Egypt and Empire: The Formation of Religious Identity After Rome 9789042940314, 904294031X

Table of contents :
Cover
Title
Copyright
Table of contents

Citation preview

BRITISH MUSEUM PUBLICATIONS ON EGYPT AND SUDAN 11

EGYPT AND EMPIRE The formation of religious identity after Rome

edited by

Elisabeth R. O’CONNELL

PEETERS

EGYPT AND EMPIRE

BRITISH

MUSEUM

PUBLICATIONS

ON

EGYPT

AND

EGYPT AND EMPIRE

The formation of religious identity after Rome

edited by

Elisabeth R. O’CONNELL

PEETERS LEUVEN – PARIS – BRISTOL, CT 2022

SUDAN

11

Cover illustration: Detail of a c. AD 300 wall painting from the imperial cult chapel at Luxor temple depicting Roman senators gathering below the tetrarchs’ thrones. Reproduced by permission of the American Research Center in Egypt. This project was funded by the United States Agency for International Development (USAID). Photo by Arnaldo Vescovo, February–March 2009.

A catalogue record for this book is available from the Library of Congress. ISBN 978-90-429-4031-4 eISBN 978-90-429-4032-1 D/2022/0602/18 © 2022, Peeters, Bondgenotenlaan 153, B-3000 Leuven, Belgium No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage or retrieval devices or systems, without the prior written permission from the publisher, except the quotation of brief passages for review purposes.

For Ethan Asher (b. 2015) and Ayla Seren (b. 2018)

TABLE OF CONTENTS Contributors ...................................................................................................................................................... Preface ..............................................................................................................................................................

IX XI

INTRODUCTION Elisabeth R. O’CONNELL One God: Egypt and empire in the First Millennium AD .............................................................................. I. EMPIRE AND RELIGIOUS IDENTITY: DEFINITIONS,

3

SOURCES AND APPROACHES

Roger S. BAGNALL Egyptian religious identities under Roman imperial rule: Critical reflections ................................................

61

Darlene L. BROOKS HEDSTROM Artefacts, archaeology and the archaeologist: Late Antique Christian material culture and the history of archaeology in Egypt ........................................................................................................................................

71

II. CALENDAR AND COSMOLOGY John P. COOPER ‘We desire to know which is the true religion’: Inter-communal rivalry and the verdict of the Nile in an episode from the History of the patriarchs of Alexandria ..............................................................................

111

On BARAK Synchronization and its discords: Calendric reform and imperial politics at the end of the nineteenth century ..............................................................................................................................................................

133

III. TOPOGRAPHY AND COSMOLOGY Katherine BLOUIN What’s faith got to do with it? A diachronic perspective on empire, land and religion ................................

151

Maja KOMINKO Beyond empire: Kosmas Indikopleustes, the Church of the East and the Indian trade in the sixth century AD

173

Yossef RAPOPORT Where did the Christians go? Peasants and tribesmen of the Fayum, AD 1060–1240...................................

199

IV. CUSTOMS AND APPEARANCE Cäcilia FLUCK Textiles in burial practice in Roman and Late Antique Egypt: Continuity and change .................................

217

Thelma K. THOMAS Fashioning ascetic leadership: The enduring tradition of mantles of authority in portraits of Egyptian monastic fathers .................................................................................................................................................

247

V. LANGUAGES AND SCRIPTS Sofía TORALLAS TOVAR Greek and Egyptian: Did linguistic policies exist in Graeco-Roman Egypt? .................................................

279

Arietta PAPACONSTANTINOU Arabic: Language of empire and language of Egypt .......................................................................................

293

TABLE OF CONTENTS

VIII VI. MINORITIES

AND MAJORITIES

David NIRENBERG Egypt, empire and Judaism 650 BC–AD 650 ...................................................................................................

313

Petra SIJPESTEIJN Visible identities: In search of Egypt’s Jews in early Islamic Egypt ..............................................................

325

Roberta Mazza Rethinking persecutions: P. Ryl. III 469 and the Manichaeans in Egypt ........................................................

335

VII. UNIVERSAL AND LOCAL HISTORIES Phil BOOTH Severan Christians between the Roman and Arab empires ..............................................................................

351

Hugh KENNEDY The Muslim elite of early Islamic Egypt ..........................................................................................................

