manuscript figs. 4, 7, 13, 20, and 24 respectively. For example, Soucek, Illustrated manuscript figs. 1, 3, 6, and 10. Soucek, Illustrated manuscript figs. 3, 10, 13, and 17 respectively. Soucek, Illustrated manuscript fig. 21.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

485

fourteenth-century book painting

figure 24.4

5

Al-Bīrūnī, Chronology of ancient nations, 1307: The Baptism of Jesus edinburgh university library, ms. arab 161, f.140b

Iconography

5.1 Nativity Scenes In the case of the third theme for discussion in this paper, namely iconography, the obvious candidate for examination is the World history of Rashīd al-Dīn. The most obvious borrowings from Byzantine sources in this manuscript have been well analyzed by Terry Allen,55 but he focused his attention on certain Old Testament prophets, and that meant omitting several striking examples of similar adaptations elsewhere in the manuscript. The image depicting the nativity of the Prophet Muḥammad clearly owes much to Byzantine scenes of the Nativity of Christ. The centrality of the baby and the mother, the presence of several angels, and of an older man seated at a distance all testify to the close connection. But there is an element of multiple repetition, of extras brought in purely as space fillers, that distinguishes the Muslim version (see fig. 24.5). The groupings in the typical Byzantine nativity scene—the shepherds, the kings, the animals, the angels—are all rooted in the sacred Christian text,56 55 56

Allen, Byzantine sources. E.g., the Nativity mosaic in Daphni; Demus, Byzantine mosaic decoration pl. 10B.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

486

hillenbrand

figure 24.5

Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: The birth of Muḥammad edinburgh university library, ms. arab 20, f.42a

whether that is from the synoptic gospels or from the Apocrypha. They all play their part in deepening the significance of the scene and in increasing its visual interest (see fig. 24.6). But three of these four groups have, understandably enough, disappeared from the nativity of Muḥammad (see fig. 24.5). What is telling is that they have been replaced by a total of eight women, all of them, except perhaps the one who tends to Muḥammad’s mother, redundant. This cannot fail to weaken the overall impact of the scene and make it a somewhat pale reflection of the standard Byzantine depiction of the Nativity. Most of them keep at a respectful distance from the action and so do not participate in it. The place of Joseph, pointedly relegated in the Christian nativity to a mere onlooker,57 is taken by the Prophet’s grandfather, ʿAbd al-Muṭṭalib. And the two angels no longer flutter protectively above the scene but have been catapulted right into its epicenter, for one of them holds the infant and the other swings an incense burner with its restorative fragrance over the mother. Their sheer bulk makes them the center of attention. Their presence at the heart of the scene means that the birth of Muḥammad is presented and indeed celebrated as a sacred event. It is particularly noticeable that Muḥammad’s mother is demoted in this composition to a secondary role, swaddled into insignificance; she does not even hold her baby. Since the traditional account of the Prophet’s birth is shorn of this element

57

As at Daphni (see previous note).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

487

of the miraculous, it is plain that almost by sleight of hand the sanctity of the scene of Christ’s birth has been transferred to the birth of Muḥammad. And the setting, again understandably, is no longer a cave in a hillside (as in fig. 24.6) but a rather grand interior, which a central segmental arch of the standard twelfth– fourteenth-century Persian type divides into three sections. The spandrels and plinths of the arch are painted a festive scarlet.58 The circular composition of the Byzantine original, with the constituent elements disposed like the spokes of a wheel with Mary at the center59 has been rolled out into an oblong layout. Gone is the sense of depth: everything happens in the frontal plane. Three further features of Byzantine origin deserve brief notice: the chair on which ʿAbd al-Muṭṭalib sits;60 the spectacular flourish of the knotted curtain;61 and the purple porphyry marble of the columns, a detail commonly found in Byzantine evangelist portraits.62 The cave in which Jesus is born underlines the theme of poverty; the Prophet, however, comes into the world at a markedly higher social level. And the gold background which at once distances and sacralizes a typical Byzantine Nativity scene (fig. 24.6), is gone, replaced by a plain white background that keeps the scene earthbound despite the angels. So while the shadow of Byzantine iconography hangs over this scene, its realization has gone far away from the putative original.63 5.2 Scenes Related to the Baptism Much the same can be said of a scene that at first glance has no Byzantine equivalent, namely, the episode of Muḥammad’s trip as a mere boy to Syria with a caravan of merchants (see fig. 24.7).64

58 59 60 61

62 63

64

This may be intended to convey a sense of high rank; in this same manuscript the color of the royal thrones is consistently scarlet. As at the Kariye/Kahrie Cami; Underwood, Kariye Djami ii, 166. Pächt, Byzantine illumination pl. 6; Friend, Portraits figs. 96, 98, 103–104. As at the Kariye/Kahrie Cami, Underwood, Kariye Djami ii, 136, 138; cf. Sermons on the Virgin by Jacobus of Kokkinobaphos (Paris, Bibliothèque Nationale, Ms Grec 1208), Grabar, Byzantine painting 183. Pächt, Byzantine illumination pls. 8 and 11. It is worth adding that the text of the Jāmiʿ al-tawārīkh accompanying this picture offers no clues to the artist; it merely states laconically that his mother gave birth to Muḥammad “at a house that was adjacent to the Kaʿba” (Edinburgh University Library, Ms Arab 20, f. 42b). Later, the text notes “it was reported that she had said when she gave birth to him, ‘I saw light coming out of the baby’s face and it shone over the whole world, and I saw the palaces of Syria and Iraq. Light came out of him and it shot up to the heavens and reached the stars.’” The classic biography of Muḥammad by Ibn Isḥāq (Guillaume (trans.), Life 69– 70) says even less about his birth. See Ms. 20, f.43b, which tells how, when the caravan approached Busra, “there was a nearby

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

488

hillenbrand

figure 24.6

Daphni, mosaic of the Nativity of Christ, ca. 1100

As they pass by a Christian hermitage, the anchorite, Baḥīrā, stops them and announces that they have a prophet among them. One by one they pass by him, but none of them is identified as such. He finally asks if there are no more of them; they mention the boy looking after the camels, and he is brought forth. Baḥīrā’s dramatically outflung arm identifies him instantly as the prophet to be. The hermit himself wears the cowl of a prophet, while a disciple beside him acts as the necessary witness to a theophany, a manifestation of God to man.65

65

hermitage in which a monk, named Baḥīrā lived. Baḥīrā was a virtuous man who studied many books where he read about the Prophet’s description. While he was on the rooftop of the hermitage, he threw a glance on the caravan, and he saw a single cloud that was as big as a shield, [that] had come down and overshadowed the Prophet […] singling him out from the rest of the company.” Cf. Ibn Isḥāq, Guillaume (trans.), Life 79–81. Nordström, Ravennastudien 63 and 98 (the oak at Mamre was venerated as a living witness to Abraham’s encounter with the three angels); Grabar, Fresque visigothique 583.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

489

fourteenth-century book painting

figure 24.7

Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: Baḥīrā recognizes Muḥammad as a prophet edinburgh university library, ms. arab 20, f.43b

The picture reads in the most natural manner from right to left. But it springs a surprise, a real coup de théâtre, in that its centerpiece is neither the hermit nor the boy but rather a group of five kneeling camels and their baggage, captured in a pose that irresistibly suggests worship.66 Their gaze is directed to the far end of the picture where the diminutive Muḥammad stands, his eyes meekly downcast, unevenly flanked by ten tall men. This arrangement cannot fail to recall a Byzantine image of the Baptism (see fig. 24.8).67 Here, too, the central figure, in this case Jesus, is flanked by other figures, and the voice from heaven announcing “This is my beloved Son, in whom I am well pleased”68 takes physical form as an outstretched hand, a shaft of light, and the Holy Spirit descending in the form of a dove. All three of these crucial elements find their equivalent in the Ilkhanid painting. Above Muḥammad a semicircle of blue sky has opened, and an angel holding a scroll and a bottle is bending down to anoint him. Meanwhile, those flanking Jesus, namely John 66

67 68

Here again, as in the case of the setting for the birth of Muḥammad, there is no basis in the text for these camels. A pun may be intended here, since a secondary meaning of baḥīra is a she-camel. Pächt, Byzantine illumination pl. 1 6; Demus, Byzantine mosaic decoration pls. 12A–B. Matthew 3:17.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

490

hillenbrand

figure 24.8

Daphni, mosaic of the Baptism of Christ, ca. 1100

the Baptist and angels, are replaced by the Arabian merchants. It is a bold and unmistakable calque of a Byzantine baptismal scene, a neat transference, reuse, and adaptation of a refined, indeed streamlined, Byzantine festival icon to serve Islamic religious purposes. The defeated pagan river god in the Byzantine scenes of the Baptism69 obviously has no place in the Muslim reworking of the model. More than that, the image contrives to suggest that even animal creation acknowledges the prophetic status of the boy, and it is a Christian holy man who authenticates that status. All in all, this composition achieves a great deal in the limited space available. It adds up to something much more than a clever adaptation of Christian iconography for Muslim purposes; it signals a growing confidence in the treatment of Muslim religious themes and the ability of Islamic iconographers to strike out on their own. 69

As in the Baptistery of the Orthodox in Ravenna: Grabar, Byzantium painting 123 (color plate).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

491

As it happens, this is not the only reference in Ilkhanid painting to the Baptism. But the image in the Bīrūnī manuscript, which purports to depict that scene, in plain fact does nothing of the kind (see fig. 24.4). Indeed, it could be described as an impudent sleight of hand. True, a Byzantine original lurks in the background, perhaps a more simplified version that is restricted to Jesus, John the Baptist, and the river god.70 But the Bīrūnī painter goes still further and reduces the dramatis personae to two. More than that, he systematically dilutes the meaning of this core Gospel episode. This is a desanctified Baptism, with no reference to God and no witnesses to the theophany. In the standard Byzantine formula, the angels around Christ underline the sanctity of the occasion; the defeated and miniaturized river god makes the Baptism a victory over paganism, and divine approval in the threefold symbolism of hand, light, and dove is concentrated on the central chord of the whole composition.71 But the Persian artist here has no respect for the event, no feel for it, no awareness of its significance. The process begins with John the Baptist, portrayed precisely as Jesus, in Matthew’s Gospel, says he is not— “a man clothed in soft raiment.”72 Gone is the wild look, the dishevelled hair, the pelt of wild animals, the leather belt and sandals. This John the Baptist is a carefully barbered fashion plate, from his turban to his embroidered boots, who has been dressed by Armani, and his diet has obviously been much more substantial than locusts and wild honey. Jesus himself is in profile, making eye contact with John the Baptist rather than the viewer.73 And above all, he is not being baptized but being helped with his clothing while his slippers float away downstream. The timing is awry—this is the moment after the Baptism, and even the dove of the Holy Spirit, now transmogrified into some East Asian bird, is arriving late. This is no festival icon; instead, it is a banal anticlimax. All the meaning has been drained out of the scene, and this is driven home, as already noted in another context, by the fact that John has a halo and Jesus does not. It is hard to see all this as an accident; after all, for Muslims, the Baptism of Jesus blasphemes the very idea of tawḥīd, that God is One, not two or three.74

70 71 72 73 74

As in the Arian Baptistery in Ravenna; Rice, Byzantine art 155. As at Daphni; Demus, Byzantine mosaic decoration pl. 12B. Matthew 7:18. This is a very serious solecism by Byzantine standards; see Demus, Byzantine mosaic decoration 8 and 65. For further commentary on this scene, see Soucek, Illustrations 145 and 147; and Hillenbrand, Non-Islamic faiths 311–312.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

492

hillenbrand

figure 24.9

Rashīd al-Dīn, Jāmiʿ al-tawārīkh, 1314: The Annunciation edinburgh university library, ms. arab 20, f.22a

5.3 Scenes Related to the Annunciation The next scene is much more straightforward, for it illustrates something much closer to a direct copy. And yet it is only half a copy, for those Byzantine artists who interpreted the Annunciation on the basis not of the Gospel accounts, especially that of Luke, but instead followed the second-century apocryphal Protevangelium of James, the so-called Infancy Gospel, which recounts: “She took her pitcher and went forth to draw water. And behold a voice which said unto her: Hail, thou that is full of grace, the Lord is with thee; blessed art thou among women. And she looked to the right and to the left that she might see whence this voice came; and being seized with a trembling she returned into her own house.”75 Only on her return home did the angel greet her; but in several Byzantine versions of the scene those two events are conflated, and she hears the angel as she is filling her pitcher at the well.76 The painter of this scene in the World history of Rashīd al-Dīn goes decisively further than this (see fig. 24.9). 75 76

Lightfoot, James, Swete et al., Excluded books 36. E.g., at the Kariye/Kahrie Cami; see Underwood, Kariye Djami i, 82 (quoting the Protevangelium); and ii, 147; and in St. Mark’s, Venice—see Demus, Mosaic decoration pl. 12B.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

493

In accordance with the account in the Quran,77 the spirit (rūḥ) sent by God appears to Mary not as an angel but in the form of a man, literally “a wellproportioned man”—he has no wings. But once again the painter conflates things. Instead of following the Quranic text, which testifies that Mary withdrew from her family to an eastern place and was in seclusion,78 he follows the tradition of the Infancy Gospel and shows her walking to the well, pitcher in hand. The well is now a pool of water in a rocky crevice, and, in a poignant touch that reveals her modesty, she casts her eyes down and touches her cheek with her open palm. Her rigidly extended little finger suggests her distress. The man gestures vigorously toward her as he offers reassurance. Both are similarly dressed in glowing identical colors, with a deep blue outer garment over a flowing red robe. Between them they take up most of the frontal plane and virtually the entire height of the picture space. The “eastern place” is rendered as an awesome backdrop of beetling crags and steep ravines in shades of sepia, their curvilinear contours carefully modeled. The debt to a Chinese mountain landscape, which the painter perhaps encountered in simplified form in a textile,79 is unmistakable, and the same can be said of the background to the encounter between Gabriel and Muḥammad later in the manuscript, another example of the propensity of these artists to reuse certain formulae and thereby save time. In both cases, sublime, towering mountain peaks offer an objective correlative to the portentous events unfolding in the foreground. Seldom has the Annunciation been stripped down to its essentials with such powerful, arresting simplicity. The depiction of the Annunciation in the Bīrūnī manuscript offers an intriguing counterpart to this dramatic image (see fig. 24.10).80 Although closely contemporary with the Rashīd al-Dīn image, it is strikingly different in its emphasis. It follows not the Quranic text but that of the Infancy Gospels, and specifically the later part of the very briefly described scene of the Annunciation. The relevant passage runs as follows: “She returned into her own house, setting down her pitcher, and took the purple and sat down on her seat and began to weave it. And behold, an angel of the Lord stood before her, saying: Fear not, Mary, for thou hast found favour before the Lord of all things,

77 78 79 80

Quran 19:17. See Stowasser, Mary, for a thorough examination of how she is presented in the Quran. For this episode, in which there is no mention of her drawing water from a well, see Quran 19:17–21. E.g., Kadoi, Cloud patterns 24, fig. 7. Soucek, Illustrated manuscript 147–148; Hillenbrand, Non-Islamic faiths 310–311; Kadoi, Islamic Chinoiserie 150, 152–153.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

494

hillenbrand

figure 24.10 Al-Bīrūnī, Chronology of ancient nations, 1307: The Annunciation edinburgh university library, ms. arab 161, f.141b

and thou shalt conceive of His word.”81 Accordingly, she is depicted wearing a wimple, seated cross-legged on a bolster, and holding a spindle. This refers to the sustained emphasis in the Infancy Gospel on the choice of Mary as a devout descendant of David to spin the thread for the new Temple veil or curtain. It was colored purple and scarlet,82 the royal and most expensive colors, foreshadowing the kingship of Jesus and the shedding of his blood; this was the veil torn at the moment of his death, symbolizing the new covenant between God and man. Her own head covering takes up these same colors. She herself will now become, in the words of Proclus, archbishop of Constantinople (d. 446), “the loom of the flesh of God.”83 It is most unlikely that the second Bīrūnī painter was aware of these theological subtleties; and in any event he has imparted his own distinctive take on this ancient iconography by his treatment of Gabriel. This is Gabriel as he might have been imagined by a Buddhist painter, with his almond eyes, apple cheeks, long smoothly brushed black hair, flame halo, and

81 82 83

Lightfoot, James, Swete et al., Excluded books 36. Lightfoot, James, Swete et al., Excluded books 35; for a depiction of this event in a mosaic in the Kariye/Kahrie Cami, see Underwood, Kariye Djami i, 76–78; and ii, 130–134. Constas, Weaving the body of God 180–183.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

figure 24.11

495

Rashīd al-Dīn, Jāmiʿ al-taw̄ arīkh, 1314: Muḥammad receives the first revelation from Jabrāʾīl edinburgh university library, ms. arab 20, f.45b

floridly billowing sash, emerging from a landscape whose intersecting colored horizons and clumps of grass pay an untutored homage to Chinese landscape conventions. So Byzantine iconography has been sinicized, and the two furthest borders of Ilkhanid Iran merge in a single image. Another kind of Annunciation altogether, but one with Byzantine echoes, is the scene where Muḥammad, while meditating on Mount Ḥirāʾ, sees the approach of the Angel Gabriel, a being so gigantic that his head is higher than the heavens and his feet lower than the earth, a hyperbole the painter has contrived to render by breaking the upper and lower frame by the angel’s crown, wing, and foot (see fig. 24.11).84 The text around this image uses the words of the Prophet himself: “Jabrāʾīl appeared and said ‘O Muḥammad, you are the Messenger of God.’ My fear grew so intense that I was about to throw myself off the mountain, so he showed himself to me and said ‘I am Jabrāʾīl and you are the Messenger of God’ and he said ‘Recite: in the name of your Lord, Who created man from a clot of congealed blood’ […].”85 This tremendous figure advances on the dumbstruck Prophet 84 85

In the words of Ibn Isḥāq, Gabriel was “in the form of a man with feet astride the horizon.” Guillaume (trans.), Life 106. Jāmiʿ al-taw̄ arīkh, Edinburgh University Library, Ms Arab 20, f.45b.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

496

hillenbrand

figure 24.12

Daphni, mosaic of the Annunciation, ca. 1100

and extends a pointing finger with the command “Recite!” (iqrā—whence Quran, “that which is recited”). Thus the scene captures the very moment of the first revelation, and the painter has extracted the maximum drama from it. The tension is electric; one can feel the dynamism, the energy, of the empty space between them. Muḥammad, so traumatized that his tongue sticks to his palette, clasps his knees as he looks down and makes the sign of protection with his right hand. Once again the setting is an austere recreation of a Chinese mountainous landscape, this time dwarfed by the two figures. Despite this Chinese setting, the obvious source of inspiration here is once again Christian, probably a Byzantine Annunciation (see fig. 24.12). Byzantine artists had devised various formulae for this scene over the centuries, and they struck different notes of wonder and awe.86

6

Conclusion

How does the evidence presented in this paper bear on the general question of how borders operated in early fourteenth-century Iran? It seems clear that there were no religious or cultural barriers preventing free interplay in the

86

For a selection of these, see Cormack, Byzantine art pls. 85–86, 106, and 124.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

497

visual arts. On the contrary, painters and other artists readily borrowed ideas from Byzantium as well as western Europe and China. So the borders were porous. And Persian painters in particular then reworked those ideas in ways undreamt of by artists using them in their original context. That change of context was decisive. Artists working within their native tradition—be it Byzantine, western European, or Chinese—were understandably constrained by past precedent. Persian painters knew no such inhibitions in their use of such alien elements—everything was fair game to them. In that sense their lack of an established local tradition on which to draw worked to their advantage. This survey should have made at least one thing abundantly clear: there is no such thing as a consistent attitude to Byzantine art in the two principal manuscripts discussed in this paper. This is not in the least strange. It is only to be expected that individual artists would respond very differently to the foreign artistic traditions that they encountered, whether that encounter was by chance or design. Some Persian artists would be drawn to one particular aspect of a foreign tradition, while quite different aspects might appeal to their other colleagues. Thus, in broad terms the two masters who illustrated the Bīrūnī manuscript consistently used the halo, but in such idiosyncratic fashion that the word “abused” might be more accurate. Similarly, the artists illustrating the Rashīd al-Dīn manuscripts readily purloined Byzantine iconography when it suited them, while avoiding direct quotations from the close-focus details of those very compositions. Thus, the borrowings from Byzantine art never stuck out like a sore thumb. On the contrary, they were seamlessly integrated into a visual tapestry that was instantly recognizable as Islamic. This easy assimilation is a tribute both to the open-mindedness of the Persian painter and to his sense of his own identity.

Personal Note It is a pleasure to dedicate this paper to Hugh Kennedy, a thoroughly collegial and stimulating friend of some 50 years’ standing; he is one of an all-too small body of historians of the medieval Islamic world who is fully attuned to the value of material culture.

Bibliography Allen, T., Byzantine sources for the Jāmiʿ al-tawārīkh of Rashīd al-Dīn, in Ars Orientalis 15 (1985), 121–136.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

498

hillenbrand

Allsen, T.T., Commodity and exchange in the Mongol empire: A cultural history of Islamic textiles, Cambridge, 1997. Baskhanov, M., A. Baskhanova, P. Petrov, and N. Serikoff, Arts from the land of Timur, Paisley 2012. Baumer, C., The church of the east: An illustrated history of Assyrian Christianity, trans. M.G. Henry, London and New York 2006. Berenson, B., Due dipinti del decimosecondo secolo venuti da Constantinopoli, in Dedalo 2 (1921), (unpaginated). Blair, S., A compendium of chronicles: Rashid al-Din’s illustrated history of the world, London 1995. Blair, S.S., Ascending to heaven: Fourteenth-century illustrations of the Prophet’s miʿrāǧ, in K. Dévényi and A. Fodor (eds.), Proceedings of the colloquium on paradise and hell in Islam. Keszthely, 7–14 July 2003. Part One, special issue, The Arabist: Budapest studies in Arabic 28–29 (2008), 19–35. Boyle, J.A., The Il-Khans of Persia and the Christian west, in History today 23 (1973), 554–63. Boyle, J.A., Appendix, in I. de Rachewiltz, Papal envoys to the great khans, Stanford 1971, 213–214. Budge, E.A.W., (trans.), The chronography of Gregory Abûʾl-Faraj 1225–1286 the son of Aaron, the Hebrew physician commonly known as Bar Hebraeus being the first part of his political history of the world, i, English translation, London 1932. Cabrol, F., and H. Leclercq, Dictionnaire d’archéologie chrétienne et de Liturgie, xii, Paris 1935, cols. 1272–1312. Constas, N.P., Weaving the body of God: Proclus of Constantinople, the Theotokos, and the loom of the flesh, in Journal of early Christian studies 3 (1995), 169–194. Cormack, R., Byzantine art, Oxford 2000. Curatola, G., Le grotte de Behestan nella Khamse: Descrizione del sito, in Quaderni del Seminario di Iranistica, Uralo-Altaistica e Caucasoligia dell’Università degli Studi di Venezia 5 (1979), 7–11. Curatola, G., The viar dragon, Soltaniye iii, in Quaderni del Seminario di Iranistica, Uralo-Altaistica e Caucasoligia dell’Università degli Studi di Venezia 9 (1982), 71–88. Dawson, C., Mission to Asia, Toronto, Buffalo, and London 1980. Demus, O., Byzantine mosaic decoration: Aspects of monumental art in Byzantium, London and Henley 1948. Demus, O., The mosaics of Norman Sicily, London 1949. Demus, O., Byzantine art and the west, New York 1970. Demus, O., The mosaic decoration of San Marco Venice, ed. H.L. Kessler, Chicago and London 1988. De Rachewiltz, I., Papal envoys to the great khans, Stanford 1971. Ettinghausen, R., Persian ascension miniatures of the fourteenth century, in Convegno

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

499

di Scienze Morali, Storiche e Filologiche. Symposium on orient and occident during the middle ages May 27–June 1, 1956. Accademia Nazionale dei Lincei 12 (1957), 360–383. Ettinghausen, R., Arab painting, Geneva 1962. Evans, H.C., with B. Ratliff, Byzantium and Islam: Age of transition 7th–9th century, New York 2012. Fischel, W., Jews in the economic and political life of mediaeval Islam, London 1968. Friend, A.M., Jr., The portraits of the evangelists in Greek and Latin manuscripts, in Art studies 5 (1927), 115–147. Galavaris, G., Alexander the great conqueror and captive of death: His various images in Byzantine art, Revue d’Art Canadienne/Canadian art review 16 (1989), 12–18 and 74–77. Grabar, A., Une fresque visigothique et l’iconographie du silence, in Cahiers Archéologiques 1 (1945), 124–128. Grabar, A., Byzantine painting, Geneva 1954. Grabar, A., Byzantium from the death of Theodosius to the rise of Islam, trans. S. Gilbert and J. Emmons, London 1966. Grabar, O., Islamic art and Byzantium, in Dumbarton oaks papers 18 (1964), 69–88. Grabar, O., and S. Blair, Epic images and contemporary history, Chicago 1980. Gray, B., The world history of Rashid al-Din, London 1976. Gruber, C.J., The Ilkhanid book of ascension: A Persian Sunni devotional tale, London and New York 2010. Guillaume, A., The life of Muhammad: A translation of Ibn Ishaq’s Sirat Rasiul Allah, Oxford 1955. Harrak, A., Le monastère de Mar-Behnam à la période atabeg, xiiie s.: l’art au service de la foi, Paris 2018. Hillenbrand, R., The symbolism of the rayed nimbus in early Islamic art, in Cosmos: The yearbook of the traditional cosmological society 2 (1986), 1–52. Hillenbrand, R., Images of Muhammad in al-Biruni’s chronology of ancient nations, in R. Hillenbrand (ed.), Persian painting from the Mongols to the Qajars: Studies in honour of Basil W. Robinson, London 2000, 129–146. Hillenbrand, R., Non-Islamic faiths in the Edinburgh Biruni manuscript, in R. Hillenbrand, A.C.S. Peacock, and F. Abdullaeva (eds.), Ferdowsi, the Mongols and the history of Iran: Art, literature and culture from early Islam to Qajar Persia. Studies in honour of Charles Melville, London 2013, 306–315. Hillenbrand, R., The lure of the exotic: The Byzantine heritage in Islamic book painting, in Z. Chitwood and J. Pahlitzsch (eds.), Ambassadors artists, theologians: Byzantine relations with the Near East from the ninth to the thirteenth centuries, Mainz 2019, 193–213. Hillenbrand, R., The holy ark of Isfahan: An unknown masterpiece from Mongol Iran, London 2019

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

500

hillenbrand

Hillenbrand, R., The great Mongol Shahnama (in press). Jahn, K., Täbris, ein mittelalterliches Kulturzentrum zwischen Ost und West, in Anzeiger der Österreichischen Akademie der Wissenschaften, phil.-hist. Klasse 105 (1969), 201–211. Jahn, K., Wissenschaftliche Kontakte zwischen Iran und China in der Mongolenzeit, in Anzeiger der Österreichischen Akademie der Wissenschaften, phil.-hist. Klasse 106 (1969), 199–211. James, D., A masterpiece of Arab painting: The ‘Schefer’ Maqāmāt manuscript in context, London 2013. Jolivet-Lévy, C., La peinture à Constantinople au xiiie siècle: Contacts et échanges avec l’Occident, in J.-P. Caillet and F. Joubert (eds.), Orient et occident méditerranéens au xiiie siècle: Les programmes picturaux [Actes du colloque international organisé à l’École française d’Athènes les 2–4 avril 2009], Paris 2012, 21–40. Kadoi, Y., Cloud patterns: The exchange of ideas between China and Iran under the Mongols, in Oriental art 48 (2002), 25–36. Kadoi, Y., Buddhism in Iran under the Mongols: An art-historical analysis, in T. Gacek and J. Pstrusinska (eds.), Proceedings of the ninth conference of the European society for Central Asian studies, Newcastle-upon-Tyne 2009, 171–180. Kadoi, Y., Islamic Chinoiserie: The art of Mongol Iran, Edinburgh 2009. Kouymjian, D., Chinese motifs in thirteenth-century Armenian art: The Mongol connection, in L. Komaroff (ed.), The legacy of Genghis Khan, Leiden and Boston 2006, 303–324. Kouymjian, D., The intrusion of East Asian imagery in thirteenth-century Armenia: Political and cultural exchange along the silk road, in P. Forêt and A. Kaplony (eds.), The journey of maps and images on the silk road, Leiden 2008, 119–133. Krawulsky, D., The Mongol Īlkhāns and their vizier Rashīd al-Dīn, New York, Oxford, and Wien 2011. Lightfoot, J.B., M.R. James, H.B. Swete, et al. (trans.), Excluded books of the New Testament, London 1927. Lowden, J., The Octateuchs: A study in Byzantine manuscript illustration, University Park, PA 1992. Mathews, T.F., and A. Walker, The Armenian gospels of Gladzor: The life of Christ illuminated, Los Angeles, 2001. Mathews, T.F., and R.S. Wieck (eds.), Treasures in heaven: Armenian illustrated manuscripts, New York and Princeton 1994. Melville, C.P., Pādshāh-i Islām: The conversion of Sulṭān Maḥmūud Ghāzān Khān, in C.P. Melville (ed.), Pembroke papers i: Persian and Islamic studies in honour of P.W. Avery, Cambridge 1990, 169–177. Mepisashvili, R., and V. Tsintsadze, The arts of ancient Georgia, trans. A. Jaffa, London 1979.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fourteenth-century book painting

501

Morgan, D.O., Rashīd al-Dīn Tabīb, in ei2, viii, 443–444. Mostaert, A., and F.W. Cleaves, Trois documents mongols des archives secrètes vaticanes, in Harvard journal of Asiatic studies 15 (1952), 419–506. Müller-Wiener, W., Bildlexicon zur Topographie Istanbuls: Byzantion-KonstantinopolisIstanbul bis zum Beginn des 17. Jh., Tübingen 1977. Nordström, C.-O., Ravennastudien: Ideengeschichtliche und ikonographische Untersuchungen über die Mosaiken von Ravenna, Stockholm 1953. Olschki, L., Guillaume Boucher: A French artist at the court of the khans, Baltimore 1946. Pächt, O., Byzantine illumination, Oxford 1952. Petech, L., Les marchands italiens dans l’empire mongol, in ja 250 (1962), 549–574. Prazniak, R., Siena on the silk roads: Ambrogio Lorenzetti and the Mongol global century, 1250–1350, in Journal of world history 21 (2010), 177–217. Richard, J., Les Mongols et l’Occident: Deux siècles de contacts, in 1274. Année charnière; mutations et continuités (Lyon-Paris 1974). Colloques internationaux du c.n.r.s., no. 558, Paris 1977, 85–96. Richard, J., Une ambassade mongole à Paris en 1262, in Journal des Savants (1979), 295– 303. Rice, D.T., The art of Byzantium, London, 1959. Rice, D.T., Byzantine art, Harmondsworth 1968. Rice, D.T., The illustrations to the “world history” of Rashid al-Din, Edinburgh 1976. Roux, J.-P., Histoire de l’empire mongol, Paris 1993. Ryan, J.D., Christian wives of Mongol khans: Tartar queens and missionary expectations in Asia, in jras series 8, 9 (1998), 411–421. Ryder, E.C., The Despoina of the Mongols and her patronage at the Church of the Theotokos ton Mougoulion, in Journal of modern Hellenism 27 (2010), 71–102. Sachau, C.E., The chronology of ancient nations (Athār-ul-bākiya of Albīrūnī) or “Vestiges of the past,” London 1879. Scarcia, G., The “Vihar” of Qonqor-olong: Preliminary report, in East and west 25 (1975), 99–104. Smart, A., The dawn of Italian painting, Oxford 1978. Soucek, P.P., An illustrated manuscript of al-Bīrūnī’s chronology of ancient nations, in P. Chelkowski (ed.), The scholar and the saint, New York 1975, 103–165. Soucek, P.P., Armenian and Islamic manuscript painting: A visual dialogue, in T.F. Mathews (ed.), Treasures in heaven: Armenian art, religion and society, New York 1998, 115–131. Spuler, B., Die Mongolen in Iran: Politik, Verwaltung und Kultur der Ilchanzeit 1220–1350, Leiden 19854 (rev. and enlarged ed.). Stowasser, B.F., Mary, in J.D. McAuliffe (ed.), eq, iii, 288–296. Teteriatnikov, N., The place of the nun Melania (the lady of the Mongols) in the Deesis program of the inner narthex of Chora, Constantinople, in Cahiers archéologiques 43 (1995), 163–180. - 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

502

hillenbrand

Underwood, P.A., The Kariye Djami. i. Historical introduction and description of the mosaics and frescoes. ii. The mosaics, New York 1966. Vryonis, S., Jr., The decline of medieval Hellenism in Asia Minor and the process of Islamization from the eleventh through the fifteenth century, Berkeley and Los Angeles 1971. Whitfield, R., and A. Farrer, with contributions by S.J. Vainker and J. Rawson, Caves of the thousand Buddhas: Chinese art from the silk route, ed. A. Farrer, London 1990.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 25

Exploring Europe through Medieval Islamic Folk Literature Niall Christie

1

Introduction

In his biography of Saladin, al-Nawādir al-sulṭāniyya wa-l-maḥāsin al-Yūsufiyya, the religious scholar and author Bahāʾ al-Dīn b. Shaddād (539–632/1145–1234) tells us that, while they were traveling from Ascalon to Acre in late 584/early 1189, Saladin said to him of his enemies, “I have it in mind that, when God Almighty has enabled me to conquer the rest of the coast, I shall divide up the lands, make my testament, take my leave and set sail on this sea to their islands to pursue them there until there no longer remain on the face of the earth any who deny God—or die [in the attempt].”*,1 This is a provocative statement for modern historians, as it raises the question of exactly what Saladin imagined he would find his enemies’ homelands to be like. We know that Saladin was highly educated, and hence we might suggest that he was aware of accounts of the Frankish lands written by Muslim geographers, but at the same time his reference to islands is more immediately reminiscent of some of the descriptions of Frankish lands found in Arabic folk literature. This, then, encourages us to consider how far Muslim views of their opponents’ homelands were more informed by what we might call “popular culture,” as expressed in folk tales such as the Arabian Nights and the sīra (epic) literature, rather than by the works of more academic writers, such as the Muslim geographers. In what follows, we will not be offering a definitive answer to this question, which is likely impossible. However, as a contribution to the discussion we will be conducting an exploration of depictions of the land of the Franks in Arabic folk literature in order to draw out some themes that might have

* The author would like to thank Letizia Osti and Maaike van Berkel for giving him the opportunity to contribute to this volume, and for their helpful and efficient guidance throughout the process. 1 Bahāʾ al-Dīn b. Shaddād, Nawādir al-sulṭāniyya 22; Bahāʾ al-Dīn b. Shaddād, Rare and excellent 29.

© Niall Christie, 2022 | doi:10.1163/9789004525245_026

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

504

christie

informed Muslim imaginings of what it was like. In the process, it is the hope of this author to respond to Hugh Kennedy’s encouragement to historians to give greater attention to what we can learn from the entertaining side of our sources. Characters mentioned below are drawn from “the rich variety of human nature; the brave and loyal, the shrewd and calculating, the cruel and corrupt and the foolish and besotted,” and hopefully in examining their stories this chapter will provide some food for thought and avoid being too “difficult, problematic and, yes, rather dull.”2 In conducting our enquiry, we will be drawing information, as indicated above, from folk literature, especially the Arabian nights and some sīra works. The Arabian nights perhaps needs no introduction; it is a collection of tales, dating in its earliest known manuscript from at least the third/ninth century, that has become so well-known in the popular consciousness that expressions such as “Aladdin’s cave” and “genie in a bottle” are in common use.3 The sīra literature, also known as sīra shaʿbiyya (popular epic), is less well-known; these are long epic works, written in prose, rhymed prose, verse, or a mixture of these, and usually centering on the adventures of a particular hero or group of heroes, that were performed orally in public to (principally male) audiences. They might incorporate fictionalized versions of historical events or persons as part of their content. As a genre, they can be traced back as early as the sixth/twelfth century, though they undoubtedly circulated in oral tradition earlier, and they have continued to be told and retold in various versions up to the modern day, as well as starting to be the subject of television shows.4 Our study of the folk literature will be supported by works from a number of other genres, most of which would have been circulating among the educated classes of the time. These include Muslim geographical works in particular; Muslim scholars read and built upon classical works of geography, especially that of Ptolemy (ca. 100–170), whose division of the world into seven horizontal bands, or climes, they followed. They also inherited their classical sources’ view that the climate in which an individual lived affected their physical and mental characteristics, as we will see.5 Where appropriate, we will also draw on other works by Muslim authors, such as travelogues, poetry, and autobiographical works.

2 Kennedy, Court of the caliphs xxiii and xx. 3 The body of scholarship on the Arabian nights is likewise immense; see Marzolph, Arabian nights bibliography. 4 For a useful introduction to the sīra literature, see Kruk, Warrior women 1–13. 5 Hillenbrand, Crusades 268–270.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

2

505

Caveats

Before beginning our discussion, there are some caveats that should be addressed. First, specifically on Saladin’s words above: as we have noted, Saladin was a highly educated man who may have been aware of works from the Muslim geographical tradition, and in such works Britain, the island from which Richard i the Lionheart (r. 585–595/1189–1199) came, makes very occasional appearances (though often with some confusion between Britain and Brittany).6 One might therefore be tempted to suggest that Saladin was referring specifically to the island origins of the principal leader of his opponents in the so-called Third Crusade. However, assuming that this account is genuine and that Bahāʾ al-Dīn’s dating is correct, Saladin made the statement over a year before Richard arrived at Acre in Jumādā i 587/June 1191. Saladin could not therefore have anticipated that Richard would become the main leader of the crusading expedition that would so frustrate him in the future, and thus is unlikely to have been thinking specifically of Britain when he made the statement recorded by his biographer.7 Second, in our understanding of the term “Franks,” we will be adopting the rather blurry definition current in medieval Arabic writing at the time, meaning that the Ifranj (as they were called) were defined as the inhabitants of western Europe, normally but not always distinguished from the Rūm (Byzantines). The Arabian nights tale of “King ʿUmar b. al-Nuʿmān and his family” refers to Franks coming from “Faransīs wa-l-Nimsā wa-l-Dūbra wa-Jawarna wa-Bunduqiyya waJanuwīz” (France, Austria, Dubrovnik, Jawarna, Venice, and Genoa),8 a description that gives some sense of just how broad this definition could be, as well as a salutary reminder that we cannot expect the sources that we examine to be consistent either internally or between each other in their descriptions of the Franks and their lands. Thus, we must remember that we will be dealing with broad-brush, often inconsistent, generalizations about Ifranja (the land of the Franks) rather than more specialized expert discussions of specific countries. Finally, we must acknowledge the rather problematic nature of the source material under discussion here. The folk stories that were circulating in the Muslim world at the time were transmitted primarily orally for a long time before they were written down, and as such were subject to mutation, adaptation, and multiplication into various different versions; thus, it is difficult to assess which version (or versions) of a story a given individual might have been 6 Christie, Levantine attitudes 10–12. 7 The author is grateful to George Christie for having drawn attention to this question. 8 Alif [sic] laila i, 488; Arabian nights i, 422.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

506

christie

familiar with. These stories are also difficult, if not impossible, to date precisely; clear internal references to historical events might provide an earliest date for at least a part of a tale, references in other works to the tale in question might help with a general dating, and clearly dated manuscripts (often of late provenance) might give us an indication of when one particular version of the tale crystallized, but that normally only gives us a broad time period that any text that we have might be seen as stemming from; for example, one of the best known Arabic sīra works is Sīrat al-amīra Dhāt al-Himma, which refers to events from the first/seventh to fourth/tenth century, is referred to by other authors in the sixth/twelfth century and appears in its first surviving manuscript only in 833–834/1430. This gives a potentially long period in which this story might have begun to circulate, though in this case the early sixth/twelfth century seems likely.9 At the same time, we cannot simply write off this material, as it contains much that is informative about the attitudes and preoccupations of its audiences, despite the difficulties that we have tracing its creation and dissemination.10 It is fair to say that what we will be examining here is a collection of themes and ideas that might have informed popular Muslim impressions of Frankish lands, though the precise image that any individual Muslim of the time might have had would have varied, depending on their levels of credulity and education.

3

Land of the Franks

Let us start with the land of the Franks itself, beginning with the Arabian nights. As we saw earlier, according to the tale of “King ʿUmar b. al-Nuʿmān and his family,” the Franks come from a variety of countries. In another part of the tale we are told that the Franks live on a number of islands, including one called “Jazīrat al-Kāfūr” (the Isle of Camphor).11 It is not clear if the countries listed previously are considered to be on these islands. In the tale of “Nūr al-Dīn and Miriam the sash-maker” we are told that Ifranja itself is “a large city—the home of many marvels, crafts and different plants—resembling Constantinople” and found in the lands of the Rūm.12 At the same time, it enjoys mixed relations with the “kings of the islands,” who on the one hand seek the eponymous Frankish Princess Miriam’s hand in marriage through diplomacy but on the other seek 9 10 11 12

Epic of the commander 3–4. Kruk, Warrior women 25. Alif laila i, 381; Arabian nights i, 330. Alif laila iv, 300, 321, and 324; Arabian nights iii, 385, 403, and 405.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

507

to steal her father’s horses.13 Both the tale of “ʿAlāʾ al-Dīn Abū l-Shāmāt” and the “Adventures of Sultan Baybars” set their scenes in Frankish lands, for the most part in the city of Genoa, providing limited information about Frankish lands elsewhere.14 Information in the sīra literature is similarly limited, if slightly more copious.15 Sīrat ʿAntar includes the Frankish King Līlimān, the Lord of the Isles, who, we are told, is not a Byzantine subject but rules a state that is 40 days from the Byzantine court and four months’ journey from one end to the other. The hero ʿAntar defeats Līlimān’s Franks in battle, and they withdraw to the Isle of Camphor, which we have also seen mentioned above.16 The action of Sīrat al-Ẓāhir Baybars ranges widely in geography, with some episodes taking place in various European locations, including Genoa, Macedonia, Rome, “Catalan,” Barcelona, Portugal, “the islands of England” (ruled by multiple kings), “the islands of al-Manīʿa” (Germany?), and also an unnamed Frankish island on which a stone bird emits a call that attracts ships.17 In Qiṣṣat Abī Zayd al-Hilālī we hear of Ghādir, brother to the Frankish King al-Ghurūr, who “lives in the islands,”18 while one version of Qiṣṣat al-Zīr states that the pre-Islamic Himyaritic King Ḥassān b. Tibān ruled a kingdom stretching from China in the east to the “lands of the Franks and the Christians” in the west.19 Sīrat al-amīra Dhāt al-Himma gives us a range of information. We are told that “Aqyūdash” is the most distant place in Frankish territory,20 while a battle is fought at the “city of Ifranja.”21 An episode in the western Mediterranean tells us of a talismanic statue bearing a key in Cadiz, as well as tales of a golden calf whose bellow signals the downfall of the Frankish rulers, and a treasure house containing images foretelling the arrival of the heroes al-Baṭṭāl and Ẓālim, who

13 14

15 16 17

18 19 20 21

Alif laila iv, 300, and 338; Arabian nights iii, 386 and 416. See for example Alif laila ii, 118–123; Arabian nights i, 877–883; and Tausend und eine Nacht iv, 826–840. It is worth noting that the “Adventures of Sultan Baybars” in Tausend und eine Nacht is likely an interpolation of a retelling of the Sīrat al-Ẓāhir Baybars and is thus probably not a genuine Arabian nights tale; however, it contains enough thematic elements of the original Sīrat al-Ẓāhir Baybars to be included in this study. On this topic, see Marzolph, van Leeuwen, and Wassouf, Arabian nights encyclopedia i, 123 and 247; and ii, 736. For an excellent discussion of Europe in the sīra literature, see Lyons, Land of war. Lyons, Land of war 42; and Lyons, Arabian epic iii, 70 (sec. 81). Lyons, Arabian epic iii, 111 (sec. 38), 112–114 (secs. 39–41), 126–127 (sec. 57), 129–132 (secs. 61– 63), 136–137 (sec. 68), 141–154 (secs. 75–93), 157–158 (secs. 98–99), 161–162 (sec. 104), 167–168 (sec. 115), 199–200 (sec. 162), 217 (sec. 189), 224–225 (sec. 200), and 234 (sec. 213). Lyons, Arabian epic iii, 299 (sec. 8). Lyons, Arabian epic i, 12; and iii, 651 (sec. A1). Lyons, Arabian epic iii, 358 (sec. 57). Lyons, Arabian epic iii, 361 (sec. 59).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

508

christie

defeat the Franks.22 Meanwhile a Panjak or Venetian city named Kandafrah is ruled by a fire-worshipping king named Daqyādūrus.23 This sīra also tells us that the Byzantine Empire comprises 17 climes with 300 islands in it, including the island of Qamrān in the “furthest sea,” ruled by the subordinate King Luke, but Luke’s son Baḥrūn conquers 160 islands and besieges the Byzantine Emperor Michael in Constantinople.24 We also meet Qarāqūnā, an “island king” who defies the Byzantine emperor from his formidable fortress, the Castle of the Tooth.25 Meanwhile, several Frankish kings lead expeditions to the east, as well as against each other.26 Thus, we have a number of accounts that provide a somewhat vague and inconsistent description of the lands of the Franks. They consist of a number of different countries; there seem to be a lot of islands, sometimes with multiple rulers; and they have unclear or mixed relations with the Byzantines and each other. This vagueness is something that these stories share with the works of Muslim geographers, who also provide inconsistent accounts of the lands of the Franks and their relations to the lands of the Byzantines.27 For example, the geographer al-Iṣṭakhrī (fl. ca. 340/951), his continuator Ibn Ḥawqal (d. after 367/978), and the anonymous author of the Persian Ḥudūd al-ʿālam (written in 327/982) all describe the country of Ifranja as being part of Byzantine territory, and by the same token two more geographers, Ibn Khurdādhbih (fl. third/ninth c.) and Ibn al-Faqīh (fl. 290/903), specifically describe Rome as being a Byzantine city.28 The traveler and geographer al-Masʿūdī (before 280– 345/893–956) gives two different descriptions in two different works. In his Kitāb al-Tanbīh wa-l-ishrāf he states that the Franks, Slavs, and other northern peoples live in a cold, damp land of snow and ice, which has turned them into blue-skinned, red-haired brutes, a reflection of the classical assumptions about the effects of the climate on humans that we mentioned above. We learn that these people share the same language and king with the Byzantines and

22

23 24 25 26 27 28

Lyons, Arabian epic iii, 372 (sec. 69). Similar predictions of the Muslims as the foretold conquerors of Iberia appear in a number of historical and geographical works; see for example Ibn Khurdādhbih, Kitāb al-Masālik 157. Lyons, Arabian epic iii, 475 (sec. 148). Lyons, Arabian epic iii, 412 (sec. 101) and 424 (sec. 111). Lyons, Arabian epic iii, 416 (sec. 104). Lyons, Land of war 45. On the presentation of Europe in the Muslim geographical tradition, including as it relates to the crusades, see Hermes, [European] other; and Christie, Illusion. See al-Iṣṭakhrī, Viae regnorum 9; Ibn Ḥawqal, Opus geographicum 14; Ḥudūd al-ʿālam 158; Ibn Khurdādhbih, Kitāb al-Masālik 104; and Ibn al-Faqīh, Kitāb al-Buldān 149.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

509

Greeks, and al-Masʿūdī also tells us that Rome is the Frankish capital.29 Meanwhile, in his better-known Murūj al-dhahab wa-maʿādin al-jawhar, he describes the Franks as fierce warriors, and states, “Ifranja is completely unified in one kingdom. There is no competition between them about that, nor is there any factionalism. The name of the capital of their kingdom at this time is Paris. It is a great city.”30 In light of our discussion it is clear that the mixed relations that we see in the folk tales between the Franks, each other, and the Byzantines, are reflected in the lack of consistency that we see in the geographical works. Al-Masʿūdī’s comments are especially engaging in this regard because his comment in the Murūj forms a direct contrast to the depiction of the Franks that we have seen in the folk tales, especially in the tale of “Nur al-Din and Miriam the sash-maker” and Sīrat al-amīra Dhāt al-Himma; while al-Masʿūdī gives us an image of a unified Frankish kingdom, the Frankish lands that we see in these tales are divided and at times fight with each other and the Byzantines. Every student of history knows that somewhat variable interactions between regions were a feature of Europe throughout the whole of the Middle Ages, but it is probably safer to assume that in the tales the mixed relations that we see between the various Christian rulers are intended first and foremost to assist the narrative rather than to provide a truthful depiction. However, it is interesting to see that the tales inadvertently provide depictions of Frankish territories that are in some ways more accurate than that found in al-Masʿūdī’s intentionally more scholarly work. Naturally, Muslim knowledge of Europeans evolved over the centuries, particularly after the latter came to the east on crusades, but it is striking that even though the Muslim historical sources from the sixth/twelfth and subsequent centuries demonstrate a greater awareness of the Franks who were active in the Levant; the knowledge that Muslim writers express about the homelands from which the Franks came continues to be sketchy. We might take, as an example, the description of the Frankish lands found in the Muʿjam al-buldān of the geographer Yāqūt b. ʿAbdallāh al-Ḥamawī (574 or 575–626/1179–1229): Afranja: A great nation with extensive countries and many kingdoms. They are Christians, and they trace their origins to their ancestor, whose name was Afranjush, and they speak faranak [Frankish]. Afranja is next to Rome. The Rūm and they are located to the north of al-Andalus, towards the east of Rome. The capital of their kingdom is Nūkabarda [Lombardy],

29 30

Al-Masʿūdī, Kitāb al-Ttanbīh 23–24, 83, and 182. Al-Masʿūdī, Murūj al-dhahab ii, 145.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

510

christie

which is a great city, and they have about 150 cities. Before the appearance of Islam, the first of their countries from the direction of the Muslims was the island of Rhodes, opposite Alexandria in the middle of the sea of Shām.31 Yāqūt also gives us an extended and at times fanciful description of the city of Rome, largely and explicitly based on the works of earlier writers, for whose errors he disclaims responsibility.32 As both Carole Hillenbrand and Bernard Lewis have noted, even writers who demonstrate a more detailed knowledge of the geography of the Frankish homelands still provide descriptions of their inhabitants that are nebulous; for example, even Muḥammad b. Muḥammad al-Idrīsī (d. 560/1164–1165 or 571/1175), who served at the court of the Norman rulers of Sicily and hence had access to sources of information that were not available to his peers, while he gives a reasonably accurate account of the physical layout of countries such as England and France, still depicts their inhabitants as little more than energetic and courageous occupants of cold countries, a stereotypical description that al-Masʿūdī would have found familiar.33 The improvements in Muslim knowledge of the geography of the Frankish lands are reflected in the folk tales; for example, Sīrat al-Ẓāhir Baybars, dating from probably the ninth/fifteenth century,34 demonstrates a much deeper awareness of the states of Europe than the earlier Sīrat al-amīra Dhāt alHimma. Yet given the presence of the crusaders in the eastern Mediterranean region at the time these tales were circulating, and the striking lack of more detailed information about Europeans that we find in the later geographical works, it is not surprising that the folk literature’s depictions of Europe and its inhabitants primarily reflect other, more important concerns than an attempt to provide a clear picture.

4

Churches and Monasteries

By the time these tales were recorded, it was widely known that the Franks were Christians.35 When we look for Christian elements in these stories, the

31 32 33 34 35

Yāqūt b. ʿAbdallāh al-Ḥamawī, Muʿjam al-buldān i, 270. Yāqūt b. ʿAbdallāh al-Ḥamawī, Muʿjam al-buldān iii, 113–117; on this point, see also Lewis, Muslim discovery 143–144 and 177–178. Hillenbrand, Crusades 271–273; Lewis, Muslim discovery 147–149. Kruk, Warrior women 164. Christie, Muslims and crusaders 84.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

511

first thing that strikes us is how often action that takes place in Frankish lands occurs in churches, monasteries, and similar institutions. We see this in both the Arabian nights and the sīra literature. Starting again with the former, in both the tale of “ʿAlāʾ al-Dīn Abū l-Shāmāt” and that of “Nūr al-Dīn and Miriam the sash-maker,” when the hero is captured by the Franks he ends up (by rescue from imminent execution in each case) working in a church, and it is in this church that his deliverance is engineered.36 Likewise, it is while she is en route to an island monastery that Miriam the sash-maker is first captured, leading to the slavery that eventually brings her to Nūr al-Dīn.37 In the “Adventures of Sultan Baybars,” when the Genoese Princess Mariam is abducted by pirates and they stop for water on Sardinia, it is to a monastery that they go, and it is in this monastery that she gives birth to a son.38 Although not strictly speaking in Frankish lands, it is also at a Christian building, this time a convent, that the Muslim Prince Sharkān meets, is entertained by, and falls in love with the Armenian/Byzantine Princess Abrīza in the tale of “King ʿUmar b. al-Nuʿmān and his family.”39 Churches and monasteries play similar roles in the sīra texts, though it is worth noting that pinning down the exact locations of these institutions is more difficult, as they are not always clearly specified; indeed, as Malcolm Cameron Lyons notes, in the sīra literature “facts of history and geography are inevitably distorted, jumbled or misinterpreted.”40 It is also worth noting that in these epics churches and monasteries in both Frankish and Byzantine territory have similar characteristics and play similar roles to those that we find in the Arabian nights, as will be apparent from examples cited here. Again, we see them appearing as places of imprisonment or enslavement. In Sīrat al-Ẓāhir Baybars, ʿAbd al-Ṣalīb, the grandson of the king of Portugal, takes a group of Muslims pilgrims prisoner and forces the young men to cut wood, the women to nurse children, the old men to tend pigs, and the children to serve at a monastery.41 In Sīrat al-amīra Dhāt al-Himma we hear, for example, of monasteries that serve as prisons, both in the Byzantine territory near Antioch and near the city of Ifranja, and of a church that serves the same purpose 36 37 38 39 40 41

Alif laila ii, 117–122, and iv, 308–15; and Arabian nights i, 878–881, and iii, 393–398. Alif laila iv, 300–301; and Arabian nights iii, 386. Tausend und eine Nacht iv, 838–839. Alif laila i, 359–374; and Arabian nights i, 311–332. Lyons, Land of war 42. Lyons, Arabian epic iii, 80 (sec. 4). In an epistle on slaves, the Baghdadi Christian physician al-Mukhtār b. ʿAbdūn b. Buṭlān (d. 458/1066) advises his readers that the best choice for a slave girl to nurse a baby is a Frank, suggesting that this view of the appropriate role for women captured from the enemy went both ways; see Nawādir al-makhṭūṭāt i, 383.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

512

christie

in a Frankish city in Iberia.42 Though not explicitly described as prisons, seven churches in Constantinople are nonetheless especially dangerous, being filled with traps that include a giant snake, a deadly fountain and pool of quicksilver, cannon, mechanical figures with swords, and a dungeon that threatens to fill with water and drown the heroes who are exploring it.43 Monasteries also continue to be places where children are born and raised. In Sīrat al-Ẓāhir Baybars the aforementioned ʿAbd al-Ṣalīb, after his birth, is nursed by a “mangy bitch” at a monastery.44 Meanwhile, like her Arabian nights counterpart, the Genoese Princess Miriam gives birth to her son ʿArnūs in an island monastery, this time en route from Jaffa to Genoa.45 On a wider level, the significance of churches as symbols of Christian presence is perceptible not only in their prominence in the narratives but also, in Sīrat al-amīra Dhāt al-Himma, in the fact that when Muslim armies raid Christian territory, these buildings are particularly singled out for destruction, suggesting that erasing these from the landscape is being presented as a way of de-Christianizing and Islamicizing it.46 Thus it seems that when our narrators wish to conjure up images of Christian lands for their listeners, one of the first devices they employ is to place significant portions of the action in churches and similar buildings, clear symbols of Christianity that are distinct from Islam. Within this, the actual functions of these buildings are often divorced from reality, and enslavement, imprisonment, and child-rearing are prominent themes. Thus they are used to further the storylines above all else, mostly disregarding attempts at realism; indeed, as Remke Kruk has argued of the sīra literature in particular, entertainment is the objective above all else. However, these tales nonetheless perpetuate stereotypes and negative images of the Franks that undoubtedly had an impact on their listeners’ perspectives.47

5

Impurity and Enslavement

Yet what of the Christian faith itself? The faith that we see depicted in these tales is one that is clearly intended to be seen unfavorably in comparison to Islam. There is a particular emphasis on impurity and (to a lesser degree)

42 43 44 45 46 47

Lyons, Arabian epic iii, 358 (sec. 57), 361–362 (sec. 59), and 372–373 (sec. 69). Lyons, Arabian epic iii, 225 (sec. 200). Lyons, Arabian epic iii, 80 (sec. 4). Lyons, Arabian epic iii, 109–111 (secs. 37–38). Lyons, Arabian epic iii, 310 (sec. 8), and 465 (sec. 140). Kruk, Strijders voor het geloof?

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

513

enslavement. We are told that ʿAlāʾ al-Dīn Abū l-Shāmāt is overburdened with tasks in the church in Genoa, on pain of death, including handling the excrement of a blind cripple. He is also empowered to go out into the street and indenture anyone he wishes into service at the church. When he is told to leave the church for the night, his employer’s first suggestion is that he should spend the night at a wine shop.48 Thus, the narrative combines compelled service, excrement, and wine in a short passage, thus associating Frankish Christianity with impurity, as well as arbitrary enslavement. Excrement also makes an appearance in the tale of “King ʿUmar b. al-Nuʿmān and his family,” when we are told that the Byzantines use the excrement of the Christian patriarch to make incense, eye ointment, and treatments for the sick.49 Themes of Christian impurity and enslavement also appear in the sīra literature; we have already seen multiple cases of Christian buildings that served as prisons for Muslims as well as, in Sīrat al-Ẓāhir Baybars, the Portuguese Prince ʿAbd al-Ṣalīb enslaving Muslim pilgrims; it is worth recalling that the old men among these slaves are forced to look after pigs, classic symbols of impurity in Islam. Themes of impurity appear again when the Genoese Princess Miriam has to ward off the attentions of a lecherous priest in Jerusalem who, after having preached in support of eating pork and drinking wine, seeks to get her to sleep with him, and again when Baybars’ arch enemy, the wicked Christian Juwān, magically takes on Baybars’ shape and demands both wine and the opportunity to seduce another man’s wife.50 In the meantime, we are told that the pope has a daughter and multiple sons, the hypocrisy of which would not be lost on listeners aware of the Catholic Church’s requirement for clerical celibacy.51 Similar symbols of impurity appear in Qiṣṣat Abī Zayd al-Hilālī when the hero Abū Zayd enters a monastery, only to be offered pork and wine; subsequently the Frankish King Ghurūr seeks to take the daughter of King Nuʿmān of the city of Kawākib by force, promising his vizier that he will give the young woman to him as a reward after he is done with her, planned sexual abuse that he does not achieve, being instead thwarted and killed by Abū Zayd.52 In Sīrat al-amīra Dhāt al-Himma wine and sexual laxity, in particular, are markers of Christian identity, with pigs also making occasional appearances. Perhaps the most striking epitome of this is the Queen of Georgia, “a formidable woman who eats a whole pig each day for breakfast and another in the evening, and is

48 49 50 51 52

Alif laila ii, 117–118; and Arabian nights i, 878–879. Alif laila i, 491; Arabian nights i, 424–425. Lyons, Arabian epic iii, 109 (sec. 37), and 169 (sec. 118). Lyons, Arabian epic iii, 130–131 (sec. 62), and 151 (sec. 89). Lyons, Arabian epic iii, 298–299 (secs. 6–7).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

514

christie

old, ugly, bloodthirsty, a drunkard and a nymphomaniac. When her lovers tire, she squeezes their heads until their eyes fall out of their sockets.”53 Later the villain ʿUqba persuades the nephew of the King of Aden to convert to Christianity by pointing out that its followers are allowed to drink wine, while Princess Zanānīr of Qalʿat al-Shams, who has previously been enthusiastic about both using her sexual charms to convert people to Christianity and drinking wine, is prepared to give up the latter (and presumably the former) for the Muslim hero Luʾluʾ, though she only converts to Islam later, after having been instructed to in a dream.54 Dire consequences associated with drinking wine are outlined early on in the epic when a Byzantine general, having set out to convert the Kaʿba into a church, gets drunk and has a dream of a giant shape that turns his head backward on his shoulders. He wakes with his head indeed turned around, returns to Constantinople, brays three times like a donkey, and dies.55 Use of wine by Christians more directly backfires when a group of them capture and bind alBaṭṭāl. They pour wine dregs over him, but the wine loosens his bonds, and he is able to slip them and escape by night.56 Meanwhile, another example of sexual laxity appears in the city of Karfanās, where al-Baṭṭāl and his companion Madhbahūn find the princess of the city hiding in an idol in the shape of a bird, and she immediately offers herself to Madhbahūn, claiming that such behavior has been permitted by her religious advisor, a monk named Saqrāq.57 Thus, we see the sīra literature peppered with symbols of Christian impurity in various forms, with wine and sexual looseness being particularly prominent.

6

Murder and Idols

Christianity is also critiqued in other ways in the folk literature, and in the process we see depictions of the faith that present its followers as bloodthirsty and idolatrous. Examples of the former that are specific to the Franks appear in the Arabian nights tale of “Nūr al-Dīn and Miriam the sash-maker”: the title characters spend two periods, unwillingly, in Ifranja. The first time, when it is discovered that his daughter Miriam is no longer a virgin, the king’s advisors instruct him to behead 100 Muslim prisoners as a purification rite, something

53 54 55 56 57

Lyons, Arabian epic iii, 315 (sec. 12). Lyons, Arabian epic iii, 468 (sec. 143) and 382–391 (secs. 75–82). Lyons, Arabian epic iii, 310 (sec. 8). Lyons, Arabian epic iii, 347 (sec. 48). Lyons, Arabian epic iii, 452 (sec. 131).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

515

that he undertakes immediately.58 Then, during their second sojourn in Ifranja, we see the Frankish king’s vizier declaring that he will slit the throats of 30 Muslims as a sacrifice to the Messiah.59 Large-scale killing of Muslims also appears in the tale of “ʿAlāʾ al-Dīn Abū l-Shāmāt,” when the King of Genoa orders the execution of every Muslim from Alexandria who comes within reach in an attempt to avert a prophecy forecasting his death at the hands of a prisoner from the city, although in this case there is no explicit link made to Christianity.60 In Sīrat al-amīra Dhāt al-Himma we hear that ʿUqba has sacrificed many Muslims in an underground chapel, while the Christian monk Bāḥa is in the habit of boiling his Muslim victims in cauldrons, sometimes alive.61 In Sīrat al-Ẓāhir Baybars, Juwān teaches the cannibal King of Rhodes that it is permitted to eat Muslim flesh, but not that of Christians.62 Turning to idolatry, the same sīra tells us that after Juwān has finally been tracked down and executed, his head is interred at a purpose-built church, which also includes a golden statue of him with emerald eyes.63 Admittedly, we are not explicitly told in this case that the head or the statue is subsequently worshipped, but other sīra works make their associations of idols and Christianity clearer. Abū Zayd, after having refused the pork and wine offered to him in the monastery in Qiṣṣat Abī Zayd al-Hilālī, as mentioned above, also takes the opportunity the following night to destroy the monastery’s idol, and he tells the monks that a saint flew away with it to Jerusalem.64 The bird-shaped idol in Karfanās from Sīrat al-amīra Dhāt al-Himma, mentioned above, is located in a church, and the king’s daughter speaks from it at the behest of the monk Saqrāq, convincing her father and others that it is a magical bird that prophesies. For Saqrāq, this trick goes disastrously wrong when the princess, at Madhbahūn’s bidding, has the bird command that the duplicitous monk and his followers be killed.65 The same epic also tells us of an old man who is found prostrating himself before a bronze idol at a cave in Buḥayrat al-Falak.66 Although, strictly speaking, heresy (from a Muslim point of view) rather than idolatry, the Muslim heroes of the epic are also appalled to hear of the beliefs 58 59 60 61 62 63 64 65 66

Alif laila iv, 308–309; Arabian nights iii, 393. Alif laila iv, 322; Arabian nights iii, 403. Alif laila ii, 117–118 and 122; Arabian nights i, 878 and 881–882. Lyons, Arabian epic iii, 402 (sec. 91) and 466 (sec. 141). Lyons, Arabian epic iii, 182 (sec. 136). Lyons, Arabian epic iii, 225 (sec. 201). Lyons, Arabian epic iii, 298 (sec. 6). This scene recalls a similar destruction of idols by the prophet Abraham in Q 21:51–66. Lyons, Arabian epic iii, 452 (sec. 131). Lyons, Arabian epic iii, 390 (sec. 81).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

516

christie

of the queen and people of the far-distant land of al-Qarāzima, who assert that the Messiah is God Himself, not just the Spirit of God.67 An additional subtext that appears in the folk literature occasionally is the idea that Christians in general are willing to allow themselves to be motivated by superstitions and questionable claims, rather than placing their faith directly in the one God. We have already seen one example of Christian gullibility, with the bird idol at Karfanās in Sīrat al-amīra Dhāt al-Himma. The mass killing of Muslims by the King of Genoa in the Arabian nights tale of “ʿAlāʾ al-Dīn Abū l-Shāmāt,” also testifies to a superstitious belief that one can avert fate (which of course fails, as he is indeed killed by ʿAlāʾ al-Dīn, who is a prisoner from Alexandria). In the former epic we also see another case of Christian gullibility when the wily al-Baṭṭāl temporarily convinces the Byzantines that he is the Messiah through a number of tricks, including having servants go to Byzantine territory and pretend to be sick, so that he can miraculously “cure” them; and smearing the faces of other servants with a mixture to make them exude a heavenly glow.68 Franks also occasionally seem to be aware of their version of Christianity being especially inadequate; in the Arabian nights retelling of the “Adventures of Sultan Baybars,” Baybar’s eventual ally Shīḥa is brought up as a prince in Genoa, but his adopted father sends him to a monastery in the Middle East to learn about the Christian religion, implying that it is not possible to gain a proper education in the faith in Europe.69 As a general, if not terribly surprising, observation, we see many more conversions from Christianity to Islam, rather than the other way round, in the folk tales, as those who are converted come to appreciate the truth of Islam over the falsehoods and deficiencies of Christianity; thus, our storytellers also use realization of inadequacies and subsequent conversion as another means to emphasize the superiority of their faith.

7

Parallels in Historical Texts

Parallels with the criticisms of Christianity that we have seen in the previous two sections may be found in a number of historical works from the crusad-

67

68 69

Lyons, Arabian epic iii, 403 (sec. 92). The implication seems to be that this is seen as an especially unusual belief in this instance, when of course this is actually true of most branches of Christianity. This does raise the question of what the writers of the sīra thought that “mainstream” Christians believed. Lyons, Arabian epic iii, 417–418 (sec. 105). Tausend und eine Nacht iv, 765.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

517

ing period. For example, issues of both Christian falsehood and impurity arise in the Riḥla (travelogue) of the Andalusian Muslim traveler Muḥammad b. Jubayr (540–614/1145–1217); in his description of Acre, he states that “Unbelief and unpiousness there burn fiercely, and pigs […] and crosses abound. It stinks and is filthy, being full of refuse and excrement.”70 Likewise, pigs and the bloodthirsty nature of Christianity are reflected in the work of an anonymous poet quoted by the Mamluk historian Ibn Taghrī Birdī (prob. 812–874/1409/10– 1470), who describes the crusaders using pigs’ blood in worship.71 Meanwhile, the Shayzarī amīr Usāma b. Munqidh (488–584/1095–1188) makes use of wine imagery in his well-known depiction of a Frankish wine-seller who returns home, finds someone in bed with his wife, and shows only limited jealousy, thus combining the drinking of forbidden substances with sexual laxity to create a negative depiction of the Franks in his Muslim readers’ minds.72 One striking difference between the historical sources and the folk literature is their depictions of what the Christians worship. As Carole Hillenbrand has demonstrated, in the historical sources the cross is a central feature of Muslim descriptions of Christianity, the primary mark of Christian identity.73 However, in the folk tales the emphasis is much more on Christians worshipping idols rather than crosses, as we have seen in the depictions above. This is not to say that the historical texts lack references to Christian “idols”—icons and statues—or that the folk literature does not include crosses. Usāma b. Munqidh, for example, describes a Frank who showed him an icon of Mary and Jesus, saying, “This is God when He was young,” while the poet Ibn alQaysarānī (478–548/1085–1154) declares that he would like to be transformed into an icon of St. George so that Frankish women might kiss him.74 Meanwhile, in the folk tales Christian names sometimes refer to the cross, such as the Portuguese Prince ʿAbd al-Ṣalīb (“servant of the cross”) above; Christian characters swear by the cross, and both Muslims and Christians at times execute (or seek to execute) their enemies by crucifixion.75 But there is a much greater

70 71 72

73 74 75

Ibn Jubayr, Rihla 303; Ibn Jubayr, Travels 318. Ibn Taghrī Birdī, Nujūm al-zāhira v, 151–152. Usāma b. Munqidh, Usāmah’s memoirs 135–136; Usāma b. Munqidh, Book of contemplation 148. For a discussion of Frankish impurity in its various forms, see for example Hillenbrand, Crusades 274–319 and 347–351. On the tale of the cuckolded husband, see also Christie, Muslims and crusaders 120. Hillenbrand, Crusades 304–307. Usāma b. Munqidh, Usāmah’s memoirs 135; Usāma b. Munqidh, Book of contemplation 147– 148; Ibn al-Qaysarānī, Shiʿr 310. See for example Alif laila iv, 322; Arabian nights iii, 403; and Lyons, Arabian epic iii, 409 (sec. 98). On this topic see also Hillenbrand, Crusades 308.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

518

christie

emphasis in the folk literature on Christians as idolaters in a sense that more greatly recalls the idol-worshipping pagans of pre-Islamic Arabia, which is presumably intended to more sharply demarcate the villains and heroes of each work for dramatic effect.

8

Conclusion

The depictions of the Frankish lands and their inhabitants in the Arabic folk tales, while in many cases resembling depictions that we find in other sources such as geographical works or historical chronicles, in general exaggerate the Franks’ negative qualities and therefore present most of them (except those who see the light and convert to Islam) as suitable villains of the stories that are told about them; thus they are worthy opponents for the Muslim heroes, and victories over them are both clearly justified and reassurances of the power of good to defeat evil. Realism and narrative consistency are on the whole subordinated to the requirements of plot and drama, making these works that seek to entertain first and instruct second. How far the information that they contain was believed to be true by their audiences is a question that can never be fully answered, but Saladin’s words at the start of this chapter suggest that they had at least some impact on their listeners’ viewpoints. In any case, they still give us an insight into the popular stereotypes and images found within the literary traditions with which most people would have been familiar. In a by-now well-known story, Usāma b. Munqidh tells us that among the Franks was a knight with whom he became so close that the latter used to call Usāma “my brother.” However, when his Frankish friend suggested that he might take Usāma’s son off to Europe to learn to be a knight, Usāma was horrified, describing his friend’s statement as “words that would never come from a truly rational head!”76 Of course, we will never know exactly what image of the Frankish lands entered Usāma’s own head upon hearing this suggestion, but if his views were at all influenced by the popular images of Ifranja that we have highlighted above, then his reluctance is entirely understandable.

76

Usāma b. Munqidh, Usāmah’s memoirs 132; Usāma b. Munqidh, Book of contemplation 144. Usāma does provide one explanation of why he objected to his friend’s suggestion, but the precise intention of the Arabic is unclear; see Christie, Muslims and crusaders 123.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

exploring europe through medieval islamic folk literature

519

Personal Note The author would like to express his immense gratitude to Hugh Kennedy for his invaluable guidance and support, first as the author’s MLitt and PhD supervisor, and subsequently as a greatly esteemed colleague in the academic community. As a tribute to Hugh, this chapter consciously seeks to expand on the work the author completed under his supervision.

Bibliography Sources Alif [sic] laila or book of the thousand nights and one night, ed. W.H. Macnaghten, 4 vols., London 1839–1842. The Arabian nights: Tales of the 1001 nights, trans. M.C. Lyons, 3 vols., London 2010. Bahāʾ al-Dīn b. Shaddād, al-Nawādir al-sulṭāniyya wa-l-maḥāsin al-Yūsufiyya (sīrat Ṣalāḥ al-Dīn), ed. J. al-D. al-Shayyāl, Cairo 1964. Bahāʾ al-Dīn b. Shaddād, The rare and excellent history of Saladin: Or al-nawādir alsulṭāniyya waʾl-maḥāsin al-Yūsufiyya, trans. D.S. Richards (Crusade texts in translation 7), Aldershot 2001. Epic of the commander Dhat al-Himma, ed. and trans. M. Magidow (Medieval feminist forum: Subsidia 9), Iowa City 2018. Ḥudūd al-ʿālam, trans. V. Minorsky (gms, new series 11), London 1970. Ibn al-Faqīh, Kitāb al-Buldān (bga 5), ed. M.J. de Goeje, Leiden 1967. Ibn Ḥawqal, Opus geographicum (bga 2), ed. M.J. de Goeje, Leiden 1967. Ibn Jubayr, Riḥla (= The travels of Ibn Jubayr) (gms 5), ed. W. Wright and M.J. de Goeje, Leiden 1907. Ibn Jubayr, The travels of Ibn Jubayr, trans. R.J.C. Broadhurst, New Delhi 2001. Ibn Khurdādhbih, Kitāb al-Masālik wa-l-mamālik, ed. M.J. de Goeje (bga 6), Leiden 1967. Ibn al-Qaysarānī, Shiʿr Ibn al-Qaysarānī, ed. ʿA.J.Ṣ. Muḥammad, Amman 1991. Ibn Taghrī Birdī, al-Nujūm al-zāhira fī mulūk Miṣr wa-l-Qāhira, ed. M.ʿA. al-Q. Ḥātim, 12 vols., Cairo 1963. al-Iṣṭakhrī, Viae regnorum (bga 1), ed. M.J. de Goeje, Leiden: 1967. al-Masʿūdī, Kitāb al-Tanbāh wa-l-ishrāf (bga 8), ed. M.J. de Goeje, Leiden 1967. al-Masʿūdī, Murūj al-dhahab wa-maʿādin al-jawhar, ed. C. Barbier de Maynard and P. de Courteille, vols. 1–2, Beirut 1965–1966. Nawādir al-makhṭūṭāt, ed. ʿA. al-S. Hārūn, vol. 1, Cairo 2001. Tausend und eine Nacht: arabische Erzählungen, trans. G. Weil, 4 vols., Pforzheim 1841.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

520

christie

Usāma b. Munqidh, The book of contemplation: Islam and the crusades, trans. P.M. Cobb, London 2008. Usāma b. Munqidh, Usamāh’s memoirs entitled Kitāb al-Iʿtibār, ed. P.K. Hitti, Princeton, NJ 1930. Yāqūt b. ʿAbdallāh al-Ḥamawī, Muʿjam al-buldān, ed. F.ʿA. al-ʿA. al-Jundī, 7 vols., Beirut 1990.

Studies Christie, N., An illusion of ignorance? The Muslims of the Middle East and the Franks before the crusades, in A.J. Boas (ed.), The crusader world, Abingdon 2016, 311–323. Christie, N., Levantine attitudes towards the Franks during the early crusades (490/1096 –564/1169), PhD diss., University of St Andrews 1999. Christie, N., Muslims and crusaders: Christianity’s wars in the Middle East, 1095–1382, from the Islamic sources (Seminar studies), Abingdon 2020. Hermes, N.F., The [European] other in medieval Arabic literature and culture, ninth– twelfth century ad (The new middle ages), New York 2012. Hillenbrand, C., The crusades: Islamic perspectives, Edinburgh 1999. Kennedy, H., The court of the caliphs: The rise and fall of Islam’s greatest dynasty, London 2004. Kruk, R., Strijders voor het geloof? Bouillon in Mekka en Baybars in Rome, in S. Houppermans, R. Kruk, and H. Maier (eds.), Rapsoden en rebellen: Literatuur en politiek in verschillende culturen, Amsterdam 2003, 61–80. Kruk, R., The warrior women of Islam: female empowerment in Arabic popular literature, London 2014. Lewis, B., The Muslim discovery of Europe, London 2001. Lyons, M.C., The Arabian epic: Heroic and oral story-telling, 3 vols., Cambridge 1995. Lyons, M.C., The land of war: Europe in the Arab hero cycles, in A.E. Laiou and R.P. Mottahedeh (eds.), The crusades from the perspective of Byzantium and the Muslim world, Washington, DC 2001, 41–51. Marzolph, U., The Arabian nights bibliography, 2021, available online: http://wwwuser​ .gwdg.de/~umarzol/arabiannights.html (last accessed August 23, 2021). Marzolph, U., R. van Leeuwen, and H. Wassouf (eds.), The Arabian nights encyclopedia, 2 vols., Santa Barbara, CA 2004.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 26

The Lordship and Bishopric of Banyas in the Latin Kingdom of Jerusalem (1126–1164) Alan V. Murray

Banyas is a most ancient city, situated on a spur at the foot of Mount Lebanon. In ancient times, in the days of the Israelite people, it was called Dan, and was the northern limit of their possessions, just as Beersheba was the southern limit. Thus, whenever the extent of the Promised Land is described, it is said to be “from Dan to Beersheba.” It was enlarged by Philip, the son of Herod the Elder, who was tetrarch of Iturea and the region of Trachonitis; to honor Tiberius Caesar and to perpetuate the memory of his own name, he called it Caesarea Philippi, as can be read in the Gospel of Luke. It is also called Paneas, but our Latins corrupted the name, just as they do with almost all other cities, and called it Belinas. To the east it borders on the territory of Damascus, where two tributaries of the Jordan originate. It is the city of which one may read in the Gospels that “Jesus came into the quarters of Caesarea Philippi, and he asked his disciples,” and so on, and it was here that Peter, foremost among the apostles, received the keys to the kingdom of Heaven from the Lord as a reward for his excellent confession.1 Thus, the historian William of Tyre, writing in the kingdom of Jerusalem in the 1180s, described the town of Banyas (Bāniyās), situated north of the Sea

1 Guillaume de Tyr, Chronique 876–877: Est autem Paneas urbs antiquissima, ad radicem Libani famosissimi promuntorii sita, antiquissimis temporibus et Israelitici populi diebus Dan dicta, terminus a septentrionali plaga possessionis eorum, sicut et Bersabee ab austro: unde quotiens longitudo Terre Promissionis describitur, dicitur a Dan usque Bersabee. Diebus autem Philippi, senioris Herodis filii, qui erat tetrarcha Ituree et Traconitidis regionis, sicut in Luca legitur, qui eam ampliavit, in honorem Tiberii Cesaris et in perpetuam sui nominis memoriam Cesarea Philippi dicta est. Dicitur autem et Paneas, sed nostri Latini, corrumpentes nomen sicut pene omnium aliarum urbium, Belinas vocant. Est autem ab oriente agro Damasceno contermina, iuxta quam fluentorum suorum Iordanis habet originem. Hec est illa civitas, de qua in evangelio legitur quod venit Iesus in partes Cesaree Philippi et interrogavit discipulos et cetera, ubi et Petrus apostolorum princeps ob egregie confessionis meritum regni claves celestis domino tradente suscepit. Translations from William’s chronicle are by the author.

© Alan V. Murray, 2022 | doi:10.1163/9789004525245_027

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

522

murray

of Galilee in the foothills of Mount Hermon, on the occasion when it was captured by Nūr al-Dīn, the Muslim ruler of Syria, in 1164, finally ending the Frankish occupation of the site.2 Banyas was indeed a settlement of great antiquity, which owed its fame and prosperity to the abundant waters that gushed forth from the mountain rock and flowed south into the Jordan. In the Hellenistic period the spring came to be a cultic site associated with the nature god Pan, from whom it took its first recorded name. Despite being officially renamed as Caesarea Philippi by Philip the Tetrarch, son of Herod the Great, in the first century ce, the settlement evidently continued to be known by the local population by the Greek name Paneas, for it was this form that survived when Arabic became the main language of the region, with the typical transformation of the initial unvoiced consonant [p] to voiced [b].3 Like William, the chronicler Fulcher of Chartres was aware of the identification with the biblical Caesarea Philippi, but both writers normally used the written form Paneas, as did the royal chancery of Jerusalem, although as William indicates, the town was known as Belinas in the vernacular of the largely French-speaking western settlers in Palestine.4 Banyas occupied a strategic position on the road from Damascus to the port of Tyre, some 40 kilometers to the west, and came to be a bone of contention between the Muslim emirate and the kingdom of Jerusalem. In the spirit of Hugh Kennedy’s work on castles in the principalities of Outremer, this essay aims to elucidate the place of Banyas as a frontier fortification, lordship, and bishopric in the Latin kingdom, considering its strategic significance and the circumstances in which it passed between Frankish and Muslim control.5

1

The Frankish Acquisition of Banyas (1129)

Banyas came under Frankish control during the reign of Baldwin ii of Jerusalem (1118–1131) as a direct consequence of political upheavals in the emirate of Damascus. The Franks already controlled most of Palestine and had made numerous raids into the Ḥawrān, the fertile and populous Damascene territory

2 Banyas is situated in the Quneitra Governorate of Syria, but with the rest of the Jawlān/Golan Heights has been under Israeli occupation since 1967. It is not be confused with the port of Baniyas in the Tartus Governorate. 3 Berlin, Archaeology of ritual; Seetzen, Unter Mönchen und Beduinen 122–123. 4 Fulcher of Chartres, Fulcheri Carnotensis Historia Hierosolymitana 336–339. 5 Kennedy, Crusader castles.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

523

banyas in the kingdom of jerusalem

east of the Jordan known to the Franks as the Terre de Suète. After Baldwin’s accession, however, Frankish incursions became more organized and greater in magnitude. The king led invasions in 1119, 1121, and 1125, and in 1126 marched against the city of Damascus itself, but withdrew after an indecisive battle.6 During this period Damascus was ruled by the Turkish atabeg Ẓahīr al-Dīn Tughtagin, who had seized power in 1104 after the death of the Seljuq king (malik) Shams al-Mulūk Duqāq.7 Tughtagin was more active than Duqāq had been in opposing the Franks, and in 1111 and 1124 he used Banyas as a base in attempts to relieve Frankish sieges on Tyre. The coastal city owed allegiance to the Fatimid Caliphate in Egypt, but it was a key conduit for the export trade of Damascus, and after the first siege, the caliph allowed Tughtagin to establish a protectorate over the port. Yet Tughtagin was unable to prevent the capture of Tyre by the Franks in 1124, and from that time his emirate came under increasing pressure from the kingdom of Jerusalem.8 In Dhū l-Qaʿda 520/November–December 1126 Tughtagin gave refuge to one Bahrām, a missionary of the Nizārī or Bātinīyya sect, on the recommendation of the Artūqid amīr Najm al-Dīn Ilghāzī, the ruler of Aleppo. The Nizārīs had carried out assassinations of several Sunni leaders in Syria and Iraq, and it seems plausible that Ilghāzī was trying to rid himself of a potentially dangerous group of revolutionaries by palming their leader off on the unsuspecting Tughtagin. Bahrām enjoyed the support of al-Mazdaqānī, Tughtagin’s vizier, but the many Ismaʿili followers who flocked to join him in Damascus were regarded with disdain and apprehension by the majority Sunni population. A solution to the growing tension presented itself when Bahrām asked to be given a fortress as a refuge for himself and his followers. Tughtagin handed over Banyas to the Nizārīs, presumably hoping that they would leave Damascus and devote their energies to combating the Franks.9 Possession of Banyas gave Bahrām and his followers a security they could never enjoy in the great cities of Syria, and it also offered prospects for proselytizing and expansion in the region. Bahrām became especially interested in Wādī al-Taym, a valley extending north of Banyas between Mount Hermon and the Lebanon mountains. At some point in 522 (January 6–December 24, 1128) he procured the murder of Baraq b. Jabal, a local leader, and led an invasion of the valley, but the murdered man’s clan rallied, taking the Nizārīs’ camp by surprise and killing most of them, including Bahrām. The consider6 7 8 9

Murray, Baldwin of Bourcq 114–116, 136, 177–180. El-Azhari, Saljuqs of Syria 171–175; Mouton, Damas et sa principauté 32–34. Murray, Baldwin of Bourcq 138–139, 157–161. Ibn al-Qalānisī, Damascus chronicle 179–190; Ibn al-Athīr, Chronicle i, 260–261.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

524

murray

able number of Nizārīs who had remained in Damascus continued to affront the Sunni townspeople, and Tāj al-Mulūk Buri, who had succeeded his father Tughtagin on Ṣafar 8, 522/February 12, 1128 now resolved to take action against this increasingly disruptive element within the population. On Ramaḍān 17, 523/September 3, 1129 he had al-Mazdaqānī seized and beheaded. As word of the execution spread, the aḥdāth (city militia), followed by much of the Sunni populace, slaughtered as many Nizārīs and their adherents as they could find.10 Bahrām had been succeeded as leader in Banyas by a Persian named Ismāʿīl, who feared an attack on his base and handed over the town to the Franks. Ibn alQalānisī and Ibn al-Athīr give the impression that the Franks were encouraged to undertake a fresh attack on Damascus only after gaining control of Banyas, but in fact Baldwin ii had been planning a major campaign against the city for at least two years. The envoys sent to the west in 1127 to negotiate a marriage for his heir, Melisende, were also charged with recruiting participants for a crusade, and crusaders were arriving in Palestine throughout 1128 and 1129.11 It is doubtful whether the Franks could have moved so quickly to seize Banyas after the massacre of the Nizārīs unless they were already making preparations to act. William of Tyre states that the transfer to the Franks involved a treaty that promised compensation to the Nizārīs, but this must have taken some time to negotiate, especially since the two parties had no previous contact. A more plausible chronology is that by the summer of 1129 Bahrām and his supporters feared that action would be taken against them by Buri and had consequently started negotiations with Baldwin ii; indeed, Ibn al-Athīr states that Buri decided to remove al-Mazdaqānī precisely because the vizier had been in contact with the Franks. Baldwin evidently secured some sort of agreement with Ismāʿīl before acting and took control of Banyas in Dhū l-Qaʿda 523/October–November 1129.12 His force of vassals and crusaders marched toward Damascus, but Buri’s troops overran Frankish detachments that had been sent south towards the Ḥawrān to forage for supplies. On hearing of this defeat, the king ordered a general withdrawal on December 6.13

10 11 12 13

Ibn al-Qalānisī, Damascus chronicle 183–187, 191–195; Ibn al-Athīr, Chronicle i, 277–278; Christie, Ibn al-Qalanisī. Riley-Smith, First Crusaders 244–245. Ibn al-Qalānisī, Damascus chronicle 195; Ibn al-Athīr, Chronicle i, 277–279; Sibṭ b. al-Jawzī, Extraits du Mirât ez-Zemân 567–568; Mouton, Damas 133–134. Murray, Baldwin of Bourcq 197–199.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

525

banyas in the kingdom of jerusalem

2

The First Frankish Lordship (1129–1132)

King Baldwin ii’s crusade failed to reach Damascus, but the acquisition of Banyas gave the Franks a forward base that could be used against the emirate in the future. William of Tyre usually refers to Banyas as a city (urbs or civitas), but this designation is probably a reflection of its status as a bishopric in his own time rather than an indication of a large population.14 Although the area enclosed by the walls amounted to an extensive 50 hectares, this was a consequence of the fortifications being sited to make optimal use of the topography. The narrow northern side of the town was overlooked by crags above the former sanctuary of Pan, while the western and southern walls were constructed along the Banyas River and one of its tributaries. The eastern wall, which covered the gap between the southern watercourse and the crags, was further protected by a fosse in front of it.15 Many of the surrounding areas, especially the slopes of Mount Hermon, were covered in forests, which offered hunting for deer and wildfowl. There was also ample grazing for domestic animals, while the wellwatered plains to the south produced cereals and citrus fruit.16 The king bestowed Banyas on Rainer Brus, a knight—probably a Norman— who had been a trusted associate of his since 1125. The town was later recorded as being a rear-fief belonging to the lordship of Beirut, and it is likely that this was the original arrangement for the new lordship, as argued by Rheinheimer.17 Even with its surrounding countryside, Banyas was probably regarded as being too small and isolated to form a lordship dependent on the crown, and constituting it as a rear-fief meant that Rainer’s own lord would be obliged to come to its defense in case of an attack. These arrangements, however, raise the question why Banyas was attached to the lordship of Beirut, some 80 kilometers to the northwest, rather than Tiberias, 50 kilometers to the south. Most Frankish acquisitions in the Ḥawrān had been incorporated into the lordship of Tiberias, held by William of Bures, a nobleman from the Ile-de-France. Baldwin ii had personally appointed him as lord shortly after his accession, and William remained one of the king’s most trusted supporters; he acted as regent during the king’s captivity in 1123–1124 and was one of

14 15 16 17

Guillaume de Tyr, Chronique 684–686, 876. Ibn al-Qalānisī, Damascus chronicle 216–217; Prawer, Latin kingdom of Jerusalem 284. For the topography and fortifications, see Graboïs, La cité de Baniyas 48–55. Usāma B. Munqidh, Book of contemplation 76–77, 203–204; Le Strange, Palestine under the Moslems 418–419. Guillaume de Tyr, Chronique 656–657; Mayer, Angevins versus Normans 12–13; Rheinheimer, Das Kreuzfahrerfürstentum Galiläa 186–187.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

526

murray

two envoys sent to France to negotiate a marriage for Baldwin’s eldest daughter and heir, Melisende. However, William had been in charge of the foraging force that had been overrun by Buri’s troops during the crusade of 1129. Beirut was held by Walter Brisebarre, another trusted supporter of Baldwin ii, whom the king had appointed to the lordship in 1125.18 By making Banyas dependent on the lordship of Beirut rather than Tiberias, Baldwin ii may have been punishing William for the debacle that led directly to the Frankish withdrawal. It is also possible that Fulk, count of Anjou, who had arrived in Outremer to marry Melisende, had a hand in the appointment, as argued by Mayer.19 Further evidence for an administrative connection between Banyas and Beirut can be extrapolated from a story told by Ibn al-Qalānisī. He relates how Shams al-Mulūk Ismāʿīl, who had succeeded his father Buri as atabeg on Rajab 21, 526/June 7, 1132, received a complaint from some Damascene merchants that the lord of Beirut had seized many valuable bales of linen from them. These goods were presumably being sent to Beirut for export, and there was no justification for their confiscation in peacetime. Ismāʿīl tried to gain restitution for the merchants, but when a prolonged diplomatic correspondence failed to produce a settlement, he marched to Banyas. He arrived on Ṣafar 1, 527 (December 12, 1132) with his ʿaskar of cavalry, the infantry aḥdāth from Damascus, and sappers from Khurasan, who were famed for their expertise in siege warfare. Ismāʿīl could claim to be pursuing his grievance with the lord of Beirut, who was the superior of the lord of Banyas, but the elaborate preparations for the siege suggest that he wanted to recover the defensive position lost three years before, and may have been looking for a convenient pretext to break the truce then in force.20 At this time, Rainer Brus was absent from his lordship, having gone to support King Fulk besiege the port of Jaffa. A constitutional crisis had arisen when Hugh, count of Jaffa, a cousin of Queen Melisende, led a rebellion because he feared a diminution of her rights by her consort Fulk. Ismāʿīl’s attack was evidently timed to exploit a situation where most of the king’s vassals had gone to southern Palestine and would be unable to respond quickly to an attack in the north of the kingdom.21 The Khurasani sappers engineered a collapse of the wall, allowing the Damascenes to rush through the breach and open the

18 19 20 21

Murray, Baldwin of Bourcq 114–116, 178–181. Mayer, Crusader principality 12–13. On Walter Brisebarre, see Mayer, Wheel of fortune 860–877. Ibn al-Qalānisī, Damascus chronicle 215–216; Purton, History 53, 124. Guillaume de Tyr, Chronique 654. On Hugh of Jaffa and the constitutional crisis, see Murray, Baldwin ii and his nobles 79–85.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

527

banyas in the kingdom of jerusalem

gates. Most of the Franks then retreated into the citadel, but realizing there was no prospect of relief, they surrendered and were taken off to captivity in Damascus. Ibn al-Qalānisī reports that the Franks “were greatly astonished that Banyās should have been taken with such ease and in so short a time in spite of the strength of its fortifications and the numbers of its defenders.”22 Two years later the Frankish captives were released when King Fulk agreed on a new truce with Ismāʿīl of Damascus. They included the unnamed wife of Rainer Brus, but William of Tyre claims that her husband later discovered that during her captivity she had not preserved the sanctity of marriage as was expected of a noble woman. Rainer repudiated her and she was obliged to take vows in a convent in Jerusalem. The story is puzzling. A period spent in foreign captivity made it difficult for any woman to maintain her reputation for chastity, but in that case one wonders why Rainer initially took his wife back only to cast her off later. The chronicler goes on to relate that after her death, Rainer married Agnes, a niece of William of Bures, lord of Tiberias. As argued earlier, William may have expected that Banyas would be attached to the lordship of Tiberias when it was taken from the Nizārīs, but he can only have been disappointed when it was constituted as a rear-fief of the lordship of Beirut. A possible explanation for this strange tale is that Rainer’s accusation against his wife was a pretext that paved the way to a second marriage, which he contracted to defuse any hostility between himself and the lord of Tiberias.23

3

Second Frankish Lordship (1140–1164)

Within a decade the Franks gained an opportunity to recover Banyas as a consequence of the rise of ʿImād al-Dīn Zangī, the atabeg of Mosul, who had taken control of Aleppo in 1128 and subsequently campaigned against both Franks and other Muslim polities in Syria. In Damascus, the amīr Ismāʿīl was becoming increasingly erratic and despotic in his governance. On Rabīʿ ii 14, 529/February 1, 1135 he was murdered with the connivance of his mother, Ṣafwat al-Mulk Zumurrud, and replaced by his brother, Shihāb al-Dīn Maḥmūd. Surprisingly, Zumurrud subsequently agreed to a marriage with Zangī, presumably hoping to preserve the inheritance of her family through this alliance. However, 22 23

Ibn al-Qalānisī, Damascus chronicle 216–218; Sibṭ b. al-Jawzī, Extraits du Mirât ez-Zemân 569. Guillaume de Tyr, Chronique 656–657.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

528

murray

Zangī had already used marriages as a way of lending legitimacy to his conquests, and there was no guarantee that he would not use this one as a means to gain power in Damascus. In Shawwāl 533/June 1139 Maḥmūd was murdered by some of his slaves. The authorities installed yet another son of Buri, Jamāl al-Dīn Muḥammad, but real power was exercised by Muʿin al-Dīn Unur, who was appointed as atabeg for the nominal ruler.24 In 1139 the Damascene city of Baalbek surrendered to Zangī, who then demanded that Muḥammad give up Damascus to him in exchange for compensation elsewhere. When his demands were refused, Zangī advanced his forces to Darayya at the end of the year and imposed a blockade on Damascus. When Muḥammad died of a lingering disease on Shaʿbān 8, 534/March 29, 1140, Unur and the Damascene authorities quickly installed his young son Ābaq as ruler. They then approached King Fulk to seek military assistance against Zangī, offering a large monthly subsidy and the return of Banyas in exchange. The governor of Banyas, Ibrāhīm b. Turghut, had previously declared his allegiance to Zangī, so the concession of the town to Fulk would not represent a major loss for the Damascenes if it helped them to ward off the threat.25 Fulk’s vassals responded enthusiastically to the proposal from Unur. The promised subsidy of 20,000 dīnār per month was a significant incentive in itself, but as William of Tyre observed, “the most powerful condition, and the one which made everyone favourable to the agreement, was its final clause, concerning the city of Banyas.”26 When the Franks marched toward Damascus, Zangī lifted the siege. The Franks and their Damascene allies now proceeded toward Banyas, arriving in May 1140. Fulk had been able to enlist the support of Raymond ii, count of Tripoli, and Raymond of Poitiers, prince of Antioch. On his journey south the latter encountered a force under Ibrāhīm b. Turghut, who had gone to raid the territory of Tyre. The Muslims were routed and Ibrāhīm was killed. Deprived of their governor, the garrison of Banyas resisted as best they could, but Unur had great seasoned beams brought from Damascus to construct a massive siege tower from which the besiegers’ missile weapons could reach most of the interior of the town. With supplies running low and no sign of relief from Zangī, the garrison agreed to Unur’s surrender terms. The Muslim inhabitants were allowed to leave with their moveable goods for destinations

24 25 26

Ibn al-Qalānisī, Damascus chronicle 228–252; El-Azhari, Zengi 61–82. Ibn al-Qalānisī, Damascus chronicle 245, 256–262; El-Azhari, Saljuqs 242–252; El-Azhari, Zengi 82–85. Guillaume de Tyr, Chronique 684: Preterea et addite conditiones causam reddunt amplius favorabilem idque potissimum omnes trahit in hanc partem procliviores, quod de urbe Paneadense in fine verborum et adiunctum.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

banyas in the kingdom of jerusalem

529

of their choosing. In keeping with his agreement, Unur handed the town over to Fulk, who restored it to its former lord, Rainer Brus.27 With Banyas restored to Frankish rule, Fulk and his prelates established a new Latin bishopric there, under the jurisdiction of the archiepiscopal see of Tyre. The first bishop was Adam, formerly archdeacon of Tyre. It is likely that the new ecclesiastical foundation had been planned before the capture of the town, since the Frankish forces were accompanied by Fulcher, archbishop of Tyre (later patriarch of Jerusalem) and the papal legate Alberic, cardinal bishop of Ostia. Banyas appears in several twelfth-century pilgrim guides, but it did not attract pilgrims in significant numbers even after it came under Frankish control; it was simply too distant from the main pilgrim routes, which led south from Acre and Tyre, the main ports of arrival, along the coast or through Samaria to Jerusalem. However, as William of Tyre was aware, it was famous as the place where Christ conferred the keys of heaven and the power of binding and loosing on Simon Peter, a passage which was used to justify the papacy’s claim to Petrine supremacy (Mt 16:19). The Latin church hierarchy generally believed that locations with significant biblical associations warranted episcopal status, even where no bishoprics had previously existed under the Greek Orthodox Church, so this in itself was reason enough for the new foundation. However, the enhanced provision of spiritual care may also have been a consequence of the need to attract settlers to the now depopulated town, especially given that Banyas was situated so far from any other major Frankish settlement.28 The late Ronnie Ellenblum persuasively argued that the Latin principalities of Outremer did not have fixed external borderlines. While the boundary between the kingdom of Jerusalem and the county of Tripoli to its north was agreed between both parties, the frontiers of the kingdom with neighboring Muslim powers were not clearly defined, and fluctuated according to the degrees of control and influence that it exercised.29 Nevertheless, at a local level—particularly during times of truce—it must have been necessary to agree to demarcations for the practical purposes of determining agricultural use, grazing rights, and revenues. It is possible to glimpse such arrangements in an anecdote recorded by Usāma B. Munqidh, which can be dated to the period between the Frankish recovery of Banyas in June 1140 and the death of King

27 28 29

Ibn al-Qalānisī, Damascus chronicle 259–262; Ibn al-ʿAdīm, Extraits 682; Guillaume de Tyr, Chronique 685–688. Guillaume de Tyr, Chronique 688–690; Mayer (ed.), Die Urkunden 320, 452 (nos. 138, 244); Wilkinson et al., Jerusalem pilgrimage 70, 112, 152. Ellenblum, Were there borders.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