361

CONTRIBUTORS Roger S. Bagnall retired from his posts as New York University Professor of Ancient History in 2017 and as Leon Levy Director of the Institute for the Study of the Ancient World in 2016. Prior to this he had a long career at Columbia University, where he taught for thirty-three years. He has led the fields of ancient history and papyrology, authoring or co-authoring thirty-eight books, editing nine more, and writing thirty-nine book chapters and over 225 articles. In October 2016 he received an honorary doctorate from the Université de Paris-Sorbonne. On Barak is Associate Professor in the History of the Modern Middle East at Tel Aviv University and specializes in the introduction of science and technology into non-Western settings. He has a joint PhD in History and Middle Eastern and Islamic Studies from New York University and did post-doctoral study at Princeton University. His published books include On time: Technology and temporality in modern Egypt (2013) and Powering empire: How coal made the Middle East and sparked global carbonization (2020). Katherine Blouin is Associate Professor in Roman History at the University of Toronto. Her research interests include the Nile Delta, multiculturalism and environmental history, as well as the ways in which imperialism intersects with classics, papyrology and Egyptology. She is the author of Triangular landscapes: Environment, society and the state in the Nile Delta under Roman rule (2014). Phil Booth is A. G. Leventis Lecturer in Eastern Christianity at the University of Oxford. His publications include Crisis of empire: Doctrine and dissent at the end of Late Antiquity (2013). Darlene L. Brooks Hedstrom is the Myra and Robert Kraft and Jacob Hiatt Chair in Christian Studies at Brandeis University, where she is Associate Professor in the departments of Classical Studies and Near Eastern and Judaic Studies. Her publications focus on the archaeology and history of monastic settlements in the Byzantine Near East with a particular focus upon Egypt. She is currently the Senior Archaeological Consultant for the Yale Monastic Archaeology Project and former Director of Archaeology for the Yale Monastic Archaeology Projects in Wadi Natrun and in Sohag, Egypt. Her most recent book is The monastic archaeology of Late Antique Egypt: An archaeological reconstruction (2017). John P. Cooper is Senior Lecturer in Arabic Studies and Islamic Material Culture at the University of Exeter. He has a PhD in Maritime Archaeology and has a particular interest in the maritime technologies and cultural landscapes of the Nile and the Medieval Islamic world. He is the author of the book The Medieval Nile: Route, navigation and landscape in Islamic Egypt (2014). Cäcilia Fluck is a Curator at the Skulpturensammlung und Museum für Byzantinische Kunst, Staatliche Museen zu Berlin. She is co-convener of the Textiles of the Nile Valley Study Group and has co-edited ten volumes in its publication series and others. She has co-curated several exhibitions, including ‘Josef Strzygowski und die Berliner Museen’ (2012–2013) and ‘Ein Gott: Abrahams Erben am Nil. Juden, Christen und Muslime in Ägypten von der Antike bis zum Mittelalter’ (2015). Hugh Kennedy is a Professor of Arabic at SOAS University of London. Prior to this post he was a Professor of Middle Eastern History at the University of St Andrews. Hugh specializes in the area of the political, economic and social history and archaeology of the early Islamic Middle East c. 600–950. He is Fellow of the Royal Society of Edinburgh (FRSE), Fellow of the British Academy (FBA) and Fellow of the Royal Asiatic Society. He is the author of The great Arab conquests: How the spread of Islam changed the world we live in (2007) and The caliphate: A Pelican introduction (2016). Maja Kominko manages the Cultural and Open Access grants of the Arcadia Fund, a foundation working to support endangered culture and nature. She holds a PhD in Byzantine History from Oxford University. Her published

X

CONTRIBUTORS

articles cover a range of subjects, including Late Antique intellectual history, art history and cartography, and cultural exchange in the Medieval Levant, particularly as documented in illustrated Syriac manuscripts. Her latest book is The world of the Kosmas: Illustrated Byzantine codices of the Topography (2013). Roberta Mazza is Lecturer in Graeco-Roman Material Culture at the University of Manchester. She is the papyrologist affiliated to the John Rylands collection; her research focus is Late Antique Egypt. David Nirenberg is Deborah R. and Edgar D. Jannotta Distinguished Service Professor of Medieval History and Social Thought at the University of Chicago, where he is also Dean of the Division of the Social Sciences. His books have focused on how Jewish, Christian and Islamic societies have interacted with and thought about each other. These include Communities of violence: Persecution of minorities in the Middle Ages (1996), Anti-Judaism: The Western tradition (2013), Neighboring faiths: Christianity, Islam and Judaism, medieval and modern? (2014) and Aesthetic theology and its enemies: Judaism in Christian painting, poetry and politics (2015). Elisabeth R. O’Connell is Byzantine World Curator at The British Museum. She received her PhD from the University of California, Berkeley (2007). Her research focuses on aspects of Late Antique social history and archaeology in Egypt. She is editor of Egypt in the First Millennium AD (2014), Abydos in the First Millennium AD (2020) and co-editor of Egypt: Faith after the pharaohs (2015), which accompanied the BM exhibition of the same title (2015–2016). Arietta Papaconstantinou is an Associate Professor at the University of Reading. Her research is focused on the religious, social and economic history of Egypt and the Near East during the transition from the Roman empire to the Caliphate. Her edited and co-edited volumes include Conversion in Late Antiquity: Christianity, Islam and beyond (2015), The multilingual experience in Egypt from the Ptolemies to the Abbasids (2010) and Becoming Byzantine: Children and childhood in Byzantium (2009). Her current project is entitled ‘Credit, debt, and dependence in early Islamic Egypt and Southern Palestine’. Yossef Rapoport is Professor in Islamic History at Queen Mary University of London. He studied at the universities of Tel Aviv, Princeton and Oxford, before joining Queen Mary in 2008. His research interests include the history of women, slaves, peasants and everyday life as well as Islamic Medieval maps. He has several publications in these fields, including Marriage, money and divorce in Medieval Islamic society (2005) and The Book of curiosities (2008), which is also available online. He is a member of the editorial board of Islamic Law and Society. Petra Sijpesteijn is Professor of Arabic at Leiden University. She received her PhD from Princeton University (2004). She was the founding president of the International Society for Arabic Papyrology (2001–2014). Since 2014 she has been director of the Leiden University Centre for the Study of Islam and Society (LUCIS) and in 2017 started an ERC Consolidator Grant project, ‘Embedding conquest: Naturalizing Muslim rule in the early Islamic empire (600–1000)’. Her recent publications include Documents and the history of the early Islamic world with A. T. Schubert (2014) and Shaping a Muslim state: The world of a mid-eighth-century Egyptian official (2013). Thelma K. Thomas is an art historian specializing in Late Antique, Byzantine and Eastern Christian art and material culture. She is Associate Professor at the Institute of Fine Arts, New York University. She is the author of Designing identity: The power of textiles in Late Antiquity (2016), Textiles from Karanis, Egypt, in the Kelsey Museum of Archaeology (2001) and Late Antique Egyptian funerary sculpture: Images for this world and the next (2000). Her current projects focus on textiles, dress, representation and sacred space. Sofía Torallas Tovar is Associate Professor in the Departments of Classics and Near Eastern Languages and Civilizations at the University of Chicago. She has published widely on the social, religious and intellectual history of Graeco-Roman Egypt, and has edited and co-edited four volumes of the Montserrat papyri. She is also editing the ostraca excavated in Aswan by the Swiss Institute, and collaborating on the editions of the Coptic Gospel of Mark and the collected works of Shenoute.