530

murray

Fulk on November 10, 1143. Usāma relates that during a time of truce, the lord of Banyas had seized flocks of sheep belonging to Muslims from the emirate of Damascus. They had the right to graze sheep in the woods around Banyas and were able to recover their flocks, which suggests that the Frankish lord recognized that he had been in the wrong. However, it was the lambing season and many newly born animals had died, but the Frankish lord refused to pay compensation. Usāma, who was in the service of Unur in Damascus, was able to gain an audience with King Fulk and put the grievances of the pastoralists to him. Fulk asked several knights who were in attendance to give a judgement on the matter. They decided that the lord of Banyas should pay for the lambs that had been lost, and Usāma was able to negotiate compensation of 400 dīnār.30 Bishop Adam was present along with most of the Jerusalem episcopate at the Council of Palmarea near Acre in 1148, when King Baldwin iii and the leaders of the Second Crusade took the decision to launch an attack against Damascus, a campaign that ended in abject failure. Adam remained in office until at least 1160. By contrast, Rainer Brus evidently died (or possibly had been removed from his lordship) before 1153.31 His successor in Banyas was Humphrey ii, lord of Toron and constable of the kingdom of Jerusalem. William of Tyre refers to the lordship as Humphrey’s “hereditary possession” (urbem Paneadem, que eius erat hereditaria possessio), which would imply that he had married the heiress of Rainer Brus, although we have no independent evidence for such a connection. Humphrey held a lordship in Upper Galilee, centered on the castles of Toron (Tibnīn) and Chastel Neuf (Qalʾat Hunīn) between Tyre and the Jordan. However, his office of constable meant that he could not always devote his attention to the defense of Banyas. In 1153 Baldwin iii besieged Ascalon, the last Fatimid enclave in Palestine, accompanied by Humphrey and most of the kingdom’s forces. Nūr al-Dīn, who had succeeded his father Zangī in 1146, tried to exploit the constable’s absence by attacking Banyas on Ṣafar 19, 548 (May 16, 1153), but Ibn al-Qalānisī implies that strife broke out among his forces, causing them to retire in disorder.32 Within four years the defense of Banyas had become too onerous for Humphrey of Toron. With the permission of the king, he granted half of the town and its surrounding countryside to the Order of the Hospital of St. John of Jerusalem. Humphrey and the Hospitallers were each to be responsible for the 30 31 32

Usāma B. Munqidh, Book of contemplation 76–77. Guillaume de Tyr, Chronique 760–761; Mayer (ed.), Die Urkunden 474 (no. 258). Ibn al-Qalānisī, Damascus chronicle 315–16; Guillaume de Tyr, Chronique 796–797; Kennedy, Crusader castles 40–43.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

banyas in the kingdom of jerusalem

531

lordship and pay half of the costs of its defense. The last formulation is telling, as it implies that the costs of holding Banyas outweighed the revenue that the lordship yielded in rents, tolls, and the profits of justice; it would seem that Humphrey had been using his income from Toron to subsidize his tenure of Banyas.33 The hospital, by contrast, was by this time increasingly receiving bequests and eleemosynary grants from pilgrims in Palestine as well as benefactors in western Europe. It was still at an early stage in a process of militarization, which would see its transformation from a purely charitable organization into a military-religious order along the lines of the Templars, but even at this time its officers would have regarded the defense of the Holy Land as a suitable purpose for such funds.34 The Hospitallers’ attempt to take possession of their new property reveals some of the problems involved in defending and supplying a frontier lordship far inland, which was effectively cut off from the main contiguous Frankish territory. William of Tyre observed that the town “was situated close to enemy territory, so that no one could enter or leave it without danger unless they traveled with a strong company or went by secret routes.”35 In the spring of 1157 the Hospitallers assembled a convoy of packhorses and camels loaded with provisions, weapons, and money. They were escorted by a number of Templars, secular knights, mounted sergeants, and foot soldiers, some of whom were evidently intended to remain as a garrison. Nūr al-Dīn, who had seized control of Damascus in the meantime, was determined to oppose this reinforcement. As the column approached Banyas on Rabīʿ i 13, 552/April 25, 1157 it was attacked by a detachment of his troops, who slaughtered most of the Franks and made off with the pack animals and supplies. Nūr al-Dīn then decided to attack the town in the expectation that both provisions and morale would be low. From Damascus he brought his ʿaskar, the aḥdath, many jihādī volunteers, and considerable siege equipment. They broke through the outer walls and the defenders retreated into the citadel. Humphrey of Toron offered to surrender on terms, but Nūr al-Dīn refused, scenting a complete victory. However, the defenders were able to hold out until mid-June, when they were relieved by a force under King Baldwin iii. Nūr al-Dīn withdrew, and now confident that the danger had 33

34 35

Guillaume de Tyr, Chronique 826–827: de consensu domini regis cum fratribus Hospitales ex equo dividit, ita ut totius civitatis et suburbanorum omnium fratres predicti habentes dimidium, impensas tam utiles quam necessarias ex dimidio ministrarent, civitatis pro parte altera debitam gerentes sollicitudinem. Riley-Smith, Knights Hospitaller 27–37. Guillaume de Tyr, Chronique 826–827: Erat autem civitas in confinio hostium posita eisque valde contermina, ita ut nulli nisi in manu forti aut clandestino nimis itinere accedenti accessus vel discessus inde posset esse nist cum periculo.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

532

murray

passed, the king omitted to reconnoitre his route home. As they crossed the Jordan at Jacob’s Ford the Franks were ambushed by Nūr al-Dīn’s troops and suffered a devastating defeat. Baldwin fled to the castle of Ṣafad some 12 kilometers to the west, but while few Franks were killed, a far greater number were taken prisoner. They included the royal marshal Odo of Saint-Amand, the master of the temple, and the noblemen Hugh of Ibelin, John Gotmann, and Rohard of Jaffa. After the destruction of the Hospitallers’ convoy, the defense of Banyas had proved doubly costly.36 William of Tyre claims that Baldwin iii summoned masons and laborers to carry out repairs before leaving the town, whereas Ibn al-Qalānisī reports that the walls and dwellings were so ruined that the inhabitants despaired of the town being rebuilt to a defensible state. The short period of time that the king stayed in Banyas makes it doubtful whether anything but the most basic and necessary repairs could have been carried out, and it seems more likely that the rebuilding mentioned by William took place much later. Nūr al-Dīn marched back from Jacob’s Ford to attack Banyas again; its inhabitants immediately retreated into the citadel, suggesting that they did not trust that the outer defenses would hold for long. In the meantime, Baldwin iii had begun to assemble fresh forces and summon assistance from Tripoli and Antioch. When he realized the numbers of troops that the king was likely to bring, Nūr al-Dīn abandoned the siege.37 William relates that the Hospitallers then tried to withdraw from their agreement and return their share of Banyas to its lord. However, a royal charter issued at Acre on October 4, 1157 indicates that they were prevailed upon to rethink their decision. This document was a confirmation by Baldwin iii, his mother Melisende, and his brother Amalric, count of Ascalon, of donations to the Hospital of Jerusalem by Humphrey ii of Toron with the consent of his son Humphrey (iii) and his daughters. The first donation comprised half of the castle (and by implication, the lordship) of Banyas, which was precisely what Humphrey had already granted earlier that year. However, the document details an additional donation, comprising half the castle of Chastel Neuf, together with houses and a hospice in Toron. There was a key distinction between these two donations. The grant of half of Banyas was made with the permission of Walter ii Brisebarre, lord of Beirut, on whose fief the lordship depended (concessione Gualterii Berythensis, de cuius feodo movet). The half of Chastel Neuf and the smaller properties, by contrast, came from Humphrey’s

36 37

Guillaume de Tyr, Chronique 826–832; Ibn al-Qalānisī, Damascus chronicle 331–337. Guillaume de Tyr, Chronique 832–833. Ibn al-Qalānisī says nothing about this second siege.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

banyas in the kingdom of jerusalem

533

own fief of Toron, which was held directly from the crown. The evidence of the charter suggests that, after the events of the intervening summer, Humphrey was obliged to offer additional inducements from his lordship of Toron in order to persuade the Hospitallers to take on shared responsibility for the defense of Banyas.38 The new Frankish condominium in Banyas lasted until the reign of Baldwin iii’s brother and successor, Amalric (1163–1174), whose main policy was the conquest of Egypt, which he attempted on five occasions.39 As constable, Humphrey ii of Toron was obliged to serve on these expeditions, and in 1164 he left Banyas under the command of one of his knights named Walter of Quesnoy (Galterius de Quaisneto). Nūr al-Dīn seized the opportunity to besiege the town. William of Tyre claims that Walter was negligent, hinting that he was bribed to surrender, but it is more likely that defensive capabilities and morale were at a low ebb. Both the lord and the bishop of Banyas were absent, and the population had still not recovered from the losses sustained in 1157. Within a few days the walls and towers had been severely damaged by projectiles and undermining by Nūr al-Dīn’s troops. On October 18, 1164 the inhabitants agreed to surrender on condition of being allowed to depart with their moveable goods.40 The capitulation ended the Frankish possession of Banyas. Since 1129 the frontier town had changed hands four times in a period of only 35 years. It was the most precarious lordship in the entire kingdom of Jerusalem, for its exposed location and proximity to Damascus ensured that it would be a regular target of Muslim attacks, particularly if the kingdom’s main forces were occupied far away, as occurred in 1132, 1153, and 1164. It was probably only the rivalry between the Zangids of Aleppo and the Burids of Damascus that enabled the Franks to retain possession for so long.

4

Epilogue: The lost Lordship

The Franks continued to hope that Banyas might be recovered. In 1169 the patriarch of Jerusalem, Amalric of Nesle, sent a general letter appealing for assistance from the clergy and laity of the west. He listed numerous reverses suffered

38 39 40

Mayer (ed.), Die Urkunden 450–452 (no. 244). Murray, Place of Egypt. Guillaume de Tyr, Chronique 877. William wrongly dates the event to the calendar year 1167, but giving the correct second regnal year of Amalric, which began on February 18, 1164. The placename Quesnoy and variants, which derive from the French dialect term quesne (oak), occur several times in Normandy and Picardy.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

534

murray

by the Franks over the previous two decades, ending with the loss of Banyas, which he described as “the key and gate and protection of the entire land of Jerusalem” (clavis et porta et defensaculum tocius terre Jherosolimitane).41 In a separate letter to King Louis vii of France sent in the same year, the patriarch stressed the holiness and strategic importance of Banyas, on the site of Caesarea Philippi, which had recently been taken by the Turks. To give weight to his appeal, the letter was entrusted to one John, bishop of Banyas, who had been in office at least since 1165, although he was unable to reside in his diocese.42 Although King Amalric had been obliged to give up his plans for conquests in Egypt, he was alert to opportunities for territorial gains on the frontiers of his kingdom. When he heard of the death of Nūr al-Dīn in May 1174 he quickly “assembled the entire strength of the kingdom” (universas convocans regni vires) and marched to besiege Banyas. Nūr al-Dīn’s widow asked for a truce, offering a large sum if Amalric abandoned the action. The king continued the siege for two weeks without making progress and eventually agreed to desist in exchange for the money offered and the release of 20 Frankish captives.43 This was Amalric’s last offensive action; on his way back to Jerusalem he fell ill with dysentery and died on July 11, 1174. Once Saladin seized control of most of Nur al-Dīn’s domains, Amalric’s son and successor Baldwin iv was forced into a purely defensive strategy to counter the twin threats from Egypt and Syria. In 1179 the king was overseeing the construction of a new fortification at Jacob’s Ford when he heard that Muslims had come with their herds into the forest near Banyas in search of pasturage. Baldwin made a swift night march north in the hope of falling on an easy target, but in the morning his company was surprised by the enemy. Humphrey ii of Toron was wounded while trying to protect the king and carried back to Jacob’s Ford where he died on April 11, 1179.44 Humphrey ii’s grandson and successor Humphrey iv cherished little hope of recovering the lost lordship and surrendered his rights in it along with Toron and Chastel Neuf to Baldwin iv in exchange for a money-fief amounting to 7,000 bezants per year.45 On October 17, 1186 the throne passed to Guy of Lusignan, husband of Baldwin iv’s sister Sibyl, despite the opposition of a con-

41 42 43 44 45

Delaville Le Roulx (ed.), Cartulaire générale i, 279–280 (no. 404). Röhricht (ed.), Regesta Regni Hierosolymitani no. 463; Mayer (ed.), Die Urkunden 541, 550 (nos. 310, 314). Guillaume de Tyr, Chronique 956–997. Guillaume de Tyr, Chronique 998–1,000. Guillaume de Tyr, Chronique 1,012.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

535

banyas in the kingdom of jerusalem

siderable section of the nobility. Within a week, on October 21, Guy granted Humphrey’s former fiefs to the seneschal Joscelin iii of Courtenay, titular count of Edessa; the text of the grant stated hopefully that Joscelin would receive Banyas “when the Lord in his mercy returned it to Christendom” (cum dominus sua miseratione christianitati reddiderit).46 Joscelin had built up a large agglomeration of fiefs, rights, and revenues in northern Galilee, known collectively as the seigneurie de Joscelin. He was thus the king’s most powerful ally in the north of the kingdom, who might be in a strong position to defend Banyas if the town were recovered.47 Yet this proved to be a vain hope. Less than a year later Saladin defeated King Guy and his allies at the Horns of Haṭṭīn on July 4, 1187 and went on to overrun almost all of the kingdom of Jerusalem. From this point there was little prospect of the recovery of Banyas, but over 60 years later the crusaders were still aware of its strategic significance. After the defeat of his crusade against Egypt in April–May 1250, King Louis ix of France sailed with his depleted forces to Acre and occupied himself in shoring up the defenses of the remaining Frankish strongholds in Palestine. While he was at Tyre the king sent a force of crusaders, Frankish barons, and knights of the military orders to Banyas, then held by the Ayyubids of Damascus. They stormed the outer defenses but made no attempt to hold the city, as they were unable to take the castle at Subeibe, built by the Ayyubids on a high ridge northeast of the town in 1228–1230 to defend the approaches to Damascus. Overlooked by this new Muslim stronghold and far from the nearest friendly territory, Banyas was untenable by the Franks.48

Bibliography Sources Delaville Le Roulx, J. (ed.), Cartulaire générale de l’ordre des Hospitaliers de Saint-Jean de Jérusalem, 1100–1310, 4 vols., Paris 1894–1906. Fulcher of Chartres, Fulcheri Carnotensis Historia Hierosolymitana (1095–1127), ed. H. Hagenmeyer, Heidelberg 1913. Guillaume de Tyr, Chronique (Corpus Christianorum Continuatio Mediaevalis 63–63A), ed. R.B.C. Huygens, 2 vols., Turnhout 1986. Ibn al-ʿAdīm, Extraits de la chronique d’Alep par Kemal ed-Din, in Recueil des historiens des Croisades: Historiens orientaux, 5 vols., Paris 1872–1906, iii, 577–732. 46 47 48

Mayer (ed.), Die Urkunden 796–799 (no. 473); Murray, Women 147–51. Mayer, Die Seigneurie de Joscelin 171–181. Joinville, Histoire de Saint Louis 239–44; Kennedy, Crusader castles 184.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

536

murray

Ibn al-Athīr, The chronicle of Ibn al-Athīr for the crusading period from al-Kāmil fī ltaʾrīkh, trans. D.S. Richards, 3 vols., Aldershot 2006–2008. Ibn al-Qalānisī, The Damascus chronicle of the crusades extracted and translated from the chronicle of Ibn al-Qalānisī, trans. H.A.R. Gibb, London 1932. Joinville, Histoire de Saint Louis, ed. N. de Wailly, Paris 1881. Mayer, H.E. (ed.), Die Urkunden der lateinischen Könige von Jerusalem: Diplomata regum Hierosolymitanorum, 4 vols., Hannover 2010. Röhricht, R. (ed.), Regesta Regni Hierosolymitani, 2 vols., Innsbruck 1893–1904. Sibṭ b. al-Jawzī, Extraits du Mirât ez-Zemân, in Recueil des historiens des Croisades: Historiens orientaux, 5 vols., Paris 1872–1906, iii, 511–570. Usāma B. Munqidh, The book of contemplation: Islam and the crusades, trans. P.M. Cobb, London 2008.

Studies Berlin, A., The archaeology of ritual: The sanctuary of Pan at Banias/Caesarea Philippi, in basor 315 (1999), 27–45. Christie, N., Ibn al-Qalanisī, in A. Mallett (ed.), Medieval Muslim historians and the Franks in the Levant (The Muslim World in the Age of the Crusades: Studies and Texts 2), Leiden 2014, 7–28. El-Azhari, T., The Saljuqs of Syria during the crusades, 463–549a.h./1070–1154a.d., Berlin 1997. El-Azhari, T., Zengi and the Muslim response to the crusades: The politics of jihad, London 2016. Ellenblum, R., Were there borders and borderlines in the middle ages? The example of the Latin kingdom of Jerusalem, in D. Abulafia and N. Berend (eds.), Medieval frontiers: Concepts and practices, Aldershot 2002, 105–120. Graboïs, A., La cité de Baniyas et le château de Subeibeh pendant les croisades, in Cahiers de civilisation médiévale 13 (1970), 43–62. Kennedy, H., Crusader castles, Cambridge 1994. Le Strange, G., Palestine under the Moslems: A description of Syria and the Holy Land from a.d.650 to 1500 translated from the works of the mediaeval Arab geographers, London 1890. Mayer, H.E., Angevins versus Normans: The new men of King Fulk of Jerusalem, in Proceedings of the American philosophical society 133 (1989), 1–25. Mayer, H.E., The crusader principality of Galilee between Saint-Omer and Bures-surYvette, in Itinéraires d’Orient: Hommages à Claude Cahen, Bures-sur-Yvette 1994, 157–167. Mayer, H.E., Die Seigneurie de Joscelin und der Deutsche Orden, in J. Fleckenstein and M. Hellmann (eds.), Die geistlichen Ritterorden Europas, Sigmaringen 1980, 171– 216.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

banyas in the kingdom of jerusalem

537

Mayer, H.E., The wheel of fortune: Seignorial vicissitudes under Kings Fulk and Baldwin iii of Jerusalem, in Speculum 65 (1990), 860–877. Mouton, J.-M., Damas et sa principauté sous les Saljoukides et les Bourides (468–549/1076 –1154): Vie politique et religieuse, Cairo 1994. Murray, A.V., Baldwin of Bourcq: Count of Edessa and king of Jerusalem, 1100–1131, Abingdon 2022. Murray, A.V., Baldwin ii and his nobles: Baronial factionalism and dissent in the kingdom of Jerusalem, 1118–1134, in Nottingham medieval studies 38 (1994), 60–85. Murray, A.V., The place of Egypt in the military strategy of the crusades, 1099–1221, in E.J. Mylod et al. (eds.), The Fifth Crusade in context: The crusading movement in the early thirteenth century, Abingdon 2016, 117–134. Murray, A.V., Women in the royal succession of the Latin kingdom of Jerusalem (1099– 1291), in C. Zey (ed.), Mächtige Frauen? Königinnen und Fürstinnen im europäischen Mittelalter (11.–14. Jahrhundert), Ostfildern 2015, 131–162. Prawer, J., The Latin kingdom of Jerusalem: European colonialism in the middle ages, London 1972. Purton, P., A history of the early medieval siege c. 450–1220, Woodbridge 2009. Rheinheimer, M., Das Kreuzfahrerfürstentum Galiläa, vol. 1, Frankfurt am Main 1990. Riley-Smith, J., The first Crusaders, 1095–1131, Cambridge 1997. Riley-Smith, J., The Knights Hospitaller in the Levant, c. 1070–1309, Basingstoke 2012. Seetzen, U.J., Unter Mönchen und Beduinen: Reisen in Palästina und angrenzenden Gebieten, ed. A. Lichtenberger, Stuttgart 2002. Wilkinson, J., with J. Hill and W.F. Ryan (eds.), Jerusalem pilgrimage 1099–1185, London 1988.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 27

Fortresses and Frontiers: Castles and Northern Syria in the Sultanate of Cairo Angus D. Stewart

My first encounter with Hugh Kennedy was as a first-year undergraduate student taking a survey course on medieval history. His lectures were among the highlights of the program: informative, clearly delivered, revealing a great breadth of knowledge and interest, and always well illustrated. This was, of course, in the days of 35mm film slides; on occasion, a slide might be loaded incorrectly into the carousel. Hugh’s response to such moments was especially appreciated by the class: he would climb onto the lowest of the—steeply raked—wooden desks and use them as an impromptu staircase to get to the projector at the top of the theater. Hugh’s lectures were serious, and he encouraged an interest not merely in the details of events but also in the importance of architecture and material culture—but he also demonstrated a sense of fun, and the importance of enthusiasm. Among the slides I saw that year were images that were later published in Crusader castles (1994). In this essay I would like to consider how Kennedy, and other scholars, have discussed the roles of castles. Specifically, I will consider how castles placed on political frontiers can be seen as distinct from those elsewhere, with their own special purpose as part of some program of “border defense.” After looking at how this debate relates to the topic of “crusader” castles—the castles built by the Frankish occupiers of the Levantine coast—I will turn to look at a different case: the fortresses of the Mamluk sultans of Cairo in Syria, and how these might relate to their northern frontier. By looking in detail at one example, Qalʿat al-Rūm, across the period of the sultanate, I will consider what the roles of such a fortress and its commander, or nāʾib, may have been.

1

Crusader castles: Context and Purpose

While in some ways a departure from his normal research areas, Crusader castles was based not merely on Hugh’s considerable enthusiasm for the topic but also on serious and extensive research. The historiography of the subject is clearly presented—to an extent pointing in the direction of the late Ronnie

© Angus D. Stewart, 2022 | doi:10.1163/9789004525245_028

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

539

Ellenblum’s Crusader castles and modern histories (2007). While the structure of each of the main chapters is built on discussions of specific sites, this is in no sense a gazetteer. These surveys of individual castles are useful in themselves, but they also contribute to the developing understanding of the whole topic. Hugh’s great breadth of interest is again demonstrated—rather than confining himself to the kingdom of Jerusalem, or the most impressive sites, he discusses a wide variety of sites from across the whole of Frankish Outremer (and beyond), based on personal experience as well as familiarity with the scholarship. His judgements have aged well—he anticipates, for example, the great revelations of the excavation at Jacob’s Ford/Le Chastellet, just beginning at the time of writing.1 He very ably combines architectural surveys, the literary record, and the results of archaeological investigations, incorporating different perspectives seamlessly.2 The castles are not viewed in isolation: the wider context, in western Europe but also, importantly, across the Middle East, is discussed. The practice of fortification building in both west and east is presented at the outset as a necessary background for the development of “crusader” castles. The perspective is broad, covering France, England, the Low Countries, and southern Italy on the one hand; and Byzantium, the Islamic world—al-Andalus as well as Syria and Egypt—and Armenian work, on the other.3 The theme of the relationship of crusader castles with those in western Europe is considered again in a “postscript”; even here, comparisons are explicitly made with work in Islamic Syria. Towers at Château-Gaillard, Dover, and Coucy are compared, not merely with near-contemporary work at Crac des Chevaliers and Chastel Pelerin but also at Damascus and Bosra. The conclusion proposed is that, in terms of developments in castle design, “experiment and experience” were the determining factors, “rather than architectural influences from the other end of the Mediterranean.”4 The reader is primed for this integration of work in the wider region by a chapter devoted to “Muslim castles” of the time.5 Study of the

1 Kennedy, Crusader castles 57: “the site is shortly to be excavated, which should shed light on this very precisely dated structure.” See Ellenblum, Crusader castles 258–274; and the website of the excavation project, http://vadumiacob.huji.ac.il/. 2 This is a point worth stressing, and certainly shows sympathy with modern castellogical studies; there is no sense of the artificial hierarchies of scholarship asserted by, for example, Allen Brown, Historian’s approach. 3 Kennedy, Crusader castles 11–20. On Armenian fortifications, the foundational work of Edwards, Fortifications, is extensively cited; but see also Pringle, Crusader castles. 4 Kennedy, Crusader castles 186–189. For a complementary view, see France, Fortifications east and west. 5 Kennedy, Crusader castles 180–185.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

540

stewart

fortresses of the central Islamic lands in this period has developed considerably in the last quarter century, and Hugh himself encouraged these developments. For example, in 2003 he organized a conference in Aleppo, held in the restored fifteenth-century Throne Hall of the citadel gatehouse, which gathered together many of the scholars working to expand our understanding of sites across Bilād al-Shām. Some of the papers presented there were collected in Muslim military architecture in greater Syria: From the coming of Islam to the Ottoman period (2006). Crusader castles incorporates subtle readings from contemporary written sources, both Arabic and western European. These are used to provide insight into not only the appearance of castles but also how they were understood and experienced. For example, William of Tyre’s account of the bombardment of Kerak in 579/1183 is used to illustrate the significance of “psychological warfare” as well as military tactics.6 Just as Hugh uses discussions of individual sites in order to build more general arguments, in examples such as these he raises serious questions about the rationale for castle building and the roles castles could play. The discussion of Blanchegarde is especially interesting. Hugh ignores William of Tyre’s explanation for the castle-building program around Ascalon as being for defense against raids from that city’s Egyptian garrison; rather, he quotes the parts of William’s chronicle that describe how the castle became an economic focus for its region. Here he follows the line of argument of Raymond Smail—that these castles represent a strategy of conquest rather than defence—but contributes some original subtle reading of the key text.7 This is by no means an isolated example. William of Tyre again is used to present Toron as the “administrative centre of a flourishing district.”8 Albert of Aachen’s chronicle is used to support an account of King Baldwin i of Jerusalem’s “ambitious expansionist policy” into Oultrejourdain being intended to control the commerce passing through the region.9 After considering Fulcher of Chartres’s description of Baldwin ii building a castle near Beirut in a fertile

6 Kennedy, Crusader castles 50–52. This account also relates to his interpretation of the vaulted halls found at Kerak as providing, among other things, “shelter from in-coming missiles”; cf., for example, the discussion of Margat/Marqab, 179. 7 Kennedy, Crusader castles 32; Smail, Crusading warfare 211–213; Ellenblum, Frankish rural settlement 15–18, provides an overview of the debate, and extensive references to the text of William of Tyre, of which, see especially William of Tyre, Chronicon xiv:22, ed. Huygens 659– 660. See also Hoch, Crusaders’ strategy. By contrast, Boas and Maeir present the building of the castles around Ascalon, such as Blanchegarde, in simple “defensive” terms, Frankish castle 3–4. 8 Kennedy, Crusader castles 40. 9 Kennedy, Crusader castles 23–25.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

541

region in order to force the local Muslim peasantry to pay taxes, Hugh comments that “a clearer statement of the reasons for castle building would be hard to find.”10 This material is supplemented by the translation of the De Constructione Castri Saphet in an appendix, describing the refoundation of the castle at Ṣafad by the Templars with the assistance of the bishop of Marseille, Benedict of Alignan. This text was almost certainly a variety of fundraising treatise, and as such, the rationale and role presented for the castle—as well, perhaps, as its appearance—are clearly idealized: this is what its audience would expect of a functioning, practical castle. The castle is, indeed, presented in defensive terms: it is “like a shield for the Christians as far as Acre,” a “bulwark and obstacle” placed in the way of potential enemy raiders (“except in very great numbers”); the lands thus protected “could be worked freely.” This is, however, only part of the picture. More attention is given to the offensive potential of the castle, as a “base for attack” for “sallies and raids into the land of the Saracens as far as Damascus”; and as a consequence the land as far as that city had become “like a desert for fear of the castle.” The pamphlet also seeks to justify the castle as a practical endeavor by describing its economic vitality, both as the site of a flourishing agricultural and commercial settlement and also as the center of a prosperous lordship. The conclusion of the text provides a further justification for donations for the knights at Ṣafad, and perhaps the most powerful one in the crusading context. Not only has the building of the castle allowed the Christian faith to be preached freely in the region, and “the blasphemies of Muhammad” to be condemned, but its presence has allegedly facilitated the access of pilgrims to many sites associated with the Gospels. In short, Ṣafad was refounded “to confound, to weaken and to hold back the infidel and to expand, multiply and comfort the faithful.”11 “Defense” is certainly part of the justification for castle building, just as crusades themselves were always presented explicitly in defensive terms, reflecting the Augustinian theoretical context; but practical, expansionist, and aggressive rationales for the castles are equally present in the literary record. These themes are a notable feature of the discussion of Crac des Chevaliers. The lands granted by the count of Tripoli to the Hospitallers were extensive and potentially prosperous, but were exposed to potential attacks from rival princes to the east, lacking “natural defenses.” The previously fortified site— Hiṣn al-Akrād—was chosen for the main Hospitaller base, and subsequently 10 11

Kennedy, Crusader castles 30. Kennedy, Crusader castles 190–198. The edition of the text used is that of Huygens, Un nouveau texte; an alternative English translation has since been made by Barber and Bate, Templars 84–93. On the text and on Ṣafad, see Pringle, Review.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

542

stewart

saw much development. The “defensive role” of this castle was revealed by the experience of several invasions of the county. Saladin’s raid of 575–576/1180 is used as an example: the raiders were left unchallenged, as the Franks retreated inside their fortresses, such as the Hospitallers in Crac; when the coast was clear they could re-emerge to “revive their ravaged lands which could not be protected from being damaged by the enemy but could be protected from being permanently occupied.” Crac was again by-passed by Saladin in 584/1188. The region was “the only significant inland area […] which remained continuously in Frankish hands” from before the battle of Haṭṭīn (583/1187) to the mid-thirteenth century, suggesting that “the value of a great castle could not be more clearly illustrated.”12 One might object that the exceptional nature of this example suggests that this is true for this great castle alone; but in any case, this is not the aspect of Crac that is most stressed. Kennedy turns the argument on its head: Crac’s exposed position was a positive advantage. For one thing, “the neighbouring Muslim settlements could be raided or put under tribute”: the Hospitallers of Crac, at times supported by other Frankish powers, were able to terrorize neighboring lands and to extract tribute from the princes of Homs and Hama, and even from the Nizārī Ismaʿilis of Maṣyāf. For another, Crac’s apparent vulnerability “could also be used to attract pious donations,” from nobles back in western Europe—such as Wladislav ii of Bohemia, a veteran of the Second Crusade—or from those visiting in person—such as Andrew ii of Hungary at the time of the Fifth Crusade. The point is made clearly that the greatness of the castle at Crac reflected not just Hospitaller income from their own lands but also pious donations, and the protection money paid by neighboring Muslims. Of course, not even a fortress as strong as Crac could hold out for ever, especially after being isolated by the conquest of nearby castles, and without the possibility of external relief—and against a determined besieger, as was Sultan al-Ẓāhir Baybars (658–676/1260–1277) in 669/1271.13

2

The Strategy of Crusader Castles: From Rey to Ellenblum

The preceding discussion relates to debates in the wider field of castellology. Much ink continues to be spilled, for example, on arguments relating to the purpose of castles and castle building in medieval England. While some would see

12 13

Kennedy, Crusader castles 146–147. Kennedy, Crusader castles 150.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

543

castles primarily in terms of military strategy and “defense,” others point to the role of the castle as residence, as regional focus, and as symbol of lordly power.14 Such debates are less a feature of discussions of “crusader castles,” but this is perhaps because a strategic role may be taken for granted. After all, the groundbreaking work of Paul Deschamps, Les Châteaux des Croisés en Terre Sainte, contained volumes subtitled La défense du royaume de Jérusalem (1939) and La défense du Comté de Tripoli et de la Principauté d’Antioche (1977)—the implication is clear.15 This lingering tendency is discussed in detail by Ellenblum in his Crusader castles and modern histories, especially with regard to the idea of “borders and their defence.”16 Ellenblum is in general dubious about the utility of thinking in terms of borders in the modern sense for the twelfth-century kingdom of Jerusalem (his period and area of concern).17 He presents such thinking as relating to what he characterizes as a “colonialist discourse,” often concerned with “natural” borders.18 Castles are fundamental to this discussion. While some fortresses are seen, by scholars such as Emmanuel-Guillaume Rey, Deschamps, and those following them, as an integral part of a frontier defending the Latin states, others less obviously on “borders” represent the existence of staggered “lines of defense” (or defended Outremer’s infrastructure). Ellenblum suggests that the “idea of a line of castles is based on the assumption that the area behind the castle is safe, and that danger is to be expected from a specific direction.”19 Behind the border marked and guarded by the line of fortresses was the enlightened Franco-Syrian colonial society imagined by this scholarship. Even some commentators who lacked sympathy with this perspective seem to echo the analysis—Ellenblum identifies what he terms a

14

15 16 17

18

19

For a particularly extremist account of the strategic role of castles in Anglo-Norman England, see Beeler, Castles and strategy; but also, much more recently, see Prior, A few wellpositioned castles. For a recent outbreak of the conflict over the role of castles, specifically in fourteenth-century England, see Platt, Revisionism; Creighton and Liddiard, Fighting yesterday’s battle. See the discussion in Kennedy, Crusader castles 7. This is the title of chapter 8, Ellenblum, Crusader castles 105–117. See, especially, Ellenblum, Crusader castles 118–145; cf. Ellenblum, Were there borders. This aspect of Ellenblum’s argument has been convincingly critiqued by Pringle, Castles and frontiers 233–239. Problems relating to the study of “frontiers” in medieval societies are discussed by Abulafia, Seven types of ambiguity; see also Standen, Nine case studies. With regard to the connections between the study of “crusader” castles and modern colonialism, it is noteworthy that among the major works of the pioneering scholar in the field, Emmanuel Rey, is his Les colonies franques de Syrie aux xiime et xiiime siècles, Paris 1883—the relationship between the past and Rey’s present is clear. Ellenblum, Crusader castles 134–145.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

544

stewart

“Zionist” approach, followed by scholars such as Joshua Prawer, who interpret the Latin kingdom through the prism of their own experience of contemporary Israel.20 One problem with such an interpretation of Frankish castle building is that it presupposes a consistent strategic overview for the Frankish rulers, and a sense of conscious long-term strategic planning, which does not seem to reflect the reality of the history. This approach was challenged by Smail—described by Ellenblum as presenting an “anti-colonialist” analysis—who saw no evidence of an overarching system to Frankish castle building. Castle building, as in the west, more likely reflected the whims of individual lords; and castles certainly played only a limited role in “border defense.”21 Rather, and as in the west, they were centers of lordship.22 In themselves, fortresses were unable to seal off a border or to prevent enemy incursions; in practice, Smail points out, the garrisons of fortresses on the border of the Latin kingdom might withdraw to join with the kingdom’s main field army at times of general mobilization—that is, assembling away from the border.23 Ellenblum, like Kennedy, is closer in spirit to Smail than to Rey or Deschamps. He points out that for much of the twelfth century, castle building in the Latin kingdom was much more likely to be away from areas especially vulnerable to attack; castles were more likely to be built in the relative security and prosperity of the years around the middle of the century. However, there is some retreat from this argument: in what Ellenblum characterizes as the “third generation,” roughly 564–583/1168–1187, castle building was more likely to be focused on the frontier, in response to what is termed the “Muslim offensive.”24 Having argued forcefully against castles offering “lines of defense,” here we see something close, with the new larger castles of the period interpreted as reflecting a conscious strategy of defense. Furthermore, “this massive investment justified itself after the battle of Haṭṭīn. Belvoir, Ṣafad, Hunin, Karak and Montreal were the last Frankish outposts to hold out despite the sieges mounted against them by Ṣalāḥ al-Dīn. None of them fell immediately to the Muslim forces like the […] castles of the earlier generations, and some of them even survived for another year and a half or more.”25 The poten20 21 22 23 24

25

Ellenblum, Crusader castles 113–117. Smail, Crusading warfare 204–209. Smail, Crusading warfare 214. Smail, Crusading Warfare 208–209; Ellenblum, Crusader Castles 111. Ellenblum, Crusader castles 161–164. On the other hand, Ellenblum presents the building of Jacob’s Ford in 574–575/1179, in the middle of this generation, and a very large castle like the others mentioned, as being an aggressive move into “Muslim territory” (comparable to the refoundation of Ṣafad in the mid-thirteenth century): 143. Ellenblum, Crusader castles 181.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

545

tial objection to this is that the kingdom did actually fall, and so eventually did all these fortresses. If these really were aimed at border defense, then they failed spectacularly, hardly suggesting that the investment in them can be or was seen as “justified”—though it might be said that these sieges at least held up Saladin’s progress. The point to be made here is that, whether or not people in medieval Syria thought in terms of mutually recognized linear political boundaries, the role of fortresses was limited. As Denys Pringle has concluded, “castles or fortifications on their own could not defend frontiers, and were never intended to do so.”26 Nevertheless, the idea of a connection between fortresses and frontier defense persists.

3

Fortresses in the Sultanate of Cairo

The eventual successors of Saladin in Syria and Egypt, the “Mamluk” sultans of Cairo, ended the Frankish presence in the Levantine littoral. From their assumption of power in Syria in 658/1260, periodic campaigns gradually reduced the Frankish lands, capturing the fortresses and cities, until in 690/1291 the new sultan, al-Ashraf Khalīl (689–693/1290–1293), dealt the coup de grâce: Acre itself was taken, and the remaining holdings on the coast abandoned. While coastal sites were often slighted, the sultans maintained occupation of more inland conquests, and this included some of the castles, where damage inflicted in the sieges could be quickly repaired. It can be difficult to disentangle work carried out under Frankish rule from that of their successors—the cover image of Crusader castles, of the southern point of the castle of Margat/Marqab, shows postconquest work more prominently than Hospitaller. Crac des Chevaliers itself has been described by Benjamin Michaudel as being not only “parangon de la fortification croisée” but also “la quintessence de l’architecture militaire mamelouke.”27 The sultans of Cairo were notable fortress builders, and not merely at previously Frankish sites. Much work was carried out at the major urban citadels of the sultanate—especially, of course, at the citadel of Cairo founded by Saladin, but also in the key cities of Syria. In Syria, the destruction inflicted during the Mongol occupation of 658/1260 and in subsequent raids meant much renovation work was necessary, both in the cities and at fortresses in the countryside. The damage left by the Mongols is often described in very general terms in contemporary accounts, and scholars

26 27

Pringle, Castles and frontiers 239. Michaudel, Le Crac des Chevaliers 46, 68.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

546

stewart

have expressed some suspicions of exaggeration.28 In some cases—such as the fortresses taken from the Franks or from the Armenians—the damage to be repaired had been inflicted by the Mamluk forces themselves.29 While some refortification work was carried out swiftly, some more ambitious programs continued for decades.30 Work at Crac des Chevaliers continued until the reign of Sultan al-Manṣūr Qalāwūn (678–689/1279–1290), with the addition of the new square-based tower on the south front in 684/1285.31 At al-Ṣubayba, east of the Jordan on Mount Hermon, the repairs took 15 years.32 The renovation of the great citadel of Aleppo was not ceremonially completed until 691/1292— though work may only have commenced in earnest after the Mongol invasion of 680/1281.33 Moreover, fortification work was not limited to the early years of the sultanate but continued for its entire existence. While the sultans’ primary focus was on Cairo and Egypt, nevertheless, work in Syria was often undertaken on sultanic initiative. At Aleppo, though some major work was begun by the local commanders, various sultans contributed to the further development of the citadel: notably, al-Muʾayyad Shaykh (815–824/1412–1421), al-Ashraf Barsbay (825–841/1422–1438), al-Ẓāhir Jaqmaq (842–857/1438–1453), al-Ashraf Qāytbay (872–901/1468–1496), and al-Ashraf Qānṣūh (906–922/1501–1516).34

4

Fortresses in Syria: In the Interior and on the Northern Frontier

It is tempting to view the fortification work of the sultans of Cairo in Syria through the prism of “frontier defense.” Northern Syria remained a frontier zone for the duration of the sultanate, and often a contested one. The sultanate was forged in the conflict with the Ilkhans and their satellites, though the last decade of the Ilkhanate was among the most peaceful in the region, following the peace agreed between al-Nāṣir Muḥammad and Abū Sāʿīd in 723/1323.35 Thereafter, the collapse of Chinggisid hegemony in the region presented the sultans of Cairo with opportunities, as they could interfere in the various prin28 29 30 31 32 33 34 35

With regard to the example of al-Ṣubayba, see the discussion in Ellenblum, Who built Qalʿat al-Subayba? 110–111; Raphael, Muslim fortresses 132. For a discussion of the siege of Crac des Chevaliers in 669/1271, and the subsequent repairs, see Michaudel, Le Crac des Chevaliers. Raphael, Muslim fortresses 131. Michaudel, Le Crac des Chevaliers 60. Raphael, Muslim fortresses, 131. Combe, Sauvaget, and Wiet, Répertoire xiii, 107–108; Gonnella, Introduction 114. Gonnella, Introduction. Amitai, Resolution.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

547

cipalities to their north and east; but these were also a threat to the control of northern Syria.36 Timur’s invasion, though spectacularly destructive, was an interlude; the fifteenth century saw increasing threats from “tribal” confederations such as the Qaramanids, Aqquyunlu, Qaraquyunlu, and Dulghadirids; and by its end the Ottoman Empire was increasingly threatening.37 For most of the period of the sultanate, then, the Anatolian-Euphrates frontier remained an area of concern; and, throughout this period, the sultans maintained fortresses and garrisons across the region. Reuven Amitai has argued that the sultans of Cairo very quickly developed a “realistic and coherent frontier strategy,” which was consistently implemented, in order both to strengthen their control of the northern Syrian borderlands and to provide “greater protection to the Syrian heartland.”38 This policy went alongside the reorganization of the sultan’s army and the wider strategic infrastructure of the sultanate; it was not merely defensive but was intended “to bring the war into Mongol-controlled territory.” The policy was therefore both proactive and reactive: on the one hand, irregular forces based in the border zone, such as Bedouin, were employed alongside more regular Mamluk forces for raids, and spies were despatched into Ilkhanid territory; on the other, news of Mongol attacks, perhaps sent by these spies, provoked a scorchedearth response in the region to deny the Mongols fodder and supplies, and also a swift mobilization of the sultan’s forces based further south in Syria or even in Egypt. Fortifications, Amitai argues, played a key role in this multifaceted policy. Sultan Baybars established the fortresses at al-Bīra and al-Raḥba to supervise important crossings of the Euphrates; later in the 660s/1260s he acquired fortresses on key routes toward the Armenian kingdom, at Baghras (Frankish Gaston), and Bahasnā, the latter only temporarily. These castles are presented in active terms: they were early warning points, but also staging points for incursions into enemy lands. This raiding activity against infidels was an important part of the sultans’ self-presentation as ghāzīs, like the heroes of the old Abbasid-Roman frontier. Moreover, the key fortresses played a similar ideological role, as “symbols of Mamluk authority in this frontier region.”39 Nevertheless, the frontier policy established thus by Baybars was “primarily to repel the frequent Mongol incursions over the border”; that is, as a program of defense. The importance of the key fortresses is stressed by a reference

36 37 38 39

The immediate successor dynasties are summarized by Melville, End of the Ilkhanate 323– 327; see also Wing, Decline of the Ilkhanate. Wing, Submission; Woods, Aqquyunlu; Venzke, Case; Muslu, Ottomans and Mamluks. Amitai, Northern Syria 129. Amitai covers similar ground in Holy war 18–27. Amitai, Northern Syria 132–135.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

548

stewart

to the Egyptian civil servant and historian Shāfī b. ʿAlī describing al-Bīra as “the lock of Syria.”40 As Amitai concludes, however, “in the long run, the best protection for the very north of Syria was provided by the deterrent effect of repeated Mamluk raids” into the lands of Ilkhanid satellites.41 The network of border fortresses was, nevertheless, maintained long after the reign of Baybars; it expanded and contracted with the vicissitudes of the times.42 Sultans continued to show personal concern for the border castles, as when al-Ẓāhir Barqūq (784–791; 792–801/1380–1389; 1390–1399) appointed new governors (nāʾibs) for some key border fortresses—Qalʿat al-Rūm, Edessa, and Malaṭya—in 797/1394. Kate Raphael suggests that this may have been aimed at reinforcing the border defenses in the face of the threat posed by Timur. She argues that this behavior is to be expected from a highly centralized polity such as the Cairo sultanate, citing the opinion of the military historian John Keegan that a marker of such centralized states is a concern for “strategic defenses at the periphery.”43 Sultans were, as we have seen, also concerned to maintain a network of fortresses across the whole of Syria, away from the northern frontier. At the same time as his reshuffling of the border castellans, Barqūq appointed nāʾibs elsewhere, for Tripoli, Ṣafad, and Kerak—it may be misleading to single out his appointments on the frontier as being especially significant.44 “Interior” fortresses may perhaps be seen in a similar way to the interpretation of “crusader” castles forming “lines of defense.” Raphael suggests that specific “interior” fortresses may be understood in terms of defence against the crusader threat— so the occupation of Qāqūn (Frankish Caco) is understood as relating to the threat posed by the Templars of Athlīt (Chastel Pelerin).45 This does not, of course, explain the continued use of such castles long after the Franks had been extirpated and Athlīt slighted—Qāqūn still possessed, according to the civil servant al-Qalqashandī, a “fine fortress” in the early fifteenth century.46 40 41 42 43 44

45

46

Amitai, Northern Syria 135; cf. Smail’s warnings about the use of such metaphorical language to describe the roles of castles, Crusading warfare 204. Amitai, Northern Syria 138. See, for example, the list of fortresses in the region provided by Qalqashandī, Subḥ iv, 119– 128. Raphael, Muslim fortresses 208–211. Al-Maqrīzī, Sulūk iii, 2, 824; Ibn Taghrībirdī, Nujūm xii, 59–62. This is also noted by Raphael, Muslim fortresses 211, who questions whether the appointment of new nāʾibs to the border castles “was directly connected with the turbulent events that were to follow, or if it was a routine shuffle of governors.” Raphael, Muslim fortresses 155–156. Of course, if castles like this are seen through the prism of “border defense,” then perhaps the work on the northern frontier could be interpreted as being less of a special case. Qalqashandī, Subḥ iv, 100 (cited also by Raphael, Muslim fortresses 155).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

549

Looking in the other direction, Bethany Walker has convincingly interpreted Transjordan as a frontier province in itself, with its great fortresses continuing to have significance—though not necessarily as “border defense.”47 This points to the variety of roles that could be held by castles across the sultanate: as outposts of the sultan’s government, as watch-posts over potentially unruly populations, as prisons, as granaries and storehouses for military equipment, as stage-posts on the barīd, the sultan’s communications network—even, potentially, as centers of wealth and lordship for members of the ruling elite.48 As with Frankish castles, this multiplicity of roles points to the danger of interpreting castles primarily in terms of strategy against external threats. This is demonstrated by Raphael’s problem with the reconstruction of Ṣafad: “it is difficult to explain the Mamluk decision to rebuild this fortress […] [its] reconstruction had little to do with the larger concept of the Sultanate’s defense.”49 On the other hand, if one rejects an inextricable link between fortress policy and frontier defense, a variety of interpretations are possible. Furthermore, it may not be helpful to categorize castles as being concerned with “defense” or with something else merely on grounds of their relative proximity to a notional frontier. This problem can perhaps be seen from the accounts of al-Ashraf Qāytbāy’s Syrian tour of 882/1477—made shortly after his commissioning of a mighty new citadel for Alexandria. According to the summary account of the chronicler Ibn Iyās, who gives us the Cairene perspective, the sultan’s journey was to the distant ends of the empire; as well as visiting the major cities and granting audiences to provincial officials and other notables, he examined a great number of fortresses, as far as the banks of the Euphrates.50 By contrast, the sultan’s secretary, Ibn al-Jīʿān, who accompanied the expedition, provides a more detailed itinerary, revealing Qāytbāy’s concern for infrastructure in general—bridges and caravanserais as well as fortresses.51 Furthermore, while some of those visited can indeed be seen as “frontier” fortresses, such as al-Bīra or Qalʿat al-Rūm, others were deeper into sultanic territory. Some still retained strategic significance, such as Baghras in a pass through the Amanus Mountains, but others were regional centers of government, such 47 48

49 50

51

Walker, Jordan. On fortresses and the barīd, see Raphael, Muslim fortresses 210; as storehouses, see Raphael, Muslim fortresses, 162; and Yovitchitch, Forteresses 66; Amitai, Arabic inscription 114–116, discusses al-Ṣubayba as a center of lordship for a favored amīr, Bilik al-Khaznadār. Raphael, Muslim fortresses 147. Ibn Iyās records that the sultan’s stated intention was to “inspect personally the governors and the fortresses [al-nuwwāb wa-l-qilāʿ]” of the Syrian territories, Badāʾiʿ iii, 135. See also Petry, Twilight 77–79. Ibn al-Jīʿān, al-Qawl, in Devonshire, Relation.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

550

stewart

as Ṣafad, where Qāytbāy ordered repairs. Castles, along with their associated settlements, remained important for the administrative structure of the sultanate. This was clearly the case for Tripoli or Ṣafad, whose governors frequently appear in lists of major officials, but perhaps also at a lower level. While its castle is not mentioned by Ibn al-Jīʿān, it is interesting that Qāytbāy stayed at Qāqūn. Of course, the continued existence of castles across the interior of Syria does not mean that there were not castles on the northern frontier that may have had distinct roles relating to their position in this region. Indeed, contemporary scholars might use these fortresses as the markers of the frontier. For example, in his geographical treatise, Abū l-Fidāʾ describes the northern “limits of Syria […] from Balīs beside the Euphrates, through Qalʿat Najm, alBīra, Qalʿat al-Rūm, Sumaysāṭ, Ḥiṣn Manṣūr, Bahasnā, Marʿash, and thence by the land of Sīs [Cilicia] to Tarsus and the Mediterranean sea”—clearly here he is tracing the political boundary of the sultanate “in our days.”52 The Egyptian civil servant al-Qalqashandī’s chancery manual groups all the main fortresses of the province of Aleppo together, including some away from the border, but the northern frontier does seem to be defined through the main fortresses, such as Abulustayn (al-Bistān), Malaṭya, Karkar, Kakhtā, Qalʿat al-Rūm, and al-Bīra.53 This is not, however, to suggest that the role of these fortresses was one of passive defense for the sultanate. It is interesting to consider the behavior of the garrisons of these castles in the face of invasions of the sultanate. In the early decades—the period where Amitai identifies a “coherent frontier strategy”—some fortresses on the Euphrates frontier, such as al-Bīra and al-Raḥba, withstood Mongol assaults—though they were unable to prevent the crossing of major forces into Syria.54 Other garrisons, however, seem to have withdrawn south to meet up with the main army of the sultanate, abandoning fortresses that in some cases had been originally obtained through determined efforts—Bahasnā, for example, was taken by Baybars in 666/1268,

52 53

54

Abū l-Fidāʾ, Taqwīm al-buldān 225. Qalʿat Najm is near Manbij in Syria; the others are all in Turkey—Birecik, Rumkale, Samsat, Adıyaman, (Eski) Besni, and Marash. In Turkey: Elbistan, (Eski) Malatya, Gerger, (Eski) Kâhta. Al-Qalqashandī, Subḥ vii, 327, 329; elsewhere other fortresses, such as Tarsus, Adana, and especially Bahasnā, or their governors, are listed: iv, 119–128, 226–229; vii, 172–175; etc. Cf. al-ʿUmarī, al-Taʿrīf 232–234. Amitai, Northern Syria 129, 134. Amitai stresses that the morale of the defenders was maintained by the reassurance that the sultan would not simply abandon them but would send relief when possible. This was also the case for Arjuwash, commander of the citadel of Damascus in 699/1300, who held out against the besieging Mongols despite the pleas of the local population, his resolve stiffened by messages sent from the sultan, al-Nāṣir Muḥammad, al-Yūnīnī, Dhayl 151–157; Amitai, Mongol occupation.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

551

but was soon back in Armenian hands, before al-Ashraf Khalīl was able to reclaim it in 692/1293. Abū l-Fidāʾ reports that several castles were evacuated in the face of Ghazan’s invasion in 699/1299–1300.55 One of these was Tall Ḥamdūn (Toprakkale), which had been acquired with Bahasnā in 692/1293; it was regained after a long siege in 703/1304.56 A century later this policy was not followed—perhaps through lack of preparation—in response to Timur’s invasion, and several fortresses are noted as having been taken, with much destruction—such as Bahasnā, where Timur refused to accept the surrender of the nāʾib in order that he might display the might of his siege weapons.57 Whether abandoned or garrisoned, such fortresses could not prevent major invasions, but even lesser forces proved difficult to resist. For example, in 821/1418 the Aqquyunlu leader ʿUthman Qarā Yuluk crossed into the sultanate; the Qaraquyunlu Qarā Yūsuf came in pursuit of his rival and plundered the towns of ʿAyntāb and al-Bīra—the town citadel might provide refuge to the citizens but by itself could not prevent the looting. At the same time a group of Türkmen had appeared further south at Ṣāfīta; they engaged in plundering and defeated Barsbay, the provincial governor of Tripoli, whose inhabitants fled. Despite the network of fortresses, one gets the impression that the border could be porous. It is interesting that it was the nāʾib of the fortress at Kakhtā who brought the first news of all of this to the sultan in Cairo, along with a leading amīr from Aleppo, the provincial capital—reporting intelligence was seen as more important than attempting to repulse invaders.58 While this was news of a major incursion, it does seem that these outposts of the sultanate were centers of intelligence gathering, and linked into the barīd network.59

55

56 57 58

59

Abū l-Fidāʾ, Mukhtaṣar iv, 57: of those castles previously taken from the Armenians in the lands south of the River Jayḥān (Chahan), only Ḥajar Shughlān was left in the sultan’s hands. See, e.g., al-Jazarī, in Sauvaget, Chronique, 26–27; Ibn al-Dawādārī, Kanz viii, 340; alMaqrīzī, Sulūk i, 784–785; al-ʿAynī, ʿIqd iv, 301. Ibn Taghrībirdī, Nujūm xii, 215–218, describes the lack of planning. For the siege of Bahasnā, see Ibn ʿArabshāh, ʿAjāyib 174; ʿAlī Yazdī, Zafarnāma 162. Ibn Taghrībirdī, Nujūm xiv, 67–74. Earlier in the year Sultan Shaykh made new appointments to the governorates of various Syrian fortresses, such as Ṣafad, Marqab, etc.; one does not get the impression that Syria was being especially neglected at this point. Nor was this incursion exceptional. One of the purposes of Sultan Qāytbāy’s tour in 882/1477, mentioned above, was to inspect the damage inflicted by the raids of a later Aqquyunlu ruler, Uzun Ḥasan. While the citadel of al-Bīra remained strong, the town was ruinous, and much of the population fled: Ibn al-Jīʿān, al-Qawl, in Devonshire, Relation 15– 16. For an account of the road network in the region that linked the fortresses to each other and to the provincial capital, see al-Qalqashandī, Subḥ xiv, 383–384.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

552

stewart

An example of this that also reveals the nāʾibs of the frontier castles working in concert is when the governors of Qalʿat al-Rūm, al-Bīra, Kakhtā, and Malaṭya sent news to Cairo of the resolution of a conflict among the Qaraquyunlu.60 Just as the defense of the sultanate against major invasions rested more on the sultan’s field army than on fortifications, more minor threats were dealt with aggressively. On occasion the justification or explanation provided for a conquest beyond the frontier was the need to root out a nest of brigands— as when Baybars took the fortress of Kaynūk (al-Ḥadath) in 672/1273.61 An attack on Malaṭya in 715/1315 was explained by the inhabitants’ adherence to the Ilkhan, and the inappropriate intermixing there of Christians and Muslims; the aim at this point was not the conquest of the town, but its elimination as a threat. A further explanation for the campaign is provided: the people of Malaṭya had interfered with (and occasionally defeated) the sultan’s own raiding parties, sent out from his border fortresses—Qalʿat al-Rūm, Bahasnā, Kakhtā, Karkar, and others.62 This suggests that, on the one hand, it was more efficient to deal with potential threats beyond the frontier—that is, not to wait for them to launch their own raids—and, on the other, that leading raids into enemy territory was among the key roles of the nāʾibs of border fortresses. This does not necessarily make them distinct from other provincial governors in Syria: we often encounter these, too, in the more major expeditions launched by the sultan. Indeed, the attack on Malaṭya in 715/1315 was led by the nāʾib of Damascus; a century later, in 822/1419, the expedition sent into Anatolia under the command of Ibrāhīm, son of Sultan Shaykh, involved the nāʾibs of Damascus, Tripoli, Ḥamā, and Ṣafad, among others.63 Nevertheless, we may expect that the activity of border garrisons to have been less extraordinary— and for that reason not so often recorded in our sources. While the historical material available from the Mamluk period is exceptional in its range, much reflects the centrality of Egypt to the sultanate. While the names of the border fortresses and, on occasion, their nāʾibs appear in the chronicles, this is often, for example, in the context of a sultan’s reorganization of his provincial government. Governors seem to have changed office frequently; these did not become “marcher lordships”; they may not have always been particularly desired appointments.