PREFACE Most of the contributions in this volume took shape at a two-day colloquium timed to coincide with the British Museum exhibition Egypt: Faith after the pharaohs (29 October 2015–7 February 2016). Entitled ‘Egypt and empire: Religious identities from Roman and modern times’ (9–10 December 2015), the colloquium formed part of the public programming in support of the exhibition and was generously sponsored by the Leverhulme Trustfunded research project, ‘Empires of faith’ (2013–2018), led by Jaś Elsner. The exhibition focused on the period from 30 BC, when Egypt officially became part of the Roman empire until the end of the Fatimid period in AD 1171. It introduced audiences to Roman Egypt, where most people continued to worship many gods; then traced the following centuries, marked by a transition to a majority Christian population by the fifth century, then to a Muslim majority in the eleventh century, with Jewish and other minority communities periodically thriving alongside. By highlighting moments of conflict and coexistence, the exhibition provided some of the background for understanding current events. The exhibition was developed within its own institutional and geopolitical contexts. Both the exhibition and associated book were collaborations between the Staatliche Museen zu Berlin (SMB) and the British Museum (BM). Informal discussion between the SMB and BM began in 2010, when Frederike Seyfried suggested a post-pharaonic exhibition highlighting the co-existence of Jews, Christians and Muslims, with contemporary relevance. The lead curators at the SMB were Cäcilia Fluck of the Skulpturensammlung und Museum für Byzantinische Kunst and Gisela Helmecke of the Museum für Islamische Kunst and, at the BM, the present author, each supported by many colleagues across the SMB and BM, respectively.1 Discussions progressed over the course of the years following the 25 January 2011 Revolution in Egypt, with the BM iteration approved for development in 2013 and a scheduled exhibition opening date assigned for autumn 2015. It was the last exhibition scheduled to open under the BM Director Neil MacGregor (2002–2015) before he transitioned to his role as one of three founding directors for Berlin’s Humboldt Forum. The BM exhibition was generously supported by the Blavatnik Family Foundation and opened by British Museum Trustee, novelist and commentator Ahdaf Soueif. From the beginning, the two exhibition displays were conceived very differently. The SMB narrative took a thematic approach, whereas the BM’s display was chronological. Decisions were informed as much by the differing audiences expected in Berlin and London as by the curators’ own sensibilities. The SMB’s exhibition was mounted in the Bode Museum, which houses the Skulpturensammlung und Museum für Byzantinische Kunst (Sculpture Collection and Museum for Byzantine Art). Visitors pay to enter the Museum, but temporary exhibitions are free once inside. The context of viewing the SMB exhibition was thus already situated in a museum dedicated to the First Millennia AD. SMB’s themed exhibition was generally structured around daily life, ending with gravestones of men and women of different faiths placed together, in order to emphasize what is shared. By contrast, the BM offers free entry to all, but temporary exhibitions require payment. The permanent collection houses hand axes over a million years old, contemporary artworks and much in between, requiring temporary exhibitions to situate themselves firmly in time and space. The BM exhibition was structured broadly according to epoch: the Roman period, when Christianity emerged and Judaism developed; the Byzantine period, when Christianity spread; and the Umayyad, Abbasid and Fatimid periods when Islam was introduced and, in time, widely adopted, while Jewish and Christian communities periodically flourished. The exhibition book did double duty for both the SMB and BM shows, with both chronological and thematic sections. The title of each exhibition was likewise selected on the basis of the two institutions’ different audiences. The SMB’s title, ‘One God: Abraham’s legacy on the Nile’, emphasized the shared heritage of Judaism, Christianity and Islam. At the BM, the exhibition curator’s preferred title, ‘One God: Egypt and empire’, was rejected as at once too provocative (‘One God’) and generic (‘empire’). Title selection is always a fraught enterprise, requiring a fine balance between content, marketing and other concerns. The title ‘Egypt: Faith after the pharaohs’ was chosen to capitalize upon interest in Egypt and familiarity with the term ‘pharaohs’ and thus to attract the largest audience possible.

1

For full acknowledgements, see the exhibition book, Egypt: Faith after the pharaohs (London, 2015).

XII

PREFACE

Exhibitions are, of course, storytelling of a particular kind. They are structured by space and time. Visitors are led on a journey that typically lasts between 45 and 60 minutes. Exhibitions must operate at multiple levels, visually and verbally, connecting with the viewer emotionally, intellectually and aesthetically. The story told in this exhibition was already complicated—the establishment of rabbinic Judaism and the successive emergence of Christianity and Islam in Egypt in the First Millennium AD. The narrative also required deep sensitivity toward members of each living faith. The panel and label texts, both constrained by word-count, needed to satisfy general audiences, as well as subject specialists. In exhibitions, there are, of course, no footnotes and no credits aside from lenders and funders named on a panel on the way to the gift shop. The colloquium showcased research that had informed the development of the BM exhibition. It also provided the opportunity to ask each presenter to address subjects beyond what could be conveyed in the exhibition and to bring together scholars working in different disciplines. For general exhibition audiences, the colloquium sought to share new research on Roman through modern Egypt, and to showcase the full range of scholarship that a specialization on Egypt affords, due to the extraordinary range and abundance of surviving sources. For their participation, I warmly thank the presenters at the colloquium and contributors to the volume. Some presenters are not represented in this book (Jaś Elsner, Alison Gascoigne, Will Hanley, Judith Kindinger and Ben Outhwaite) with the unfortunate consequence that Medieval and Modern Egypt as well as archaeology and art history are not as well represented as the history of Roman and Late Antique Egypt. Another contribution was subsequently commissioned in order to partially redress the balance between history and archaeology (Darlene L. Brooks Hedstrom). I also thank the invited chairpersons who led and fostered discussion at the colloquium: Sophie Lunn-Rockliffe, Amandine Mérat, Venetia Porter, Dominic Rathbone, Mariam Rosser-Owen, Roberta Tomber and Susan Walker. I am grateful to Jaś Elsner and JD Hill for their support in funding the colloquium as part of the ‘Empires of faith’ project. I thank staff in the Department of Ancient Egypt and Sudan (now Egypt and Sudan) for help with the logistics of organizing and running the colloquium, foremost Tania Watkins, Eirini Koutsouroupa and Neal Spencer. Numerous other colleagues kindly advised on aspects of the book at various stages: Doris Abouseif, Alan Bowman, Alberto Camplani, Malcolm Choat, Susanna Elm, Miriam Frankel, Johannes den Heijer, Matthew Gordon, Brendan Haug, Erica Hunter, Karel Innemée, Ceyda Karamursel, Yaacov Lev, Maria Mavroudi, Volker Menze, Maged Mikhail, Ian Moyer, Lucian Reinfandt, Christina Riggs, Giovanni Ruffini, Gesa Schenke, Peter Sheehan, Ruiha Smalley, Ellen Swift, Marie Vandenbeusch, Naïm Vanthieghem, John Peter Wild, Philip Wood and Greg Woolf. I am responsible for any errors in judgement or fact that may remain. I thank the British Museum Scholarly Publications Fund for supporting this volume. The copyeditor, Carolyn Jones, has provided consistency and good humour over the course of the politically tumultuous years over which this volume has come together. Bert Verrept and the excellent team at Peeters have steered the volume through this period, which culminated in the 2020 global health crisis. I thank Louise Bertini and Andreas Kostopoulos at the American Center for Research in Egypt for permission to use the image of a detail of a wall-painting from the c. AD 300 imperial cult chapel in Luxor temple for the cover. This volume is dedicated to my two children, who were born during the preparation of the colloquium and the editing of this volume, respectively.

Elisabeth R. O’Connell

INTRODUCTION

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD Elisabeth R. O’CONNELL

In the course of the First Millennium AD, rabbinic Judaism, Christianity and Islam developed the systems of belief and practice that most of their adherents would recognize today (Fowden 2014, 54). Across Eurasia and North Africa, empires rose and fell (PreiserKapeller, Reinfandt and Stouraitis 2020), each adopting a universalizing faith which distinguished it broadly from its neighbours (Elsner 2020). So too did their visual cultures develop in relation to the religious iconography of traditional cults and, in time, to each other (Elsner 2020). In Egypt, our sources are particularly rich due to the land’s arid climate and the unparalleled survival not only of stone, ceramic and metalwork, but also of organic material such as textiles, wood and manuscripts found on papyrus, parchment and paper. Among these latter, everyday documents in particular convey the day-to-day experience of faith communities living under empire. Today, they are the closest approximation we have to overhearing the conversations of Jews, Christians and Muslims in the First Millennium AD.1 Egypt provides both a case study and a privileged vantage point, offering exceptional access to the everyday lives of past individuals and their communities living under successive empires. The First Millennium coincides with Egypt’s transformation from a self-governed state into a province of successive empires. The modern study of Egypt began at a time when it was again an imperial province and, later, a colonial protectorate of European empires (e.g., Reid 2002 and 2015; Carruthers 2015; for Late Antiquity, more generally, Elsner 2020), even as it colonized Sudan (Powell 2003; Sharkey 2003). Thus both Egypt’s history in the First Millennium and the writing of that history are beset with the biases of modern scholars’ own religious, ethnic, class and other affiliations. This volume brings together over a dozen of the world’s leading specialists to explore the dialectical

1

The case is strongest for private letters. Although often written by amanuenses and governed by formulae, the letter did not ‘simply substitute for face-to-face spoken communication: it

interplay between empire and religious identity through a series of case studies from Egypt. Evidence from Egypt suggests that it was precisely in the context of empire that ‘religious identity’ emerged as a distinctive marker used in ways similar to ethnonyms. Thereafter, like other categories of peoples grouped by citizenship, ethnonym, profession or (other) association, adherents of a particular faith could be identified for official purposes. Using the unrivalled abundance and variety of material culture surviving from Egypt, this volume explores the formation, renegotiation and reconstitution of religious identities from the Roman period forward. Whereas Egypt’s ‘pharaonic’ millennia (c. 3000– 30 BC) have been studied as a coherent whole, the last two millennia are often studied as fragments. The volume offers a different approach by covering together periods that are usually treated separately in different academic disciplines. Before outlining the contributions to this volume, this introductory chapter provides a short chronological overview introducing the major developments in the history, archaeology and visual culture of Egypt in the First Millennium AD. While general, it is intended to provide a framework within which the more detailed and nuanced contributions to this volume might be situated. Egypt, empire and religious identity Egypt’s cultural and economic significance, both as a producer and a hub for trade between the Mediterranean, Arabia, East Africa and the Indian Ocean, established it as a geopolitical force since the earliest times. As it became a part of successive empires, rather than the seat of empire, its agricultural yield and prime strategic position made it the object of competition. Under Roman rule, Egypt’s wealth flowed first to

augmented and extended it: performing functions that oral communication could not’ (Choat 2017, 72 and for recent bibliography, 17–18).