60 61 62 63

Al-Maqrīzī, Sulūk iv, 1, 417. Ibn ʿAbd al-Ẓāhir, Rawḍ 417; Amitai-Preiss, Mongols 131–132. Abū l-Fidāʾ, Mukhtaṣar iv, 74. Ibn Taghrībirdī, Nujūm xiv, 84–86.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

553

fortresses and frontiers

5

A Fortress on the Euphrates Frontier: Qalʿat al-Rūm

As a conclusion to this discussion, I will look in detail at one of these frontier fortresses, Qalʿat al-Rūm, and at what can be learned of its nāʾibs.64 Of the twenty or so governors of Qalʿat al-Rūm that can be identified, some only appear in passing—as when, in 791/1389, one Āqbāy al-Ashrafī appears in a list of Syrian governors who had fled to Egypt in the face of Barqūq’s restoration of power.65 The activities of the castellans in the region must be pieced together. Al-Qalqashandī’s chancery manual provides some sense of the sort of individual who might receive such an office. While the northern frontier was part of the province governed by the nāʾib of Aleppo, it was normal for the governors of the castles to be appointed directly by the sultan—they were, then, his “deputies” rather than those of the governor of Aleppo, even if the provincial hierarchy was clear.66 According to al-Qalqashandī, the nāʾibs of Qalʿat al-Rūm were senior amīrs—commanders of a thousand—and overall in the seventh or eighth rank of the notables of the sultanate—equivalent to, for example, the governors of Homs or Jerusalem. Al-Bīra seems to have had similar prestige, but other castellans, even at somewhere like Bahasnā, might have had a lower rank.67 Al-Qalqashandī often groups these frontier fortresses together and describes the road system connecting them to each other and Aleppo; the geographical relationship of Qalʿat al-Rūm to this network is stressed by Abū l-Fidāʾ, who positions it three stages north of ʿAyntāb, a stage to the west of alBīra, and also to the east of Sumaysāṭ, and south of Edessa.68 Some individual nāʾibs of Qalʿat al-Rūm do indeed seem to have been figures of importance—three, for example, were former mamlūks of al-Ẓāhir Barqūq. Others seem to have held the position as part of a wider career in Syrian provincial administration, such as Sayf al-Dīn Ṭuqtamur b. ʿAbdallāh al-Kalatāyī, who died in Aleppo in 787/1385, and was buried in a madrasa of his own foundation

64 65 66

67 68

Some of the following discussion is also covered, in more detail, in Stewart, Rum kale. Ibn Taghrībirdī, Nujūm xi, 359. For a discussion of the significance of the title of nāʾib, see Gibb, Nāʾib; Holt, Structure, 56. Northrup, From slave 215–216, discusses the role of the nāʾib al-qalʿa, pointing out that while the sultan appointed the castellans directly, they were still under the supervision of the provincial nāʾib. For a detailed study of the niyāba of Damascus, see Van Steenbergen, Office. The title might also be used for rulers beyond the sultanate who nevertheless accepted a notional position as “deputy of the sultan”; see Wing, Submission 379. Al-Qalqashandī, Ṣubḥ iv, 226; vii, 189; viii, 220, 229; xii, 168. Al-Qalqashandī, Ṣubḥ xiv, 429; Abū l-Fidāʾ, Taqwīm 268–269 (his sense of direction is about 90° out).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

554

stewart

there, having been (at various times) nāʾib of al-Bīra, Sinjār, and Qalʿat al-Rūm, and then chief chamberlain (ḥājib al-ḥujjāb) in Tripoli and finally Aleppo.69 An office on the northern fringe of the sultanate was not always an attractive one. One nāʾib of Qalʿat al-Rūm, Mughulbāy al-Bajāsī, bought himself a more junior but presumably more desirable position at Tripoli in 856/1452; another, Āq Qujā al-Aḥmadī, recorded as nāʾib in 823/1420, was an amīr of 40 in Cairo in the following year.70 It may well be that in periods when the sultan was paying special attention to the frontier more significant figures were appointed to the office, while at other times it was left to the governor of Aleppo to fill the post—after Mughulbāy’s departure a junior figure from Aleppo, Nāṣir al-Dīn Muḥammad, was inserted. While most of Qalʿat al-Rūm’s governors had Turkish or “mamlūk” names, there are three or four with Arabic-Islamic names, like this Muḥammad; it is possible that this reflects membership of the body of the awlād al-nās, the descendants of mamlūks, and possibly therefore lower status. This is certainly suggested by a reference to Abū Bakr b. Bahādur al-Bābīrī alJaʿbarī as nāʾib in 820/1417; as with Nāṣir al-Dīn Muḥammad, he may well have been a short-term appointment made to fill an empty position—we only learn of him in the report of his replacement by a more significant figure, ManklīKhujā al-Sayfī, by Sultan Shaykh in the course of his visit to the region.71 Some sense of the role of the nāʾib is provided by this episode: the sultan arrived at Qalʿat al-Rūm on the night of Rajab 6/August 19, and “the next morning the sultan went down to the maydān, after having restored the condition of the fortress. He granted 500 dinars to its nāʾib and gave a nafaqa [grant of money] to its baḥrīyya [garrison].”72 These gifts seem to have been made in a public ceremony; on the one hand, the sultan may have been attempting to buy the loyalty of the garrison, but on the other the sum provided to his deputy may have been intended for the necessary repairs identified. That this was not a profitable office may be why some sought employment elsewhere, but at the sultanate’s zenith, in the early fourteenth century, things may have been different. The nāʾib arrested in 710/1310 after Sultan al-Nāṣir Muḥammad’s restoration, Ayāz al-Shamsī, was “taken with his money, which was a million dirhams.”73 The nāʾib through the latter part of al-Nāṣir Muḥammad’s reign, Sayf al-Dīn Jarkas, who died in office in 745/1344/5, was, according to the biographer al-Ṣafadī,

69 70 71 72 73

Ibn Taghrībirdī, Manhal vi, 419–420. Ibn Taghrībirdī, Ḥawādith 378; al-Maqrīzī, Sulūk iv:1, 517–518, 565–566. Ibn Taghrībirdī, Nujūm xiv, 53; al-Maqrīzī, Sulūk iv:1, 410. Al-Maqrīzī, Sulūk iv:1, 411–412. Al-Maqrīzī, Sulūk ii:1, 87.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

555

fortresses and frontiers

proverbial for his wealth.74 The impression given by Abū l-Fidāʾ’s description of the place, from about the same period, is of a prosperous town grown beyond its walls, with gardens and fruit trees; but it may be that the governor’s wealth came also through tolls on traffic on the Euphrates, or through border raiding.75 By the less stable fifteenth century the role of the frontier governor may have been less attractive. While Qalʿat al-Rūm itself seems to have withstood the assaults of Timur, and its strength attracted the refugee populations of other towns such as al-Bīra and ʿAyntāb, there is not the same impression of prosperity.76

6

The Fortress and the Frontier: The Example of Qalʿat al-Rūm

The building and maintenance of castles has often been presented as relating in some way to the defence of borders. This is the case in medieval texts as much as in modern scholarship; the relationship is a debated aspect of the study of the Frankish castles in the Latin east, and also with regard to castles in the sultanate of Cairo. This is not the whole picture, of course; just as Hugh Kennedy demonstrates in Crusader castles, texts from the Mamluk period can also reveal a variety of roles for fortresses and their garrisons. In its heyday, in the period before the Black Death, we get a sense of Qalʿat alRūm as the sort of thriving castle town described in the De Constructione Castri Saphet. As with Ṣafad under the Templars, the rationale for the Mamluk occupation of Qalʿat al-Rūm is presented in exalted terms: religious, defensive, and aggressive. In letters proclaiming its capture in 691/1292, Sultan al-Ashraf Khalīl described his forces as mujahidin; their victory “closed the doors of the Armenians and Tatars,” limiting their possibilities for raiding; by contrast, breaking the locks of the fortress meant opening the gate of the Euphrates to the Muslim sultan, as the first step, God willing, of the conquest of the Ilkhanate. The sultan announced his intention to become ruler of lands extending from the sunset to the sunrise.77 Such grandiloquence exaggerates the significance of a single border fortress, however well-built or strategically sited. Its conquest neither

74 75 76 77

Al-Ṣafadī, Aʿyān ii, 149. Abū l-Fidāʾ, Taqwīm 268–269. Ibn ʿArabshāh, ʿAjāyib 118–126; Ibn al-Jīʿān, Qawl, in Devonshire, Relation 15–16. Al-Nuwayrī, Nihāyā xxxi, 145–146. Cf. Khalīl’s inscription on the citadel of Aleppo; Combe, Sauvaget, and Wiet, Répertoire xiii, 107–108. On the conquest of Qalʿat al-Rūm in 691/1292, see Stewart, Qalʿat al-Rūm; Raphael, Muslim fortresses 103–104, 187–188; Danielyan, Arabic sources 245–264; Stewart, Rum kale.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

556

stewart

prevented further Mongol attacks, nor led to the destruction of their realm. The example of Qalʿat al-Rūm may, however, give us a sense of the role of a fortress on the sultanate’s Syrian frontier—as a refuge for local populations; as a base for raiding and intelligence gathering; as an economic focus; and as a symbol of the power of the sultan himself.

Bibliography Sources Abū l-Fidāʾ, Taqwīm al-buldān, ed. J.T. Reinaud and W. MacGuckin de Slane, Paris 1840. Abū l-Fidāʾ, al-Mukhtaṣar fī akhbar al-bashar, 4 vols., Istanbul, 1869–1870. ʿAlī Yazdī, Zafarnāma, trans. J. Darby, The history of Timur-Bec, 2 vols., London 1723. al-ʿAynī, ʿIqd al-jumān fī tārīkh ahl al-zamān, ed. M.M. Amīn, 4 vols., Cairo, 1987–1990. Ibn al-Dawādārī, Kanz al-durar wa-jāmi al-ghurar, vol. 8, ed. Ulrich Haarmann, Freiburg 1971. Ibn al-Jīʿān, al-Qawl al-mustaẓraf fī safar mawlānā al-Malik al-Ashraf, in R.L. Devonshire (trans.), Relation d’un voyage du Sultan Qâitbây en Palestine et en Syrie, in Bulletin de l’Institut Français d’Archéologie Orientale 20 (1920), 1–43. Ibn ʿAbd al-Ẓāhir, al-Rawḍ al-zāhir fī sīrat al-Malik al-Ẓāhir, ed. ʿA.ʿ A. al-Khuwayṭir, Riyadh 1976. Ibn Iyās, Badāʾiʿ al-zuhūr fī waqāʾiʿ al-duhūr, ed. M. Mostafa, 3 vols., Cairo 1984. Ibn Taghrībirdī, Ḥawādith al-duhūr fī madā al-ayyām wa-l-shuhūr, ed. M.K. ʿIzz al-Dīn, Beirut 1990. Ibn Taghrībirdī, al-Manhal al-ṣāfī wa-l-mustawfī baʿda al-wāfī, 13 vols., Cairo, 1984– 2009. Ibn Taghrībirdī, al-Nujūm al-zāhira fī mulūk miṣr wa-l-qāhira, 16 vols., Cairo 1929–1972. Ibn ʿArabshāh, ʿAjāʾib al-maqdūr fī nawāʾib Tīmūr, ed. J. Golius, Historia Tamerlanis Arabice, Leiden 1636. al-Jazarī, Ḥawādith al-zamān, extracts and summaries in J. Sauvaget, La Chronique de Damas d’al-Jazari (années 689–698), Paris 1949. al-Maqrīzī, Kitāb al-Sulūk li-maʿrifat duwal al-mulūk, ed. M.M. Ziyāda and S.ʿA.-F. Āshūr, 4 vols., Cairo 1934–1972. al-Qalqashandī, Subḥ al-Aʿshā fī ṣināʿat al-inshāʾ, 14 vols., Cairo 1913–1920. al-Ṣafadī, Aʿyān al-ʿaṣr wa-aʿwān al-naṣr, ed. ʿA.A. Zayd et al., 6 vols., Beirut 1998. al-ʿUmarī, al-Taʿrīf bi-muṣṭalaḥ al-sharīf, Cairo 1895. William of Tyre, Chronicon, ed. R.B.C. Huygens, Turnhout 1986. al-Yūnīnī, Dhayl mirʾat al-zaman, in Li Guo, Early Mamluk Syrian historiography: alYunini’s Dhayl mirʾat al-zaman, 2 vols., Leiden 1998.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

557

Studies Abulafia, D., Introduction: Seven types of ambiguity, in N. Berend and D. Abulafia (eds.), Frontiers in the middle ages, Aldershot 2002, 1–34. Amitai-Preiss, R., Mongols and Mamluks: The Mamluk-Ilkhanid war, 1260–1281, Cambridge 1995. Amitai, R., Holy war and rapprochement, Turnhout 2013. Amitai, R., An Arabic inscription at al-Subayba (Qal‘at Namrud) from the reign of Sultan Baybars, in M. Hartal et al., The al-Subayba (Nimrod) Fortress: Towers 9 and 11, Jerusalem 2001, 109–123. Amitai, R., Northern Syria between the Mongols and Mamluks: Political boundary, military frontier and ethnic affinity, in D.J. Power and N. Standen (eds.), Frontiers in question: Eurasian borderlands c. 700–1700, London 1999, 128–152. Amitai, R., The Mongol occupation of Damascus in 1300: A study of Mamluk loyalties, in M. Winter and A. Levanoni (eds.), The Mamluks in Egyptian and Syrian politics and society, Leiden 2004, 21–41. Amitai, R., The resolution of the Mongol-Mamluk war, in R. Amitai and M. Biran (eds.), Mongols, Turks, and others: Eurasian nomads and the sedentary world, Leiden 2005, 359–390. Barber, M., and K. Bate (eds.), The Templars: Selected sources, Manchester 2002. Beeler, J.H., Castles and strategy in Norman and early Angevin England, in Speculum 31 (1956), 581–601. Boas, A.J., and A.M. Maeir, The Frankish castle of Blanche Garde and the medieval and modern village of Tell es-Safi in the light of recent discoveries, in Crusades 8 (2009), 1–22. Brown, R.A., An historian’s approach to the origins of the castle in England, in Archaeological journal 126 (1969), 131–146. Combe, É., J. Sauvaget, and G. Wiet (eds), Répertoire chronologique d’épigraphie arabe, 18 vols., Cairo 1931–1991. Creighton, O., and R. Liddiard, Fighting yesterday’s battle: Beyond war or status in castle studies, in Medieval archaeology 52 (2008), 161–169. Danielyan, G., Arabic sources on the history of the Armenian Catholicosate of Hromkla, in A. Bozoyan (ed.), Cilician Armenia in the perception of adjacent political entities, Erevan 2019, 184–266. Edwards, R.W., The fortifications of Armenian Cilicia, Washington DC 1987. Ellenblum, R., Crusader castles and modern histories, Cambridge 2007. Ellenblum, R., Frankish rural settlement in the Latin Kingdom of Jerusalem, Cambridge 1998. Ellenblum, R., Three generations of Frankish castle-building in the Latin Kingdom of Jerusalem, in M. Balard (ed.), Autour de la première croisade, Paris 1996, 517–551. Ellenblum, R., Were there borders and border-lines in the middle ages? The example of

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

558

stewart

the Latin Kingdom of Jerusalem, in N. Berend and D. Abulafia (eds.), Frontiers in the middle ages, London 2002, 105–119. Ellenblum, R., Who built Qalʿat al-Subayba?, in Dumbarton oaks papers 43 (1989), 103– 112. France, J., Fortifications east and west, in H. Kennedy (ed.), Muslim military architecture in greater Syria, Leiden 2005, 281–294. Gibb, H.A.R., Nāʾib, 1. In pre-modern usage, in ei2, vii, 915. Gonnella, J., Introduction to the citadel of Aleppo, in S. Bianca (ed.), Syria: Medieval citadels between east and west, Turin 2007, 103–138. Hoch, M., The Crusaders’ strategy against Fatimid Ascalon and the “Ascalon project” of the Second Crusade, in M. Gervers (ed.), The Second Crusade and the Cistercians, New York 1992, 119–128. Holt, P.M., The structure of government in the Mamluk Sultanate, in P.M. Holt (ed.), The eastern Mediterranean lands in the period of the Crusades, Warminster 1977, 44–61. Huygens, R.B.C. (ed.), Un Nouveau Texte du traite “De constructione castri Saphet,” in Studi medievali 3 (1965), 355–387. Kennedy, H., Crusader castles, Cambridge 1994. Kennedy, H. (ed.), Muslim military architecture in greater Syria: From the coming of Islam to the Ottoman period, Leiden 2006. Melville, C., The end of the Ilkhanate and after: Observations on the collapse of the Mongol world empire, in B. de Nicola and C. Melville (eds.), The Mongols’ Middle East: Continuity and transformation in Ilkhanid Iran, Leiden 2016, 309–335. Michaudel, B., La Fortification comme sceptre des Ayyoubides et des Mamelouks dans le Bilād al-Shām et en Egypte à l’époque des croisades, in Bulletin d’études orientales 57 (2006–2007), 51–64. Michaudel, B., Le Crac des Chevaliers, quintessence de l’architecture militaire mamelouke, in Annales islamologiques 38 (2004), 45–77. Michaudel, B., The use of fortification as a political instrument by the Ayyubids and the Mamluks in Bilād al-Shām and in Egypt (twelfth–thirteenth centuries), in msr 11 (2007), 55–67. Milwright, M., The fortress of the raven: Karak in the middle Islamic period (1100–1650), Leiden 2008. Muslu, C.Y., The Ottomans and the Mamluks: Imperial diplomacy and warfare in the Islamic world, London 2014. Northrup, L.S., From slave to sultan: The career of al-Mansūr Qalāwūn and the consolidation of Mamluk rule in Egypt and Syria (678–689ah/1279–1290ad), Stuttgart 1998. Petry, C.F., Twilight of majesty: The reigns of the Mamlūk sultans al-Ashraf Qāytbāy and Qanṣūh al-Ghawrī in Egypt, Seattle 1993. Platt, C., Revisionism in castle studies: A caution, in Medieval archaeology 51 (2007), 83–102.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

fortresses and frontiers

559

Pringle, D., Castles and frontiers in the Latin east, in A. Jotischky and K.J. Stringer (eds.), Norman expansion: Connections, continuities and contrasts, Farnham 2013, 227–239. Pringle, D., Crusader castles and fortifications: The Armenian connection, in C. Mutafian (ed.), La Méditerranée des Arméniens, xiie–xve siècle, Paris 2014, 353–374. Pringle, D., Review of Reconstructing the castle of Safad, in Palestine exploration quarterly 117 (1985), 139–149. Prior, S., A Few well-positioned castles: The Norman art of war, Stroud 2006. Rabbat, N., The citadel of Cairo: A new interpretation of royal Mamluk architecture, Leiden 1995. Raphael, K., Muslim fortresses in the Levant: Between crusaders and Mongols, London 2011. Smail, R.C., Crusading warfare 1097–1193, Cambridge 1956. Standen, N., Introduction B. Nine case studies of premodern frontiers, in D.J. Power and N. Standen (eds.), Frontiers in question: Eurasian borderlands c. 700–1700, London 1999, 13–31. Stewart, A.D., “One of the most glorious fortresses”: Rum kale in the Sultanate of Cairo, in S. Redford (ed.), Ortaçağ’dan Günümüze Rumkale/Rumkale from the middle ages to the present, Gaziantep 2022. Stewart, A.D., Qalʿat al-Rūm/Hromgla/Rumkale and the Mamluk siege of 691ah/1292 ce, in H. Kennedy (ed.), Muslim military architecture in greater Syria, Leiden 2006, 269–280. Tonghini, C., Shayzar i: The fortification of the citadel, Leiden 2012. Van Steenbergen, J., The office of Nāʾib al-Salṭana of Damascus: 741–784/1341–1382, a case study, in U. Vermeulen and J. Van Steenbergen (eds.), Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras iii, Leuven 2001, 429–448. Venzke, M.L., The case of a Dulgadir-Mamluk iqṭāʿ: A re-assessment of the Dulgadir principality and its position within the Ottoman-Mamluk rivalry, in Journal of the economic and social history of the Orient 43 (2000), 399–474. Walker, B.J., Jordan in the late Middle Ages: Transformation of the Mamluk frontier, Chicago 2011. Wing, P., Submission, defiance, and the rules of politics on the Mamluk Sultanate’s Anatolian frontier, in jras 25 (2015), 377–388. Wing, P., The decline of the Ilkhanate and the Mamluk Sultanate’s eastern frontier, in msr 11 (2007), 77–88 Woods, J.E., The Aqquyunlu: Clan, confederation, empire, Salt Lake City 1999. Yovitchitch, C., Forteresses du Proche-Orient: l’Architecture militaire des Ayyoubides, Paris 2011.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 28

The Sasanian Fort of Pānkān Balázs Major

1

Introduction

Sasanian sites of military origin are relatively scarce compared to the more numerous and much better studied remains from the Roman and Byzantine areas. For this reason, the recently identified site at the present-day village of Pānkān in the territory of the Kurdistan Regional Government of Iraq, which was a fortified structure of Sasanian origin, is of considerable importance. The rectangular enclosure with projecting corner and interval towers shows clear elements of a fortified site, while field surveys show that this building did not stand alone. It is also evident that life continued at the site and its immediate environs after the Muslim conquest, too. The present chapter concentrates on the main architectural features of the site and its dating based on the ongoing research of the Kurdish-Hungarian Archaeological Mission (kham).*

2

The Site

Pānkān is situated at the meeting point of the great Mesopotamian plains and the mountain ranges of northern Iraq. It is 37 km north-northeast of Erbil (ancient Arbela) and 90km to the east-northeast of Mosul (antique Ninive).

* The Kurdish-Hungarian Mission is the joint research project of the Institute of Archaeology of Pázmány Péter Catholic University (ppcu), the Department of Archaeology of Salahaddin University of Erbil, and the Directorate of Antiquities of Erbil under the auspices of the Directorate General of Antiquities of the Kurdistan Regional Government. The project centered on Dwīn castle and its environs and was started under the directorship of Dr. Zidan Bradosty and recently is led by Dr. Aziz Zebari (on the Kurdish side) and Dr. Balázs Major and Bendegúz Takáts (on the Hungarian side). Its main supporter was the mol (Hungarian Oil and Gas Public Limited Company) and the research grant program (kap) of Pázmány Péter Catholic University, and recently the research grant of the Ministry of Human Resources of Hungary. It is being realized within the project framework entitled: Archaeology Research on the Contacts between Hungary and the East (Our Eastern Heritage, ppcu History and Archaeology Interdisciplinary Research Team; tkp2020-nka-11), with the support of Thematic Excellence Program, National Research, Development and Innovation Office. We owe our sincere thanks

© Balázs Major, 2022 | doi:10.1163/9789004525245_029

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

figure 28.1

561

Valley of the Kargh River with the rectangular enclosure of Pānkān in the foreground photo b. takáts

The site is strategically placed in the southern foreground of the gorge through which the Kargh River (local name Pīra Bakir) breaks through the Razwān Mountains at the Baroj Tījī cliffs (see fig. 28.1). Having passed this gorge only 2km from Pānkān, it collects the waters of a few temporary tributary streams and flows into the Great Zāb, 10 km to the northwest of Pānkān. The fortified enclosure itself sits close to the edge of the ravine of the Kargh River, from where there is a good view over the river valley and one can see as far as the mountain ranges of Akrī behind the Great Zāb to the northwest. According to our local colleagues, the ruins at Pānkān were already included in the Iraqi archaeological inventory as the remains of an Ottoman caravanserai in the 1960s. The locals, who call it Khān Dīrūk (or sometimes Qalʿat al-Jinn), consider it as being of very ancient origin. A study published in 2018 identified the place as a sixth-century monastic foundation under the name Dayr al-Juṣṣ, which was destroyed in the 9th century, after which it was only transformed into a fortification in the nineteenth century by having the corner towers added to it.1 to the Consulate General of Hungary in the Kurdistan Regional Government, to Mr. Mala Awat, Kayfi Ali, Nader Babakr, Dr. Zidan Bradosty, Dr. Aziz Zebari, Yadgar Argoshi, and all the Kurdish and Hungarian team members for their enthusiastic work. 1 Badrī, Discovery 340–341.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

562

figure 28.2

major

Fort of Pānkān from the northeast photo b. takáts

The first kham field surveys commenced in November 2014, and besides basic documentation of the architectural remains above the surface, they started extended field surveying around the site and executed a geomagnetic2 and a Remotely Piloted Aircraft System (rpas) based survey3 in the southern and western foreground of the enclosure. The results of the initial surveys showed that the enclosure was mostly a single-phase structure, with the scanty remains of rounded towers on its corners and a number of interval towers on its southern and western sides (see fig. 28.2). The pottery showed clearly that this fort and its environments were used from the Sasanian into the early Islamic period and seem to have been largely abandoned by the Ottoman period. That the remains are preserved to a considerable degree under the present ground level was indicated by the two elongated rooms in the north-eastern corner of the enclosure (R100 and R101). They were partially unearthed by clandestine excavations almost to their ori-

2 Directed by Dr. Gábor Bertók (Institute of Archaeology—ppcu). 3 Done by Bendegúz Takáts (Institute of Archeology—ppcu).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

563

ginal ground level, and this enabled the estimation of the remaining height, which is about 3m. Charcoal samples collected from the mortar of these walls, and forming the base of ams radiocarbon analyses, showed that the date of construction of R100 was between 430–560ce. Fieldwork in 2015, 2016, and 2017 continued with the documentation of the architectural remains, and the archaeological field surveys prepared the way for the excavations in 2018 and 2019. The excavations concentrated on the gate area and the western ranges of the fort and added considerably to our knowledge of the architectural details and construction sequences of the site. The resulting picture emerging from this fieldwork shows the foundation of a fortified enclosure in the late Sasanian period, which was partially surrounded by a settlement, both of which were in use in the early Islamic period and might have vanished in the thirteenth century, possibly as a result of the depopulation following the Mongol invasion.

3

The Fort

3.1 The Enclosure The fort is a roughly regular square shaped enclosure with its northern and western walls being 68.2m, the southern one 62.8m and the eastern wall 71.7 m in length (see fig. 28.3). Although the outer walls have been mostly razed to the present-day ground level, it is clearly visible from the mortared stones on the surface that the enclosure had round towers at its corners and was also equipped with a number of interval towers that were possibly able to provide flanking fire, covering the enclosure walls between them. With the help of the corner towers and the relatively numerous imprints of the enclosure wall on the surface, in the form of mortar-bonded stones or regular stone scatters, it is possible to follow the general outline of the enclosure walls. These seem to deviate at some points from the strictly straight line but show the architects’ preference for it. The estimated thickness of the wall seems to be rather thin, especially compared to some walls of the buildings in the interior. The small segment of the enclosure wall unearthed in excavation trench 2019/2 clearly verified that the curtain wall on the western side was only 1 m thick. Here, at wall section w324, it could be observed that the wall was set within a shallow foundation of only 22cm of depth in the extremely hard soil on the interior side at room R300. Above the compacted earthen floor of the interior the wall remains stood 190cm high. On the outer side of the western wall, 50 cm below the top of the wall, remains the start of a plastered sloping base, and a kind of glacis structure could be observed. It might have continued for another 140 cm

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

564

figure 28.3

major

Plan of the fort b. major, c. chlela, and b. takáts

downwards if the terrain level outside the fort was roughly identical with that of the rooms in the interior. How high this glacis-supported wall might have been we can only guess. If one considers the estimated 8–9 m height of the roughly contemporary Tammīsheh and Gorgān walls, which had a wall thickness of 3m,4 then the enclosure walls of Pānkān with their 1 m thickness might have been closer to 6m. The walls are of solid construction, equipped with a glacis, and they were intended to be supported by the thick-walled buildings

4 Priestman, Architecture of the Tammīsheh wall 258.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

565

on their interior. Without these latter structures, the existence of any useful chemin de ronde on this wall thickness would have been quite unlikely. Because of the dilapidated state of the remains, it is almost impossible to make a precise estimate of the size of the corner and interval towers, and the work is further complicated by the construction method of the fort. Its walls were built with the local rubble stone with a rounded shape, as most of it was collected from the nearby riverbed. The middle-sized stones were set in a very thick gypsum mortar, and they both were poured into formworks or moulds. Even in the case of adjacent buildings, it was done individually, like in the case of the towers, which were attached to the already-existing wall structures in a second step. The same construction methodology was typical in the region as late as the nineteenth century, too, as exemplified by the nearby late Ottoman-period castle of Dwīn. It is also apparent at several towers that they were constructed in this “agglutinated way.” First, a rounded core structure was built with a small radius, then around this core structure, a second layer of a gypsummortar, rubble stone wall was erected. In the case of the southwestern corner tower (T3), one can observe the circular core shape of the tower with a diameter of about 1.2m, around which a 1.36m thick crust of wall was added. If there was no third “wall crust” around T3, then its diameter can be estimated to have been about 5.1m. Although it is likely that the rest of the corner towers were of the same dimension, this can be ascertained only by excavations, as the present visible remains indicate the existence of rounded corner towers of only 3.5m (T6), 3.74m (T7), and 2.7m (T8) in diameter. If they had these rather small diameters and employed such construction methods, the corner towers were almost certainly filled, at least on their ground levels, with no rooms inside. The question of the interval towers on the curtain walls is an even more uncertain issue. There are clear traces of two on both the western (T4 and T5) and the southern walls (T2 and T3). The architects evidently tried to have the towers evenly distributed on both wall stretches, also taking into account that the southern wall had a gate-tower opening in its center. Both field surveys and the aerial photos show two extensive stone scatters along the northern and eastern enclosure wall, too, but on the surface, no clear evidence of in situ mortar-bonded stones were found. Furthermore, the two considerable stone scatters along the eastern wall that might indicate the former presence of the supposed interval towers (T 12 and T13) are also not evenly distributed. No stone accumulation of similar size can be observed along the northern wall on the surface. All along the scanty remains of the northern wall there is only one place where there is a piece of wall core jutting out from its line (at w128),

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

566

figure 28.4

major

Detail of northern enclosure wall (w129) showing the remains of the formwork casting photo b. takáts

and it might indicate the existence of an interval tower. In this case, we can expect another interval tower similar to T5 on the eastern half of the northern wall, which otherwise does not show any signs of a gate or other structures that would interrupt the sequence of interval towers observed in case of the other walls. There was also one anomaly observed along the middle of the northern wall in the surviving stretch of wall w129. The remaining 7.5 m long wall here was 1.2m thick and was cast in a formwork with segments measuring 1.3 and 1.5m (see fig. 28.4). Whether it can be taken as an indicator of the former existence of some kind of subsidiary structure is still a question. Sounding no. 2019/7 at tower T5 makes it very probable, that the interval towers were U shaped (see fig. 28.5). T5 had a straight northern side wall projecting from the enclosure for 2.2m. The curve of the semicircular western side seems to have projected a further 1.2m, thus the tower itself seems to have jutted out from the façade of the enclosure wall for a distance of a minimum of 3.4 m. The estimated width of the tower was around 5 m, and as such, it might have been a “real” flanking tower, not just a simple tower buttress unable to provide effective flanking for the adjacent walls.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

567

the sasanian fort of pānkān

figure 28.5

Northwestern cluster of buildings in the fort b. major, c. chlela, and b. takáts

3.2 The Gate The traces of what was the main gate of the fort were found in the middle of the southern wall, and the main target of seasons 2018 and 2019 was the excavation of this structure (see fig. 28.6). The unearthed remains show that the structure was executed in several phases, (see fig. 28.7) but the first ones seem to have been part of a single construction period, like in the case of the corner and interval towers. The archaeological data retrieved here provided clear proof of the prolonged use of the fort. The first element (no. 1 on fig. 28.7) to have been constructed was the southern enclosure wall, with a “buttress tower” on each side of the gate opening. These buttresses had the shape of a quarter segment of a circle. The radius could be measured on the completely excavated western buttress, which is a bit deformed, and thus the length varies between 1 m and 1.5 m, depending whether one takes the measurement on its eastern or northern side. The width of the gate opening between the two buttresses was 3.8 m. In a second phase

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

568

figure 28.6

major

Excavated gate from the south photo b. major

(no. 2 on fig. 28.7), the width of the gate passage was narrowed by the addition of a 40cm thick wall on each side that framed the gateway. Their incurving southern ends served as the jambs for the gate, which became about 2.9 m wide. Without these side jambs, it would have been quite hard to close the wide gate effectively, so it is almost certain that this was also part of the original construction period. In a third phase (no. 3 on fig. 28.7), a thickening of the enclosure wall and the buttress towers took place, with the attachment of a new layer of wall to the already existing one from the outside. This new addition was 80 cm thick in front of the curtain wall, which became 1.8 m thick altogether. The buttress towers received a roughly 1.15m thick layer of wall around them. Thus, the final diameter of the buttress towers came to around 3.4 m, a structure that was more harmonizing with the dimensions of the corner and interval towers. For this reason, and also because of the “agglutinating” construction style already observed in the case of the towers of the enclosure walls, it is very likely that this addition was only a third construction phase in a single-period building project. If this was the case, it is probable that the final construction around the gate was not intended to create two individual tower buttresses, but it is more likely that the enlarged quarter segments were forming the base of a single gate tower. For the former case, a good example could be the gate of the Umayyad enclosure of Qaṣr al-Kharāna (Jordan) a mere two centuries later. For the latter

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

figure 28.7

569

Construction periods of the gate b. major

case, the semicircular gate tower, one might pick the gate tower of the Umayyad main qaṣr at Jabal Says (Syria). Given the fact that the inner sides of the buttress towers are in line with the opening itself, the possibility that the third-phase structure was a single gate tower with a rounded outer façade is more probable. This thickening of the wall of phase 3 is likely to have been only limited to the stretch of the southern wall between interval towers T2 and T9, as there were no signs of such constructions in the sounding of 2019/3 on the western wall. Phase 3, with its slightly sloping bases, was in all likeliness employed to prepare the base of a more considerable gate tower structure and support the wall sections in its immediate vicinity. In the foreground of the western edge of the western buttress tower, the excavations revealed the remains of a thin wall that joined the enclosure wall

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

570

major

perpendicularly and in all probability was nearly contemporary to the thickening of the wall (no. 4 on fig. 28.7). As its width was only one course of rubble stones, and it was plastered from only the northern side, while the other was attached to the construction debris, it is very likely that this was a simple revetment wall coating the western side of the foreground of the gate to some height. It might have been built to demarcate the northern side of a shallow trench area in front of the gate. The phasing of the construction work is clearly reflected by the floor of the gate passage, too, which was also inserted in consecutive phases. The first section of the floor only covered the area of the gate tower itself (no. 5 on fig. 28.7), and the rest were only added when the walls of the buildings of the interior were already completed. The floor itself consisted of large rubble stones sunk in a thick layer of mortar. After the gate structure and its floor were ready came the addition of two thick wall stretches (no. 6 on fig. 28.7) in the continuation of the gate passage toward the interior of the enclosure. The walls are only about 1.7m long, but the western one is 1m thick, while the eastern one opposite to it measures 1.3m in thickness. As in their continuation toward the north, the excavations found the walls (no. 7 on fig. 28.7) considerably thinner (only around 0.4m); it is very likely that these thick walls of phase 6 were added to the gate so it would be able to hold a brick barrel vault with a north-south axis that covered the entrance passage. The northern part of the rubble-stone and plaster floor (no. 8 on fig. 28.7) was completed only after the walls of phases 6 and 7 were standing, and the thick division line between floor sections no. 5 and no. 8 clearly shows that they were not done simultaneously. That the passage of the gate was originally covered by a brick vault was shown by the huge quantities of fallen brick of the Sasanian type (see fig. 28.8). These bricks usually measure approximately 30× 35 cm in area, with an average thickness of 6cm, and they frequently have finger-drawn diagonal lines on their surfaces. It is not rare to see complete palm imprints on their surfaces, too. The remains of the brick vault covered the area between phases 6 and 2 and seem to have extended into the zone of the phase-3 additions, too. This indicates that all the aforementioned elements of the passage were planned to be executed in a single construction period divided into different phases. The vaulting of the gate opening constructed in phase 3 might be taken as another sign of a single tower structure above the gate, which was supported by the vault spanning the passage between the two buttress tower–like structures. The abovementioned construction phases were followed by a rather long period of decay, during which debris of more than 35 cm thick accumulated in the passageway of the gate. This took place gradually, as is reflected by the numerous ash layers scattered in the section walls. Some of them are rather

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

figure 28.8

571

Sasanian-period bricks photo b. major

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

572

major

short, which indicates that they were “camp fires” rather than layers of destruction. Radiocarbon dating of the samples collected there will hopefully give a clearer dating of this period. We see the return of some limited construction activities at a later date. The opening of the gate was narrowed by two brick walls (no. 9 on fig. 28.7) attached to the southern end of the gate jambs constructed in phase 2. The walls were constructed of bricks differing from the Sasanian ones, as these are more reddish, their material is lumpier, and the finger-drawn lines are also missing. They were bonded with mortar of inferior quality compared to the earlier constructions, and it also contains pebbles and smaller stones. The same period also saw the construction of a threshold (no. 10 on fig. 28.7) made of rubble mortar, with the reuse of the fragment of an architrave and four Sasanian bricks. Behind the brick wall narrowing on the eastern side of the gate passage a bench-like structure (no. 11 on fig. 28.7) was constructed of rubble mortar. Its remains are 45 cm high, and it seems to have lasted until the northern end of the thick wall was constructed during phase 6. It measured 45cm in depth, thus it was seemingly adjusted to the new entrance width, which became only 1.2 m. A small extension (no. 12 on fig. 28.7) to the brick wall of phase 9 of the western side of the gateway was also added, apparently in the same period and also built of bricks. Chronologically, the next event that left clear traces in the stratigraphy was the collapse of the brick vault over the passage of the gate. It covered the abovementioned debris layers, and above it, thick deposits of rubble accumulated with time. The traces of meager construction activities were found in the upper ranges of the rubble. A simple floor pavement of broken brick and debris was visible on the top of the western wall of phase 6. On the eastern wall of the same phase, a 35cm wide groove might have been carved as the foundation of a brick wall, with some of its fragments still visible (no. 13 on fig. 28.7). These were rather simple structures constructed on top of the debris layers, about 130cm above the original Sasanian plaster floor. They in all probability made use of the shade that was provided by the still-standing remains of the enclosure walls. 3.3 Buildings in the Interior Inside the perimeter of the enclosure the remnants of former structures are all under the ground level, save the remains of a handful of division walls. Because of terrain difficulties, the possibility of geophysical surveys is very limited, so information was primarily gained from architectural surveys and limited excavations. The orthomosaic plan and terrain models based on rpas-based survey indicate that about three quarters of the interior area of the fort was covered by buildings (see fig. 28.9).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

figure 28.9

573

Terrain model of the fort b. takáts

Precise mapping of the fragments of mortar-bonded in situ stones on the surface prove that almost the whole northern half of the interior of the enceinte was covered by buildings of some kind. The southern half had a line of structures 10m wide behind the southern wall and running along its full length. The line of stone scatters also shows that there was a row of rooms behind the western wall, with an estimated width of 6.5 m. A similar row of buildings can be envisaged behind the eastern wall, but very likely with a more considerable, 10m width. If the whole northern half of the 4,663 m2 interior area of the enceinte was built over, and we subtract the area of the abovementioned line of buildings beside the walls of the southern half of the fort, we are left with 1,120m2 for the courtyard. This is slightly less than a quarter of the area of the whole fort. There is also a possibility for the existence of lesser courtyards in the northern half, but it is impossible to ascertain now. The clear evidence of recent bulldozing activities on the surface of the northern half of the fort also raises caution. It is noteworthy that the interior did not seem to have been planned with complete symmetry. While the arrangement of the rooms behind the western wall was perpendicular to it, most of those on the eastern wall ran parallel to it in at least two rows. It is also obvious from both the standing and excavated remains that the buildings were not constructed in a single phase but were

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

574

major

figure 28.10 Northeastern cluster of buildings in the fort b. major, c. chlela, and b. takáts

attached to each other. The gypsum-mortar and rubble-stone walls seem to have been mostly raised with the help of timber formworks and were “molded” individually, one after the other. The thick mortar covered most of the façades, but as excavation trench 2019/1 in the interior of the western room behind the gateway showed, the façades of the rooms were plastered and covered with a whitewash. 3.3.1 Northeastern Cluster of Buildings The best-preserved constructions above ground in the whole fort are found in the northeastern half of the enclosure (see fig. 28.10). Four of them (no. 100, no. 101, no. 102, and no. 103) had enough remains to calculate their original area, while a number of other wall remains show that they were part of a larger set of interconnected buildings. Although at first glance they seem to show some kind of regular arrangement, they also were not arranged according to a symmetrical plan.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

figure 28.11

575

Buildings R100 and R101 from the south, with the imprint of the floor tiles that once covered R101 in the middle part of the picture photo b. major

Two of them, nos. 100 and 101, stand almost to their full ground-floor height (see fig. 28.11). Their good state of conservation might owe something to the fact that they are sunken about 1m below the average floor level of the fort, which stood at 526.37m above sea level. Both rooms are rather narrow, with an elongated rectangular plan, and both were covered by a barrel vault of rubble stone bonded by gypsum mortar casted in a formwork. The first room to be “molded” was R101, which in turn was in all probability attached to the rear of the eastern wall. The room was approximately 7.3×2m, with a narrow entrance on the northern side. Near the southern end of the western sidewall the remains of an arched niche with an estimated area of some 60×60cm can be observed. Built parallel to the neighboring room R100, it is probable that this was not the only niche in the western wall of the room. The western wall (in which the niche is positioned) was preserved intact, and it is clearly visible that the top of the mostly vanished barrel vault attached to it preserved the imprints of a large brick pavement. The bricks were similar in size to the Sasanian-type bricks seen at the gate. It is almost certain that the brick-paved floor was the floor of a covered space, a first-floor room above R101. As the ground-floor room seems to have been around 3m high, the estimated 6m height of the enclosure wall, with two levels of rooms behind, does seem to be reasonable.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

576

figure 28.12

major

Interior of R100 looking south, with the niches visible on the left photo b. major

Room R100 was attached to R101 in a second phase (see fig. 28.12). While its length was about 6.3m, it was very narrow, only 1.25 m. It seems that its 82cm wide original doorway opened in the southern end; however, it was later walled up from the outside, and the new approach seems to have been broken in its western wall. Here we find three arched niches in the western wall and at its northern end a rather irregular-shaped smaller one, too. The best-preserved and almost unearthed niche is the third one from the south, which is 83cm wide and 106cm high, with a depth of about 60 cm. It was the second niche from the south through which the new doorway seems to have been broken into room R1004. It seems that R100 differed from R101 not only in its size but also in its roofing system. The northern end of the room was covered with a mould-cast barrel vault of a slightly apsidal form; however, the longer middle stretch of the barrel vault had a steeper angle, which started from a purposely built stepping in the vault. There are no traces of the possible existence of a first floor above room R100. Given the narrowness and presence of the niches of this partially sunken room, R100 might be regarded as an area with a storage function. Only excavations can reveal the real extent of R1004, which ran parallel to R100. Also, little is seen of room R1005, which shared its northern wall (w213) with R1004 and seems to have preserved its short stretch of eastern wall (w212),

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

577

too. It might have lasted until w1021 in the east, and thus would have measured 5.5×2m. It had a narrow doorway opening at the western end of its northern wall (w213) and a similar door on its western wall (w212). The remains above them indicate that this room was also covered with a vault. The building complex that contained R103 and R102 and had several attached buildings lay further to the south. R103 was a rectangular room, measuring 9 ×2.5m, with its longer eastern wall attached to the eastern enclosure wall. The room had its main doorway on the south and is likely to have had a door linking it to the neighboring room to the west. This room R102 measured 8.5 × 3.7 m and was attached secondarily to room R103 from the west. Some scanty remains indicate that another three rooms of unknown dimensions were attached to its elongated western side. They were arranged perpendicularly to R102. Another room, R1005, also perpendicular to the eastern enclosure wall, was built to the north of the cluster of these five rooms, and its estimated 3 m width was covered with a barrel vault, too. It seems that this was the first building of the cluster to be constructed, which was then followed by R103, then R102, and finally R1006– R1008. Although their width varied, all three northern rooms (R100, R101, and R1104) seem to have ended in the same line both in the north and the south. Furthermore, it is also probable that the northern end walls of these rooms formed a continuous line with wall fragments w189–192, w1020, w193, and w196–197. The southern cluster of rooms (R103, R102, and R1006–1008) also seem to have shared a common end wall in their south that ran in line with w229 and w1023. If we take them as the northern and southern end of the cluster of buildings in this part of the fort, their distance measures about 29.5 m. If we take the same distance from the eastern enclosure wall, then we find that the straight wall remains of w208 and w215, with their postulated continuation w1010, running from north to south, might have formed the western border of this regular block of buildings. They are about 30m from the eastern enclosure wall. Wall fragments w209 and w210 might have been the remains of buildings filling the western side of this roughly 30×30m square. 3.3.2 The Northwestern Cluster of Buildings (See Fig. 28.5) Standing remains above ground level did not exist in this part of the fort, so most of what we know at the moment comes mainly from the excavation season of the autumn 2019, the continuous field surveys, and the study of the aerial photography. Excavations targeted the area behind the interval tower T5 on the western enclosure wall. Behind it a continuous line of rooms seems to have run with an internal width of 4.7m for most of the western side of the fort. This regularity seems only to have been broken in one area of the northwestern cluster,

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

578

figure 28.13

major

Excavation trench 2019/3 in the area of room R300, looking west photo b. major

immediately behind the northwestern corner tower. Here a smaller room seem to have been present, very likely because of the presence of the tower. A somewhat similar arrangement of smaller rooms behind the corner tower could be observed in the so-called Palace of Hisham building in Ruṣāfa (Syria)5 from the Umayyad period. Surface observations and the northwestern edge of excavation trench 2019/2 revealed the size of one of the rooms adjacent to the western enclosure wall. Room R301 had an internal area of 4.7×2.6m. The adjacent room from the south (R300) had the same width from east to west and was evidently much longer than R301; however, the excavations in 2019 did not find its southern wall. If its main doorway on its eastern wall was positioned in the center of the room, room R300 could not have been much longer than the currently excavated 6 m. The sounding that cut the center of the room on an east-west axis also shed some light on its history (see fig. 28.13). The room itself did not seem to have had any plastered or paved floor. It was dug into the hard and very compact ground soil. Close to its western wall (which was the enclosure wall itself) and eastern wall, a thick layer of mortar spill was

5 Sack, Resafa-Sergiupolis/Rusafat Hisham 39.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

579

the sasanian fort of pānkān

found that flowed into the room from the direction of the side walls while they were being constructed. These inclinations of construction debris on its sides were finally leveled with compact earth and seem to have been used without pavement, as mentioned above. The room also preserved on its western walls the imprint of a brick structure. Above this earthen floor a thick (sometimes more than 10cm) burnt layer was spread almost all along the floor. The deterioration of the site was also reflected in the debris of fallen wall plaster and mortar that accumulated beside the feet of the eastern and western walls of the room. Some activity during this period is reflected in the circle built of rubble stones in this debris layer, which possibly served as a fireplace. This debris of about 18cm thick was then covered by the 50–60cm thick layer of the collapsed vault, which contains a huge quantity of bricks that testify to the presence of brick vaulting in the Sasanian period. These bricks had the typical finger-drawn diagonal lines on their surface. The uppermost 1m thick layer consisted mostly of the debris of the fallen walls. From room R300 a 2m wide door opened into area S1001.

figure 28.14

Possible courtyard S1001 in excavation trench 2019/2 photo b. major

It measures 8.8×2.9m. As the floor of this area was constructed in a similar way to that of the gateway passage, namely having large rubble stones filled with mortar, it is very likely that S1001 was an open courtyard, as opposed to the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

580

major

covered rooms, which seem to have had earthen floors. This might point to the possibility that the interior arrangement of the clusters contained elements of the so-called bayt type, rooms arranged around a courtyard. The arrangement of the few rooms and door openings visible in the better-preserved northeastern cluster of buildings also hints at the possibility of different rooms opening from central spaces, some of which might have been uncovered courtyards. The area immediately to the east of S1001 has only been subject to limited excavations (2019/5), so we can only make some general remarks about it at the moment. The most important features in the unearthed part were the two stone-lined circular storage pits sunk into the ground level.

figure 28.15

Excavation trench 2019/5, with the openings of the two stone-lined storage pits (con100 and con101) photo b. major

They are about 1m apart and seem to have been in a covered area demarcated by a half-rounded (w332) and a round pilaster (w333), with walls attached to them. Their position makes it likely that the area immediately bordering them to the south might have been a courtyard (S1002). Both storage pits have a bottle-shape narrowing toward their circular mouth that just jutted above the estimated floor level. They were constructed of small rubble stones and large pebbles bonded with thick lime mortar. The western pit (con100) had a depth of more than 2m and was 2m wide close to its bottom. The

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

581

eastern one (con101) also had the remains of a preserved brick-paved curb. The brick itself was of the Sasanian type, with finger-drawn diagonal lines on its surface. None of the pits have been completely excavated yet, but even this showed some differences in their use. The western one (con100) had a thick ashy layer starting at a depth of 195cm, and above it one could see a rather uniform filling of rubble stone and earth. If this filling took place at one certain time, then this must have been at a rather recent date, as testified by the two perfectly preserved bronze bowls found at a depth of 97 cm from the rim of the pit. The inscription on one of them can be dated to the late Ottoman period. As opposed to the western pit, covered only with lime plaster, the interior of the eastern one (con101) also received a 0.5 cm thin, fine clay plastering. This was done to enable some fire-related activities. It is clearly shown by the intensely burnt clay, which looks like the interior surface of the traditional tannūr ovens used for baking. The relatively large amount of simple pottery fragments and the increasing number of animal bones proceeding toward the 1m depth of the excavated zone also indicate food-preparation activities in a later period in the pit. However, their original function is most likely to have been granaries for the fortified enclosure.

4

Around the Fort

Field surveys conducted around the fort showed clear evidence of human habitation, especially on the southern side in the foreground of the gate. Here, huge quantities of unglazed ceramics were retrieved, largely from the early Sasanian to the early Islamic periods (see fig. 28.16). Most of them belong to the “Fine creamware” type6 and were often ornamented with incised or stamped7 decorations. Also, a considerable number of the so-called honeycomb ware was present, which was regarded as typical Sasanian pottery, but is nowadays dated to the eight century and afterward.8 The nearly total absence of glazed pottery or the lack of clay pipes, so characteristic of neighboring sites of the Ottoman period, show that the settlement was abandoned by the late medieval and Ottoman periods. While the brick, and even the rooftile pieces found scattered around the fort could have originated 6 See Nováček, Melčák, Starková, and Amin, Medieval urban landscape 194–195. 7 Tamm et al., Ausgrabungen in Gird-I Kazhaw 112. Late-Sasanian ceramic with similar stamped motif was found to the northwest of the fort. 8 Kennet, Sasanian and Islamic pottery 80; Kennet, Decline of eastern Arabia 92.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

582

major

figure 28.16 Sample of fine creamware shards a. kocsis

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

583

the sasanian fort of pānkān

as spolia from the enclosure, the large amounts of adobe and storage jar shards clearly attest to the presence of houses in the foreground of the gate and to a lesser extent on the western side of the fort, too. The geophysical survey made in autumn 2014 in a 50 × 50 m grid resulted in some features that might be explained as the presence of possible adobe structures in the southern foreground of the fort. They are not aligned to the site and thus could predate the enclosure, but this is hoped to be verified by soundings.9 Coins were retrieved in large numbers during the surveys and were mapped with gnss. This also helps to draw the limits of possible human habitation around the fort (see fig. 28.17). It is interesting that while the number of coin finds inside the enclosure was rather low, even in the excavated areas, their occurrence was much higher around the fort, especially in the western foreground of the northern wall. On the southern edge of the supposed settlement a human burial was also excavated, with a considerable coin hoard in it. This contained 40 silver coins, ranging from the late Sasanian to the early Abbasid periods.10

5

Historical Sketch

The enclosure of Pānkān shows clear signs of military design, so it is quite probable that it was established as such at this strategically important crossroad. The radiocarbon dating gave a rather wide timespan, between 430 and 560, for its foundations, and the earliest datable coins at the moment are from the end of the sixth century. The construction of a fort, which might have served as a supply base for the Sasanian army, would most likely have occurred after Emperor Anastasius constructed the new Byzantine base of Dara in 505, which resulted in a new series of wars between the two powers.11 The complete lack of any decoration and the simple beaten earth floors of the hitherto excavated rooms, combined with such strongly utilitarian structures as the large storage pits, also hint at a simple military base. The earliest dated Sasanian coins are all from the time of Khusrau ii (590– 628). It is rather surprising that the number of Byzantine coins found is relatively high. They constituted of 19 folles and seven bronze coins that are likely to have belonged to the follis type. About half a dozen coins belong to the early

9 10 11

The survey was conducted by our colleague, Dr. Gábor Bertók. Takáts, Medieval burial. Maksymiuk, Geography of Roman-Iranian wars 22–23.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

584

figure 28.17

major

Map of the metal and ceramic finds of the 2016 survey r. lóki

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

585

the sasanian fort of pānkān

folles of Iustinian.12 Given the close vicinity to the east of Mosul, it is tempting to think that the large quantity of Byzantine coins might be connected to the physical presence of Byzantines or their troops. The only such attested case could have been in autumn 627, when Heraclius is said to have spent the second half of November in this region preparing for the Battle of Ninive with the Sasanian forces of Rhahzadh.13 What seems to be a half follis of Emperor Phocas (602–610) struck in Kyzicus might also date from this period.14 It is not infrequent to find a monastery inserted into an abandoned military base in the late antique period. This northern part of Mesopotamia had at least one good nearby example of such a structure, which was discovered in Bāzyān.15 By the time the enclosure at Pānkān was established, the initial hostility of the Sasanians toward Christianity ceased, and from the second half of the sixth century relations were normalized.16 Still, there are no real indications yet of any strong Christian presence at the site. The sole bronze pendant cross and another bronze cross of the “Nestorian type,” about 10 cm, were found on the surface of the site without any stratigraphic context. The bronze object (see fig. 28.18) and the handle of another identical one, the almost perfect parallel of which was found at Bāzyān and identified as an “incense burner,”17 can also not be taken as firm proof, for example, similar objects have been also variously interpreted as oil lamps18 or cosmetic mortars.19 It is more probable that after the Sasanian period the fort was used as a kind of caravanserai on the ever-important road from Arbela to the Great Zāb. This seems to be supported by the relatively high number of early Islamic coins, including silver ones found at the site. The surveys and the excavations also resulted in the finding of several heavily corroded lead castings. They were probably lead seals that can be taken as a good sign of commercial activity. The latest datable medieval coin is from the time of the Zangid atabeks of Mosul (1126–1234), and as pottery that could be dated to the later periods (especially the completely missing glazed types) is absent from the site, we can assume that it was abandoned by the end of the thirteenth century. This does not mean that life completely disappeared from the region, as some lead balls from the

12 13 14 15 16 17 18 19

Langó, Byzantine coins. Howard-Johnston, Last great war of antiquity 307. Langó, Byzantine coins. Déroche, Site of Bazyan 11–18. Dignas and Winter, Rome and Persia 225–226. Déroche, La fouille de Bazyan 34. Keall and Keall, Qalʾeh-i Yazdigird pottery plate Pl. via. Allan, Nishapur 37–38.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

586

major

figure 28.18 Bronze cosmetics mortar photo b. takáts

late Ottoman period and the two bronze bowls in the already half-filled western storage pit testify. However, the Sasanian enclosure seems to have been largely in ruins by then.

6

Conclusion

The Pānkān fort is a rare and relatively well-preserved example of the lessersized forts of the Sasanian period. The majority of the firmly dated examples were larger military camps along the great wall of Gorgān, which together with the lesser ones are also usually completely buried.20 The number of smaller fortified enclosures must have been much more considerable in this western part of the Sasanian territories21 than known now, and the published examples elsewhere show that they were not infrequent in other areas of the empire, too.22

20 21

22

Priestman, Forts, fortlets and watchtowers. For the pre-Islamic period in northern Iraq, an extensive collection of sites was given by Ibrahim, Pre-Islamic settlement 48–88; many of which might have been intended as part of an in-depth defence system, like Khirbat Jaddala, see Foietta, Khirbet Jaddalah 267– 272. See the smaller, lesser-investigated sites of allegedly Sasanian date around Uruk in southern Iraq. Finster, Sasanidische und Frühislamische Ruinen 164–167; or the late Sasanian

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

587

The use of these structures or their characteristic architectural forms might have continued well into the Muslim era in the region.23 The influence of Sasanian architecture on early Muslim architecture, especially on palaces, has been studied to some extent,24 but the Romano-Byzantine impact is usually more frequently dealt with.25 The fortified enclosure in Pānkān shows a number of elements that were practically missing in the forts of the eastern provinces of Rome and Byzantium but are not infrequent in the architecture of the Umayyad quṣūr, the most iconic structures of the new Islamic empire. One such example could be the semicircular gate towers or gate tower flanked by buttress towers in a shape of a quarter segment of a circle, for which we have numerous Umayyad examples, like the eastern qaṣr of Umm al-Walīd (Jordan) or the western qaṣr of Khān al-Zabīb (Jordan),26 in addition to the earlier mentioned examples. This form also appears in early Abbasid enclosures, like the “Nordostkomplex” in Raqqa27 (Syria), and seems to have been applied at lesser-scale structures and even examples of domestic architecture, like in the case of building no. 5 (the “caravanserai”) and residential building no. 1 in Jumayra (United Arab Emirates) dated to the ninth–eleventh centuries.28 Another rather oriental feature seen here in Pānkān was the filling of at least two-thirds of the interior with buildings. Military barracks and other similar structures in the Romano-Byzantine territories were almost exclusively restricted to the rear of the walls of the castra, leaving most of the interior of the enclosure free for a courtyard. Here we also have indications of possible lesser courtyards inserted between the buildings. Heavily overbuilt courtyards can be seen at many Umayyad quṣūr of the Levant from the enclosure in Bālis (Syria)29 to the iconic Qaṣr al-Mushatta (Jordan).30 There are examples further to the east at Tulūl al-Shuʿayba (Iraq) or even as far as the fort of Akyrtash (Kazakhstan).31 Later, the early Abbasid qaṣr of Ukhaydir in Iraq combined both

23

24 25 26 27 28 29 30 31

fort of Fulayj (Oman) that covered 30 × 30 m, with circular corner towers and circular tower buttresses beside its single gate. Jahwari et al., Fulayj 724–727. See the impressive collection of potential sites in the area of Kona Makhmūr, enumerated in Nováček et al., Medieval urban landscape 102–159. On the continuity of layout and architectural forms, see Kennet, On the eve of Islam 115. See for example, Stern, Notes sur l’architecture 82–97; Bier, Sasanian palaces 56–66. Genequand, Umayyad castles. Genequand, Trois sites omeyyades 139–140, 147. Siegel, Al-Raqqa/al-Rafiqa 412. Qandil, Recent discoveries 318. Leisten, For prince and country(side) 386–387. Perlich, Qasr al-Mschatta. Northedge, Umayyad desert castles 256.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

588

major

features. It might be a coincidence, but the overall dimensions of the Pānkān fort are strikingly close to the 70×70m average area of the Umayyad castles of the “classic type.”32 It is hoped that the ongoing research of the KurdishHungarian Archaeological Mission will yield more data that might add to our knowledge on the material culture of the period.

Personal Note My first reading from the works of Prof. Hugh Kennedy was on the Crusader castles, back in 1996 as a university student, and it had a major influence on my work on the medieval castles of the Levant. Later, I relied intensively on his studies on Islamic history while teaching in the Arabic Department of our university. At these early times I did not imagine that beside his inspiring works I would also have the chance for close personal contact with him. Yet he was the opponent of my PhD at Cardiff University. I feel very lucky that I have the chance to know Hugh in person, and I am honored to have been invited to contribute to this volume dedicated to him.

Bibliography Al-Jahwari, N., D. Kennet, S. Priestman, and E. Sauer, Fulayj: A late Sasanian fort on the Arabian coast, in Antiquity 92 (2018), 724–741. Allan, J.W., Nishapur: Metalwork of the early Islamic period, New York 1982. Badrī Tawfīq, B., The discovery of the lost site of Dayr al-Juṣṣ in the village of Dīri Brīsh in Māwarān region, in Proceedings of the first international conference of Rwandz 19– 20 May 2016, Soran 2017, 339–347 (in Arabic). Bier, L., The Sasanian palaces and their influence in early Islam, in Ars orientalis 23 (1993), 56–66. Déroche, V., La fouille de Bazyan (Kurdistan Irakien): Un monastère Nestorien? in Routes de l’Orient revue d’archéologie de l’Orient ancien 1 (2014), 31–45. Déroche, V., and N. Amin Ali, The site of Bazyan: Historical and archaeological investigations, in K. Kopanias and J. MacGinnis (eds.), The archaeology of the Kurdistan region of Iraq and adjacent regions, Oxford 2016, 11–18. Dignas, B., and E. Winter, Rome and Persia in late antiquity: Neighbours and rival, Cambridge 2007.

32

Genequand, Umayyad Castles 8.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

the sasanian fort of pānkān

589

Finster, B., and J. Schmidt, Sasanidische und Frühislamische Ruinen im Iraq (Baghdader Mitteilungen 8), Berlin 1976. Foietta, E., Khirbet Jaddalah and its land: A study of the military landscape in the eastern part of the kingdom of Hatra (2nd–3rd cent. ad), in Thiasos 10 (2021), 255–272. Genequand, D., Umayyad castles: The shift from late antique military architecture to early Islamic palatial building, in H. Kennedy (ed.), Muslim military architecture in Greater Syria from the coming of Islam to the Ottoman period, Leiden 2006, 3–25. Genequand, D., Trois sites omeyyades de Jordanie centrale: Umm al-Walid, Khan alZabib et Qasr al-Mshatta, in K. Bartl and M. Abd al-Razzaq (eds.), Residences, castles, settlements: Transformation processes from late antiquity to early Islam in Bilad alSham (Orient-Archäologie 24), Rahden 2008, 125–152. Howard-Johnston, J., The last great war of antiquity, Oxford 2021. Ibrahim, J.K., Pre-Islamic settlement in Jazirah, Baghdad 1986. Keall, E.J., and M.J. Keall, Qalʾeh-i Yazdigird pottery: A statistical approach, in Iran 19 (1981), 33–80. Kennet, D., Sasanian and Islamic pottery from Ras al-Kaimah (Society for Arabian studies monographs 1), Oxford 2004. Kennet, D., On the eve of Islam: Archaeological evidence from eastern Arabia, in Antiquity 79 (2005), 107–118. Kennet, D., The decline of eastern Arabia in the Sasanian period, in Arabian archaeology and epigraphy 18 (2007), 86–122. Langó, P., Byzantine coins from the Sasanian site of Pānkān, forthcoming. Leisten, T.S., For prince and country(side): The Marwanid mansion at Balis on the Euphrates, in K. Bartl and M. Abd al-Razzaq (eds.), Residences, castles, settlements: Transformation processes from late antiquity to Early Islam in Bilad al-Sham (OrientArchäologie 24), Rahden 2008, 337–394. Maksymiuk, K., Geography of Roman-Iranian wars, military operations of Rome and Sasanian Iran, Siedlce 2015. Northedge, A., The Umayyad desert castles and pre-Islamic Arabia, in K. Bartl and M. Abd al-Razzaq (eds.), Residences, castles, settlements: Transformation processes from late antiquity to early Islam in Bilad al-Sham (Orient-Archäologie 24), Rahden 2008, 243–260. Nováček, K., M. Melčák, L. Starková, and N.A.M. Amin, Medieval urban landscape in northern Mesopotamia, Oxford 2016. Perlich, B., Qasr al-Mschatta: Ein Palast, in J. Cramer et al. (eds.), Qasr al-Mschatta ein frühislamischer Palast in Jordanien und Berlin, Petersberg 2016, 285–293. Priestman, S., The architecture of the Tammīsheh wall, in E. Sauer, H.O. Rekavandi, T.J. Wilkinson, and J. Nokandeh (eds.), Persia’s imperial power in late antiquity: The great wall of Gorgān and the frontier landscapes of Sasanian Iran, Oxford 2013, 252– 272.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

590

major

Priestman, S., Forts, Fortlets and watchtowers on the Gorgān wall, in E. Sauer, H.O. Rekavandi, T.J. Wilkinson, and J. Nokandeh (eds.), Persia’s imperial power in late antiquity: The great wall of Gorgān and the frontier landscapes of Sasanian Iran, Oxford 2013, 178–243. Qandil, H., Recent discoveries at Jumeirah, in D.T. Potts, H. al-Naboodah, and P. Hellyer (eds.), Archaeology of the United Arab Emirates: Proceedings of the first international conference on the archaeology of the u.a.e., London 2003, 318. Sack, D., Resafa-Sergiupolis/Rusafat Hisham—neue Forschungsansätze, in K. Bartl and M. Abd al-Razzaq (eds.), Residences, castles, settlements: Transformation processes from late antiquity to early Islam in Bilad al-Sham (Orient-Archäologie 24), Rahden 2008, 31–44. Siegel, U., Al-Raqqa/al-Rafiqa—die Grundrisskonzeption der frühabbasidischen Residenzbauten, in K. Bartl and M. Abd al-Razzaq (eds.), Residences, castles, settlements: Transformation processes from late antiquity to early Islam in Bilad al-Sham (OrientArchäologie 24), Rahden 2008, 403–412. Stern, H., Notes sur l’architecture des chateaux omeyyades, in Ars Islamica 11–12 (1946), 72–97. Takáts, B., A medieval burial and its coin hoard in Pānkān (Iraki Kurdistan), forthcoming. Tamm, A., J. Fassbinder, I. Hofmann, C. Fink, P. Borsdorf, R. Davtyan, E. Schmalenberger, L. Stier, B. Einwag, and A. Otto, Ausgrabungen in Gird-I Kazhaw (Iraqi-Kurdistan) 2015–2017, in Mitteilungen der Deutschen Orient-Gesellschaft zu Berlin 150 (2018), 89– 146.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 29

Negotiating the North: Armenian Perspectives on the Conquest Era Tim Greenwood

Over the last 50 years, the History attributed to Sebēos has attracted a good deal of scholarly attention.*1 Meticulous research undertaken in the 1960s and 1970s underpinned the creation of a critical edition, published by Abgaryan in 1979, and it was from this that Thomson prepared his translation for publication alongside Howard-Johnston’s commentary in 1999. Thirty years before, Kaegi had noted that Sebeos, “the late seventh-century Armenian historian,” interpreted the appearance of the Arabs through the prophecies of Daniel 7 and called for further research.2 This arrived in 1977 via Crone and Cook’s Hagarism, in which the “Armenian Chronicle written in the 660s and ascribed to Bishop Sebeos” was given a brief but prominent role in the opening chapter.3 Its inclusion in this “wonderfully provocative” book, as it has recently been termed by Vacca,4 established the History attributed to Sebēos firmly within the contours of the longstanding debate on the use of non-Islamic works for studying the nature and development of Islam in the formative period, whether in conjunction, in comparison, or instead of Arabic-Islamic sources. The many twists and turns of this debate will not be treated here, beyond noting that all historical compositions reflect the intellectual, social, and cultural contexts in which they were created; even contemporary sources tell their own stories in their own ways and for their own purposes, reporting, reshaping, and reimagining as required. But there is little doubt that this exposure launched

* A version of this paper was delivered at the conference Negotiation in conquest: Wars, treaties and recollections of the rise of the caliphate convened by Petra Sijpesteijn and Nynke van der Veldt at the University of Leiden, September 12–14, 2019, under an erc-funded project, Embedding conquest: Naturalising Muslim rule in the early Islamic empire (600–1000). I would like to thank the convenors for their kind invitation to attend that conference and acknowledge its role in shaping this paper. 1 Sebēos, Patmut‘iwn; Thomson and Howard-Johnston, History. Since the translation includes the page numbers of Abgaryan’s edition, only that edition will be cited. 2 Kaegi, Initial Byzantine reactions 146–149. 3 Crone and Cook, Hagarism 6–8. 4 Vacca, Fires of Naxčawan 324.

© Tim Greenwood, 2022 | doi:10.1163/9789004525245_030

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

592

greenwood

this Armenian composition into broader scholarly discourses. By virtue of its date of compilation—now generally accepted as 655 ce with brief updating scholia extending its coverage to the conclusion of the first fitna in 41/661— its unaltered state, and its remarkable breadth of historical vision, the History attributed to Sebēos is now considered to be one of the principal sources for the study of the seventh-century Middle East.5 It features in many historiographical surveys and has also contributed to a wide range of specific studies, including recent research into eschatology, governance, and construction activity in Jerusalem, and the composition of Sasanian royal history.6 Its popularity shows no sign of waning. The recent attention paid to the History attributed to Sebēos, however, has not resulted in a raft of new studies analyzing Armenia in the conquest era. Manandyan’s reconstruction of the series of campaigns and counterattacks that characterize the years after 19/640 remains essential reading, but it is now over 70 years old.7 In 1982, Martin-Hisard supplied an outline narrative that is notable for its contention that Armenians benefited from the new world order since their territories were no longer partitioned between Rome and Persia: “les conditions de la domination arabe étaient incontestablement plus légères que celles de tout autre régime antérieur.”8 Ter-Łevondyan’s 1986 article dealt briefly with the first raids but the principal focus of his research was caliphal Armenia in the eighth and ninth centuries.9 Kaegi’s chapter “Byzantium, Armenia, and Armenians” from 30 years ago remains the latest sustained treatment, and its perspective is self-evident.10 There has been no Armenian analogue to Robinson’s analysis of the conquest of Khuzistan, and recent surveys by Kennedy and Haldon do not offer specific studies of Armenia in the conquest era.11 The situation may now be changing. In 2017, Vacca published a groundbreaking monograph examining the construction of the caliphal north, primarily in terms of its Sasanian legacy as represented through Arabic and Armenian literature composed in the later ninth and tenth centuries.12 Her research also draws on the Armenian History of Łewond, which is treated as a work of the late eighth century and accorded particular value. Vacca sets out to analyze 5 6 7 8 9 10 11 12

Howard-Johnston, Witnesses 70–102. Shoemaker, Prophet has appeared 62–72; La Porta, Sense of an ending 364–72; Hoyland, History 15–18. Manandyan, Invasions arabes. Martin-Hisard, Domination 216. Ter-Łewondyan, L’Arménie. Kaegi, Byzantium 181–204. Robinson, Conquest; Kennedy, Great Arab conquests; Haldon, Empire. Vacca, Non-Muslim provinces.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

593

“how people reading and writing in Arabic and Armenian wanted the North to be understood, not to describe the North as it actually was.”13 Indeed, Vacca observes judiciously that “Writing the history of the conquest- and Sufyanidera North wie es eigentlich gewesen […] needs to start with a close look at the expectations, goals, and concerns embedded in our sources,” and this monograph does much to stimulate that research.14 Shunning narrative in favor of a thematic approach allows Vacca to examine Arabic, Armenian, and Georgian sources in comparison and explore how and why traditions about the same episode changed over time, without having to prefer one over another. Vacca is interested in their individual representations of the past rather than trying to establish what happened. As a result, her research expands our knowledge of ninth- and tenth-century attitudes to, and perceptions of, the history of the caliphal north and Armenia’s place within it. At the same time, however, it reduces—at least for the present—the history of Armenia in the conquest era to “a few simplified generalisations that find support in both Armenian and Arabic historical traditions”: the arrival of Muslim troops in the Rāshidūn period; the peace treaties that left the north as “a tributary neighbor, loosely affiliated on and off with one of its two powerful neighbors, the Caliphate or Byzantium”; and the incorporation of Armenia as a caliphal territory following the reforms undertaken during the caliphate of ʿAbd al-Malik at the turn of the eighth century.15 This study accepts that the sources for the history of Armenia in the conquest era present multiple challenges. They were composed at different times and in different social, cultural, and intellectual contexts, shaping the past in various ways and for various purposes. None of them are simple vehicles for the preservation of neutral reports on what happened; all are freighted with individual meanings. Nor should we accept that contemporary sources necessarily offer greater historical accuracy than later accounts; as observed above, they possess their own interpretative frameworks and conform to attitudes and genres of the day. Furthermore, the tendency to classify sources on the basis of language has served to conceal important connections between them. By way of illustration, although the Armenian character and language of the History of Łewond imparts an impression of otherness for modern scholarship, closer examination reveals that this composition was influenced by both Armenian and non-Armenian historical traditions.16 Far from being conceived 13 14 15 16

Vacca, Non-Muslim provinces 18. Vacca, Non-Muslim provinces 19. Vacca, Non-Muslim provinces 38. Greenwood, Reassessment of the History 102.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

594

greenwood

and written in historiographical isolation, therefore, this work was a product of cross-cultural engagement and acculturation. In other words, simply being composed in Armenian does not guarantee independence from other literary cultures; the Armenian witnesses may not be the independent controls they have sometimes been treated as. It is certainly not the intention here to offer a new narrative of Armenia in the conquest era. As Vacca observed, the current state of research on the individual sources precludes such a work. At the same time, however, there may be ways of expanding the brief outline supplied by Vacca, particularly if we are prepared to accept that correspondence between the Arabic-Islamic and Armenian historical traditions, while significant, is not the only criterion that may be applied. Other histories of conquest-era Armenia may be constructed if we employ different variables to determine selection. This study introduces one such approach. It focuses on contemporary Armenian-language sources, some familiar, others little known, and proposes that these could be used collectively to construct such a history. Two case studies are then presented. The first analyzes relations between Armenia and Iran in the 630s and 640s; the second reassesses the treaty established in 653ce between Muʿāwiya and T‘ēodoros, lord of Ṙštunik‘ and preserved exclusively in the History attributed to Sebēos. If the first illustrates the potential of this material, the second reveals some of its challenges.

1

Armenian Sources for the Conquest Era

Let us start by acknowledging that far from being a single event, the ArabIslamic conquest of Armenia was a complex process that spanned seven decades, from the first attested raid, probably in the autumn of 640 ce, down to the eclipse, although not exclusion, of Byzantine interests and connections across Armenia in the years after 705ce. This complexity takes many different forms. It is impossible to speak of a single Armenian experience over this period of time. While some of the remote, mountainous districts of Armenia escaped much of the conflict, those situated on the major routes through Armenia experienced multiple campaigns, often in quick succession, as the struggle for control of these routes—and the urban centers and fortifications along them—ebbed and flowed. This complicates attempts to establish a definitive chronology, especially in the 640s and 650s. It is also clear that raiding parties advanced from both Syria and northwestern Iran at different times and almost certainly independently of one another. Moreover, Armenia was a world of local lordships characterized by tension and conflict within as much

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

595

as between the individual noble houses. It should come as little surprise to discover that local lords fought for and against the raiders from the start. We learn from the History attributed to Sebēos that the first raiding party came from Syria in the autumn of 640ce and entered the region of Tarōn through the Bitlis pass.17 It was guided by Vardik, the prince of Mokk‘. Since this route from Syria into Armenia passed through the district of Mokk‘, Vardik’s actions may have been determined by self-preservation as much as anything else, although we have no way of knowing what motivated his decision to assist rather than resist. It is likely that the Armenian elite fractured during every campaign thereafter, even if the specific details largely elude us. We can also see that Armenian nobles switched sides as the conflict unfolded and circumstances changed. By way of illustration, Hamazasp Mamikonean was the son-in-law of T‘ēodoros, lord of Ṙštunik‘, one of the leading figures in Armenia after 628ce, and was with him on the island of Ałt‘amar in Lake Van in 653 ce, after T‘ēodoros had deserted Constans ii and transferred allegiance to Muʿāwiya.18 Barely two years later, however, Hamazasp switched back to Constans ii, receiving the title of curopalates to denote his role as principal client.19 Conversely his nephew Mušeł Mamikonean submitted to Constans ii in the city of Karin/Theodosiopolis in 653 as the emperor rushed eastward to shore up his support following the defection of T‘ēodoros but had gone over to the Ishmaelites (as they are termed in the text) by early 655, reportedly because four of his sons were being held hostage.20 Hamazasp’s brother was also being held hostage but evidently this was not enough to sway his allegiance. This level of detail is exceptional. It reveals the vulnerability of the elite at this time, individuals twisting this way and that to protect themselves and perhaps obtain advantage over rivals. The zig-zag patterns of submission and betrayal by Hamazasp and Mušeł Mamikonean look like mirror-images of one another, although it is not clear which of them took the lead and whether this compelled the other to respond. Such family dynamics are almost always hidden from view, but it is likely that other noble houses were equally split. Mindful of the other complexities noted above, it is never going to be possible to work out exactly what happened and establish a comprehensive narrative history of Armenia in the conquest era. On the other hand, we do possess a range of contemporary Armenian sources that offer different insights. Where should one start? 17 18 19 20

Sebēos, Patmut‘iwn 138. Sebēos, Patmut‘iwn 169. Sebēos, Patmut‘iwn 175. Sebēos, Patmut‘iwn 165, 173, 175.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

596

greenwood

As Howard-Johnston and others have shown, the two richest Armenian historical compositions for the study of Armenia in the conquest era are the History attributed to Sebēos and the History of Ałuank‘/Caucasian Albania attributed variously to Movsēs Dasxuranc‘i or Movsēs Kałankatuac‘i.21 Since the first of these histories has been discussed above and is the more familiar of the two, this study will focus on the latter. Although this work reached its present form in the tenth century, it is now widely accepted that it is a compilation that preserves material dating from much earlier, including the seventh century. There is, however, no consensus on the number or nature of its underlying components, nor when they were pieced together. Akopyan identified four postulated sources: a history of Viroy, the Catholicos of Ałuank‘, who died in 629; a History to 684; a panegyric on prince Juanšēr; and an account of the mission of Bishop Israyēl to the Huns and their leader Alp‘ Iłit‘uēr in the winter of 681–682 ce.22 Howard-Johnston also proposed four clusters of material, although he defined the sources somewhat differently.23 He also maintained that all the material had been collected together by a single compiler active in the late seventh century in a work he titled the History to 682. While the individual components remain contested, prevailing scholarly opinion holds that they were all composed in the seventh century.24 Admittedly, its value for studying Armenian history has been questioned. Garsoïan, among others, noted the Albanian focus of the material and queried its relevance.25 Nevertheless, its record of contemporary conditions across the Caucasus offers useful comparison, and it also contains important insights into conditions inside Armenia. By way of illustration, when Prince Juanšēr switched allegiance from Constans ii to Muʿāwiya, probably in 664–665ce, he traveled to and from Damascus via Grigor Mamikonean, the leading prince in Armenia at the time and Muʿāwiya’s principal client.26 It is striking that Juanšēr is not described as attending upon an Arab governor or ostikan. Arguably, this was because no such figure had been appointed; it is only in the last decade of the seventh century that Armenian sources indicate a permanent Arab-Muslim presence in the north. Finally, although most discussion of the History of Ałuank‘ for the study of this era has centered on the 21

22 23 24 25 26

Howard-Johnston, Witnesses 70–128. For once, the remarkable collection of early medieval Armenian ecclesiastical correspondence and documentation known as the Girk‘ T‘łt‘oc‘ or Book of letters has nothing to contribute. So vital for earlier and later eras, it lacks records dating from the century after 608 ce. Akopyan, Albania-Aluank 197–207. Howard-Johnston, Witnesses 108–113; Howard-Johnston, Caucasian Albania 357–358. Zuckerman, Khazars 407–410. Garsoïan, Interregnum xiii. Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn ii.27.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

597

contents of Book ii, it is worth noting that Book iii also preserves valuable documentary records from the first decade of the eighth century.27 These reveal the circumstances and the decisions of the Council of Partaw/Bardh‘a, convened in 705ce, and attest interventions by both the Armenian Catholicos Ełia and ʿAbd al-Malik. Therefore, the History of Ałuank‘ has much to contribute to our understanding of conquest-era Armenia. On the other hand, the third Armenian historical narrative usually consulted when approaching the conquest of Armenia, the History of Łewond, merits more circumspect treatment.28 Whether it is a product of the late eighth century or the late ninth century, this composition is separated from the start of the conquest era by at least 150 years.29 Recent research has confirmed that its version of events in the 640s and 650s derives from the narrative preserved in the History attributed to Sebēos.30 Rather than supplying a simple précis, however, it constitutes a sustained reworking of that material in which the chronological sequence of the raids has been altered and much of the specific detail suppressed. Conversely, the description of the assault on Duin has been developed and elaborated. It has been proposed that these changes had two principal aims: to establish from the outset the untrustworthiness of Byzantium in its dealings with Armenia; and to represent the experience of the citizens of Duin during the conquest era as normative for all Armenians. Forgetting Byzantium and reimagining the role of Duin lie at the heart of Łewond’s record of the conquest era. His History constitutes a much more sophisticated work than has previously been acknowledged. Many studies in the past have utilized Łewond’s account when establishing their own conquest narratives, but they have treated it as a simple record of what happened. These will remain embedded in the scholarship for many years to come, but their reconstructions need to be handled with caution. Two of the three historical compositions, therefore, should be placed at the heart of any study of seventh-century Armenia. There are, however, several other sources that offer valuable insights, and this study will introduce three of them. Perhaps the least studied comprises a set of Armenian church canons recording the decisions of a council held in Duin “in the fourth year of

27 28 29 30

Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn iii.3–11. Łewond, Patmabanut‘iwn. For a recent French translation, see Martin-Hisard, Łewond vardapet. A new English translation by Vacca and La Porta is eagerly awaited. For the traditional dating, see Martin-Hisard, Łewond vardapet 237–260; Greenwood, Reassessment of the History 104–121. Łewond, Patmabanut‘iwn cc. 2–8; Greenwood, Reassessment of the History 133–142 and 150–153.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

598

greenwood

Constans, emperor and pious king of the Romans […]” (644–645ce).31 This addresses a series of pastoral rather than theological issues, including the seizure of church property by the azatk‘, a term defining those members of the lay elite who enjoyed exemption from certain taxes. Three of the canons seem to capture something of the conquest era. Canon 12 contemplates with dismay the billeting of the azatk‘ and cavalrymen in village churches and religious communities and their pollution of the hallowed places of God with “minstrels and dancers.”32 Evidently these sacred spaces had been occupied and transformed into sites of entertainment, of storytelling, performance, and dance, presumably accompanied by feasting. The implication is that quite apart from the depredations of raiding parties, local society had also been disrupted by the actions of Armenian soldiers. Canon 7 acknowledges that many men and women had been taken captive when the country was seized by enemies and sets a seven-year rule for those wanting to remarry, with financial penalties and penance for those who remarried before the seven years had elapsed; it also admits the possibility that captives might return.33 In addressing the issue of lay confiscation of church property, Canon 9 observes that the soil and water belonging to the holy church had also been declared to be free (i.e., exempt from certain taxes), just as that belonging to the azatk‘.34 It further explains that during the lordship of the Persians, although the houses of the priests had been established in the diwan—here meaning state archives created for fiscal and administrative purposes—this had only been for the purpose of providing services (hask‘) to the royal estate, and it is implied they were otherwise free. It seems therefore that Armenian communities were being threatened in various ways, with people taken captive, churches requisitioned, and church land confiscated. The solutions offered by the bishops looked to precedents from the past, when Armenia had been under Persian hegemony. Evidently this past was still meaningful and relevant; it had not yet been cast aside. Two other bodies of evidence merit brief comment. First, there is a small corpus of nine seventh-century inscriptions commemorating the foundation of churches between 629 and ca. 695 ce.35 Three of them are dated by reference to the regnal years of Heraclius and on the basis of the epithets applied to the emperor, it seems that these dates derived from the protocols employed

31 32 33 34 35

Hakobyan, Kanonagirk‘ ii, 200–215, at 200: Յաﬕ չորրորդի Կոստանդիանոսի կայսեր բարեպաշտ թագաւորի Հոռոմոց […]. They lack a published translation. Hakobyan, Kanonagirk‘ ii, 212: գուսանաւք եւ վարձակաւք. Hakobyan, Kanonagirk‘ ii, 205–206. Hakobyan, Kanonagirk‘ ii, 209–210: եւ ազատ էին սրբոյ եկեղեցւոյ հող եւ ջուր. Greenwood, Corpus.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

599

in imperial correspondence.36 These inscriptions therefore attest the receipt and retention of imperial correspondence in Armenia. Another inscription, at Aruč, is dated by reference to the 29th year of Constans ii, which corresponds to 669–670ce.37 This church was founded by Grigor Mamikonean, the same figure who was identified above as the principal client of Muʿāwiya. This choice of chronological marker is therefore intriguing. Furthermore, five of the inscriptions, including the latest in the sequence at T‘alin, reveal that the individual founders chose to identify themselves, and sometimes others, by reference to specific Byzantine honorific titles.38 This evidence challenges Łewond’s projection of an ineffective and distant Byzantium, quickly forgotten. Instead, it seems that generations of Armenians during the conquest era obtained recognition from Byzantium, together with the gifts and wealth attendant on that relationship. They preferred to use, or were untroubled by the use of, imperial regnal years for dating purposes. A modest collection of ten seventh- and early eighth-century colophons complements the epigraphic evidence.39 They attest similar forms of engagement, including the grant and use of Byzantine titles, the use of Byzantine regnal years, and translation activities in Constantinople. Their potential value may be illustrated through one colophon preserved in a thirteenth-century manuscript containing a collection of 26 homilies and seven letters of Basil of Caesarea.40 Having recorded 18 homilies, at fol. 156a, the manuscript contains the following: “Davit‘ Tarōnec‘i, a translator, turned these homilies from Greek into the Armenian language, in the city of Damascus, at the command of Hamazasp curopalates and lord of the Mamikoneans.” The reference to Hamazasp Mamikonean as curopalates dates Davit’s presence in Damascus to between 655 and ca. 660ce, and so within a decade of Juanšēr’s visit to the same city, discussed above. One can only speculate why Damascus, Muʿāwiya’s center of operations and place of residence, was chosen as the place to look for a translation of these works. It is not a recognized or otherwise attested site of Armenian translation activity, and we may suspect that Davit‘ had other reasons for being there at the behest of Hamazasp. Although Hamazasp had already switched sides twice by this time, and was the leading client of Constans ii, it seems he continued to maintain a connection with Damascus. It is only when this colo-

36 37 38 39 40

Greenwood, Corpus A.4 (Ałaman), A.5 (Bagawan), and A.7 (Mren), and 44–47. Greenwood, Corpus A.11 (Aruč) and 48–50. The dating of the Council of Duin to his fourth year was noted above. Greenwood, Corpus A.4, A.7, A.8 (Naxčawan), A.11, and A.12 (T‘alin). Mat‘evosyan, Hišatakaranner nos. 23–25, 28, 29, 31–34; Hovsep‘ean, Yišatakarank‘ no. 17. Mat‘evosyan, Hišatakaranner no. 23. The manuscript is M822 (Matenadaran).

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

600

greenwood

phon is analyzed in the context of the evidence from the History attributed to Sebēos that its wider implications for the study of the conquest era are revealed. This illustrates how individual contemporary Armenian sources can be used in comparative analysis to shed new light on Armenia in the conquest era. Let us now turn to the two case studies.

2

Armenia and Iran in the 630s and 640s

Despite the widely recognized hegemony of Sasanian Iran across four-fifths of the districts of historic Armenia from the second quarter of the fifth century— punctuated by brief periods of Roman ascendency, notably between 590 and 607ce—there has been little research into Iranian engagement with Armenia after the demise of Khusro ii in February 628. Instead, attention has been focused on two other perspectives: the expansion of Roman influence and control into Armenia during the reigns of Heraclius and then Constans ii, and the advent of the Arab-Islamic raids. This study contends that members of the Armenian elite continued to engage with the world of Iran and that elements within the Iranian elite continued to engage with Armenia, at least until the early 640s. These relationships are charted in several densely packed passages preserved in the History attributed to Sebēos. The first of these reveals that Khusro ii’s successor, Kavad ii, appointed Varaztiroc‘ Bagratuni as marzpan of Armenia as well as recognizing him as tanutēr, head of the Bagratuni house with control over the ancestral lands, acting in effect as the head of a family trust.41 Indeed, the notice in the text offers an important gloss on the legal significance of being appointed as tanutēr, commenting that Kavad “sent him to Armenia [with authority] over all his ancestral possessions.”42 This occurred in spring or summer 628, before Kavad’s untimely death in September. His appointment as marzpan is confirmed by an inscription at Bagaran which records that the church was completed on October 8, 629, while Varaztiroc‘ was marcpan [sic.] and aspet of Armenia.43 His tenure therefore extended for at least one year and did not end with Kavad’s death. On his arrival in Armenia, the History attributed to Sebēos tells us that Varaztiroc‘ set about appointing a successor to Komitas as Catholicos of Armenia, just as his father Smbat had

41 42 43

Sebēos, Patmut‘iwn 128–129. Sebēos, Patmut‘iwn 129: արձակէ ի Հայս հանդերձ աﬔնայն ընչիւք հայրենի. Greenwood, Corpus A.3. The variant form of marcpan is unique.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

601

negotiating the north

overseen the appointment of Abraham, bishop of Ṙštunik‘, as Catholicos on his return to Armenia in 607ce.44 We learn that K‘ristap‘or, a relative of Abraham, was appointed through the direction or leadership of T‘ēodoros, lord of Ṙštunik‘. This is the first reference to T‘ēodoros, and it is significant that he is depicted acting in concert with Varaztiroc‘. We discover in a later passage that T‘ēodoros and Varaztiroc‘ had been brought up together at the court of Khusro ii and greeted one another with tears of friendship when they met in Constantinople after Varaztiroc‘ was brought back from exile.45 Arguably, therefore, when Varaztiroc‘ was sent to Armenia as marzpan in 628, he was accompanied by, or renewed his relationship with, T‘ēodoros; it seems clear that they worked together. If this is correct, T‘ēodoros’s political career began in service to Sasanian Iran; later, he transferred allegiance, first to Constans ii and finally to Muʿāwiya. This career profile matches that of Juanšēr, prince of Ałuank‘, although the chronologies are different. The History attributed to Sebēos does not reveal exactly how long Varaztiroc‘ remained marzpan of Armenia, nor if a successor was appointed. We are told that he refused to submit to the great prince of Atrpatakan, Xoṙox Ormizd/Farrukh Hormizd.46 This title appears to be an Armenian rendering of the Middle Persian spāhbed kust ī Ādurbādagān, one of the four principal military commands of the late Sasanian era, now securely attested through sigillographic evidence.47 Farrukh Hormizd briefly served as hramatar (mp framādār), the chief minister of Queen Bor/Bōrān following her accession in 630, but he was murdered within a year.48 Nor did Varaztiroc‘ submit to Farrukh Hormizd’s son Ṙostom/Rustam, who succeeded him as spāhbed of Ādurbādagān. Sebēos’s narrative refers to a great altercation between Varaztiroc‘ and Rustam, although no further details are provided.49 Nevertheless, it seems highly likely that these events occurred during the civil war that convulsed the Iranian state and that Varaztiroc‘ and Rustam were supporting different claimants. Intriguingly, the Roman commander Mžēž Gnuni is alleged to have antagonized the situation, slandering Varaztiroc‘ to Rustam and suggesting that he should be removed; there is no way of knowing if this was the case. Rustam sent his brother (later named in the narrative as Xoṙoxazat/Farrukhzād) to go and winter in Duin as darik‘pet (mp darīgbed, head of the palace) in the course

44 45 46 47 48 49

Sebēos, Patmut‘iwn 129 (Varaztiroc‘), 100 (Smbat). Sebēos, Patmut‘iwn 143. Sebēos, Patmut‘iwn 132: իշխանին ﬔծի որ յԱտրպատական աշխարհին. Gyselen, Géographie 269–277. Sebēos, Patmut‘iwn 130. Sebēos, Patmut‘iwn 132.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

602

greenwood

of which he was instructed to seize the aspet Varaztiroc‘.50 Farrukhzād’s precise responsibilities remain unclear, but underlying this appointment is the assumption that some part of Armenia was under Rustam’s control. When Varaztiroc‘ was warned of his imminent arrest—apparently he was informed that he would be seized the following day, implying that he was resident in the city of Duin when Farrukhzād took up his appointment—he fled with his wife and children westwards to the district of Tarōn. Following assurances of safe conduct, he then went to meet Heraclius in person in northern Syria. In the winter of 630–631, Heraclius was in Hierapolis (modern Manbij) negotiating with Athanasius, the miaphysite patriarch of Antioch, and this supplies a credible date and location for this meeting.51 Following the meeting, Varaztiroc‘ and his family traveled to Constantinople, where he was honored with residences, silver thrones, and riches, but he was not sent back to Armenia. We do not know if he was accompanied by T‘ēodoros, lord of Ṙštunik‘; indeed, the latter’s whereabouts, actions, and loyalties remain unknown for the rest of the decade. This first phase therefore opens in 628ce with the appointment of Varaztiroc‘ as marzpan, at a time when there was a recognized šahanšah. It ends in early 631 with Varaztiroc‘ withdrawing from Armenia and seeking refuge with Heraclius. It would, however, be wrong to assume that Iranian connections with Armenia ended at this time. The History of Ałuank‘ offers important insight into a second phase. It records the circumstances in which a young Albanian prince, Juanšēr, went to war in the service of the Sasanian šahanšah Yazdgird iii in the autumn of 637. It tells us that when he reached “the common ground where thousands were mustered” with his contingent of Albanian cavalry, he discovered that he had arrived before the prince of Siwnik‘ and the sparapet of Armenia.52 Armenian, Siwnian, and Albanian forces were assembled in one location, placed under the command of Rustam, the spāhbed of Ādurbādagān, and led south, first to Ctesiphon and then on campaign to confront the Arab-Islamic invaders. It is only through the History attributed to Sebēos that we learn that the Armenian general who attended on Rustam was Mušeł

50

51 52

Sebēos, Patmut‘iwn 132. For the title, see Gyselen, Géographie 264–265. The title darīgbed is remarkably similar in meaning to the Greek curopalates. For another contemporary reference, presenting the darīgbed as the leading figure at the royal court, see Greenwood, Reassessment of the life 165. Howard-Johnston, Last great war 375. Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn ii.18: Որոյ առեալ զգունդն ի հաւրէն նախքան զիշխան Սիւնեաց ե զսպարապետն Հայոց յառաջէր հասանել ի համաչխարհականն յայն բիւրաւոր ժողովս:

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

603

Mamikonean, with 3,000 fully armed men, and that he was joined by Grigor lord of Siwnik‘ with a further 1,000 troops.53 In the subsequent battle, at Qādisiyya on January 6, 638, Rustam, Mušeł, and Grigor were all killed, together with two of Mušeł’s nephews and one of Grigor’s sons, while Juanšēr was seriously wounded in the fighting.54 It seems therefore that this general mobilization of Caucasian military power was an organized and familiar process. Following a call to arms, contingents were raised and sent to a common meeting place for onward transit under Persian command. Even if this is not the case, and the description reflects what should have happened or what the author imagined to have happened, the passages show that Armenians, Siwnians, and Albanians fought and died for Yazdgird iii in early 638. The History of Ałuank‘ maintains that Juanšēr fought for seven years and only returned home after the defeat at Nihāvand, in late 642 or early 643. Evidently, members of the Armenian and Albanian elite were still integrated into the political structures and military processes of Sasanian Iran. Despite the internal conflicts at the start of the 630s, at least some features of the Iranian state were still being exercised in Armenia and Albania in 637. Yet even the disaster at Qādisiyya does not mark the end of Iranian engagement with Armenia, as a third passage in the History attributed to Sebēos reveals.55 This third phase in the narrative is entangled and some elements are presently unresolved—was T‘ēodoros the Armenian prince who is titled the Greek commander the same person as T‘ēodoros, lord of Ṙštunik‘, or were there two figures called T‘ēodoros?56 The dating of events is also hard to establish. At some point early in the reign of Constans ii—usually identified as during his fifth year, so 645–646, but perhaps as early as 642–643—one T‘umas was sent to Armenia. He united all the Armenian princes and went to the prince of the Medes to negotiate peace with him on their behalf; evidently there was already a pact of some kind between this figure and the Roman emperor. The prince of the Medes should almost certainly be identified as Rustam’s brother and successor as spāhbed of Ādurbādagān, Xoṙoxazat/Farrukhzād. It is less clear whether this title, prince of the Medes, reflects a self-designation, an expression of the territory under his control—the historic region of Media—or an eschat-

53 54 55 56

Sebēos, Patmut‘iwn 138. Sebēos, Patmut‘iwn 138; Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn ii.18. Sebēos, Patmut‘iwn 143–145. Most commentators have seen two figures, but it is striking that while T‘ēodoros the Greek commander (զաւրավարն Յունաց) is depicted interceding with Constans ii for Varaztiroc‘ to be released from exile, it is T‘ēodoros, lord of Ṙštunik‘, who embraces Varaztiroc‘ warmly when they meet in Constantinople after his release.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

604

greenwood

ological interpretation of his identity devised by the author of the passage, as the inferior successor to Nebuchadnezzar described in Daniel 2.57 According to the passage, T‘umas promised Farrukhzād he would arrest T‘ēodoros, lord of Ṙštunik‘, then acting as prince of Armenia, and have him conveyed to Constantinople; this duly happened. However, on his arrival, T‘ēodoros was exonerated. Two features of this episode stand out: first, the hostilities, which are otherwise unreported, pitted the prince of the Medes against Armenian princes— probably under the leadership of T‘ēodoros, since he is titled prince of Armenia—but did not involve the Romans, who seem to have been developing relationships with both parties; and second, the authority of the prince of the Medes, at least in Armenia, collapsed very soon after this. This is supported by the decision to appoint Varaztiroc‘ Bagratuni as curopalates and prince of Armenia despite the fact that he had recently fled from Constantinople and taken refuge in Armenia.58 As we have noted above, Varaztiroc‘ had gone to Heraclius a decade earlier when warned that Farrukhzād was about to arrest him in Duin. His appointment as principal Roman client while Farrukhzād was still active in Armenia would have been, at the very least, inflammatory. If, however, Farrukhzād had been forced to pull out of Armenia and focus his attention elsewhere, it is possible that Varaztiroc‘ became an attractive choice to use as a client, someone through whom Roman influence could be extended eastward into regions that had, until that point, been under Farrukhzād’s hegemony. Varaztiroc‘ had experience governing Persian Armenia as marzpan a decade before, as well as ties of friendship with other Armenian nobles, some of which went back to his upbringing in Ctesiphon. Constans ii and his advisers were looking for someone who was unlikely to go over to Farrukhzād, and Varaztiroc‘ Bagratuni was such a figure. Furthermore, after Varaztiroc‘ death, Constans ii appointed T‘ēodoros, lord of Ṙštunik‘, as commander of Armenia, someone who had recently opposed Farrukhzād, to the extent that the latter had engineered his removal. Evidently, control over Armenia remained divided down to this time, with some parts still under Iranian control, albeit a regional power based in Ādurbādagān rather than an imperial power based in Ctesiphon. The appointments of Varaztiroc‘ and then T‘ēodoros by Constans ii therefore mark

57

58

Sebēos, Patmut‘iwn 143: իշխան Մարաց. See also Sebēos, Patmut‘iwn 172, describing the land of Media, աշխարհին Մարաց, as comprising the deep forested valleys, cliffs, and rocky places of the river Gaz and the mountains of Media, where the people of Geln and Delumn lived, in other words of Gēlān and Dēlamān/Daylam. These are the western Alburz mountains, south and west of the Caspian Sea. Sebēos, Patmut‘iwn 144.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

605

negotiating the north

the moment when Iranian political hegemony over Armenia waned and the traditional bipartite character of Armenia, divided between Roman and Persian sectors for two centuries, dissolved. The testimony of two other sources support this proposition. The History of Ałuank‘ reveals that when Juanšēr returned from fighting for the šahanšah Yazdgird iii, probably in 643ce, and retired to his own country, the “Persian commander” as he is termed, urged him to marry his sister.59 It is highly likely that this was none other than Farrukhzād. Juanšēr refused on the grounds that she was an unbeliever, and hostilities ensued, in the course of which he was forced to abandon the city of Partaw to the Persians and retreated to the hills.60 After a series of campaigns, reported in great detail but all undated, the Persian commander sought to make peace with Juanšēr, employing the great prince of Siwnik‘ as his intermediary. Arguably, it was this warfare in Caucasian Albania that pulled Farrukhzād away from Armenia and enabled the extension of Roman influence. The use of the reign of Constans ii as the chronological marker in the Canons of Duin, noted above, also provides indirect support for this Roman expansion. This is the earliest extant Armenian ecclesiastical document to be dated by reference to the reigning Roman emperor. Armenia therefore remained contested space throughout the 630s and early 640s, as Roman and Iranian interests interacted and competed for influence among the local elite. The narrative is complex and incomplete, reflecting the changing fortunes of the šahanšah and the emerging but short-lived regional power of the spāhbed of Ādurbādagān. Accepting the above reconstruction, the first Arab-Islamic raid into Armenia in autumn 640 therefore occurred while Armenia was still partitioned; and the second in summer 643 took place not only in the aftermath of the battle of Nihāvand but also in the context of Farrukhzād’s hostilities with Armenian and Albanian princes.