E. R. O’CONNELL

4

Mediterranean Sea Paraitonion

Canopus Alexandria

Gaza za S Sais

Naukratis atis

Marina el-Alamein

Thm hmouis hmo mou s DELTA DE EL A

SCE TIS

Athri Athribi hribis

Pelousion

Rhinokol R oloura

WAD AD I T UMI L AT

Siwa Oasis

Babylon

Petra

Kly lysma

Memph Memphis p

Aila

FAY U M

Arsino noe no le

SINAI

Ni

Herakleo He erakleopolis magna Bahariya Oasis

suf B ahr Yu

Oxyrhynchos

Herm mopolis magna

Ankyro Ankyronpolis Kynopolis HA VIA

DR

IAN

A

Antinoopolis

EASTE R N D E S E RT MONS PORPHYRITES TES

Koussai ai Lykopolis ykopolis olilis is LA N Antaeopolis MONS CLAUDIANUS Ant Hypsele Aphrodito p ito Panopolis Atripe Thinis Ptolemaiss Dendera Den D endera d Abydos os Koptos optos Diospolis iospolis ospolis parva

Farafra Oasis

WE ST E R N D E S E RT Dakhla

Memnoneia/Jeme me Hermonthis is

Oas Oasis

Red Sea Myos Hormos

Diospolis ma magna (Thebes) bes

Latopolis is Kharga Oasis

Apollonopolis mag agna Kysis

MONS SM SMARAGDUS S

O Ombos Phila ilae

Syene First Cataract

Lak aake ak k Nasser

Hierasykaminos rasy s Pr i n c i p a l q u a r r i e s Pr i n c i p a l d e s e r t r o a d s

(Abu Simbel))

0

Primis is ((Qasr Ibrim)

200 km

Map 1: Roman Egypt (Illustration by Claire Thorne).

Berenike ke

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Rome, then Constantinople and, later, under successive caliphates, to Umayyad and then to Abbasid territories. At the end of the First Millennium AD, its rulers once again governed from Egypt, when the Fatimids (AD 969–1171) established their capital in Cairo. A chronological review provides the opportunity to focus on transitions between chronological periods and different polities, and thus to explore continuity and change, similarity and difference. By embracing empire as a unifying framework, this volume traces Egypt’s privileged position as part of successive empires, beginning with its official annexation by Rome.2 These empires provided an institutional structure for managing and taxing trade within and beyond their borders and ensuring that their capitals and armies were provisioned. Empires also played a role in asserting or underpinning religious authority, whether it was accepted or resisted. Despite the asymmetrical balance of power, influences are multi-directional (e.g., Dietler 2010, 50–53; Tyson Smith 2014). Empires were constituted by provinces, cities, households and individuals, each with their own agency, operating at different levels. The concept of ‘religion’ as we define it today would have been incomprehensible to pre-modern individuals and their communities. The vocabulary used in the sources to describe what we would call religion has been defined, interrogated and reassessed, with some scholars arguing that the term should not be used at all (Nongbri 2013; Boyarin and Barton 2016). Others seek to refine the language of its study, to define a nuanced and self-conscious approach by adopting some of the aspects of identity theory in order to better appreciate the usefulness of ‘groupness’ (Rebillard 2012, 4–5; Rebillard and Rüpke 2015). ‘Religious identity’ thus sits alongside ethnic, professional, familial and other identities that the individual or group may privilege in a given circumstance or upon a given occasion. ‘Religious identity’ recognizes ‘religion’ as a modern construction, while at the same time preserving a category for study (Frankfurter 2015). This terminology is adopted here, if advisedly, recognizing religious identity as one aspect of social identity.

2

For definitions of ‘empire’ in archaeology and history, I rely on Alcock et al. 2001 and Bang and Bayly 2011.

5

Roman Egypt By 168 BC, Egypt had effectively become a Roman protectorate. After Octavian defeated the last Ptolemaic ruler of Egypt, Cleopatra VII, together with Mark Antony in 31 BC, Egypt officially became part of the Roman empire (Map 1; Bagnall and Rathbone 2016). Octavian, now known as Augustus, used propaganda to assert Rome’s dominance; imperial statues and coins with his image broadcast a message of military and political supremacy throughout the empire and beyond (Figs 1–2) (Zanker 1988; Opper 2014). Alexandria was one of the most important cities in the Roman world, together with Antioch, Rome and, later, Constantinople (Fig. 3). Founded in 331 BC by Alexander the Great, the city was famed for its lighthouse, great libraries, temples and public spaces, its philosophers, poets, scientists and physicians (McKenzie 2007). In the Roman period, Alexandria was at the centre of a vast network of trade extending from the Atlantic coast in the west to India in the east, with Egypt supplying grain, wine, oil, papyrus, stone and gems (Blue and Khalil 2011). People from all points of the compass mingled in the city. Sailing into the famed harbour of Alexandria, visitors saw the magnificent Caesareum, a temple founded by Cleopatra VII and completed by Augustus (Tkaczow 1993). Built to honour the divine Caesar, it was later used for the worship of Roman emperors, including Augustus. Philo of Alexandria––a Jewish philosopher writing in Greek, whose nephew became the Roman governor of Egypt–– described the spectacle of sailing into Alexandria’s harbour and of seeing the city resplendent in images of emperors, going on to describe both statues and painted portraits (Figs 4–5). When the Greek citizens of Alexandria accused Jews of failing to honour the emperor Caligula in AD 38, violence erupted. Jewish leaders were tortured and killed, their synagogues defiled with imperial statues or destroyed. Embassies representing both sides were sent to Rome to give an account of events to the emperor. In his response to the letter in AD 41, Claudius gives the Alexandrians permission to erect statues to him and

6

E. R. O’CONNELL

Fig. 1: Bronze head of Augustus, c. 27–25 BC, excavated from under the threshold of a temple in ancient Meroe, Sudan, where it was buried as a trophy after it was looted, probably from Syene/Aswan on the Roman empire’s southern frontier (BM 1911,0901.1).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

7

Fig. 2: Gold coin of Augustus, 27 BC, mint unknown. The back of the coin celebrates Augustus’ conquest, with the image of a crocodile representing Egypt and the Latin inscription Aegypt[o] capta: ‘Egypt captured’ (BM 1897,0604.4).