3

Treaty of 653ce

The second of the case studies addresses the terms of the treaty made in 653 ce between the “prince of Ismael,” Muʿāwiya, and T‘ēodoros, lord of Ṙštunik‘, until then the principal Armenian client of the Romans. The treaty is preserved in the History attributed to Sebēos and reads as follows:

59 60

Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn ii.18: զաւրապետն Պարսից. Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn ii.19.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

606

greenwood

Let this be the pact of my treaty between me and you, for as many years as you may wish. I shall not take tribute from you for one three-year period. Then you shall pay by oath, as much as you wish. You shall keep in your country 15,000 cavalry and provide sustenance from the country and I shall reckon it in the royal tribute. I shall not request the cavalry for Syria but wherever else I shall command, they shall be ready for work. And I shall not send amirs to the fortresses nor an Arab army, neither many nor a single cavalryman. An enemy shall not enter Armenia. And if Romans should come against you, I shall send you troops in support, as many as you may wish. I swear to the great God that I shall not be false.61 This passage has been analyzed most fully by Jinbashian and Vacca, although several others, including Dadoyan and Garsoïan, have also commented on it.62 There has been general consensus that it contains the terms of an agreement established between two military men in a time of conflict. It is primarily concerned with arrangements for the payment of tribute, the provision of military service, and mutual defence. It may therefore be distinguished from many other conquest-era treaties that tend to record the terms negotiated by clerical leaders on behalf of individual urban communities. Armenian clerics are not recorded acting in this way as intermediaries before the end of the seventh century.63 Even a cursory glance confirms that the passage contains a wealth of information. It refers repeatedly to tribute (sak) rather than taxation (hark) and envisages that this is levied or assessed every three years. The same term sak is used by the later sixth-century Armenian writer Ełišē when referring to the tribute paid to the Persians.64 Although the above translation retains the reading 61

Sebēos, Patmut‘iwn 164: Այս լիցի ուխտ հաշտութեան իմոյ ընդ իս եւ ընդ ձեզ՝ որչափ ամաց եւ դուք կաﬕջիք. եւ ոչ առնում ի ձէնջ սակ զերեամ ﬕ. ապա յայնժամ տաջիք երդմամբ , որչափ եւ դուք կաﬕջիք: Եւ հեծեալ կալէք յաշխարհիդ ԺԵ հազար , եւ հաց յաշխարհէն տուք , եւ ես ի սակն արքունի անգարեմ. եւ զհեծեալսն յԱսորիս ոչ խնդրեմ. Բայց այլ ուր եւ հրամայեմ՝ պատրաստ լիցին ի գործ. եւ ոչ արձակեմ ի բերդորայն աﬕրայս, եւ ոչ տաճիկ սպայ՝ ի բազմաց ﬕնչեւ ցﬕ հեծեալ: Թշնաﬕ ﬕ՛ մտցէ ի Հայս. եւ եթէ գայ Հոռոմ ի վերայ ձեր՝ արձակեմ ձեզ զաւրս յաւգնականութիւն՝ որչափ եւ դուք կաﬕջիք: Եւ երդնում ի ﬔծն Աստուած՝ եթէ ոչ ստեմ.

62

Jinbashian, Arab-Armenian peace; Vacca, Non-Muslim provinces 186–193; Dadoyan, Armenians i, 56–57; Garsoïan, Interregnum 9, 13. Surprisingly, it is not discussed by LevyRubin, Non-Muslims. Łewond, Patmabanut‘iwn, c.12 (Sahak); but on this story, see Vacca, Non-Muslim provinces 37 and 183–184; more securely, see Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn iii.4–5 (Ełia). Vacca, Non-Muslim provinces 189 and n. 35.

63

64

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

607

erdmamb (by oath), it is possible that this conceals a variant instrumental form of eream, perhaps eremamb, meaning “per three year period.”65 A connection is made between tribute and military service, with the provision of supplies for the 15,000 cavalrymen apparently being set off against the existing liability for royal tribute. The apparently mysterious reference to not requiring military service in Syria should be placed in the context of the foundation inscription on the church at Mren, which describes the anonymous curopalates—almost certainly Davit‘ Sahaṙuni—uniquely as sparapet of Armenia and Syria.66 Nothing is known about his military service in the 630s, but this extension of his command to Syria suggests that he participated in operations there. It is striking that it does not preclude military service elsewhere. The passage contains the earliest occurrence in Armenian transliteration of the Arabic word amīr. It envisages them being situated in fortresses, just as Iranian commanders had been appointed previously to Duin and, almost certainly, to Naxčawan. The Arab army is represented as comprising cavalrymen. The common enemy is identified as being Roman rather than Persian. Yet there are also mysterious elements in the treaty. The repeated phrase “as much as you wish” produces uncertainty, both in terms of the amount of tribute to be paid and the number of reinforcements to be sent. This is the opposite of what one would expect in an agreement. Jinbashian proposed an elaborate solution, based on his reconstruction of an Arabic original, which included the term ʿafw or surplus, but Vacca has observed that this seems to be more aligned with Abbasid-era norms.67 Jinbashian is right, however, in drawing attention to the perspective from which the treaty is presented. It is written in the first person, from the perspective of Muʿāwiya rather than that of T‘ēodoros. Moreover, the treaty itself and both the contracting parties are described in negative terms. T‘ēodoros is recorded as making “a pact with death and an alliance with hell”; Muʿāwiya, the “prince of Ismael,” is described as “the servant of the Anti-Christ.”68 These make it difficult to envisage the circumstances in which a copy of the treaty could have found its way into the hands of the compiler. Finally, an oath to “the great God” at the end of the treaty is repeated elsewhere in the History attributed to Sebēos, not only in the letter purportedly sent by the “king of Ismael”—Caliph ʿUthmān b. ʿAffān—to Constans ii before his attack on Constantinople in 33/654 ce, but also in the letter purportedly sent by Vahram Č‘obin to Mušeł Mamikonean 65 66 67 68

For երդմամբ, read երեմամբ. Greenwood, Corpus A.7. Vacca, Non-Muslim provinces 188. Sebēos, Patmut‘iwn 164: ուխտ ընդ մահու եւ ընդ դժոխոց դաշինս … նեռին արբանեակն.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

608

greenwood

in 590 ce—“I swear by the great God, Aramazd […].”69 This is not the only point of overlap. ʿUthmān’s letter to Constans ii contains the following promise in return for Constans’s submission: “I shall give to you forces, as many as you shall wish and I shall receive from you tribute, as much as you are able to give.”70 Vahram’s letter to Mušeł make the following offer in return for military assistance: “I shall give you the kingdom of Armenia; you shall make whoever you may wish as king for yourselves.”71 Both these letters are written from the perspective of one of the protagonists in the first person, and both contain terms giving agency to the other party to determine the number of troops required, the amount of tribute, or the identity of the future Armenian king. These points of overlap undermine the proposition that the History attributed to Sebēos preserves the form of the actual treaty. Instead, it is a literary creation, consistent with other exchanges, epistolary and oral, between leading protagonists found elsewhere in the work. The compiler is no more likely to have had access to this treaty than to historic correspondence between an Armenian prince and an Iranian usurper of sixty years before or to current exchanges between caliph and emperor. On the other hand, we still have to account for its contents. Three lines of argument, which are not mutually exclusive, may be advanced. First, it is possible that the terms were devised on the basis of prior agreements negotiated with Sasanian šahanšahs. These feature in Armenian historical tradition. Writing at the turn of the sixth century, Łazar P‘arpec‘i records that the settlement reached in 485 ce between Vahan Mamikonean and the šahanšah Vałarš/Wālaxš ii included the supply of Armenian cavalry forces for immediate service against Zareh, son of the previous šahanšah Peroz.72 The later sixth-century Armenian historian Ełišē also records military recruitment from Armenia by Yazdgird ii for campaigning outside Armenia; he notes in passing that 10,000 cavalry were sought from Caucasian Albania, a comparable figure.73 Three of the 24 mathematical problems attributed to Anania Širakac‘i and dating from the middle of the seventh century contemplate Armenian mil-

69 70 71 72 73

Sebēos, Patmut‘iwn 169: առ Աստուածն ﬔծ; 77: ի ﬔծ աստուածն Արամազդ. Sebēos, Patmut‘iwn 170: Եւ տամ քեզ զաւրս որչափ եւ կաﬕցիս, եւ առնում ի քէն սակ՝ որչափ եւ դու կարես տալ. Sebēos, Patmut‘iwn 78: ձեզ տուեալ լիցի իմ զՀայոց թագաւորութիւնն. զո՛ր եւ դուք կաﬕք՝ արասջիք ձեզ թագաւոր. Łazar, Patmut‘iwn sections 89–99 for the negotiations; 94 for the military service against Zareh. Ełišē, Vasn Vardanay 22.3 (2.63–74) for recruitment; 63 (3.81–3) for Albanian cavalry.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

609

negotiating the north

itary service on and beyond the eastern frontier of Sasanian Iran.74 The History attributed to Sebēos contains numerous passages recording the recruitment of Armenian forces in Persian or Roman campaigns; indeed, the size of the forces of the country of Armenia assembled to fight for Khusro ii against Vahram Č‘obin is recorded to have numbered around 15,000.75 Even if the Sasanianera agreements preserved in Armenian literature do not reproduce the terms verbatim, they nevertheless reflect the understanding and perceptions of contemporary writers. If they were wholly fictitious, bearing no relation to the lived experience of the audience, the willingness of that audience to accept the version of the past being represented would be compromised. The treaty between Muʿāwiya and T‘ēodoros represented in the History attributed to Sebēos therefore belongs to a long tradition of such reconstructed agreements; despite the contemporary reference to amīrs, it looked back to a Sasanian-era model. The reference to the “great God” does not denote a contemporary expression of the divine from within the nascent community of Believers but reflects the terminology found within Armenian tradition for the supreme Iranian deity, Ohrmazd/Ahuramazda. A second interpretation views the treaty as being composed in light of subsequent events, and so with the benefit of hindsight. In response to Constans ii’s visit to Armenia in 653 to reconnect with his supporters and displace or destabilize his erstwhile client T‘ēodoros, the latter is recorded retreating to the island of Ałt‘amar with his son-in-law Hamazasp Mamikonean and requesting, and receiving, 7,000 troops from the Ishmaelites.76 In autumn of the following year (654ce), Hamazasp split with T‘ēodoros, prompting the latter to appeal once again for troops from the Ishmaelites.77 It could be the case that these actions were deemed by the author of the passage to have been covered in the original treaty, hence the inclusion of an appropriate clause. But there is a third solution, namely that Hamazasp was the source of information about the agreement. He had remained loyal to his father-in-law T‘ēodoros for at least a year after the agreement with Muʿāwiya, negotiating a temporary truce with his nephew Mušeł Mamikonean only after an Ishmaelite army had been billeted in Duin and T‘ēodoros had fallen ill.78 In one of the final passages to the work, Hamazasp is praised for being “a virtuous man in all respects, one who stayed at home, a lover of reading and study” but untrained

74 75 76 77 78

Greenwood, Reassessment of the life 161–167 (problems 2, 11, and 21). Sebēos, Patmut‘iwn 77. Sebēos, Patmut‘iwn 169. Sebēos, Patmut‘iwn 172. Sebēos, Patmut‘iwn 171–172.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

610

greenwood

and inexperienced in military affairs and combat.79 Furthermore, he is the last named Armenian figure in the original composition. When taken together, these elements lend support to the proposition that Hamazasp was the sponsor of the History attributed to Sebēos. His recent breach with T‘ēodoros could explain the sudden reversal in the latter’s portrayal, from “pious Armenian commander” to the one who abandoned the divine covenant when he negotiated the treaty with Muʿāwiya, the servant of the Anti-Christ. It would certainly have been in the interests of Hamazasp to distance himself from T‘ēodoros, and the highly symbolic language used to define the two protagonists indicates one of the ways in which this was achieved. This study therefore contends that the History attributed to Sebēos does not preserve the actual treaty established between Muʿāwiya and T‘ēodoros, lord of Ṙštunik‘; instead, it supplies a credible reconstruction of what a contemporary Armenian writer believed it to have comprised. This was devised on the basis of similar agreements in the past between Armenian princes and Sasanian Iran, as remembered in Armenian historical tradition. It may also have been written with the benefit of hindsight and even, perhaps, have been shaped by the personal knowledge and experience of Hamazasp Mamikonean; the context in which it is situated implies his involvement. It seems unlikely that we will ever be able to determine the degree to which each of these three dimensions contributed to its form and content. Whatever its exact terms, it seems to have been in effect for no more than eighteen months or so. By the end of 654, an Arab army was stationed in Duin—in breach of the purported treaty—and T‘ēodoros, lord of Ṙštunik‘, had fallen ill and withdrawn again to the island of Ałt‘amar.80 He had also been abandoned by Hamazasp. The contemporary nature of the description of the treaty imbues the passage with great significance, but it is not the verbatim record that many have assumed it to be.

4

Conclusion

Although the History attributed to Sebēos is now attracting a good deal of scholarly attention, the wider history of Armenia in the conquest era continues to languish on the margins, awaiting critical assessment and re-evaluation. Vacca’s recent monograph constitutes an important exception, analyzing Armenian, Arabic, and Georgian sources for their representations of the caliphal 79

Sebēos, Patmut‘iwn 174: այր առաքինի յաﬔնայն դէմս … ընդանեսուն եւ ընթերցասէր եւ ուսուﬓասէր […]. The rhythm and rhyme of these terms is lost in translation.

80

Sebēos, Patmut‘iwn 171–172.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

611

north and Armenia’s place within it. Aspects of Armenia in the conquest era can also be recovered through careful interpretation of a spectrum of contemporary Armenian-language sources, including the History attributed to Sebēos, the History of Ałuank‘, a set of church canons dated to 645ce, and collections of inscriptions and colophons; the value of the History of Łewond for this period, however, is questioned. Adopting this methodology, two case studies are then presented. The first traces the complex relationships between elite figures in Iran and Armenia from 628ce down to ca. 643. It demonstrates that Armenia was a zone of intense political rivalry, as Persian, Armenian, Roman, and then Arab-Islamic interests clashed and competed for control. Persian engagement with Armenia did not end with the death of Khusro ii in 628, as Farrukh Hormizd, the spāhbed kust ī Ādurbādagān and then his two sons Rustam and Farrukhzād, encountered various members of the Armenian elite. It was only after the battle of Nihāvand that Farrukhzād turned his attention away from Armenia and focused on Caucasian Albania and his interests in northwestern Iran instead. The second case study reassesses the treaty between Muʿāwiya and T‘ēodoros, lord of Ṙštunik‘, as preserved in the History attributed to Sebēos. It proposes that the treaty should be approached not as an authentic document, somehow embedded in the narrative, but rather as a contemporary reconstruction of such an agreement. Several features are mirrored in earlier agreements between Armenians and Sasanian šahanšahs, as represented in Armenian literature; others occur in passages in the History attributed to Sebēos recording diplomatic proposals. If the first case study illustrates the potential of this approach, the second indicates the need for caution. It seems that the Armenian sources for the conquest era still have much to tell us.

Bibliography Sources Ełišē, Vasn Vardanay ew Hayoc‘ paterazmin, ed. E. Tēr-Minasean, Erevan 1957; repr. in Matenagirk‘ Hayoc‘, i, Antelias 2003, 515–765; R.W. Thomson (trans.), Eḷishē history of Vardan and the Armenian War, Cambridge, MA 1982. Łazar P‘arpec‘i, Patmut‘iwn Hayoc‘, in Matenagirk‘ Hayoc‘, ii, Antelias 2003, 2,197–2,375; R.W. Thomson (trans.), The history of Łazar P‘arpec‘i. Atlanta, GA 1991. Łewond, Patmabanut‘iwn, ed. G. Tēr-Vardanyan, in Matenagirk‘ Hayoc‘, v, Antelias 2007, 711–851. Movsēs Dasxuranc‘i/Kałankatuac‘i, Patmut‘iwn Aluanic‘, ed. V. Aṙak‘elyan, Erevan 1987; repr. in Matenagirk‘ Hayoc‘, xv, Antelias 2012, 25–437; C.J.F. Dowsett (trans.), The history of the Caucasian Albanians by Movsēs Dasxuranc‘i, London 1961.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

612

greenwood

Studies Akopyan, A., Albania-Aluank v greko-latinskikh i drevnearianskikh istočnikakh, Erevan 1987. Crone, P., and M. Cook, Hagarism: The making of the Islamic world, Cambridge 1977. Dadoyan, S., The Armenians in the medieval Islamic world, 3 vols., New Brunswick, NJ 2011. Garsoïan, N.G., Interregnum: Introduction to a study on the formation of Armenian identity (ca 600–750), (csco vol. 640) Louvain 2012. Greenwood, T.W., A Corpus of early medieval Armenian inscriptions, in Dumbarton oaks papers 58 (2004), 27–91. Greenwood, T.W., A reassessment of the life and mathematical problems of Anania Širakac‘i, in Revue des études arméniennes 33 (2011), 131–86. Greenwood, T.W., A reassessment of the History of Łewond, in Le Muséon 125 (2012), 99–167. Gyselen, R., La Géographie administrative de l’empire sassanide: les témoignages épigraphiques en moyen-perse (Res Orientales 25), Bures-sur-Yvette, 2019. Hakobyan, V., Kanonagirk‘ Hayoc‘, 2 vols., Erevan 1971. Haldon, J.F., The empire that would not die: The paradox of eastern Roman survival, 640– 740, Cambridge, MA 2016. Hovsep‘ean, G., Yišatakarank‘ jeṙagrac‘, Ant‘ilias 1951. Howard-Johnston, J.D., Witnesses to a world crisis, Oxford 2010. Howard-Johnston, J.D., Caucasian Albania and its historian, in R.G. Hoyland (ed.), From Albania to Arrān, Piscataway, NJ 2020, 351–369. Howard-Johnston, J.D., The last great war of antiquity, Oxford 2021. Hoyland, R.G., The “History of the kings of the Persians” in three Arabic chronicles (Translated texts for historians 69), Liverpool 2018. Jinbashian, M., Arabo-Armenian peace treaty of ad652, in Haykazean hayagitakan handēs 6 (1977–1978), 169–174. Kaegi, W.E., Initial Byzantine reactions to the Arab conquest, in Church history 38 (1969), 139–149. Kaegi, Byzantium and the early Islamic conquests, Cambridge 1992. Kennedy, H., The great Arab conquests, London 2007. La Porta, S., The sense of an ending: Eschatological prophecy and the Armenian historiographical tradition (7th–10th centuries), in Le Muséon 129 (2016), 363–393. Levy-Rubin, M., Non-Muslims in the early Islamic empire: From surrender to coexistence, New York 2011. Manandyan, H., Les invasions arabes en Arménie (notes chronologiques), in Byzantion 18 (1948), 163–195. Martin-Hisard, B., Domination arabe et libertés arméniennes (viie–ixe siècle), in G. Dédéyan (ed.), Histoire du peuple arménien, Toulouse 2007 (rev. ed.), 213–241.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

negotiating the north

613

Martin-Hisard, B., Łewond vardapet Discours historique (Centre de recherche d’histoire et civilisation de Byzance Monographies 49), Paris 2015. Mat‘evosyan, A.S., Hayeren jeṙagreri hišatakaranner 5–12dd, Erevan 1988. Robinson, C., Conquest of Khūzistān: A historiographical reassessment, in bsoas 67 (2004), 14–39. Sebēos, Patmut‘iwn, ed. G.V. Abgaryan, Erevan 1979. Shoemaker, S.J., A prophet has appeared: The rise of Islam through Christian and Jewish eyes, Oakland 2021. Ter-Łewondyan, A., L’Arménie et la conquête arabe, in D. Kouymjian (ed.), Études arméniennes in Memoriam Haïg Berbérian, Lisbon 1986, 197–213. Thomson, R.W., and J.D. Howard-Johnston, The history attributed to Sebeos (Translated texts for historians 31), 2 vols., Liverpool 1999. Vacca, A., The fires of Naxčawan: In search of intercultural transmission in Arabic, Armenian, Greek, and Syriac, in Le Muséon 129 (2016), 323–362. Vacca, A., Non-Muslim provinces under early Islam: Islamic rule and Iranian legitimacy in Armenia and Caucasian Albania, Cambridge 2017. Zuckerman, C., The Khazars and Byzantium: The first encounter, in P.B. Golden, H. Ben Shammai, and A. Róna-Tas (eds.), The world of the Khazars: New perspectives, Leiden 2007, 399–432.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

chapter 30

New Palaeoenvironmental Evidence on the Possible Impact on Agriculture of Early Arab-Islamic Raiding Activity on Crete John Haldon

1

Introduction

One of the issues Hugh Kennedy has often highlighted in his writings on the first century of Islam is the problematic nature of the Arabic source tradition, with its emphasis on prosopographical information, heroic narrative, and the centrality of the chain of transmission.1 The non-Arabic textual evidence— whether in Latin, Greek, Syriac, or Armenian—for the effects of the early Islamic raids into east Roman lands in the period from the middle decades of the seventh into the later ninth century is equally problematic, if for often very different reasons.2 It has nevertheless generally been taken to indicate considerable economic and societal disruption, even if variable in its effects from region to region.3 Yet the patchiness and character of all the written sources presents us with many questions and problems, whether of chronology or of detail. To a certain extent this evidence can be tested against archaeological data, where the dating of destruction layers or similar can be reasonably precise and associated with a particular chronological range, and where some precision in the geographical specificity of supposed impacts can be established.4 But it is usually not exact enough to confirm the details of such attacks that may on occasion be reported in the sources, and in some cases it can challenge established interpretations that are based on the literary evidence. In

1 E.g., Kennedy, Armies of the caliphs, xi–xiv, with older literature. He is, of course, not alone in this, see, e.g., Noth and Conrad, Early Arabic historical tradition; Donner, Narratives. My thanks go to Archie Dunn, Warren Eastwood, and Hugh Elton for helpful comment and criticism. 2 See, in general and most recently, Howard-Johnston, Witnesses. 3 See in particular Jankowiak, First Arab siege; more generally, Lilie, Die byzantinische Reaktion. 4 Kiousoupoulou (ed.), Byzantine cities; Böhlendorf-Arslan and Schick (eds.), Transformations; Brubaker and Haldon, Byzantium, 531–563; Zavagno, Cities in transition; Brandes, Die Städte Kleinasiens.

© John Haldon, 2022 | doi:10.1163/9789004525245_031

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

615

the last 30 or so years this situation has begun to improve, as the information derived from proxy data for land-use and settlement patterns, such as pollen— which can inform about vegetation and landscape changes—on the one hand, and proxies for climate on the other (carbon and oxygen stable isotopes, tree rings, diatoms, geochemistry), begin to be exploited by historians and archaeologists.5 More often than not, however, the date range derived from such sources has been too broad to permit a clear connection between known historical events and changes in either climate or land use. But as more data become available and as more exploratory work is undertaken at sites across the lands once comprising the eastern Roman and early Islamic world, we are beginning to be able to obtain more precise chronologies, and these can help us connect both people and environment to historical developments reported in our sources, as well as suggest causal relationships between them. One of the best examples of this to date has been the evidence from Lake Nar in Cappadocia, where sediments laid down annually can be precisely dated and show that significant and dramatic land-use changes took place at precisely the time that the region was being devastated on a yearly basis by Arab raids or substantial armies. Since these significant changes did not coincide with the robust chronology for shifts in climatic conditions in the region, it has been reasonably concluded that they were caused by human activity, and given the nature of the changes—disappearance of traditional agriculture, expansion of grasses and scrub associated with abandoned agricultural land, and expansion of woodland—together with the relevant written sources for the period, they indicate a dramatic demographic decline and are probably directly associated with the effects of warfare.6 The Lake Nar case seems relatively clear because the chronology does not depend on Carbon-14 (14C) dates alone but rather on independently dated laminated sediments; but it is nevertheless important to deploy the palaeoenvironmental evidence with great care. Merely juxtaposing climate events or changes with historically attested events in itself is no evidence for a causal connection, although there is no doubt that both climate and environment affect human social and economic organization and activity.7 But the relationship is a complex one, and simple one-to-one causal associations are rare. In this short con-

5 See esp. Izdebski, Rural economy; Haldon et al., Climate and environment. 6 England et al., Historical landscape change; Eastwood et al., Integrating palaeoecological and archaeo-historical records; Haldon, “Cappadocia.” 7 Knapp and Manning, Crisis in context; Xoplaki et al., Medieval climate anomaly.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

616

haldon

tribution I want to outline an example where the palaeoenvironmental proxy data indicates that human activity certainly affected the established landscape and agriculture, but where the precise nature of that activity remains problematic. In this case, rather than the Anatolian mainland, our example comes from the island of Crete, and the evidence from Lake Kournas in the southeast of the Chania peninsula.8 On the whole, Crete is not well-documented in terms of palaeoenvironmental proxies, so that although there is a broadly agreed outline of how the landscape evolved from the Neolithic onward, no sites until now have been able to produce reasonably precise chronologies for landscape change or anthropogenic impacts. Lake Kournas was investigated as a potential source in 2003, but recent work has produced a core from which detailed data have been extracted and which provides more defined chronologies than any other site investigated for palaeoenvironmental data on Crete so far.9

2

Historical and Archaeological Context

Crete may not be well-documented from a paleoenvironmental point of view, but Crete during the early–middle Byzantine period is no better served in respect of written sources. No east Roman historian wrote from or about Crete during the period from the sixth to tenth century with the single exception of Andrew of Crete in the first half of the eighth century, whose focus was the city and see of Gortyn.10 The result is that the island appears only occasionally in the written sources, as a destination for travelers or a target for hostile raiders, as the home of this or that churchman, or as the victim of the occupation and conquest by the Andalusian/Egyptian forces under Abū Ḥafṣ in the 820s. Our

8

9

10

Jouffroy-Bapicot et al., Olive groves. For general considerations on the history and key characteristics of the Cretan landscape, land use, and settlement, see Atherden and Hall, Human impact 183–184; Rackham and Moody, Making of the Cretan landscape; see also Raab, Rural settlement. For the Roman period, see Gallimore, Food surplus; for the older literature, see JouffroyBapicot et al., Olive groves 4, 7–10. The chronological ranges for the pollen and charcoal are +/- 30 years, but with a 95 % certainty on the dates arrived at in the analysis of the sedimented material from the core. The Kournas pollen data complements but offers greater chronological precision from that extracted from the peat bog at Asi Gonia in the province of Rethymnon in the nearby White Mountains; also with a summary of previous palynological work on Crete, see Atherden and Hall, Human impact. See Lilie et al., Prosopographie, #362; Auzépy, La carrière d’André de Crète.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

617

knowledge of its social and economic history is even sparser, and were it not for the increasing availability of archaeological data, we would have very little idea indeed of how the island and its population fared for a period of over three centuries.11 While this limited evidence suggests that Crete was frequently a target for raids and occasionally a longer invasive campaign, it also seems that Crete was spared the constant and devastating warfare and economic disruption suffered by the frontier provinces in Anatolia in the period from the 650s to the middle of the eighth century. The archaeological evidence for substantial continuity in many districts and for a system of coastal defenses and fortresses is clear, although this had little effect on the longer-term process of urban settlement retrenchment.12 Many defensive emplacements have now been fairly securely dated to the seventh and eighth centuries,13 for example, and while regional survey continues to improve upon the amount of evidence available, the established perception that the populations of coastal regions fled inland or abandoned their settlements entirely,14 while the resulting weakened demographic situation of the island was the main reason for its relatively rapid conquest in the 820s is no longer supported by the archaeology.15 New approaches, including Historic Landscape Characterization and Retrogressive Landscape Analysis, the former pioneered by Rackham and Moody in Crete, will no doubt make a significant contribution to our changing understanding of the evolution of the Cretan landscape in the late Roman and Byzantine periods and are already confirming key elements of both continuity and change.16 As a case in point, the continued use, maintenance, and construction of small bathhouses

11

12

13 14 15 16

Christides, Conquest of Crete; Herrin, Crete; for the written sources relevant to the history of the island, see Tsougarakis, Byzantine Crete; see also Detorakis, Ιστορία της Κρήτης. For attacks on Crete, with sources, see Lilie, Die byzantinische Reaktion, 66, 79; HowardJohnston, Witnesses, 221, 303. See Andrianakis, Early Byzantine acropolis; Andrianakis, Herakleion; Perna, L’Acropoli di Gortina; Di Vita, Gortina; Tsigonaki, Les villes crétoises; Poulou-Papadimitriou, Sailing; Poulou-Papadimitriou, Sea routes. See also comment and literature cited in Cosentino, From Gortyn to Heraklion, 73–83. Tsigonaki, University of Crete excavations; Tsigonaki and Sarris, Recapturing the dynamics 8. As still noted in Tsigonaki and Sarris, Recapturing the dynamics 3. Poulou and Tantsis, From town to countryside; Bintliff, Contribution of regional surface survey. Rackham and Moody, Making of the Cretan landscape; in specifically Byzantine contexts, see Green, Rural Byzantine landscapes; Crow, Turner, and Vionis, Characterizing the historic landscape; Crow and Turner, Silivri; Turner and Crow, Unlocking historic landscapes; see also Gkiasta, Historiography.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

618

haldon

provides further evidence of continuity, not only for urban but also rural contexts, a significant indication of the maintenance of key cultural traditions as well as the availability of the relevant technical knowledge and resources.17 It would be wrong, of course, to conclude from this still very limited information that the pattern of settlement, the form and function of fortified centers and towns, the relationship between town and village, and the overall demography of Crete did not change quite considerably in the course of the seventh and eighth centuries.18 In the first place, Crete shared in the general direction of changes in urbanism, in the relationship between the state and local elites, and between landlords and tenants, found elsewhere—with regional variations— throughout the eastern Roman world. Secondly, while the information about hostile activity before the conquest of the 820s is limited, the island certainly suffered economic disruption and occasional localized devastation throughout the later seventh and eighth centuries, almost certainly more often than is recorded in either the Greek or the Arabic sources. A seaborne attack by raiders described as Slav reportedly devastated the island already in 623. Attacks increased in frequency from the 640s on as warfare between the empire and the new Islamic power in the east became endemic. A significant attack, supposedly the first, involving probably also a brief occupation of (parts of) the island, is reported to have taken place in the period between 652 and 654 in the course of a series of campaigns also directed against Rhodes, Kos, and other islands. This seems to have been resisted successfully, although parts of the island are reported to have been plundered. Some years later, probably associated with a longer-term occupation of Rhodes and the major attack and blockade of Constantinople in the late 660s, a substantial Arab force landed on Crete and wintered there but left no occupying force when they departed (the chronology is confused, and this may in fact be dated to the mid-670s); while further occasional raids took place across the following century, some quite serious, resulting in the capture and destruction of several fortresses and strongholds. The Arabic sources suggest that parts of the island were held for a period of months or longer on the occasion of some of these raids.19 17 18 19

Poulou and Tantsis, From town to countryside; Poulou-Papadimitriou, Archaeological research at Loutres. See Tsougarakis, Rites of passage; Tsougarakis, Byzantine Crete, 91–153; Zanini, Creta in età protobizantina. For the “Slav” attack in 623, see Tsougarakis, Byzantine Crete 22. For the Arab attacks, see Eickhoff, Seekrieg 5, 22; Tsougarakis, Byzantine Crete 22–41; Pryor and Jeffreys, Age of the Δρόμων 25 and n. 33 (raid of 652–654), 26, 33, bearing in mind that the supposed siege or blockade of Constantinople in the years 673/4–678 has been convincingly shown to result from the conflation and misdating of events in Theophanes’s Chronographia, and the

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

619

In the second place, the archaeological evidence from both excavation and survey indicates significant adjustments in the pattern of urban settlement. As Tsigonaki pointed out, in the sixth century the Synekdemos of Hierokles lists 22 cities on the island, selected presumably for their relative importance. Of these, between 10 and 12 can be identified at different times as the seats of bishops from the episcopal Notitiae, and in at least two cases, those of Heraklion and Phoinike, the cities mentioned in an earlier list are replaced by new names, Siteia and Hierapetra. Later historians, writing in the tenth century, mention that the Arabs captured 29 “cities” when they conquered the island, by “city” referring to a range of different fortified settlements.20 Yet while there is clear evidence in the archaeology of a substantial restructuring of urban space at a city such as Gortyn, the island’s capital, this also shows that the town remained a center of population and economic activity, now re-oriented around its nearby acropolis. Several earthquakes (618–621, 666–670, and 796) substantially damaged the city, yet it seems to have recovered, and in spite of the threat from Arab raids or the pestilence described in the Life of Andrew of Crete it remained a thriving settlement. But its administrative functions seem to have been transferred to the smaller settlement at Heraklion on the north side of the island, which from the later seventh or early eighth century now became the seat of the (military) governor.21 Evidence from several other urban sites points to a similar pattern of refocusing around a fortified strongpoint or acropolis, in which key administrative and ecclesiastical activities were located. On some occasions this involved the abandonment of an original settlement and the move of the population to a more defensible location—as at Chersonisos, abandoned sometime in the later seventh century with the inhabitants relocating to Episkopi in the Pediada region.22 This is accompanied by strong indications of continuity of occupation under new circumstances, including evidence from the ceramic record of

20

21

22

blockade and naval attack(s) in fact took place in the period 667–669, see Jankowiak, First Arab siege; Howard-Johnston, Witnesses, 302–304; Theophanes, am 6166, 354; and Theophanes, Chronicle 495; Hitti, Al-Balādhurī 376 (for an attack in the first years of Caliph al-Walīd and another during the reign of Hārūn al-Rashīd, see Pryor and Jeffreys, Age of the Δρόμων 42 and n. 70); see also Lilie, Die byzantinische Reaktion 66, 79. See Honigmann, Le Synekdèmos d’Hiéroklès 649.3–651b; Darrouzès, Notitiae episcopatuum 29, 235, nos. 239–250; Tsigonaki, Desperate vicissitudes of cities 78–80. For a gazetteer of the cities on the island, see Malamut, Les îles 191–210. See Cosentino, From Gortyn to Heraklion, who modifies some conclusions in Tsigonaki, Desperate vicissitudes of cities 81–88; Morrisson and Sodini, Sixth-century economy 191– 192. Hammond, Imperial territory 120.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

620

haldon

continued commercial connections with other parts of the Aegean and eastern Mediterranean world—less intensive than in the late Roman period, differently configured and refocused around local and interregional movement of goods, but nevertheless clearly present. The picture is not dissimilar from what we know from both Anatolia and some of the Aegean islands.23 Finally, it is clear that the imperial administrative presence on the island was substantial. It may well have been responsible, if only indirectly through local officials and military personnel, for the substantial works of fortification evident at key sites around the coast as well as inland;24 by the 760s, if not already before this, a military command of Crete had been established, and the lead seals of officials, both civil and military, for the period confirm the presence of an established administrative apparatus. The lead seals include those for fiscal officials and those connected with commercial taxes, indicating the continued activity of the (efficient) imperial administrative apparatus.25 There can be little doubt that the changes resulting from shifting urban-rural relationships, changes in urban spatial organization and functions, the impact of hostile attacks, and the pressures from imperial fiscal officials for particular types of product, as well as the occasional outbreak of (probably) bubonic plague,26 all had an impact on rural settlement patterns, and probably on land use, on Crete just as it is clear they did elsewhere. From a somewhat different but nevertheless comparable context, the archaeological evidence at Ephesos for continuing occupation of parts of the classical city, along with the refocusing of the settlement around a more defensible walled core, is nuanced by clear indications from the palaeoenvironmental data of a reduction in local landscape exploitation and population.27 The written evidence for how these developments affected the Cretan landscape, what crops were grown, and what farmers produced is sparse, dependent as it has been on the occasional mentions in the sources. The study of ceramic types has improved this situation to a

23

24 25

26 27

Tsigonaki, Desperate vicissitudes of cities 89–99; Gigourtakis, Fortified Byzantine acropolis; in general, see Saradi, Byzantine cities; Saradi, Byzantine city; for trade and commerce, see Vroom, Byzantine sea trade; Haldon, Commerce and exchange; the contributions in Mundell-Mango, Byzantine trade; Cosentino, A longer antiquity? Tsigonaki and Sarris, Recapturing the dynamics 7–8. The date of the establishment of a Cretan military command is debated; Prigent, Notes sur l’évolution 398; Nesbitt and Oikonomidès, Catalogue 94ff.; Dunn, Byzantine fiscal official; Tsougarakis, Byzantine seals from Crete; Winkelmann, Byzantinische Rang- und Ämterstruktur 101; Oikonomidès, Listes 353. Tsougarakis, Byzantine Crete 25 (the evidence of the Life of Andrew of Crete). Stock et al., Human impact.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

621

new palaeoenvironmental evidence

figure 30.1

Google Image map of the island of Crete

degree, permitting archaeologists to detect shifts in diet and food preparation, for example, indicated by changing pottery shapes and types.28 Nevertheless, until now there has been little solid data about agricultural production other than the occasional passing mention, although the textual and archaeological evidence for the earlier economic situation of Roman and late Roman Crete is sufficient to demonstrate that agriculture was dominated by market-oriented wine production, with olives/olive oil and then cereals playing a secondary role.29

3

Palaeoecological Evidence

Situated in the Chania peninsula in the northwest of the island, and with an area of approximately 1km2, Lake Kournas is the only natural lake on Crete, located some 3km from the coast in the foothills of the White Mountains in the western part of the island, which has the highest levels of rainfall on Crete. The landscape around Lake Kournas, which is surrounded by a narrow band of scrub and woodland, can be divided roughly into two zones, with steeply sloping limestone scrubland dominated by typical Mediterranean phrygana or garrigue30 to the south and west, and a relatively level area characterized by habitation and cultivation—olive groves, vineyards and fields—to the northeast. The core from the lake has a chronological span of almost 10,000 years

28

29 30

See Vroom, Ceramics; and the contributions to Poulou-Papadimitriou, Nodarou, and Kilikoglou (eds.), Late Roman coarse wares; and Bonifay and Tréglia (eds.), Late Roman coarse wares. Gallimore, Food surplus; Harris, Crete in the Hellenistic and Roman economies; see also Foxhall, Olive cultivation. Diamantopoulos et al., Variation in Greek phrygana vegetation.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

622

haldon

and can be divided into 4 pollen zones, the beginning and end of each reflected significant shifts in the overall pollen signature for the catchment area. The first zone covers the period ca. 7600–6500bce, the second ca. 6500–3900 bce, and the third 3900bce–400bce. The fourth zone is of particular interest here. Dated 400 bce–1950 ce, subzone 4a (400 bce–650ce) is marked at its outset by a dramatic increase in the proportion of Olea/olive, which thereafter varies between 20 % and 35 % of the total pollen across the period and must reflect the growth of market-oriented production in the Hellenistic and Roman Mediterranean. But this growth is followed by a marked decrease in Olea at the beginning of subzone 4b (~1300 Cal bp = ~650 ce), accompanied by an increase in oak and other woodland. At the same time, the data indicate small but short-lived decreases in vine (Vitis), which increases again slightly a generation or so afterward, and also in Pistacia. Interestingly, while there is very little pollen for cereals, there is no indication of any substantial change, and the same applies for the grazing indicators. That this subzone is highlighted statistically by coniss is important, since it shows that the pollen content of this subzone is dissimilar to what came before and after.31 The increase in woodland indicators reverses from ca. 950 ce, accompanied by the appearance of pollen indicative of cultivated trees and shrubs, and an increased indication of vine cultivation as well as chestnut. The shift at the beginning and end of subzone 4b is highlighted statistically in the coniss cluster analysis and suggests that it is “real” and significant—the pollen content of this subzone is dissimilar to what came before and after and implies a substantial shift in the cultivation of olive. Unlike at Lake Nar in Cappadocia, where there is relatively high precision dating afforded by laminated sediments, the chronology at Lake Kournas depends on 14C dating, which renders it less precise. Nevertheless, the age of the 4b boundary appears to be directly linked to a 14C age with a 95% confidence margin, established on the basis of radiometric markers used to create an age-depth model.32 And although the chronology is 31

32

coniss is a stratigraphically constrained cluster analysis that provides a multivariate method for quantitative definition of stratigraphic zones and thus provides a measure of how dissimilar the pollen content is between adjacent zones: Grimm, coniss. Cal bp = “calibrated (or calendar) years before the present,” meaning that raw radiocarbon dates have been corrected, and where “the present” = 1950ce. Since atmospheric carbon fluctuates over time, adjustments to the signal, or calibrations, are made to correct for such variations, or “wiggles.” Unadjusted dates are referred to simply as rcybp = “radiocarbon years before the present.” Jouffroy-Bapicot et al., Olive groves around the lake 7–10 for technical details. The 14C age of (1295 +/- 30 yr bp) at 69.40 cm produces a calibrated age of 1232 +/- 55 Cal bp and is

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

623

far less precise, palynological data from the Asi Gonia cores echo the profile of the changes indicated by the Lake Kournas pollen analysis.33 In neither case are these changes associated with any recorded climatic shifts, so we may assume that human impacts on the landscape underlie the changes indicated by the pollen data. Olive trees require careful management, in the absence of which growth is limited or stunted and the production of flowers, and therefore pollen, decreases, along with that of the fruit itself. Given the absence of any detailed written evidence about Cretan agriculture for the late Roman and Byzantine periods, the importance of the core from Lake Kournas is obvious. It provides a relatively detailed set of data about the land cover in the region around the lake and thus about the impact of human activity on the landscape from the Neolithic onward. It confirms some important changes in agriculture on the island that took place during the early Iron Age (ca. 850 bce) as the classic Mediterranean triad of olives, vines, and cereals came to dominate the agrarian landscape. It indicates a noticeably sharp increase in Olea pollen from ca. 400 bce, with percentages indicating that there were probably olive groves in the immediate vicinity of the lake, while at the same time an increase in charcoal indicates the use of fire, most probably explained by land clearance for olive cultivation at the expense of the dominant deciduous oak in the region.34 Charcoal is found along with pollen in the sediments and reflects the amount of burning within a landscape—prior to the twentieth-century landscape, burning was most frequent at times of diminished human impact, when fuel biomass increases. The greater the human impact on the landscape, in particular with regard to the control of landscape burning, the less the percentage of charcoal found in the sediments. However, this factor must be interpreted within the broader context—as noted above, burning was and is also employed by farmers in land-clearance and vegetation management.35 In the region around Lake Kournas, the pollen data suggest strongly that olive production was a major aspect of agricultural production throughout the Roman period, and while the documentary evidence for the export of olive oil is ambiguous, archaeological evidence from nearby local centers, such as

33 34 35

the age that most probably “anchors” subzone 4b. This confirms the robustness of the 4b boundary, since it is directly linked to a 14C age. For some helpful and accessible discussion of the difficulties and challenges accompanying the dating of pollen, see Roberts, Revisiting the Beyşehir occupation phase. Kouli et al., Regional vegetation histories 77–79; Atherden and Hall, Human impact 186– 192. Jouffroy-Bapicot et al., Olive groves around the lake 9–10, 15 with figs. 6 and 7. Bakker et al., Climate, people, fire.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

624

haldon

Lappa (some 6km southeast of the lake), attest to involvement in a substantially wider Mediterranean trade network.36 These indications remain stable from the beginning of the first millennium ce and suggest a patchwork of woodland and scrub with a strong pollen signature for olive trees. This picture changes rapidly and markedly from ca. 650ce, with indications of a steep downturn in olive cultivation, although at the same time the minimal indications that the data provide for other anthropogenic cultivars in the area, such as cereals and grazing crops, seem to remain stable—perhaps an indication that little changed demographically? Together with an accompanying decrease in charcoal and in the evidence for a slight increase in biodiversity, these changes might suggest a demographic downturn or movement of people away from the immediate locality, and certainly a significant reduction in olive production. It is particularly significant that this steep downturn in olive pollen does not coincide with any evidence for shifts in climatic conditions at this time, and so the change cannot be climate related; it is equally important to note that the change is too sudden to suggest any gradual shifts in agrarian strategies in the region.37

4

Some Conclusions

It seems unlikely to be mere coincidence that the sudden and very marked reduction in olive cultivation indicated by the pollen evidence that begins from ca. 650ce took place at more-or-less exactly the same time as the first known major attack on the island in 652–654, an attack that, although the exact details are lacking, appears to have been substantial and to have wreaked considerable devastation. A cause-and-effect link seems plausible and at least presents us with a hypothesis for further testing through continuing palaeoecological analyses. Various possibilities can be considered. The disruption caused by hostile invasion and raiding may very likely have had a direct impact on production in the Lake Kournas area, leading to the abandonment of the olive groves around the lake. This is certainly suggested by the suddenness of the downturn. Yet since the indications for the other economic cultivars remain stable, an alternative may be that it was not just this but the disruption of the commercial networks that rendered the market-oriented olive cultivation in the

36 37

Bowsky and Gavrilaki, Klio’s clay 199; see also Izdebski et al., Landscape change and trade. Jouffroy-Bapicot et al., Olive groves around the lake 15–16 and fig. 7.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

625

Lake Kournas catchment economically unviable, so that after the departure of the hostile forces local producers simply did not restore production to former levels. Shifts in market demand—possibly themselves a consequence of insecurity of transport—have been suggested as reasons for a reduction in the pollen signature of other cultivars elsewhere, for example, from the Negev during the later sixth century, and similarly may have been a contributing factor to the demise of the industrial-scale cultivation of olives in the Sinope region in the early–middle seventh century.38 In contrast, since the empire had lost its major sources of grain in Egypt and, by the last decades of the seventh century, in North Africa, too, cereal production both on Crete and in other regions would have become more important than hitherto. It may be significant that the stability of indicators for cereal and livestock farming in the Lake Kournas data are consistent with the indications from a number of Anatolian sites from the mid-seventh century for the continuing importance of cereal cultivation and raising stock, aspects of the agrarian economy throughout the empire in which the state had a particular interest.39 It seems thus less likely that the decline in olive production from the midseventh century was a result of the direct impact of raiding alone. Rather, given the disruption of established commercial routes and markets following the growth in Arab sea power from the 640s and 650s onward, reduced demand and/or lack of access to established markets is likely to have discouraged the restoration of damaged groves and encouraged a retrenchment in production and a reorientation of local agricultural priorities.40 It may not be a coincidence either that, on the basis of both palaeoenvironmental and archaeological evidence, olive production in Byzantine Apulia substantially increased at about the same time, a shift in volume that has similarly been associated with the disruption of existing commercial exchange networks as a result of hostile maritime 38

39

40

Fuks et al., Rise and fall of viticulture 8–9. For Sinop the causal connections are more problematic, since the changes also coincide with the beginnings of cooler climatic conditions, and the evidence is based on survey archaeology, with no detailed palynological data; see Cassis et al., Evaluating archaeological evidence 385–388. See Haldon, Some thoughts on climate change; and Haldon, Empire that would not die 224–232; Roberts et al., Not the end of the world? It is important to note that while at many sites there was an absolute decline in the pollen percentages for all anthropogenic cultivars, cereals, and grazing, weeds increased relative to other cultivars. There is no climate-related event with which such changes can plausibly be associated. Recent work confirms the broader picture of agrarian retrenchment across central Anatolia in the period ca. 650/700–950/100; see Dönmez et al., Vegetation record. For summaries of the ceramic evidence indicative of such changes, see Poulou-Papadimitriou, Sailing in the dark; with further literature, see Poulou-Papadimitriou, Sea routes in the Aegean; and Yangaki, La céramique.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

626

haldon

activity.41 We may wonder whether this was one example of the ways in which other central and eastern Mediterranean centers of olive production filled the gap left by the declining output of districts such as Lake Kournas. While the general situation in the latter district improved during the course of the tenth century, this is not reflected in the palaeoenvironmental data for olive production, although, as we have seen, the production of other crops did recover—all this no doubt a reflection of the relocation and restructuring of commercial relationships and demand in the wider region. Whatever the original stimulus for the change, the minimal indications for olive production in the Lake Kournas record after the 650s continue until the late twelfth century, when further changes appear, marking the transfer of power in the island to Venice. To conclude: the shifts indicated in the Lake Kournas pollen data are consistent with the broader trend in the Aegean region across the same period,42 but at this particular site we can see some chronologically quite specific impacts. Until we have more complete data for Crete as a whole, as well as a better archaeological record for relating land-use changes to developments over time in the settlement demography in the hinterland of Lake Kournas, it is not possible to decide whether the decline of olive production was a direct result of hostile activity or an indirect result following the ongoing dislocation of trade networks and market demand, or a combination of both. In either case, the Lake Kournas record supports the conclusion that the onset of Arab-Islamic maritime attacks and their impact on regional and local economies was a significant factor in changing both the intensity and the pattern of production as well as the character of commercial networks. It also reinforces the point that local and subregional patterns varied, and that broad generalizations are valid only up to a certain point.

Bibliography Andrianakis, M., The early Byzantine acropolis of Chania, in A. Kavvadia and P. Damoulos (eds.), Defensive architecture in the Aegean and the medieval settlement of Anabaton on Chios, Chios 2012, 75–90 (in Greek). Andrianakis, M., Herakleion in Crete, in J. Albani and E. Chalkia (eds.), Heaven and earth: Cities and countryside in Byzantine Greece, Athens 2013, 252–263.

41 42

Stranieri, Olive cultivation. Weiberg et al., Long-term trends; Kouli et al., Regional vegetation histories; Rull et al., A unique Pyrenean varved record.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

627

Atherden, M.A., and J.A. Hall, Human impact on vegetation in the White Mountains of Crete since ad500, in The holocene 9 (1999), 183–193. Auzépy, M.-F., La carrière d’André de Crète, in Byzantinische Zeitschrift 88 (1995), 1– 12. Bakker, J., E. Paulissen, D. Kaniewski, J. Poblome, V. De Laet, G. Verstraeten, and M. Waelkens, Climate, people, fire and vegetation: New insights into vegetation dynamics in the eastern Mediterranean since the 1st century ad, Climate of the past 9 (2013): 57–87. Baron, H., and F, Daim (eds.), A most pleasant scene and an inexhaustible resource: Steps towards a Byzantine environmental history, Mainz 2017. Bintliff, J., The contribution of regional surface survey to Byzantine landscape history in Greece, in J. Poblome (ed.), Exempli Gratia. Sagalassos, Marc Waelkens and interdisciplinary archaeology, Leuven 2013, 127–139. Böhlendorf-Arslan, B., and R. Schick (eds.), Transformations of city and countryside in the Byzantine period, Mainz 2020. Bonifay, M., and J.-C. Tréglia (eds.), Late Roman coarse wares, cooking wares and amphorae in the Mediterranean: Archaeology and Archaeometry (lrcw 2), Oxford 2007. Bowsky, M., and I. Gavrilaki, Klio’s clay: Inscribed instrumenta domestica from Lappa (Crete), in Creta Antica (2010), 173–253. Brandes, W., Die Städte Kleinasiens im 7. und 8. Jahrhundert (Berliner Byzantinische Arbeiten 56), Berlin 1989. Brubaker, L., and J.F. Haldon, Byzantium in the iconoclast era c. 680–850. A history, Cambridge 2011. Cassis, M., O. Doonan, H. Elton, and J. Newhard, Evaluating archaeological evidence for demographics, abandonment, and recovery in late antique and Byzantine Anatolia, in J.F. Haldon (ed.), Society and environment in the east Mediterranean ca. 300– 1800ce. Resilience, adaptation, transformation, special issue, Human ecology 46 (June 2018), 381–398. Christides, V., The conquest of Crete by the Arabs (ca. 824): A turning point in the struggle between Byzantium and Islam, Athens 1984. Cosentino. S., A longer antiquity? Cyprus, insularity and the economic transition, in Centre d’Études Cypriotes 43 (2013), 93–102. Cosentino, S., From Gortyn to Heraklion? A note on Cretan urbanism during the 8th century, in Byzantina Symmeikta 29 (2019), 73–89. Crow, J., S. Turner, and A. Vionis, Characterizing the historic landscape of Naxos, in Journal of Mediterranean archaeology 24 (2011), 111–137. Crow, J., and S. Turner, Silivri and the Thracian hinterland of Istanbul: An historic landscape, in Anatolian studies 59 (2009), 167–181. Darrouzès, J., Notitiae episcopatuum ecclesiae Constantinopolitanae, Texte critique, introduction and notes, Paris 1981.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

628

haldon

Detorakis, Th.E., Ιστορία της Κρήτης, Athens 1986. Diamantopoulos, J., S. Pirintsos, N. Margaris, G. and Stamou, Variation in Greek phrygana vegetation in relation to soil and climate, in Journal of vegetation science 5 (1994), 355–360. Di Vita, A., Gortina di Creta: Quindici secoli di vita urbana (L’Erma di Bretschneider), Rome 2010. Dönmez, E.O., F. Ocakoğlu, A. Akbulut, C. Tunoğlu, B.A. Gümüş, A. Tuncer, T. Görüm, and M. Tün, Vegetation record of the last three millennia in central Anatolia: Archaeological and palaeoclimatic insights from Mogan Lake (Ankara, Turkey), in Quaternary science reviews 262 (2021), 106973, https://doi.org/10.1016/j.quascirev.2021​ .106973. Donner, F., Narratives of Islamic origins: The beginnings of Islamic historical writing, Princeton 1998. Dunn, A.W., A Byzantine fiscal official’s seal from Knossos excavations and the archaeology of dark-age cities, in Creta Romana e Protobizantina: Atti del Congresso Internazionale (Iraklion, 23–30 settembre 2000), i, Padova 2005, 139–146. Eastwood, W.J., O. Gümüşçü, H. Yiğitbaşıoğlu, J.F. Haldon, and A. England, Integrating palaeoecological and archaeo-historical records: Land use and landscape change in Cappadocia (central Turkey) since late antiquity, in T. Vorderstrasse and J. Roodenberg (eds.), Archaeology of the countryside in medieval Anatolia (pihans 113), Leiden 2009, 45–69. England, A., W.J. Eastwood, C.N. Roberts, R. Turner, and J.F. Haldon, Historical landscape change in Cappadocia (central Turkey): A palaeoecological investigation of annually-laminated sediments from Nar lake, in The holocene 18 (2008), 1,229– 1,245. Eickhoff, E., Seekrieg und Seepolitik zwischen Islam und Abendland, Berlin 1966. Foxhall, L., Olive cultivation in ancient Greece: Seeking the ancient economy, Oxford 2007. Fuks, D., G. Bar-Oz, Y. Tepper, T. Erickson-Gini, D. Langgut, L. Weissbrod, and E. Weiss, The rise and fall of viticulture in the late antique Negev highlands reconstructed from archaeobotanical and ceramic data, in Papers of the national academy of sciences of the USA 117 (2020), 19,780–19,791. Gallimore, S., Food surplus and archaeological proxies: a case study from Roman Crete, in World archaeology 49 (2017), 138–150. Gkiasta, M., The historiography of landscape research on Crete, Leiden 2008. Gigourtakis, N., The fortified Byzantine acropolis of Lykto, in Kritiki Estia 14 (2011–2013), 49–64 (in Greek). Green, K., Rural Byzantine landscapes of the eastern Mediterranean: New approaches to characterization and analysis, in H. Baron and F. Daim (eds.), A most pleasant scene and an inexhaustible resource: Steps towards a Byzantine environmental history, Heidelberg 2018, 35–45.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

629

Grimm, E.C., coniss: A fortran 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares, in Computers and geosciences 13 (1987), 13–35. Haldon, J.F., Some thoughts on climate change, local environment, and grain production in Byzantine northern Anatolia, in H. Izdebski and F. Mulryan (eds.), Environment and society in the long late antiquity: Steps towards a Byzantine environmental history, Leiden 2018, 200–206. Haldon, J.F., The empire that would not die: The paradox of eastern Roman survival 640– 740, Cambridge, MA 2016. Haldon, J.F., Commerce and exchange in the seventh and eighth centuries: Regional trade and the movement of goods, in C. Morrisson (ed.), Trade and markets in Byzantium, Washington DC 2012, 99–122. Haldon, J.F., “Cappadocia will be given over to ruin and become a desert”: Environmental evidence for historically-attested events in the 7th–10th centuries, in Mediterranea: Festschrift Johannes Koder, Vienna 2007, 215–230. Haldon, J.F., N. Roberts, A. Izdebski, D. Fleitmann, M. McCormick, M. Cassis, O. Doonan, W.J. Eastwood, H. Elton, S. Ladstätter, S. Manning, J. Newhard, K. Nichol, I. Telelis, and E. Xoplaki, The climate and environment of Byzantine Anatolia: Integrating science, history, and archaeology, in Journal of interdisciplinary history 45 (2014), 113–161. Hammond, N.D., An imperial territory, in L.V. Watrous et al. (eds.), The Galatas survey: The socio-economic and political development of a contested territory in central Crete during the Neolithic to Ottoman periods, Philadelphia 2017, 119–161. Harris, W.V., Crete in the Hellenistic and Roman economies: A comment, in A. Chaniotis (ed.), From Minoan farmers to Roman traders: Sidelights on the economy of ancient Crete, Stuttgart 1999, 353–358. Herrin, J., Crete in the conflicts of the seventh century, in V. Kremmydas, Chr. Maltezou, N.M. Panagiotakis (eds.), Aphieroma ston Niko Svorono, i, Rethymno 1986, 113–126. Hitti, P.K. (trans.), Al-Balādhurī, Kitāb futūh al-Buldān: The origins of the Islamic state, London 1916 and Beirut 1966. Honigmann, E., Le Synekdèmos d’Hiéroklès et l’Opuscule géographique de Georges de Chypre (Éditions de L’Institut de philologie et d’histoire orientales et slaves), Brussels 1939. Howard-Johnston, J., Witnesses to a world crisis: Historians and histories of the Middle East in the seventh century, Oxford 2010. Izdebski, A., A rural economy in transition: Asia Minor from late antiquity into the early middle ages, Warsaw 2013. Izdebski, A., T. Słoczynski, A. Bonnier, G. Koloch, and K. Kouli, Landscape change and trade in ancient Greece: Evidence from pollen data, in The economic journal 130 (2020), 2,596–2,618.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

630

haldon

Izdebski, A., and M. Mulryan (eds.), Environment and society in the long late antiquity, Leiden 2019. Jankowiak, M., The first Arab siege of Constantinople, in C. Zuckerman (ed.), Constructing the seventh century: Travaux et Mémoires 17 (2013), 237–320. Jouffroy-Bapicot, I., T. Pedrotta, M. Debret, S. Field, R. Sulpizio, G. Zanchetta, P. Sabatier, N. Roberts, W. Tinner, K. Walsh, and B. Vannière, Olive groves around the lake: A tenthousand-year history of a Cretan landscape (Greece) reveals the dominant role of humans in making this Mediterranean ecosystem, in Quaternary science reviews 267 (2021), 107072, https://doi.org/10.1016/j.quascirev.2021.107072. Kennedy, H., The armies of the caliphs: Military and society in the early Islamic state, London and New York 2001. Kiousoupoulou, A. (ed.), Byzantine cities: Research perspectives and new interpretative approaches, Rethymno 2012 (in Greek). Knapp, A.B., and S.W. Manning, Crisis in context: The end of the late bronze age in the eastern Mediterranean, in American journal of archaeology 120 (2016), 99–149. Kouli, K., M. Masi, A.M. Mercuri, A. Florenzano, and L. Sadori, L., Regional vegetation histories: An overview of the pollen evidence from the central Mediterranean, in H. Izdebski and F. Mulryan (eds.), Environment and society in the long late antiquity: Steps towards a Byzantine environmental history, Leiden 2018, 69–82. Laiou, A. (ed.), The economic history of Byzantium from the seventh through the fifteenth century, Washington DC 2002. Lilie, R.-J., C. Ludwig, T. Pratsch, I. Rochow, et al., Prosopographie der mittelbyzantinischen Zeit: Erste Abteilung (641–867), 6 vols., Berlin and New York 1999–2002. Lilie, R.-J., Die byzantinische Reaktion auf die Ausbreitung der Araber (Miscellanea Byzantina Monacensia 22), Munich 1976. Malamut, E., Les îles de l’empire byzantin (viiie–xiiie siècles) (ii. Byz Sorbonensia 8), i, Paris 1988. Morrisson, C., and J.-P. Sodini, The sixth-century economy, in A.E. Laiou (ed.), The economic history of Byzantium, Washington DC 2002, 171–220. Mundell-Mango, M. (ed.), Byzantine trade, 4th–12th centuries: The archaeology of local, regional and international exchange, Aldershot 2009. Nesbitt, J., and N. Oikonomidès, Catalogue of Byzantine seals at Dumbarton Oaks and in the Fogg Museum of Art, vol. 2: South of the Balkans, the islands, south of Asia Minor, Washington DC 1994. Noth, A., and L.I. Conrad, The early Arabic historical tradition: A source-critical study, Princeton 1994. Oikonomidès, N., Les listes de préséance byzantines des ixe–xe siècles, Paris 1972. Perna, R., L’Acropoli di Gortina: La tavola ‘A’ della carta archeologica della città di Gortina, Macerata 2012.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

631

Poulou-Papadimitriou, N., Sailing in the dark: Trade networks and commercial interaction between Crete and the neighboring islands in a period of transition, in The 8th century: Patterns of transition in economy and trade throughout the late antique, early medieval and Islamicate Mediterranean, International conference, Berlin, October 4–7, 2017. Poulou-Papadimitriou, N., Archaeological research at Loutres, Mochlos: Investigation of a bath complex of medieval Crete, available online at https://www.doaks.org/​ research/support‑for‑research/project‑grants/reports/2016‑2017/poulou (last accessed April 28, 2022). Poulou-Papadimitriou, N., Sea routes in the Aegean during the early Byzantine period: The testimony of the ceramics, in N. Zarras and E. Stefanakis (eds.), Archaeology and art in the Dodecanese during late antiquity (Eulimene 2), Rethymnon 2014, 127–152 (in Greek). Poulou-Papadimitriou, N., E. Nodarou, V. and Kilikoglou (eds.), Late Roman coarse wares, cooking wares and amphorae in the Mediterranean: Archaeology and archaeometry (lrcw 4), Oxford 2014. Poulou, N., and A. Tantsis, From town to countryside: Middle-Byzantine bath-houses in eastern Crete and their changing functions, in Land 7 (2018), http://dx.doi.org/10​ .3390/land7030107. Prigent, V., Notes sur l’évolution de l’administration byzantine en Adriatique (viii– ix siècle), in Mélanges de l’École française de Rome: Moyen Âge 120 (2008), 393– 417. Pryor, J., and E. Jeffreys, The age of the Δρόμων: The Byzantine navy ca. 500–1204, Leiden and Boston 2006. Raab, H.A., Rural settlement in Hellenistic and Roman Crete: The Akrotiri Peninsula (bar Int. Series 984), Oxford 2001. Rackham, O., and J. Moody, The making of the Cretan landscape, Manchester 1996. Roberts, N., Revisiting the Beyşehir occupation phase: Land-cover change and the rural economy in the eastern Mediterranean during the first millennium ad, in H. Izdebski and F. Mulryan (eds.), Environment and society in the long late antiquity: Steps towards a Byzantine environmental history, Leiden 2018, 53–68. Roberts, N., M. Cassis, O. Doonan, W.J. Eastwood, H. Elton, J.F. Haldon, A. Izdebski, and J. Newhard, Not the end of the world? Post-classical decline and recovery in rural Anatolia, in J.F. Haldon (ed.), Society and environment in the East Mediterranean ca. 300–1800ce. Resilience, adaptation, transformation, special issue, Human ecology 46 (June 2018), https://doi.org/10.1007/s10745‑018‑9973‑2. Rull, V., T. Vegas-Vilarrúbia, J.P. Corella, M.C. Trapote, E. Montoya, and B. Valero-Garcés, A unique Pyrenean varved record provides a detailed reconstruction of Mediterranean vegetation and land-use dynamics over the last three millennia, in Quaternary science reviews 268 (2021), 107–128.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

632

haldon

Sanders, I.F., Roman Crete: An archaeological survey and gazetteer of late Hellenistic, Roman and early Byzantine Crete, Warminster 1982. Saradi, H., The Byzantine cities (8th–15th centuries): Old approaches and new directions, in T. Kiousopoulou (ed.), Byzantine cities, Rethymno 2012, 25–45. Saradi, H.G., The Byzantine city in the sixth century: Literary images and historical reality, Athens 2006. Stock, F., M. Knipping, A. Pint, S. Ladstätter, H. Delile, A.G. Heiss, H. Laermanns, P.D. Mitchell, R. Ployer, M. Steskal, U. Thanheiser, R. Urz, V. Wennrich, and H. Brückner, Human impact on Holocene sediment dynamics in the Eastern Mediterranean—the example of the Roman harbour of Ephesus, in Earth surface processes and landforms 41 (2016), 980–996. Stranieri, G., Olive cultivation and olive products in southern Apulia (6th–11th c.), in H. Izdebski and F. Mulryan (eds.), Environment and society in the long late antiquity: Steps towards a Byzantine environmental history, Leiden 2018, 172–182. Theophanes, Theophanis Chronographia, ed. C. de Boor, 2 vols., Leipzig 1883 and 1885; C. Mango and R. Scott (ed. and trans.), The chronicle of Theophanes Confessor, Oxford 1997. Tsigonaki, C., Les villes crétoises aux viie et viiie siècles: l’apport des recherches archéologiques à Pyrgi d’Eleutherna, Annuario della Scuola Archeologica Italiana di Atene 85 (2007), 263–297. Tsigonaki, C., “Desperate vicissitudes of cities”: Historical and archaeological evidence from Gortyn and Eleutherna in Crete [4th–8th c.], in T. Kiousopoulou (ed.), Byzantine cities, Rethymno 2012, 73–100 (in Greek). Tsigonaki, C., The university of Crete excavations on the acropolis of Eleutherna [trench ii, central], in P. Karanastasi, A. Tzigounaki, and C. Tsigonaki (eds.), Archaeological work on Crete 3: Proceedings of the third meeting, Rethymno, 5–8 December 2013, vol. 2: Chania, Rethymno, Lasithi, Rethymnon 2015, 391–406 (in Greek). Tsigonaki, C. and Sarris, A., Recapturing the dynamics of the early Byzantine settlements in Crete: Old problems—new interpretations through an interdisciplinary approach, in Proceedings of the 3rd international landscape archaeology conference, Rome 2014, https://dx.doi.org/10.5463/lac.2014.5. Tsougarakis, D., Byzantine Crete: From the 5th century to the Venetian conquest, Athens 1988. Tsougarakis, D., The Byzantine seals from Crete, in N. Oikonomidès (ed.), Studies in Byzantine sigillography 2, Washington DC 1990, 137–152. Tsougarakis, D., Rites of passage. From late Roman to Byzantine Crete: Cultural development and social transformations, in E.G. Kapsomenos, M. Andreadaki-Vlazaki, and M. Andrianakis (eds.), Proceedings of the 10th Cretological congress (Chania, 1– 8 October 2006), Chania 2011, 281–307 (in Greek). Turner, S., and J. Crow, Unlocking historic landscapes in the eastern Mediterranean:

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

new palaeoenvironmental evidence

633

Two pilot studies using historic landscape characterization, in Antiquity 84 (2010), 216–229. Vroom, J., Byzantine sea trade in ceramics: Some case studies in the eastern Mediterranean (ca. seventh–fourteenth centuries), in P. Magdalino and N. Necipoğlu (eds.), Trade in Byzantium: Papers from the third international Sevgi Gönül Byzantine studies symposium, Istanbul 2016, 157–177. Vroom, J., The ceramics, agricultural resources and food, in J. Haldon, H. Elton, and J. Newhard (eds.), Archaeology and urban settlement in late Roman and Byzantine Anatolia: Euchaïta-Avkat-Beyözü and its environment, Cambridge 2016, 134–184. Weiberg, E., A. Bevan, K. Kouli, M. Katsianis, J. Woodbridge, A. Bonnier, M. Engel, M. Finné, R. Fyfe, Y. Maniatis, A. Palmisano, S. Panajiotidis, C.N. Roberts, and S. Shennan, Long-term trends of land use and demography in Greece: A comparative study, in The holocene 29 (2019), 742–760. Winkelmann, F., Byzantinische Rang- und Ämterstruktur im 8. und 9. Jahrhundert (Berliner Byzantinische Arbeiten 53), Berlin 1985. Xoplaki, E., D. Fleitmann, J. Luterbacher, S. Wagner, J.F. Haldon, E. Zorita, I. Telelis, A. Toreti, and A. Izdebski, The medieval climate anomaly and Byzantium: A review of the evidence on climatic fluctuations, economic performance and societal change, in Quaternary science reviews 136 (2016), 229–252. Yangaki, A., La céramique des ive–viiie siècles ap. J.-C. d’Eleutherna: sa place en Crète et dans le bassin égéen, Athens 2005. Zanini, E., Creta in età protobizantina: un Quadro di sintesi regionale, in D. Michaelides, P. Pergola, and E. Zanini (eds.), The insular system of the early Byzantine Mediterranean: Archaeology and history, Oxford 2013, 173–189. Zavagno, L., Cities in transition: Urbanism in Byzantium between late antiquity and the early middle ages (ad500–900) (British archaeological reports, int. series 2030), Oxford 2009.

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

Index (Abū) al-Qāsim b. Ghassān → al-Ṭāʾifī Abān b. ʿAbd al-Ḥamīd al-Lāḥiqī 84 Abān b. ʿUthmān 17 Ābaq (amīr of Damascus) 528 Abāqā 478, 480 ʿAbbād b. Ziyād b. Abī Sufyān 31 al-ʿAbbās b. al-Walīd b. ʿAbd al-Malik 38, 39 ʿAbbāsa 363, 372 Abbasid(s) 4, 15, 46, 50, 51, 63, 64, 69–71, 73, 74, 78–80, 86, 91–94, 95, 96, 98–102, 104, 110, 117–119, 121, 122, 132, 135, 141, 142, 144, 145, 149–152, 210, 213, 214, 221, 223–225, 228, 229, 234, 241, 249, 255– 258, 260, 262, 263, 264, 265, 266, 269, 271–273, 275, 277–280, 299, 300, 301, 302, 304–310, 312–317, 321, 323, 329, 331, 334, 335, 337, 340, 356, 357, 358, 363, 369, 370, 372, 375, 380, 408, 411, 428–430, 434, 436, 443, 448, 449, 455, 457–460, 462–465, 478, 547, 583, 587, 607 ʿAbd al-ʿAzīz b. Marwān 15, 18, 33, 34, 37, 38, 337, 343 ʿAbd al-Jabbār → Abū Ṭālib ʿAbd al-Jabbār ʿAbd al-Laṭīf 323 ʿAbd al-Malik b. al-Muṭṭalib b. Ḥanṭab alMakhzūmī 33 ʿAbd al-Malik b. Ḥabīb 69 ʿAbd al-Malik b. Marwān (Umayyad caliph) 18, 29, 36, 80, 85, 372 ʿAbd al-Muṭṭalib 486, 487 ʿAbd al-Raḥmān (ii) b. al-Ḥakam 84 ʿAbd al-Raḥmān (iii) b. Muḥammad 85 ʿAbd al-Raḥmān b. Abī Dhiʿb 33 ʿAbd al-Raḥmān b. al-Ḥārith al-Makhzūmī 29 ʿAbd al-Raḥmān b. ʿĪsā 414, 415 ʿAbd al-Raḥmān b. Muʿāwiya 30 ʿAbd al-Raḥmān b. al-Walīd b. ʿAbd al-Malik 38 ʿAbd al-Ṣalīb (sīra character) 511–513, 517 ʿAbd Shams 17, 18, 23, 27, 30 ʿAbdallāh al-Baghdādī 412 ʿAbdallāh al-Maʾmūn → al-Maʾmūn ʿAbdallāh b. ʿĀmir b. Kurayz 18, 31 ʿAbdallāh b. ʿAmr b. al-ʿĀṣ 37

ʿAbdallāh b. Mālik → al-Khuzāʿī ʿAbdallāh b. Sulaymān 273 ʿAbdallāh b. Ṭāhir b. al-Ḥusayn 437 ʿAbdallāh b. al-Zubayr → Ibn al-Zubayr Abel 74 Abraham 77, 388, 477 Abraham, Catholicos of Armenia 601 Abrīza (Arabian Nights character) 511 ʿAbshāmīs → ʿAbd Shams Abū l-ʿAbbas al-Aḥmad ii (Hafsid Caliph) 140, 142, 151 Abū l-ʿAbbās al-Saffāḥ → al-Saffāḥ Abū ʿAbdallāh b. Mūsā al-Hāshimī 344, 414, 417–419, 422, 423, 425, 426 Abū ʿAli al-Qālī 345 Abū l-ʿĀṣ b. Umayya 13, 17, 21–23, 34, 38 Abū l-ʿAtāhiya 56, 61, 63, 267, 268, 342 Abū Bakr al-Ṣiddīq (Caliph) 29, 50, 80, 81, 82, 258, 390 Abū Bakr b. ʿAbd al-ʿAzīz 37, 47, 73, 74, 146, 151, 179, 378, 554 Abū Bakr Muḥammad b. Yaḥyā al-Ṣūlī → al-Ṣūlī Abū Dāwūd 81 Abū l-Faraj al-Iṣfahānī 305, 345, 370 Abū Ḥafṣ 616 Abū Hālid Yazīd b. Ilyās al-ʿAbdī 96 Abū Ḥanīfa 184 Abū Ḥanīfa al-Dīnawarī 69 Abū l-Ḥasan ʿUbayd Allāh b. Yaḥyā b. Khāqān 352 Abū Hāshim b. ʿUtba b. Rabīʿa 30 Abū Hurayra, Jaʿfar 176, 177, 179, 180 Abū l-Ḥusayn al-Rāzī 71 Abū Jaʿfar al-Manṣūr → al-Manṣūr Abū Jaʿfar al-Mūsawī 280 Abū Jaʿfar al-Ṣaymarī 417 Abū Layla Isḥāq b. Muḥammad b. ʿAbd alḤamīd 95 Abū Muslim 56, 63 Abū Nāmī 145, 146 Abū Nuwās 57, 62, 64, 305, 310, 313, 314, 342, 343 Abū l-Qāsim Hishām 177 Abū Sāʿīd, Īlkhān 546 Abū Sufyān Ṣakhr b. Ḥarb b. Umayya 25

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

636 Abū Ṭālib, Abbasid ancestor 56, 265, 427 Abū Ṭālib ʿAbd al-Jabbār 47, 49, 54, 63, 64, 66 Abū Tammām 433, 434 Abū ʿUbayda b. al-Walīd b. ʿAbd al-Malik 38 Abū Yūsuf 142, 183, 184, 185, 186, 187, 188, 190, 191–194, 195, 196, 197, 198, 313, 371 Abū Zayd (sīra character) 439, 513, 515 Abyssinians 4, 7, 9 Acre 161, 503, 505, 517, 529, 530, 532, 535, 541, 545 ʿĀd 77 Adam 74, 76, 389 Adam, bishop of Banyas 529, 530 ʿAdī b. Kaʿb 21 ʿAdī b. Zayd 84 al-ʿĀdil (Abū Bakr b. Ayyūb) 159, 163, 164 Aegean (Sea, islands) 620, 626 al-Afḍal ʿAlī (b. Salāḥ al-Dīn) 162 ʿAffān b. Abī l-ʿĀṣ 19, 22, 40 Africa 98, 103, 104, 110, 146, 166, 205 al-Afshīn Khaydar b. Kāwūs al-Usrūshānī 215 Agapius of Manbij 384, 385, 386, 399, 400, 401 Agnes, wife of Rainer Brus 527 Aḥmad b. Abī Duʾād 72 Aḥmad b. ʿAlī al-Mādarāʾī 444 Aḥmad b. Ḥanbal 72 Aḥmad b. Ḥasan b. Jaʿfar 275 Aḥmad b. Maymūn 414, 419 Aḥmad b. Muḥammad al-Asadī al-Anbārī 74 Ahmad b. Muhammad al-Ṭāʾī 21, 47, 48, 83, 94, 141, 274, 303, 304, 306, 333, 341, 342, 343, 389, 414, 417, 418, 423, 425 Aḥmad b. Muḥammad b. Makhlad 442 Aḥmad b. Ṣuʿlūk 421 Aḥmad b. Yūsuf 444 Ahmed, Asad 15, 16, 21, 23, 27 Ahrimān 484 al-Aḥsāʾ 256 Ahwāz 418, 422–425 Ainsworth, William 451 ʿĀʾisha bt. ʿAbd al-Malik b. Marwān 35 ʿĀʾisha bt. ʿAbdallāh b. Muʿāwiya 37 ʿĀʾisha bt. Hishām b. Ismāʿīl al-Makhzūmī 35 ʿĀʾisha bt. Muʿāwiya 30, 31

index ʿĀʾisha bt. Muʿāwiya b. al-Mughīra b. Abī l-ʿĀṣ 33 ʿĀʾisha bt. Mūsā b. Ṭalḥa 37 ʿĀʾisha bt. ʿUthmān b. ʿAffān 32 al-Ajfar 274 Akbar (Mughal emperor) 360, 365 Akhshunwār, king of the Hephthalites 392, 393 Akrī 561 Akyrtash 587 ʿAlāʾ al-Dīn Abū l-Shāmāt (Arabian Nights character) 507, 511, 513, 515, 516 Alberic, cardinal bishop of Ostia 529 Aleppo 73, 157–161, 165–168, 523, 527, 533, 540, 546, 550, 551, 553, 554 Alexander the Great 384, 385, 389, 391, 401 Alexandria 167, 173, 174, 204, 384, 389, 399, 510, 515, 516, 549 ʿAlī b. Abī Ṭālib, caliph 6, 34, 38, 73, 80, 81 ʿAlī b. al-Jahm 47, 48, 50, 51, 63–65, 69, 70, 73, 74, 75, 76, 79, 80, 82, 83, 84, 85, 86 ʿAlī b. ʿĪsā 275, 405, 414, 420 ʿAlī b. ʿĪsā b. Māhān 370 ʿAlī b. Yalbaq 415 ʿAlid(s) 64, 73, 81, 93, 94, 101, 110, 111, 269, 270, 274, 275, 280, 436 al-ʿAliyya, Fās al-ʿAliyya 97, 98, 103 Almohad(s) 140–148, 150–152 Alp‘ Iłit‘uēr, leader of Huns 596 Ałt‘amar 595, 609, 610 Ałuank‘/Caucasian Albania 596, 597, 601– 603, 605, 608, 611 Amalric of Nesle, patriarch of Jerusalem 533 Amalric, king of Jerusalem 166, 532, 533, 534 al-Amīn (Abbasid caliph) 50–53, 57, 61, 64, 82, 83, 269, 270 Āmina bt. ʿAbd al-ʿUzza b. Ḥurthān 21, 22 Āmina bt. ʿAlqama b. Ṣafwān b. Umayya b. Muḥarrith al-Kinānī 23 Āmina bt. Saʿīd b. al-ʿĀṣ 38 Amīnaddīn, amīr 181 ʿĀmir b. Luʾayy 33 ʿĀmir b. Ṣaʿṣaʿa 34 Amman 225, 227 ʿAmr b. Abī Sufyān 25 ʿAmr b. al-ʿĀṣ al-Sahmī 38, 395, 396 ʿAmr b. Saʿīd al-Ashdaq 18, 33

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index ʿAmr b. ʿUthmān 29, 31, 40 Anania Širakac‘i 608 Anastasius 583 Anatolia 118, 125, 159–161, 168, 454, 552, 617, 620 al-Andalus 47, 70, 84, 92, 93, 98, 100, 102, 120, 145, 150, 152, 166, 174, 509, 539 Andrew (St., of Crete) 616, 619 Andrew ii, king of Hungary 542 Andronicus ii 478 al-Anṣārī, Muslim b. al-Walīd 370 al-Anṣārī al-Kūfī, Abū Yūsuf b. Ibrāhīm 371 ʿAntar (sīra character) 309, 507 Antioch 159, 160, 390, 459, 466, 511, 528, 532, 602 Aphrodito/Ishkaw 201, 202 Apulia (Byzantine) 625 al-ʿAqīq 267 Aqyūdash 507 Arabia 77, 147, 254–257, 258, 269, 270, 275, 276, 390, 518 Arbela 560, 585 Ardashir i 394 Arghūn 478 Armani 491 Armenia/Armenian(s) 131, 159, 166, 400, 413, 479, 480, 511, 539, 546, 547, 551, 555, 591–596, 597–606, 607–611, 614 ʿArnūs (sīra character) 512 Arsacids 77 Aruč 599 Arwā bt. Asīd b. ʿIlāj b. Abī Salama 21 Arwā bt. Kurayz b. Rabīʿa b. Ḥabīb b. ʿAbd Shams 22 Asad 274, 275 al-Asadī, Abū l-Ḥasan Aḥmad b. Muḥammad 47 al-Aṣbagh b. ʿAbd al-ʿAzīz b. Marwān 37 Ascalon 161, 399, 503, 530, 532, 540 Ashmūnayn 175, 177, 178 Ashnās 215, 221, 222 al-Ashraf Barsbāy 48, 546 al-Ashraf Khalīl (sultan of Cairo) 545, 551, 555 al-Ashraf Mūsā (b. al-ʿĀdil Abī Bakr) 163, 164 Asi Gonia (Crete) 623 ʿĀṣim b. ʿAbd al-ʿAzīz 37 Asmāʾ bt. Abī Jahl 27

637 al-Aṣmaʿī 61, 64 Assyrian Christians 480 Asyūṭ 176, 190, 192 ʿAṭāʾ b. Rāfiʿ 205 Athanasius, patriarch of Antioch 602 ʿĀtika bt. ʿAbdallāh b. Muṭīʿ 38 ʿĀtika bt. Yazīd b. Muʿāwiya 35, 40 Atrpatakan/Ādurbādagān 601–605, 611 al-ʿAttābī, Abū ʿAmr Kulthūm b. ʿAmr 371 Austria 505 ʿAwf 32 Awraba 95, 97, 105, 106, 109, 110 ʿAws 32 Aybak, al-Malik al-Muʿizz 133, 134, 307 ʿAyn Jalūt 458 ʿAyntāb 551, 553, 555 Ayyūb b. Shādhī 157 Ayyubid/Ayyubids 48, 119, 125, 133, 134, 157, 158–161, 162–165, 166, 167, 168, 299, 301, 308, 330, 387, 535 Azd 17, 29 Baalbek 528 al-Badandūn 52 Badr al-Dīn Luʾluʾ 322 Badr al-Kharshanī 415, 416 Bagaran 600 Baghdad 48, 50, 57, 63, 64, 70, 91, 96, 100, 140–142, 145, 164, 165, 181, 184, 198, 210, 212–215, 226, 227, 234, 241, 249, 250, 269, 299, 300, 304, 305, 310–313, 316, 320–325, 328, 329, 330, 332–335, 337, 340, 342, 347–351, 353, 354, 357, 359, 368, 371, 406, 411, 413, 415–421, 422–424, 426–430, 433, 434, 435, 436, 437, 438, 439–442, 448, 449, 466, 468 Baghras (Frankish Gaston) 547, 549 Bāḥa (sīra character) 515 Bahāʾ al-Dīn al-Zuhayr 303 Bahāʾ al-Dīn b. Shaddād → Ibn Shaddād, Bahāʾ al-Dīn al-Bahāʾ Muḥammad b. al-Qāḍī al-Jamāl Yūsuf 48 Bahasnā 547, 550, 551, 552, 553 Baḥīrā 488, 489 Bahlūl al-Matgharī 96 Bahram Gor 394 Bahrām, Nizārī missionary 523, 524 al-Baḥrayn (Eastern Arabia) 256, 258

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

638 Baḥrūn (sīra character) 508 Bajkam 416, 418, 419, 425, 426 al-Balādhurī, Aḥmad b. Yaḥyā 21, 22, 23–25, 26, 31, 33, 36, 38, 39, 84, 223, 224 Baldwin i, king of Jerusalem 540 Baldwin ii, king of Jerusalem 522, 524–526, 540 Baldwin iii, king of Jerusalem 530–533 Baldwin iv, king of Jerusalem 534 Balīkh valley 457 Bālis 451, 453, 550, 587 Balkh 357, 358, 362, 372 Balkuwārā 220, 223, 224, 228, 230 Banū Kalb 18, 29, 30, 33, 34, 41, 73 Banū Mūsā 302 Banū Sāma b. Luʾayy 70, 71 Banyas (Bāniyās) 399, 521, 522, 523–535, 599 Bar Hebraeus 326, 332 Baranī, Ḍiyāʾ al-Dīn 359, 364–374, 376–378 Baraq b. Jabal, amīr 523 Barcelona 507 Bardarash (region) 235 Barghawāṭa 91 al-Barīdī, Abū ʿAbdallāh 414, 418–420, 422– 425 Barmak 369, 370, 372, 373 barmak 357, 358 Barmakid(s) 57, 61, 63, 71, 267, 310, 314, 315, 356–359, 362–364, 365, 368–375, 377 Baroj Tījī 561 Barqūq, al-Malik al-Ẓāhir (sultan of Cairo) 124, 132, 133, 548, 553 al-Barzūya (fortress) 459 Basileios (administrator of Aphrodito/Ishkaw) 201–206 Basra 31, 94, 99, 101, 210–212, 241, 256, 329, 343, 347, 350, 351, 425, 444 al-Baṣra (Morocco) 98, 99 Bātinīyya → Nizāris al-Baṭṭāl (sīra character) 507, 514, 516 al-Battānī, Muhammad 466, 467, 468, 469 al-Bāʿūnī, al-Shams Muḥammad b. Aḥmad 48 Baybars (Mamlūk sultan, sīra character) 507, 510–513, 515, 516 Baybars, al-Malik al-Ẓāhir (sultan of Cairo) 125, 134, 299, 542, 547, 548, 550, 552 Bāzyān 585

index Bedouin(s) 127, 148, 254, 256–259, 267–276, 277, 304, 453, 547 Beirut 161, 525–527, 532, 540 Belgrade 476 Bell, Gertrude 337, 453 Bell, Richard 293 Bencheikh, J.E. 312 Benedict of Alignan, bishop of Marseille 541 Bilād al-Shām → Syria Bindoi 394 al-Bīra 547–551, 552–555 al-Bīrūnī 389, 480, 482, 485, 494 Bishr b. Marwān b. al-Ḥakam 33, 34 Bistam 394 Bitlis 595 Bor/Bōrān 601 Bosworth, C.E. 369 Boucher, Guillaume de 476 Britain 505 Brittany 505 Budandūn 439 Buddhism/Buddhist 357, 477, 478, 479, 494 Buḥayrat al-Falak 515 al-Buḥturī 433, 434 al-Būnī 173 Būrān bt. al-Ḥasan b. Sahl 58, 436 Būrī, Tāj al-Mulūk (atabeg of Damascus) 524, 526, 528 Burton, John 284, 293 Burton, Richard 315 Būṣīr 175 Buyid(s) 141, 280, 299, 315, 417 Byzantium, Byzantine(s) 50, 52, 64, 92, 104, 119, 134, 159, 163, 165, 302, 340, 344, 385, 394, 396–399, 407, 408, 410, 411, 437, 439, 451, 454, 456, 458, 459, 475, 476, 477, 478, 479, 480, 482, 483, 484, 485, 486, 487, 489, 490, 491, 492, 495, 496, 497, 505–509, 511, 513, 514, 516, 539, 560, 583, 585, 587, 592, 593, 594, 597, 599, 616, 617, 623, 625 Cadiz 507 Caesar 150, 393, 484 Caesarea (Philippi) → Banyas Cain 74

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index Cairo 46, 117, 118, 123, 133, 135, 140–142, 144, 157–159, 165–167, 168, 176, 181, 214, 254, 299, 300, 303–305, 308, 312, 313, 334, 335, 343, 347, 357, 538, 545–548, 551, 552, 554, 555 Camphor, Isle of 506, 507 Canopus 255, 470 Cappadocia 615, 622 Caspian Sea 94 Castle of the Tooth 508 Catalan(s) 507 Catholic friars 478 Caucasus 104, 168, 596 Central Asia 161, 168 Ceuta 145 Chamberlain, Michael 128 Chania 616, 621 Chastel Neuf (Qalʾat Hunīn) 530, 532, 534 Chastel Pelerin (Athlīt) 539, 548 Chersonisos → Crete China/Chinese 161, 179, 214, 227, 329, 330, 343, 347, 390, 475, 476, 478, 483, 493, 495–497, 507 Chios 483 Christian(s) 72, 84, 119, 141–143, 145, 150– 152, 159, 160, 164, 166, 173, 176, 177, 180, 293, 294, 326, 332, 384–386, 391, 397, 399, 400, 401, 464, 476–478, 479, 480, 484–486, 488, 490, 496, 507, 509–511, 512–516, 517, 518, 541, 552, 585 Cilicia 159, 160, 550 Circassians 131, 133 Claudius Ptolemy 391 Cleopatra 399 Clouston, William Alexander 315 Conermann, Stephan 307 Constans ii 400, 595, 596, 599–601, 603, 604, 605, 607–609 Constantine (City in Algeria) 145, 384, 385 Constantine i 389, 401 Constantine v 411 Constantinople 166, 389, 478, 494, 506, 508, 512, 514, 599, 601, 602, 604, 607, 618 Copernicus, Nicolaus 466, 467 Córdoba 172 Coussonet, Patrice 311, 312

639 Crac des Chevaliers (Hiṣn al-Akrād) 539, 541, 545, 546 Creswell, K.A.C. 453, 454, 460 Crete 614, 616, 617, 618, 619, 620, 621, 625, 626 Crone, Patricia 21, 591 Crusade(s), Crusader(s) 159, 161, 163, 166, 301, 505, 509, 510, 517, 524, 525, 526, 530, 535, 538, 539, 540, 541, 542, 543, 545, 548, 555, 588 Ctesiphon 210, 602, 604 Cyrus (emperor) 389 Ḍabīna 259 Daftary, Farhad 255 al-Ḍaḥḥāk b. Qays 85 Damascus 73, 140, 157–165, 166–168, 181, 254, 305, 308, 322, 347, 398, 399, 468, 521–528, 530, 531, 533, 535, 539, 541, 552, 596, 599 Daphni 488, 490, 496 Daqīqī 85 Daqyādūrus (sīra character) 508 Dār al-Ḥadīth 326 Dār al-Khilāfa 212, 215, 220, 222, 223, 226 Dār al-Qurʾān 326 Dārā 77, 583 Dārayyā 528 Darb Zubayda 262–264, 266, 269, 272, 273, 275, 276, 278–280 Darius iii 392 Ḍariyya 258–260, 279 Dashqalan River 238 Dāthin 396 Dāʾūd b. ʿAbd al-Malik 35 David 77, 388, 494 Davit‘ Sahaṙuni 607 Davit‘ Tarōnec‘i 599 Daws 17 Dāwūd b. Razīn 98, 266 Ḍayfa Khātūn (bt. al-ʿĀdil Abī Bakr) 159 Dayr al-Juṣṣ 561 Deisis 478 Dēlamān → Daylam Delhi Sultan(ate) 359, 372 Deschamps, Paul 543, 544 Despina Khātūn 478 Dohuk (city and river) 238 Dubrovnik 505

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

640 Duin → Dwīn Duqāq, Shams al-Mulūk (king of Syria) 523 Dwīn 565, 597, 601, 602, 604, 605, 607, 609, 610 East Asia 491 Eastern Mediterranean 118, 125, 510, 620, 626 Edessa 399, 535, 548, 553 Egypt/Egyptian(s) 31, 33, 38, 41, 46, 118–123, 124, 125, 127, 130, 132–136, 141, 143, 145, 148, 152, 157, 158, 160, 162–168, 171, 173– 175, 176, 178–181, 184, 190, 197, 198, 201, 254, 299, 300, 303–305, 307, 311, 317, 335, 342, 343, 352, 384, 390, 391, 394, 395, 407, 523, 533, 534, 535, 539, 540, 545–548, 550, 552, 553, 616, 625 Eli 484 Ełia, Catholicos of Armenia 597 Ełišē 606, 608 Ellenblum, Ronnie 529, 539, 542–544 England 46, 300, 360, 507, 510, 539, 542 Ephesos 620 Episkopi (Crete) 619 Erbil 560 Eski Mosul Dam 235 Euphrates River 158236, 330, 418, 451–453, 457, 547, 549, 550, 553, 555 Euphrosyne Palaiologina 478 Eutychius of Alexandria 384, 385, 386, 387, 388, 389, 390, 391–401 Eve 74, 76 al-Faḍl b. Rabīʿ 370 al-Faḍl b. Sahl 370 al-Faḍl b. Yaḥyā al-Barmakī 359, 370 Fākhita bt. Ghazwān 27 Fākhita bt. Qaraẓa 31 Farāghina 215 Faraj b. Barqūq, al-Malik al-Nāṣir 124, 309 Farīdūn 484 Fārs 40, 71, 415 al-Fatḥ b. Khāqān 352 Fāṭima, daughter of Muḥammad 397 Fāṭima bt. ʿAbd al-Malik b. Marwān 41 Fāṭima al-Fihriyya 98 Fāṭima bt. al-Walīd al-Makhzūmī 27 Fatimid(s) 141, 166, 167, 387, 523, 530 Fayḍ 261, 265

index Fayyūm oasis 190 Fazāra 18, 262, 272 Fez, Madīnat Fās 95, 97, 98, 103, 109, 145, 149 Firdawsī 85 Fīrūzshāh Tughluq (sultan) 359, 378 Foucault, Michel 129 France 141, 364, 367, 505, 510, 525, 526, 534, 535, 539 Frank(s), Land of the Franks 161, 166, 390, 503, 505–511, 512, 514–518, 522–525, 527, 528, 531–535, 542, 546, 548 Fukuyama, Francis 119, 120, 122, 128 Fulcher (patriarch of Jerusalem) 522, 529, 540 Fulk of Anjou (king of Jerusalem) 526–530 Fusṭāṭ 165, 173–176, 178–180, 190, 195, 201, 204, 214, 399 Gabriel 482, 493–495 Galileo Galilei 466 Galland, Antoine 309, 310, 312, 315, 345, 347, 351, 353, 363 Gayūmard 85 Gaza 396 Genoa 505, 507, 512, 513, 515, 516 Genoa, King of (Arabian Nights character) 515, 516 George of Pisidia 84 George Synkellos 390 George, St. 517 Georgia 480 Georgia, Queen of (sīra character) 513 Ghādir (sīra character) 507 Ghanī 260, 277 Ghaṭafān 18, 29, 272 al-Ghazālī 164, 370 Ghazan, Īlkhān 478, 479, 551 ghulāh 63 Ghurūr (sīra character) 513 Ghuta (of Damascus) 162 al-Ghuzūlī, ʿAlāʾ al-Dīn ʿAlī 307, 309, 314 Goitein, S.D. 174, 175, 180 Golden Horde 478 Gomel (site and river) 238 Gorgān wall 564 Gortyn (Crete) 616, 619 Gotmann, John 532 Grabar, Oleg 458, 459