Fig. 3: Silver gilded furniture fittings depicting Constantinople, Antioch, Rome and Alexandria, four of the most important cities of the late Roman world, personified as goddesses, or Tychai, c. AD 330–370. Found as part of the Esquiline Treasure, on the Esquiline Hill, Rome (BM 1866,1229.21–24).

8

E. R. O’CONNELL

Figs 4–5: Marble statues of the emperors Marcus Aurelius and Septimius Severus, c. late second and early third centuries, respectively. Found in 1801 on the coast of the Eastern Harbour of Alexandria, which was identified at the time as the location of the Caesareum. Both were ceded by the French to the British according to the terms of the Treaty of Alexandria and donated by King George III to the British Museum (BM 1802,0710.1–2).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

his family and to make his birthday a holiday, but he denies their request to build a temple for his worship. Importantly for the Jews, he reconfirms their status and tells the Alexandrians not to interfere with their worship and customs (P.Lond. VI 1912). But this peace did not last, and Jews were periodically exiled or killed. Imperial images were displayed throughout the empire. In Egypt, Roman emperors were also shown in the traditional roles of pharaohs (Brophy 2015; Smith and Ward-Perkins 2016). Even into the reign of Diocletian (r. AD 284–305), the Roman emperor was shown in ancient Egyptian dress standing in front of an offering table burning incense and pouring a libation before the mummified manifestation of the bull-god Buchis (Fig. 6). The last recorded Buchis-bull burial took place in AD 340, under the Christian emperor Constantius II, but its dating formula records its dedication in year 56 of Diocletian (Grenier 1983; and below, p. 45). At the same time, traditional Egyptian gods were sometimes depicted in Roman dress. The god Horus was the divine representation of the living pharaoh, usually shown as a man with a falcon’s head. After the Roman conquest, the Egyptian god was sometimes shown in Roman armour with a general’s cloak —the clothing of power—and in poses characteristic of Roman emperors (Figs 7–8). The question remains, how would such images have been understood by ancient viewers? Is Horus represented as an emperor, the emperor as a god, or both? Systems of belief were complex. In Roman Egypt there were hundreds of gods to help navigate this life and the next. People from different cultures sometimes understood one another’s gods as forms of their own deities. In this way, gods accumulated qualities and features borrowed from others. Isis, for instance, was identified with so many other gods that she became almost universal in character (P.Oxy. XI 1380; Blouin, this volume). Under the influence of Greek philosophical thought, some universalized deities came to be seen as an expression of ‘one God’—heis theos. Those given this title, such as Serapis, were not thought of as the only god, but as the foremost among the gods (Mitchell and Van Nuffelen 2010). As a family unit, Serapis, Isis and Harpokrates (or Horus-the-child) developed from the ancient Egyptian divine triad of Osiris, Isis and Horus (Fig. 9). As father, mother and child, these Egyptian gods provided a mythological model for royal succession, ensuring the country’s peace and stability when a king died (M. Smith 2017). Under the Ptolemies, the combination

9

Fig. 6: Sandstone stela of Diocletian offering to the Buchis bull, AD 288, excavated at the Bucheum in Hermonthis/ Armant. The emperor is shown in the traditional role of ancient Egyptian kings before the mummified bull-god Buchis (BM EA 1696).

god Osiris-Apis was given the features of a Greek deity and became Serapis. The cults of Serapis, Isis and Harpokrates, individually and collectively, were exported to the farthest reaches of the Roman empire and beyond (cf. Apuleius, Metamorphoses XI; Keulen et al. 2015). In Egypt, only purified temple priests were traditionally allowed to enter the innermost parts of the temple. Their appearance, with heads shaved and wearing linen garments, set them apart. By the second century, men associated with the cult of Serapis are thought to have been identifiable by three locks falling on the brow and star diadems (Figs 10–11) (Walker 2000, 59–60). The wider temple complex was a community hub, as shown by the survival of invitations sent to guests for a meal

10

E. R. O’CONNELL

Fig. 7: Bronze figure of Horus in Roman military costume, first–third century AD, Egypt (BM EA 36062).

Fig. 8: Limestone figure of Horus, first–third century, Egypt. Wearing a cloak over feathered mail armour, which doubles as falcon feathers, Horus is seated in the pose of senior Olympian deities and Roman emperors (BM EA 51100).

Fig. 9: Steatite patera, third–fourth century AD, Egypt. Used for pouring liquid offerings to the gods, this patera shows the gods Harpokrates, Serapis and Isis and is decorated with typical classical imagery, including bunches of grapes and erotes, chubby babies personifying love (BM EA 38517).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 10: Mummy portrait of a man wearing a gold diadem with a star ornament, c. AD 140–60, excavated at Hawara, near the entrance to the Fayum (BM EA 74714).

11

12

E. R. O’CONNELL

Fig. 11: Gold diadem ornament, first–third century, Egypt (BM EA 26328).

‘at the table of Lord Serapis, in the dining room of the Serapeum’. While there were regular feasts of the god, the dining room of a temple could be hired out for private parties too, which also honoured Serapis (inter alia, P.Oxy. XIV 1755, second–third century AD). During the economic crisis of the third century AD, the Roman state did not have the same resources to lavish on temples and priesthoods. Over time, and before the spread of Christianity, temple complexes began to decline (Bagnall 1993, 26–68). As temples fell out of use, they were sometimes abandoned, used for other purposes or quarried for stone. At the same time, religious practice became more personal and portable. In the second and third centuries, painted panels depicting images, or ‘icons’ (from the Greek eikones) of deities became popular, probably as less expensive alternatives to stone sculpture for religious devotion (Fig. 12; Rondot 2013). Some items combined deities and powers in ways that are surprising today. For example, gems were carved with names and images of goddesses, gods and divine powers from Greek, Roman, Mesopotamian, Persian and other cultures, sometimes also including the names of the one God, angels and figures from the Hebrew Bible and, later, Christ and the Holy Spirit (Figs 13–14; Walter 1994; Michel 2001; Spier 2007).

As well as public religious practices shared by the community or family, individuals could seek the help of ritual specialists when dealing with events in the natural, supernatural and social worlds (Wilburn 2012; Frankfurter 2017). Requests might include healing, protection or success in competition, be it in sport, business or sex. For curses, ritual practitioners could require effigy figures of the person against whom the spell was said, or something associated with that person, such as some of their hair (Ogden 1999). In this way, the practitioner was able to control the individual by acting upon a small part of him or her (Fig. 15). As well as such ‘applied magic’, handbooks compiling spells represent ‘theoretical magic’. For example, Egyptian temple priests in Thebes compiled spells for all manner of purposes and invoking the names of deities or powers from different traditions––the more the better (Dieleman 2005; Bagnall 2009a, 83–86). In the London-Leiden Magical Papyrus most of the text is written in demotic Egyptian, which runs from right to left (PDM XIV, c. AD 200–225). One spell is written in Greek, which runs in the opposite direction. In the margins some of the Egyptian words have been written in Greek script, i.e., early attempts at what would become the last stage of the Egyptian language, Coptic. Although Coptic is associated above all with the language of Christians in Egypt, it was first developed in an earlier, non-Christian, context (Choat 2010). Among the deities called upon in the London-Leiden Magical Papyrus is the Great God of Moses who was ‘revealed to him on the mountain’ (Fig. 16). Residents of Roman Egypt combined other elements of Egyptian, Greek and Roman culture, such as language, appearance and customs. Although few people could read and write, they lived in a literate society, where births and deaths were registered, and tax receipts marked the intervening years in most people’s lives (Bagnall 2011). Under the Ptolemies, people used Greek alongside the Egyptian script known as demotic for everyday writing. Under the Romans, Greek replaced demotic Egyptian as the language of everyday administration, with Latin used at its highest levels (Torallas Tovar, this volume). While literacy in Egyptian had always been restricted to a small, professional elite, Greek could be learned by anyone with enough money to buy an education, a mark of status. In learning Greek, pupils also absorbed Greek culture, including mythology, history, customs and values (Cribiore

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 12: Painted wooden panel depicting seated Harpokrates, second–third century AD, Egypt (BM 1891,0423.1).

13

14

E. R. O’CONNELL

Fig. 13: Haematite gem showing King Solomon spearing the demon Lilith, fourth century, Mediterranean. The text states ‘Seal of God’ (BM OA.9576).

Fig. 15: Wax figure, first–third century AD, Egypt. What appears to be human hair is in the navel and a papyrus roll, probably containing the spell, has been inserted into its back (BM EA 37918).

Fig. 14: Jasper gem showing the crucified Christ, second–third century AD, Eastern Mediterranean. The text names the ‘Son, Father, Jesus Christ’ and magical names (BM 1986,0501.1).

2001). Verses from Homer’s Iliad, the epic Greek poem first written down in the eighth century BC, became in time the essence of ‘Greekness’ and continued to be used to teach Greek until the end of Late Antiquity (Fig. 17). It was by far the most popular text used in Greek school exercises found in Egypt (Cribiore 1996). With Greek writing came a new technology (Clarysse 1993). Written Egyptian was traditionally ‘painted’ with a brush from right to left, but Greek was written with a reed pen from left to right. In the Roman period, many writers also adopted the pen to write demotic (see Fig. 16). Even as religious beliefs change, religious practices surrounding death tend to be

Fig. 16: Section of the London-Leiden papyrus handbook from the Theban Magical Library, c. AD 200–225. Containing spells for telling the future, curing fever and for causing madness and desire, the text is written with a reed pen. The prescriptions are in red ink and the instructions for achieving them are in black ink (BM EA 10070/2).

15

16

E. R. O’CONNELL

Fig. 17: Wooden board with iron handle for hanging, fifth century, Egypt. The text contains lines 468–73 from Book I of Homer’s Iliad (BM 1906,1020.2).

conservative. While Greeks and Romans usually either cremated or buried their dead, elite Egyptians practised mummification. In Roman Egypt funerary practices were combined in new ways. Painted plaster masks, wooden panels and shrouds covering mummies depicted people wearing fashionable Roman hairstyles, clothes and accessories, combining the quintessential Egyptian burial practice with the Roman tradition of commemorating the dead with realistic portraiture (Fig. 18; Walker 2000; Riggs 2005). In general, Romans did not interfere with the religious practices of Jews. Under the Ptolemies, Jewish communities had generally thrived, with some adopting Greek culture and language (Modrzejewski and Cornman 1995). In the first century AD, Philo of Alexandria could claim that one million Jews lived in Egypt. According to tradition, Jewish scholars were commissioned to translate the Torah into Greek for the famous library at Alexandria. But this special status was challenged in the first and second centuries by a series of Jewish revolts across the Roman empire. Several factors sparked the violence, including a special tax imposed on Jews, and a recurring expectation among the Jewish community of the arrival of a messiah—a divinely appointed human leader. Retaliation was brutal and the rebellions were crushed. In Egypt, especially after the Diaspora Revolt of AD 115 to 117, surviving