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index Great Zāb 561, 585 Greeks 390, 509 Grigor lord of Siwnik‘ 603 Grigor Mamikonean 596, 599 Guy of Lusignan (king of Jerusalem) 534 Güyük 476 Ḥabīb b. al-Ḥārith b. al-Ḥakam 32, 78, 79 al-Hādī (Abbasid caliph) 51, 52, 53, 56, 57, 61, 63 Ḥafṣa bt. ʿĀṣim b. ʿUmar b. al-Khaṭṭāb → Umm ʿĀṣim Ḥafṣid(s) 140–144, 145–148, 149–152 Hagar 77 Hajar 256 al-Ḥajjāj b. Yūsuf al-Thaqafī 40, 85 al-Ḥakam b. Abī l-ʿĀṣ 4, 5, 9, 13, 18, 19, 22, 23–25, 27, 31–33, 40 al-Hallāj 484 Halm, Heinz 255 Ḥamā 158, 165, 542, 552 al-Hamadhānī, Abū ʿAlī al-Ḥasan b. Harūn 413 Hamazasp Mamikonean 599, 609, 610 Ḥamdallāh Mustawfī 334 Hamdanid(s) 405, 416, 419, 426, 459 Ḥāmid b. al-ʿAbbās 421 Ḥanafī(s) 5, 171, 184, 323, 328 Ḥanbalī(s) 165, 171, 323, 328 Ḥanẓala b. Abī Sufyān 25 al-Ḥarbiyya 215 al-Ḥārith b. al-Ḥakam 32 al-Ḥārith b. Kaʿb 41 Ḥarrān 463–466 Hārūn al-Rashīd (Abbasid caliph) 50–53, 57, 59, 61, 63, 82, 93, 96, 183, 184, 191, 221, 263, 265, 266, 267–271, 299, 309, 310, 313, 316, 340, 341, 342–354, 357–359, 371, 448, 449, 451, 453, 454, 456–458 al-Ḥasan b. Sahl 370, 436 Ḥassān b. Tibān (sīra character) 108, 507 Hatra 394 Haṭṭīn 458, 535, 542, 544 Ḥawrān 162, 522–525 Helwan 399 Heraclius 397–399, 585, 598, 600, 602, 604 Heraklion (Crete) 619 Heraqleh 448–454, 456–461, 463–465, 467– 470

641 Hermon, Mount 522, 523, 525, 546 Herzfeld, Ernst 330, 336, 337, 453, 457 Hierapetra (Crete) 619 Hierapolis/Manbij 384, 385, 399, 602 Hierokles 619 Ḥijāz 15, 94, 95, 104, 118, 125, 145, 258, 261, 263, 272, 274, 279 Hilāl al-Ṣābiʾ 226 Hillenbrand, Carole 510, 517 Ḥimṣ 3, 9 Ḥimyar, Ḥimyarites 4, 5, 6–9 Hind bt. Muʿāwiya 30, 31 Hirawī, Muḥammad b. Ḥusayn b. ʿUmar 367, 368, 377 Hishām (i) b. ʿAbd al-Malik (Umayyad caliph) 13, 18, 20, 21, 35, 38, 40, 41, 70, 358, 443, 578 Hishām b. Ismāʿīl al-Makhzūmī 35, 37 Ḥiṣn Kayfā 164 Homs 158, 165, 542, 553 Horovitz, Josef 308 Hosios Lukas 483 Hospitallers, order 530–533, 541, 542 Hūd 77 Hudaybiyya 407 Hugh of Ibelin 532 Hugh, count of Jaffa 526 Ḥujarī troops 415 Hülegü 141, 335, 478 Ḥulwān 72 Humphrey ii, lord of Toron 530, 532–534 Humphrey iv, lord of Toron 534 al-Ḥusayn b. ʿAlī al-Nawbakhṭī 48, 94, 160, 414, 418, 424, 425 Iberia 512 Ibn ‘Awkal 175 Ibn ʿAbd al-Ḥakam 394 Ibn ʿAbd Rabbih 85, 305, 314, 315, 410 Ibn Abī l-Baqāʾ, Muḥammad b. Muḥammad b. Sulaymān 48 Ibn Abī Ḥajala, Shihāb al-Dīn Aḥmad 306 Ibn Abī l-Sāj, Yūsuf 274, 413, 414, 420–422 Ibn Abī Ṭāhir Ṭayfūr 432, 433 Ibn ʿAsākir 21, 164 Ibn al-Athīr 524 Ibn al-ʿAṭṭār 172 Ibn Baṭṭūṭa 326 Ibn al-Faqīh, Aḥmad b. Muḥammad 508

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

642 Ibn al-Farrāʾ 408–410, 412, 413, 423, 428, 429 Ibn Ḥajar 305 Ibn al-Ḥājj 174 Ibn Ḥamdān, al-Ḥasan b. Abī l-Hayjāʾ 405, 406, 416 Ibn Ḥanbal 5 Ibn Hāniʾ, al-Ḥasan → Abū Nuwās Ibn Ḥawqal, Abū l-Qāsim b. ʿAlī 508 Ibn Ḥazm 349 Ibn Ḥijja al-Ḥamawī 306, 309, 314, 315 Ibn Isḥāq 9 Ibn Jubayr, Muḥammad b. Aḥmad 141, 323 Ibn Khaldūn 109, 117–123, 125–127, 128–132, 134–136, 140–152, 344, 345 Ibn Khallikān 73, 74, 370 Ibn al-Khaṭīb, Lisān al-Dīn 47, 49, 59, 63, 64, 65, 149, 152 Ibn Khayyāṭ 106, 390 Ibn Khulayd al-ʿAbsī 260 Ibn Khurradādhbih, ʿUbayd Allāh b. ʿAbdallāh 6 Ibn al-Muqaffaʿ 315, 349, 392, 393 Ibn Muqātil, Muḥammad b. ʿAlī 424, 425 Ibn Muqla, Abū ʿAlī 422 Ibn Muqla, Abū l-Ḥusayn 427 Ibn al-Muʿtazz 85, 305, 306, 308 Ibn al-Nadīm 73, 344, 443, 466 Ibn al-Qalānisī, author 524, 526, 527, 530, 532 Ibn al-Qaysarānī, Muḥammad b. Naṣr 517 Ibn Qutayba 69, 305, 345, 369, 393, 394 Ibn Rāʾiq 416, 418–420, 423–426 Ibn Sabʿīn 143, 144, 146–148, 151 Ibn Saʿd 21, 407 Ibn Shaddād, Bahāʾ al-Dīn 160, 503, 505 Ibn Shaddād, ʿIzz al-Dīn 160 Ibn Shirzād 418 Ibn Sūdūn, ʿAlī al-Bashbughāwī 303 Ibn Taghrī Birdī, Jamāl al-Dīn Yūsuf 517 Ibn Waḥshiyya 305 Ibn Wāṣil 322, 325, 333 Ibn al-Zubayr 37, 38, 80, 85 Ibrāhīm b. al-ʿAbbās al-Ṣūlī 73 Ibrāhīm b. ʿAbdallāh 101 Ibrāhīm b. al-Aghlab 95, 96 Ibrāhīm b. al-Mahdī 58, 302, 369

index Ibrāhīm b. Turghut, governor of Banyas 528 Ibrāhīm b. al-Walīd b. ʿAbd al-Malik 40, 41 al-Ibshīhī, Bahāʾ al-Dīn Muḥammad 306, 309, 313–315 Idrīs 93, 94, 95–97, 99, 101, 102–104, 106, 108, 109–111, 385 Idrīs ii 93, 95–98, 100, 102–104, 107–109 Idrisid(s) 91, 93, 94–96, 98–101, 103, 104, 106, 107, 109–111 Ifranj/Ifranja → Frank(s) Ifrīqiya 91–93, 95, 96, 98, 100, 102, 104, 147, 171, 173, 174 Ilghāzī, Najm al-Dīn (ruler of Aleppo) 523 Ilkhanid(s) 69, 475–477, 484, 489, 491, 495, 547, 548 Imām Qaffāl 371 Imruʾ al-Qays 300 India 329, 347, 364, 375, 378, 394, 451 Indian Ocean 166, 168 Inner Asia → Central Asia Iran 30, 77, 84, 104, 141, 160, 161, 168, 243, 244, 347, 356, 373, 386, 468, 477, 478, 479, 495, 496, 594, 600, 601, 603, 609– 611 Iraq 25, 30, 31, 34, 40, 41, 69, 71, 72, 80, 94, 99, 101, 104, 159–161, 210, 234, 235, 237, 238, 241, 249, 254, 256, 261–263, 270, 271, 273, 274, 301, 308, 311, 321, 329, 337, 405, 417, 420, 434, 436, 470, 523, 560, 587 Isaac 77 Isfahan 415 al-Iṣfahānī, Ḥamza 305, 309, 313–315, 389, 414 Isḥāq b. Ibrāhīm b. Abī Khamīsa 97, 228, 273, 314, 369, 370, 374, 375, 418 Ishmael 77, 391 Ismāʿīl (Nizārī missionary) 35, 37, 65, 274, 418, 524, 526, 527 Ismāʿīl, Shams al-Mulūk (atabeg of Damascus) 526 Ismaʿili(s) 167, 254, 322, 523, 542 Israel 77, 384, 477, 544 Israyēl, bishop 596 al-Iṣṭakhrī, Ibrāhīm b. Muḥammad 464, 508 Istanbul 140, 141, 168, 357, 479 Italy/Italian(s) 150, 181, 455, 475, 539 al-Itlīdī 363

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index Iustinian 585 ʿIzza bt. ʿAbd al-ʿAzīz b. ʿAbdallāh b. ʿAmr b. ʿUthmān 38 Jabal 415 Jabal Says 569 Jacob’s Ford/Le Chastellet/Vadum Iacob 532, 534, 539 Jaʿfar b. ʿAbdallāh 396 Jaʿfar b. Ḥarb 82 Jaʿfar b. Muḥammad al-ʿĀmirī 436, 438 Jaʿfar b. Yaḥyā al-Barmakī 347, 352 al-Jaʿfarī 224, 227, 228, 230, 231 Jaffa 512, 526 al-Jāḥiẓ 71, 305–307, 309, 313, 314, 345 al-Jahm b. Badr 49, 70 al-Jahshiyārī → Muḥammad b. ʿAbdūs Jalāl al-Dīn Khwārazmshāh 321 Jāmiʿ al-Qaṣr 333 Jānibak al-Ashrafī 303 al-Jannābī, Abū l-Qāsim Saʿīd 256 al-Jannābī, Abū Saʿīd 256, 257 al-Jannābī, Abū Ṭāhir Sulaymān 256, 279 Jaques, R. Kevin 442 Jawarna 505 al-Jawsaq 220, 228, 230 al-Jazarī, Badīʿ al-Zamān 302 Jazīra 159, 160, 162, 166, 234, 236 Jerusalem (city) 77, 161, 162, 165, 166, 389, 399, 458, 513, 515, 521–523, 527, 529, 530, 532–535, 539, 540, 543, 553, 592 Jerusalem, Kingdom of 161, 521, 522, 523, 529, 530, 533, 535, 539, 543 Jerwan 235 Jesus 77, 78, 143, 299, 484, 485, 487, 489, 491, 494, 517, 521 al-Jibāl 71 Jingizid(s) 125 John iv, pope 400 John Malalas 390, 391 John the Baptist 484, 491 John, bishop of Banyas 534 John Chrysostom 206 Jordan river 243, 244, 459, 521–523, 530, 532, 546, 568, 587 Joscelin iii of Courtenay (titular count of Edessa) 535 Joseph 486

643 Juanšēr (prince of Ałuank‘) 596, 599, 601– 603, 605 Judah 77 Julian the Apostate 401 Julius Africanus 391 Jumayra 587 Jund 215, 220, 223, 225 Jundab b. ʿAmr 29 Juwān (sīra character) 513, 515 Kaʿb al-Aḥbār 3 Kaʿb b. Maʿdān 260 Kaʿb b. Qays 32 Kahrie Cami 478 Kakhtā 550–552 Kalb 18, 29, 33, 41 al-Kāmil Muḥammad (b. al-ʿĀdil Abī Bakr) 163, 164, 167 Kandafrah 508 Kanūd bt. Muʿāwiya 31 Kanza/Kathīra 96 Karakorum 475 Karfanās 514–516 Kargh river 561 Karin/Theodosiopolis 595 Karkar 550, 552 Kashgar 347 Katwa bt. Muʿāwiya → Kanūd bt. Muʿāwiya Kavad i 394 Kavad ii 600 Kay-Kāʾūs (Rum Seljuq sultan) 159 Keegan, John 548 Keller, Hans 434 Kennedy, Hugh N. 3, 12, 63, 70, 92, 97, 117, 118, 135, 141, 152, 157, 183, 210, 234, 255, 310, 316, 321, 356, 386, 429, 433, 438, 448, 470, 476, 497, 504, 519, 522, 538, 542, 544, 555, 588, 592, 614 Kepler, Johannes 466 Kerak 540, 548 Khālid b. Barmak 358 Khālid b. Yazīd b. Muʿāwiya 13, 35 Khālida bt. Abī l-ʿĀṣ 23 Khalidi, Tarif 122 Khalīfa b. Khayyāṭ → Ibn Khayyāṭ Khān al-Zabīb 587 Khān Dīrūk 561 al-Khaṣībī 413–415, 420–422 al-Khaṭīb al-Baghdādī 73, 164, 213

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

644 al-Khazir River 238 Khirbat al-Mafjar 459 Khosr river 238 Khurasan/Khurasani(s) 70, 71, 72, 301, 313, 322, 358, 370, 378, 434, 439, 526 Khushqadam, al-Malik al-Ẓāhir 133 Khusrau ii 394, 583 Khusrau ii 394, 583 al-Khuzāʿī, ʿAbdallāh b. Mālik 370, 375 Khwarizmian(s) 163, 166 Kilāb tribe (Kilāb b. ʿĀmir b. Ṣaʿṣaʿa) 34, 37, 267 Kināna 17, 23 al-Kinānī, Muḥammad b. ʿUmar 173 al-Kindī 142 Kinnis 235 Kirby, William Forsell 315 al-Kirmānī, Afḍal al-Dīn Abū Ḥāmid 367, 378 al-Kirmānī, ʿUmar (b. Azraq, var. ʿAbd alRazzāq) 367, 369, 374 kitab 435 Kournas Lake (Crete) 616, 621–626 K‘ristap‘or (Catholicos of Armenia) 601 Kruk, Remke 512 Kufa/Kufan(s) 9, 38, 40, 184, 210–212, 241, 256, 268, 272, 413, 414, 416 al-Kūfī, Abū ʿAbdallāh Aḥmad b. ʿAlī 369, 413, 417–420, 422–426, 428, 429 Kurd(s) 166 Kuthna bt. Umayya b. Abī Sufyān b. Umayya 30 Kyzicus 585 Lābarī, Abū Muḥammad ʿAbdallāh 368 Lamech 77 Lamīs bt. ʿAlī 41 Lapidus, Ira 128 Lappa (Crete) 624 Laylā bt. Suhayl 37 Laylā bt. Zabbān b. al-Aṣbagh b. ʿAmr al-Kalbī 33 al-Layth b. Saʿd 190–192, 389 Łazar P‘arpec‘i 608 Lebanon 243, 521, 523 Leisten, Thomas 459 Levant 104, 509, 587, 588 Lewis, Bernard 510

index Łewond 592, 593, 597, 599, 611 Līlimān (sīra character) 507 Loiseau, Julien 132, 480, 481 Louis ix (king of France) 144, 161, 535 Louis vii (king of France) 534 Luke 492, 521 Luke (sīra character) 508 Luʾluʾ (sīra character) 514 Lyons, Malcolm Cameron 511 MacDonald, Duncan Black 315 Macedonia 391, 507 Machiavelli 149, 150 Madelung, Wilfred 3, 27, 255 Madhbahūn (sīra character) 514, 515 Maʿdin al-Qurashī 275 Madīnat al-Fayyūm 176, 177 Madīnat al-Salām → Baghdad Maghāriba 215, 220 Maghrib al-Aqṣā (region) 91, 92, 98, 100, 111 Māhān 398 Mahdi, Muhsin 309, 310, 345, 347, 351, 353 al-Mahdī (Abbasid caliph) 51–53, 56, 61– 63, 101, 103, 263, 264, 266, 267, 302, 313, 314 Māhiya-sar, king of Ṭabaristān 373 Maḥmūd of Ghazna (Sultan) 368, 371 Maḥmūd, Shihāb al-Dīn (amīr of Damascus) 48, 371, 377, 527, 528 al-Māḥūza 221, 228, 230 Makhzūm 17, 18, 21, 27, 32, 34 Malaṭya 223, 548, 550, 552 al-Malik al-Manṣūr Qalāwūn 123 al-Malik al-Nāṣir Dāʾūd → al-Nāṣir Dāʾūd Mālik b. Anas 171 Mālik b. al-Riyāb 258 Mālikī(s) 95, 152, 173, 323, 328 Maltai 235 Malwiya 221 Mamluk(s) 117–121, 123, 124, 125, 128, 131, 132–136, 141, 143, 144, 145, 149, 152, 160, 166, 167, 198, 299–314, 317, 458, 517, 538, 545–549, 552–555 al-Maʾmūn (Abbasid caliph) 50–53, 57–59, 62, 64, 70, 71, 73, 82, 104, 269, 270, 302, 309, 314, 434–439, 443, 444, 464, 465, 469 al-Manīʿa, Islands of 507

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index al-Manṣūr, Abū Jafʿar (Abbasid caliph) 50– 53, 56, 60, 63, 101, 132, 211, 223, 241, 263, 299, 309, 313, 330–333, 335, 370, 411, 414, 448, 546 Manṣūr b. Sarjūn 346, 398, 550 al-Maqdisī (Muṭahhar) 74, 75 Maragheh observatory 468 Marco Polo 141 Maria (empress wife of Khusrau ii) 394 Maria Palaiologina 478, 480 Mariam (Genoese princess, Arabian Nights character) → Miriam Marj Rāhiṭ 85 Marouf, Najī 335 Marwān (i) b. al-Ḥakam 13, 22, 32, 33, 85 Marwān (ii) b. Muḥammad b. Marwān 20, 41 Marwān b. ʿAbd al-Malik (son of ʿĀtika) 35 Marwān b. ʿAbd al-Malik (son of Wallāda) 35 Marwān b. Abī Ḥafṣa 266 Marwanid(s) 13, 14, 17, 19, 20, 30, 32, 37, 41, 241, 242, 249, 250 Mary (mother of Jesus) 289, 484, 487, 493, 494, 517 Maryam bt. ʿAbdallāh al-ʿAwsī 32 Maryam bt. ʿUthmān 29 al-Maslāḥ 273 Masrūr 310, 311, 347 Massignon, Louis 215 al-Masʿūdī, ʿAlī b. al-Ḥusayn 6, 9, 74, 75, 266, 305, 314, 315, 370, 385, 388–390, 400, 508–510 Masʿūdī-yi Marwazī 85 Matthew 491 al-Māwardī 305 al-Mawṣilī, Isḥāq b. Ibrāhīm 370, 374 Maximian (emperor) 393 Maysūn bt. Bahdal 30 al-Māzārī 173, 174 Mazdak 394 al-Mazdaqānī, vizier of Damascus 523, 524 Māzin 27 Mecca (Makka) 17, 23, 25, 31, 52, 56, 61, 63, 77, 80, 94, 140, 142–148, 161, 167, 178, 254, 256, 261–263, 267, 269, 270, 274, 275, 280, 335 Medina 32, 35, 56, 75, 81, 85, 94, 140, 161, 167, 257, 260, 261, 263, 272, 316, 389, 397

645 Mediterranean Sea 550 Melane/Melania 478 Melisende (queen of Jerusalem) 524, 526, 532 Merv 454 Mesopotamia 210, 236, 384, 463, 585 Methuselah 77 Michael (Byzantine Emperor, sīra character) 508 Michael viii Palaiologos 478 al-Mīlawī (known as Ibn al-Wakīl) 364 Miriam (Genoese princess, sīra character) 506, 512–514 Miriam the sash-maker (Arabian Nights character) 506, 509, 511, 514 Miskawayh 413, 415–427, 429 Mitchell, Timothy 126, 129 Möngke 476 Mongol(s) 121, 141, 158–160, 162, 163, 165– 168, 299, 321, 334, 335, 475, 477–479, 484, 545–547, 550, 556, 563 Mongolia 476 Morocco 91, 93–96, 98, 99, 101, 103, 105, 108, 111 Moses 77, 385 Mosul 234–238, 241, 242, 247, 249, 250, 314, 322, 416, 422, 426, 527, 560, 585 Motzki, Harald 284 Mount Ḥīraʾ 495 Movsēs Dasxuranc‘i/Kałankatuac‘i 596 Mren 607 Muʿāwiya (i) b. Abī Sufyān 13, 18, 29, 35, 78 Muʿāwiya (ii) b. Yazīd 13, 31 Muʿāwiya b. ʿAbd al-Malik b. Marwān 35 Muʿāwiya b. Marwān b. al-Ḥakam 34 al-Muʿaẓẓam ʿĪsā (b. al-ʿĀdil Abī Bakr) 162, 164 al-Muʿaẓẓam Tūrānshāh (b. al-Ṣāliḥ Ayyūb) 164 al-Mubārak 48, 221 al-Mubarrad 314, 345 Mufāda bt. al-Zibriqān b. Badr 32 Mughal 360, 363, 365, 372, 375 al-Mughītha 274 Muḥammad al-Amīn → al-Amīn Muḥammad, Jamāl al-Dīn (amīr of Damascus) 528 Muḥammad al-Mustanṣir → al-Mustanṣir

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

646 Muḥammad al-Nafs al-Zakiyya Nakūr → al-Nafs al-Zakiyya Muḥammad b. ʿAbd al-ʿAzīz 37 Muḥammad b. ʿAbdallāh, the Prophet 19, 22, 23, 25, 27, 50, 65, 66, 75, 76, 79, 80, 81, 84, 93–96, 101–104, 110, 111, 171, 222, 258, 259, 283, 284, 289, 326, 332, 389, 397, 407, 409, 459, 479, 485–489, 495 Muḥammad b. ʿAbdallāh b. Ḥasan → alNafs al-Zakiyya Muḥammad b. ʿAbdūs Abū ʿAbdallāh alJahshiyārī 344, 370 Muḥammad b. Abī l-Sāj → Ibn Abī l-Sāj Muḥammad b. al-ʿAlqamī 324 Muḥammad b. Idrīs 100, 103 Muḥammad b. al-Jahm al-Barmakī 70, 71 Muḥammad b. al-Jahm al-Sāmī 71 Muḥammad b. Jarīr al-Ṭabarī → al-Ṭabarī Muḥammad b. Khalaf al-Nirmānī 413, 414 Muḥammad b. Marwān b. al-Ḥakam 34 Muḥammad b. Qalāwūn → al-Malik alNāṣir al-Muḥammadiyya 223 Muhanna, Elias 304, 305 al-Muhtadī (Abbasid caliph) 221, 440 Muʿizz al-Dawla 417 Mulayka bt. Awfā 23, 32 Munabbih b. Shabīl b. al-ʿAjlān b. ʿAttāb b. Mālik b. Kaʿb 25 al-Muntaṣir (Abbasid caliph) 80, 83, 224, 347 Muqawqis 395, 396, 398 al-Muqtadir (Abbasid caliph) 275, 309, 413, 414, 427, 430 Murra b. ʿAwf 23, 272 Mūsā al-Hādī → al-Hādī Mūsā al-Riḍā 59 Mūsā b. Nuṣayr 52, 59, 205, 369 Muṣʿab b. al-Zubayr 21, 80 Musaylima 484 Mušeł Mamikonean 595, 607, 609 al-Mustaʿīn (Abbasid caliph) 47, 50, 65, 74, 75, 83 al-Mustamsik 46 al-Mustanṣir (Abbasid caliph) 320–325, 326, 329, 331–335 al-Mustanṣir (Abbasid in Egypt) 299 al-Mustanṣir (Hafsid caliph) 140–148, 150, 151

index al-Mustanṣiriyya 320, 324, 325, 327, 329– 332, 334, 336 al-Mustarshid 47 al-Mustaʿṣim 48, 141, 334, 335 al-Muʿtaḍid (Abbasid caliph) 85, 466 al-Muʿtamid (Abbasid caliph) 47, 74 al-Mutanabbī 459 al-Muʿtaṣim (Abbasid caliph) 50, 71–73, 79, 215, 221, 270, 408, 410 al-Mutawakkil (Abbasid caliph) 47, 71–74, 78, 80, 83, 220–223, 231, 273, 275, 279, 299, 314, 315 al-Mutawakkiliyya 215, 219, 220, 222, 223, 227–231 Muther, Ryan 435 al-Muttaqī (Abbasid caliph) 414, 416, 417, 420, 426, 427 Muẓaffarid(s) 356, 367 Muzayna 265 Mžēž Gnuni 601 Nabhān 265, 269 Nafīsa bt. Zayd b. al-Ḥasan b. ʿAlī 38 Nahr al-Nīl 457 Nahrawān 436 Nahray b. Nissīm 175 Nāʾila bt. al-Farāfiṣa 29 Nāʾila bt. ʿUmāra 30 Najaf 74 Najd/ Najdī(s) 256–266, 268–272–280 Najm al-Dīn Ayyūb 335 Naqīʿ 259 Nar Lake (Cappadocia) 615, 622 al-Naṣībī, Taqī al-Dīn Abū l-ʿAbbās Aḥmad b. al-Mubārak 48 al-Nāṣir (Abbasid caliph) 141, 321 al-Nāṣir Daʾūd (b. al-Muʿaẓẓam ʿĪsā) 162, 322 Nāṣir al-Dawla 405, 416, 419 al-Nāṣir Muḥammad b. Qalāwūn (sultan of Cairo) 124, 132, 306, 546, 554 al-Nāṣīr Nāṣir al-Dīn Muḥammad b. Jaqmaq 305 Naṣr b. Muzāḥim 390 Naṣr b. Shabath 436–438 Navkur plain 235, 240 Naw Bahār 357 al-Nawājī, Muḥammad b Ḥasan 307, 309, 313, 314

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index al-Nawbakhṭī, al-Ḥūsayn b. ʿAlī → alḤusayn b. ʿAlī al-Nawbakhṭī, Isḥāq b. Ismaʿīl 418 Nawfil bt. ʿAbd Manāf al-Qurashī 30 Naxčawan 607 Nebuchadnezzar 77, 484, 604 Nestorian(s) 387, 478, 479, 585 Nicholas 478 Nicholas iv, pope 478 Niebuhr, C. 330, 337 Nihāvand 603, 605, 611 Ninive 560, 585 Nishapur 322 Niẓām al-Mulk 322, 334, 370, 411 Nizārī(s) 523, 524, 527, 542 Noah 77 Nogai Khān 478 North Africa 91, 92, 94, 96, 108, 110, 120, 121, 140, 142, 147, 149, 166, 167, 205, 625 North Africa 91, 94, 96, 108, 110, 120, 121, 140, 142, 147, 149, 166, 167, 205, 625 Northedge, Alastair 220, 222, 224, 227, 229, 449 Nuʿaym b. Ḥammād al-Marwazī 3–9 Nubia 168 Nuʿmān, King of Kawākib (sīra character) 513 Numayr 272 al-Nuqra 265 Nūr al-Dīn (Arabian Nights character) 310– 312, 346–353, 506, 509, 511, 514 Nūr al-Dīn Maḥmūd b. Zangī, ruler of Syria 164–166, 522, 530–534 al-Nuwayrī, Shihāb al-Dīn 198, 304–306 Ochus (emperor) 391 Octateuchs 484 Odo of Saint-Amand (marshal of Jerusalem) 532 Öljeitü 478 Olympias 475 Osti, Letizia 443, 444 Ottoman(s) 46, 119, 125, 134, 141, 152, 234, 235, 238, 244, 299, 309, 310, 313, 363, 540, 547, 561, 562, 565, 581, 586 Ouseley, Gore 356, 359–362, 364, 367, 370, 372, 378, 482, 485

647 Palestine 161, 164, 387, 390, 395, 398, 522, 524, 526, 530, 531, 535 Pan (classical deity) 522, 525 Pānkān 560–562, 564, 583, 585–588 Paris 379, 509 Partaw/Bardh‘a 597, 605 Parthian(s) 453, 459 Paul, Jürgen 128, 543 Pediada (Crete) 619 Peroz 392, 393, 484, 608 Persians 4, 7, 9, 77, 389, 390, 393, 476, 598, 605, 606 Peter, apostle 521, 529 Philip (son of Herod the Elder) 391, 521, 522 Phocas 585 Phoinike (Crete) 619 Pīra Bakir 561 Pisano brothers 475 Portugal 507, 511 Prawer, Joshua 544 Proclus 494 Procyon 470 Proust, Marcel 316 Ptolemy 466, 504 Pucelle 475 Qādisiyya 448, 449, 603 al-Qāhir, Abbasid caliph 415, 418 Qalʿat al-Jinn 561 Qalʿat al-Rūm (Hromkla, Rumkale) 538, 548–550, 552–555, 556 Qalāwūn, al-Malik al-Manṣūr (sultan of Cairo) 124, 125, 132, 546 Qalāwūnid(s) 123–125, 132, 134, 135 al-Qalqashandī, Shihāb al-Dīn Aḥmad 304, 548, 550, 553 Qamrān, Island of 508 Qānṣawh al-Ghawrī 305, 308 al-Qarʿā 274 Qarā Yūsuf (Qaraquyunlu ruler) 551 Qarāmiṭa (sing. qarmaṭī) 254–258, 275– 280, 414, 415, 421, 422 Qarāqūnā (sīra character) 508 al-Qarārīṭī 414, 418, 422 al-Qarāzima 516 Qarmatians → Qarāmiṭa al-Qāsim b. Ghassān → al-Ṭāʾifī al-Qāsim b. Jundab 262 Qaṣr al-Kharāna 568

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

648 Qaṣr al-Mushattā 587 al-Qaṭīf 256 Qāytbāy (Qaitbay), al-Ashraf (sultan of Cairo) 308, 546, 549, 550 Qinnasrīn 398 al-Qosh 244 Qudāma b. Jaʿfar 412 Quraysh, Qurashī 4–6, 7, 9, 17–19, 21–27, 30, 32–34, 36–40, 69, 70, 140, 151, 152, 265 Qurra b. Sharīk 201–207, 466 Quṣayy 77 Qusṭā b. Lūqā 385, 400 Qutayba b. Muslim 358, 392 Quṭiyya bt. Bishr b. Mālik 34 al-Rabadha 259, 272, 279 al-Rāḍī (Abbasid caliph) 405, 414–417, 419, 422 Rāfiḍite(s) 322 Rāfiqa 451 Rainer Brus (lord of Banyas) 525–527, 529, 530 Ramla bt. Abī Sufyān 22, 25 Ramla bt. Muʿāwiya 30, 31 Ramla bt. Shayba 27, 32 Ramla bt. Yazīd b. Muʿāwiya 31 Raqqa 426, 448–451, 457, 460, 465–467, 469, 587 al-Rashīd → Hārūn al-Rashīd Rashīd al-Dīn 69, 480, 483, 485, 486, 489, 492, 493, 495, 497 Rāshidūn caliphs 78 al-Rawḥāʾ 265 Rayḥāna bt. Abī l-ʿĀṣ 23 Raymond ii (count of Tripoli) 528 Raymond of Poitiers (prince of Antioch) 528 Rayne, Louise 450, 452, 457, 468 Rayy 421 al-Rāzī, Yaḥyā b. Muʿādh 94, 371 Razwān 561 Red Sea 484 Rey, Emmanuel-Guillaume 542–544 Rhahzadh 585 Rhodes 510, 515, 618 Rhodes, King of (sīra character) 515 Richard i the Lionheart 505 Robinson, Majied 15, 16, 18, 21, 236, 241, 247, 592

index Rohard of Jaffa 532 Roman Empire 20, 172, 384 Rome 384, 386, 507–510, 587, 592 Rosenthal, Franz 76, 434 Ṙostom → Rustam Rowson, Everett 307, 308 Rūm → Byzantines Ruqayya bt. al-Ḥārith b. ʿUbayd b. ʿUmar 21 Ruqayya bt. Muḥammad 27 Ruṣāfa 227, 578 Rustam/Rustamid(s) 91, 93, 104, 601–603, 611 Sabians 463–465 Sābūs 333 Sachau 451 Ṣafad 532, 541, 544, 548–550, 552, 555 al-Ṣafadī, Khalīl b. Aybak 307, 554 al-Saffāḥ (Abbasid caliph) 60, 263 Safīna 81, 82 Ṣafiyya bt. Muʿāwiya → ʿĀʾisha bt. Muʿāwiya Ṣafiyya bt. Rabīʿa b. ʿAbd Shams 21 Saḥnūn 173 Saʿīd b. al-ʿĀṣ 18, 29, 32, 33, 34, 38 Saʿīd b. Sinān 5 Sājī troops 415, 416 al-Sakhāwī 48 Saladin → Salāḥ al-Dīn Salāḥ al-Dīn, Yūsuf b. Ayyūb 145, 158, 160, 162–164, 167, 340, 503, 505, 518, 534, 535, 542, 544, 545 al-Ṣalāḥiyya 335 Ṣāliḥ (prophet) 77, 314 al-Ṣāliḥ Ayyūb (b. al-Kāmil Muḥammad) 164, 167 Ṣāliḥid(s)/Salihiyya 91, 165 Samanid(s) 356, 368, 377 Samaria 529 Samarra (Surra Man Raʾā) 210, 212–216, 219– 223, 225, 228, 249, 270, 271, 273, 305, 449, 454, 455 al-Samhūdī 264 Samīra 274 Sapor ii 393 Saqrāq (sīra character) 514, 515 Sardinia 511 al-Sarraf, S. 301 al-Sarrāj, Abū Muḥammad Jaʿfar b. Aḥmad 48, 346

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index Sarre, Friedrich 330, 336, 453 Sasanian(s) 75, 77, 79, 85, 242, 243, 249, 301, 384, 386, 389, 392–394, 458, 459, 560, 562, 563, 570–572, 575, 579, 581, 583, 585–587, 592, 600–603, 608–611 Sayf al-Dawla 416, 419, 459 Sayf al-Dīn al-Qaymarī 165 Sayf b. ʿUmar 389, 553, 554 Scheherazade → Shahrazād Schmid, H. 320, 324, 329, 332, 334 Sebēos 591, 592, 594–597, 600–603, 605, 607–611 Seljuq(s) 159, 163, 321, 322, 334, 367, 523 Seljuqs of Rūm 478 Seydi, Masoumeh 440, 441 al-Shādhiyākh 71 Shāfiʿī(s) 323, 328 Shāh Shujā, Jalāl al-Dīn Abū l-Fawāris 367, 379 Shahrazād 312, 343, 346, 352 Shahriyār 311, 312, 343, 346, 351, 354 Shākiriyya 220, 273 al-Shām → Syria Shāriʿ Abī Aḥmad 215 Shāriʿ al-Aʿẓam 215 Sharkān (Arabian Nights character) 511 Shaybān 274 al-Shayzarī, Amīn al-Dawla Abū l-Ghanāʾim 48 Shīḥa (Arabian Nights character) 516 Shiʿi(s), Shiʿite(s) 56, 63, 99, 141, 167, 254, 255, 322, 414 al-Shiʿrā al-Ghumayṣāʾ → Procyon Shīrkūh b. Shādhī 157 Shuhadāʾ bridge 329 Sibyl (queen of Jerusalem) 534 Ṣiffīn 390, 451 Sijilmāsa 91, 98, 108 Simon Peter → Peter Siteia (Crete) 619 Siwnik‘ 602, 605 Slavs 508 Smail, Raymond 540, 544 Smbat Bagratuni 600 Solomon 77, 347, 477, 484 St. Mary of the Mongols 478 Stowasser, Barbara 149 Subeibe, castle 535 Suʿda bt. ʿAbdallāh b. ʿAmr b. ʿUthmān 40

649 Sufi(s)/Sufism 143–146, 147, 148, 165, 342 Sufyanid(s) 13, 14, 37, 593 Suhayl → Canopus Sulaym 145, 272, 273 Sulaymān b. ʿAbd al-Malik 18, 35, 37, 38, 40, 41 Sulaymān b. al-Ḥasan 414, 415 al-Ṣūlī, Abū Bakr Muḥammad b. Yaḥyā 47, 73–75, 412, 414, 417, 419, 422, 426, 427, 429, 443, 444 Sumatar 464 Sunni(s) 72, 135, 141, 164, 165, 167, 254, 255, 321–323, 326, 331, 334, 335, 523, 524 al-Suyūṭī, Jalāl al-Dīn 46–50, 53, 63, 64, 66, 299, 300, 364 Synekdemos (of Hierokles) 619 Syria (Greater Syria, Bilād al-Shām) 3, 29, 31, 77, 118–120, 122–125, 127, 132, 134, 145, 157–162, 165, 166, 168, 237, 254, 256, 299, 300, 304, 317, 358, 372, 384, 399, 453, 487, 522, 523, 527, 534, 538–540, 545– 548, 550, 552, 569, 578, 587, 594, 595, 602, 606, 607 al-Ṭabarī, Muḥammad b. Jarīr 21, 27, 36, 38, 39, 271, 272, 274, 316, 329, 358, 369, 374, 389, 390, 400, 432–445, 454 Ṭabaristān 101, 372, 373 Tabrīz 475, 480 Taghrībirdī al-Maḥmūdī 303 T’ang dynasty 214 Tāhart 91, 93, 98 Ṭāhir b. ʿAbdallāh b. Ṭāhir 71 Ṭāhir b. al-Ḥusayn 437, 439 Tahirid(s) 228, 436 al-Ṭāʾif 17, 23 al-Ṭāʾifī, (Abū) al-Qāsim Muḥammad b. Ghassān 368, 369, 374, 377–379 T‘alin 599 Tamīm 279 Tamīm b. ʿAlqama 84 al-Tanūkhī 305, 309 Tarōn 595, 602 Tarsus 439, 550 al-Tawḥīdī 349, 352 Ṭayyiʾ 261, 274, 275, 279 Templars (order) 531, 541, 548, 555 T‘ēodoros (lord of Ṙštunik‘) 594, 595, 601– 605, 607, 609–611

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

650 Terre de Suète → Ḥawrān al-Thaʿlabiyya 274 Thamūd 77 Thaqīf 17, 21, 23, 25 Theodosius i 389 Theophanes the Confessor 390 Theophilus (emperor) 400, 437 Theophilus (patriarch) 389 Theophilus of Edessa 385 Thumāma bt. al-Ḥakam b. Abī l-ʿĀṣ → Umāma bt. al-Ḥakam b. Abī l-ʿĀṣ Tiberias 525–527 Tiberius Caesar 521 al-Tīfāshī, Shihāb al-Dīn Aḥmad 303 Tigris River 212, 235, 236, 238, 241, 245, 249, 311, 323, 325, 329–331, 337, 347 al-Tilimsānī, Shams al-Dīn Muḥammad 303 Timur 547, 548, 551, 555 Timurid(s) 125, 306 Toorawa, Shawkat M. 443, 444 Toqto’a 478 Toron 530–534, 540 Toueir, Kassem 448, 449, 454, 457, 458, 462 Tripoli 173, 305, 528, 529, 532, 541, 543, 548, 550–552, 554 Tsigonaki, Christina 619 Tudgha 92, 95, 96 al-Ṭughrāʾī, Muʾayyad al-Dīn 307 Tughtagin, Ẓahīr al-Dīn (atabeg of Damascus) 523, 524 Tulūl al-Shuʿayba 587 T‘umas 603, 604 Tunis 140–151 Turk(s) 122, 130–135, 166, 215, 220–223, 226, 299, 394, 534 Ṭūs 52 Tūz 274 Tūzūn 416, 417, 419, 420, 426, 427 Tyre 161, 522, 523, 528–530, 535 Ukhaydir 587 ʿUlayya bt. al-Mahdī 56, 63 Umāma bt. al-Ḥakam b. Abī l-ʿĀṣ 33 ʿUmar (i) b. al-Khaṭṭāb, caliph 31, 37, 50, 316, 348, 395, 397 ʿUmar (ii) b. ʿAbd al-ʿAzīz 18, 37, 41, 50 ʿUmar b. Faraj 272

index ʿUmar b. al-Nuʿmān (Arabian Nights character) 505, 506, 511, 513 al-ʿUmarī, Ibn Faḍl Allāh 304 Umayya b. ʿAbd Shams 12, 13, 22 Umayyad(s) 4, 5, 12–21, 29, 32, 34, 40– 42, 47, 50, 64, 78–80, 85, 92, 93, 100, 102, 141, 150, 164, 201, 207, 208, 221, 225, 241, 242, 249, 250, 259– 261, 262, 264, 265, 271, 277, 278, 304, 305, 315, 358, 372, 385, 443, 455, 457, 464, 478, 568, 569, 578, 587, 588 Umm Abān bt. al-Ḥakam b. Abī l-ʿĀṣ 33 Umm Abān bt. ʿUthmān b. ʿAffān (Umm Abān bt. Shayba) 32, 34 Umm ʿAbd al-Raḥmān bt. Muʿāwiya 31 Umm ʿAbdallāh bt. ʿAbdallāh b. ʿAmr b. al-ʿĀṣ 37, 38 Umm ʿAbdallāh bt. ʿAbdallāh bt. ʿAmr b. ʿUthmān 38 Umm ʿAmr al-Dawsī 29 Umm ʿAmr bt. ʿAbdallāh b. Khālid b. Abī l-ʿĪṣ 40 Umm ʿĀṣim Ḥafṣa bt. ʿĀṣim b. ʿUmar b. alKhaṭṭāb 37 Umm Ayyūb bt. ʿAmr b. ʿUthmān 37 Umm al-Banīn bt. ʿAbd al-ʿAzīz b. Marwān 38, 40 Umm al-Banīn bt. al-Ḥakam b. Abī l-ʿĀṣ 32, 33 Umm al-Banīn bt. ʿUyayna al-Fazarī 27 Umm Ḥabība bt. Abī Sufyān → Ramla bt. Abī Sufyān Umm al-Ḥajjāj bt. Muḥammad b. Yūsuf alThaqafī 40 Umm al-Ḥakam bt. ʿAbd al-ʿAzīz b. Marwān 18, 40 Umm al-Ḥakam bt. al-Ḥakam b. Abī l-ʿĀṣ 33 Umm Ḥakīm bt. ʿAbd al-Muṭṭalib b. Hāshim 22 Umm Khurmān (Beacon) 268 Umm Kulthūm bt. ʿAbd al-Malik b. Marwān 35 Umm Kulthūm bt. ʿAbdallāh b. ʿĀmir b. Kurayz 30 Umm Kulthūm bt. Muḥammad b. ʿAbdallāh, the Prophet 27 Umm Kulthūm bt. Muḥammad b. Rabīʿa 32

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

index Umm al-Nuʿmān bt. al-Ḥārith b. Abī ʿAmr 25 Umm al-Qāsim bt. ʿAbdallāh b. Khālid b. Asīd 32 Umm Salama bt. al-Ḥakam b. Abī l-ʿĀṣ 33, 81 Umm Shuʿayb, or Umm ʿUthmān, bt. [Saʿīd, or Shuʿayb, b.] Zabbān 32, 41 Umm Sulaymān bt. ʿĀmir b. al-Ḥarash 32 Umm ʿUthmān bt. Khālid b. ʿUqba b. Abī Muʿayṭ 32 Umm al-Walīd 587 Umm Yūsuf al-Baʿītha bt. Hāshim b. ʿUtba b. Rabīʿa b. ʿAbd Shams 25 Unur, Muʿīn al-Dīn (atabeg of Damascus) 528–530 Upper Zab River 238 ʿUqba (sīra character) 514, 515 Uruk Khātūn 478 Usāma b. Munqidh 517, 518, 529, 530 Usayd b. al-Akhnas al-Thaqafī 33 ʿUtba, female slave of al-Mahdī 27, 56 ʿUthman Qarā Yuluk (Aqquyunlu ruler) 551 ʿUthmān b. ʿAffān (caliph) 13, 18, 19, 22, 27, 28, 32, 34, 81, 607 ʿUthmān b. al-Ḥakam 23 ʿUthmān [b. Saʿīd] b. Kathīr 5 ʿUthmān b. Ṣāliḥ 394 Uzbek 478 Vahram Č‘obin 607, 609 Vałarš/Wālaxš ii 608 Valencia 145 Van Gelder, Geert Jan 86, 309 Varaztiroc‘ Bagratuni 600, 604 Vardik (prince of Mokk‘) 595 Venice 505, 626 Verkinderen, Peter 435, 441 Viroy (Catholicos of Ałuank‘) 596 Wādī al-Taym 523 Wādī Bāndaway 244, 249 Waḍī Wajj 23 al-Walīd (i) b. ʿAbd al-Malik 20, 39, 41, 69, 260 al-Walīd (ii) b. Yazīd b. ʿAbd al-Malik 41 al-Walīd b. ʿĀmir al-Yazanī 5 al-Walīd b. ʿUqba b. Abī Muʿayṭ 32

651 Walīla (Roman Volubilis) 93, 95–97, 100– 102, 104, 106, 107, 109, 111 Wallāda bt. al-ʿAbbās al-ʿAbsī 35 Walter i Brisebarre (lord of Beirut) 526 Walter ii Brisebarre (lord of Beirut) 532 Walter of Quesnoy 533 al-Wansharīsī 173 al-Wāqidī 396, 397 Waṣīf 215, 220 Wāsiṭ 333, 413, 416, 418–422, 425 al-Wāthiq (Abbasid caliph) 70, 71, 73, 74, 271–274, 279 Watt, W. Montgomery 21 al-Waṭwāṭ, Jamāl al-Dīn 303, 306 West Arabia → Ḥijāz West Asia 121, 125, 128 White Mountains (Crete) 621 White, Hayden 258 Wilkinson, Tony 457 William of Bures (lord of Tiberias) 525, 527 William of Tyre 521, 524, 525, 527–533, 540 Wladislav ii (king of Bohemia) 542 Woods, John 128 Xoṙox Ormizd/Farrukh Hormizd 601, 611 Xoṙoxazat/Farrukhzād 601–605, 611 Yaḥyā b. Aktham 58 Yaḥyā b. al-Ḥakam al-Ghazāl 84 Yaḥyā b. al-Ḥakam b. Abī l-ʿĀṣ 32 Yaḥyā b. Khālid b. Barmak 358, 359 Yaḥyā b. Maʿīn, Abū Zakariyyāʾ 370 Yalbaq (father of ʿAlī b. Yalbaq) 415 al-Yamāma 258 al-Yaʿqūbī 69, 221, 330, 369, 389, 390, 397 Yāqūt b. ʿAbdallāh al-Ḥamawī 6, 8, 9, 451, 453, 454, 457–459, 509, 510 Yathrib → Medina Yazdgird i 394 Yazdgird ii 608 Yazdgird iii 602, 603, 605 Yazdī, ʿAbd al-Jalīl 367, 374 Yazīd (i) b. Muʿāwiya b. Abī Sufyān 13, 17, 19, 30, 31, 37 Yazīd (ii) b. ʿAbd al-Malik b. Marwān 35, 38, 40, 41 Yazīd (iii) b. al-Walīd b. ʿAbd al-Malik 40, 41 Yazīd b. Abī Sufyān 25

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano

652 Yazīd b. Ḥātim 92, 95 Yazīd b. Khumayr 5 Yemen 4, 48, 148, 254 Young, G.M. 300 Yūnus b. Bukayr 9, 341, 343 Yūsuf b. Yaḥyā al-Maghāmī 40, 41, 48, 69, 74, 274, 364, 413, 415, 420–422, 427 Ẓafār 4, 6–9 Zagros Mountains 235 Zaḥḥāk 484 al-Ẓāhir Ghāzī (b. Salāḥ al-Dīn) 158 Zakarawayh b. Mirawayh 256, 257, 279 Zakkār, Suhayl 4, 255 Ẓālim (sīra character) 507 Zamzam 263, 270 zanādiqa 63 Zanānīr, Princess of Qalʿat al-Shams (sīra character) 514

index Zangī, ʿImād al-Dīn (atabeg of Mosul) 527, 528, 530 Zangid(s) 533, 585 Zareh 608 Zaynab bt. ʿAbd al-Raḥmān 32 Zaynab bt. ʿAmr al-Makhzūmī 34 Zaynab bt. al-Ḥakam b. Abī l-ʿĀṣ 33 Zaynab bt. al-Ḥasan b. al-Ḥasan b. ʿAlī 38 Zenodotium 451 Zīj al-Ṣābiʾ 466 Ziyād b. Abī Sufyān 25, 31, 210 Ziyādat Allāh i 104 Zubāla 269, 270, 280 Zubayda bt. Jaʿfar 263, 269, 305, 309, 310, 313, 341, 343, 353 al-Zubayrī, Abū ʿAbdallāh Muṣʿab b. ʿAbdallāh 21–26, 29, 33, 36, 38, 39 Zumurrud, Ṣafwat al-Mulk 527

- 978-90-04-52524-5 Downloaded from Brill.com 01/29/2024 12:42:59PM via Università Statale degli Studi di Milano