records contain far fewer references to Jews and Jewish communities (e.g., P.Oxy. IX 1205, AD 291). Besides the appearance of Jewish names in inscriptions and papyrus documents, the depiction of the menorah—a seven-branched lampstand—is one of the few ways of identifying Jews (Fig. 19). Otherwise, everyday objects that Jews made and used were generally the same as those used by non-Jews. The first Christians were, of course, Jews who believed that Jesus Christ was their messiah. His title ‘christos’ is the Greek translation of Hebrew ‘messiah’, meaning ‘anointed one’. In Alexandria, as scholars debated early Christian thought, they were hugely influenced by their inheritance of Greek philosophical debate and by the Judaism of Greek-speaking scholars such as Philo of Alexandria (Clark 1992; Rogers 2017). Their discussions and disputes helped to shape the development of Christianity. The Roman authorities, however, viewed this new religion with suspicion. While Judaism had a long history, Romans saw Christianity as a difficult, occasionally criminal sect (Bagnall, this volume). Because of their reluctance to sacrifice to the gods who protected the empire, Christians were seen as a threat and were periodically persecuted. The emperor Decius (r. AD 249–251) ruled in a time of military crisis, when the Roman empire’s northern and eastern frontiers were especially

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 18: Mummy portrait of a woman, c. AD 55–70, excavated at Hawara at the entrance to the Fayum. She wears a tunic with a darker clavus edged with gold and a mantle draped around her shoulders. Her gold ball earrings and crescent pendant necklace are types popular across the empire in the later first century AD (BM EA 74716).

17

18

E. R. O’CONNELL

threatened. In an attempt to regain divine favour, he issued an official order in January AD 250 calling upon residents of the empire to sacrifice to the gods. Although not strictly aimed at Christians, it created the first great crisis for the early Church as Christians faced exclusion from their own community if they did sacrifice, or imprisonment and even death if they did not (Luijendijk 2009, 164–67). A certificate of sacrifice (libellus), dozens of which have been found in Egypt, was required and had to be witnessed by a Roman official (e.g., P.Oxy. LVIII 3929). Decades later, the emperor Diocletian (r. AD 284– 305), an able general and administrator, dealt with rebellions and invasions and introduced political and economic policies that stabilized the Roman empire. He also persecuted groups thought to contribute to instability, Christians among others. In a series of official orders, he purged the army and civil service of Christians, ordered the destruction of books and places of worship, prevented assembly, imprisoned or exiled bishops and other clergy, and reinstated universal sacrifice to the Roman gods. Christians remembered these acts as the ‘Great Persecution’. Nevertheless, the documents display the resourcefulness of Christian individuals and families in the face of official orders. In one letter, written from Alexandria by a man named Copres to his wife, he explains that when he arrived at court to make his case, he was required to sacrifice (P.Oxy. XXXI 2601, about AD 298). Presented with this problem, he found a solution: he identified someone to take his place and sacrifice on his behalf (Bagnall 1993). In AD 311 the emperor Galerius legalized Christianity and in AD 313 Constantine restored the rights and property of Christians. Soon after, Christianity became the privileged religion of the imperial household, and wealthy and influential people throughout the empire adopted the new faith. The balance of power had changed. Although it was not yet the state religion, imperial coins began to display Christian imagery (Fig. 20). Under Constantine, disagreements between Christians became a concern of the state. In order to attempt agreement, Constantine ordered the first Church council to meet in Nicaea, near Constantinople, in AD 325. Here bishops gathered to establish a hierarchy, a canon of accepted texts, and the relationship of God the Father, the Son and the Holy Spirit, summed up in a profession of faith still used in churches today and known as the Nicene Creed. The case was not settled, and controversy continued to fester (Davis 2004).

Fig. 19: Ceramic oil lamp with a menorah, said to be from Alexandria, third–fourth century AD (BM 1886,0623.2).

Fig. 20: Reverse of a gold coin of Constantine, minted in Antioch, AD 336–37. Victory is shown advancing left holding a trophy and a palm frond, with a Christian chi-rho tucked between her and the inscription (BM R.165).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

In Alexandria, Constantine’s son and successor donated the Caesareum, the temple once dedicated to the worship of Augustus and his successors, to the Christian Church (McKenzie 2007). While previous churches were located on the margins of the city, under imperial patronage they were established in its very heart. By the end of the fourth century, imperial legislation began to penalize those who followed traditional religious practice. In AD 391/392, following an eruption of violence at the temple of Serapis in Alexandria, it was closed to public worship (Dijkstra 2011; 2020). The powerful Archbishop of Alexandria, Theophilus,

19

reportedly targeted both followers of traditional deities and Jews. Despite the dramatic accounts found in literary sources, the motivations for the mutilation or destruction of traditional images are less clear. For example, a bust of Germanicus, great-nephew of Augustus, was probably carved around the time of his death in AD 19, shortly after a visit to Egypt (Fig. 21). Years later, after the establishment of Christianity, someone inscribed a cross on his forehead. Either at the same time or separately, someone hacked the statue across the nose, right ear and neck (Kristensen 2012, 39–40, 55; 2013). We

Fig. 21: Basalt portrait bust of Germanicus, c. AD 14–20, Egypt. Centuries after the bust was carved, a cross was incised upon the forehead and the nose, right ear and neck were mutilated by hacking, probably in or after the fourth century AD (BM 1872,0605.1).

20

E. R. O’CONNELL

can only speculate as to why the cross was added––did it neutralize spirits thought to dwell in images, or perhaps ‘Christianize’ a popular member of the imperial family? Or was it, perhaps (following the ancient tradition of marking slaves with tattoos) intended to mark Germanicus as a ‘slave of God’, just as we are told some Christians marked themselves (C. Jones 1987; Elm 1996)? Egypt in a Christian empire In AD 330 the Roman emperor Constantine consecrated a new city strategically located at Byzantium. This ‘New Rome’, or Constantinople, eventually became the capital of an Eastern Roman or Byzantine empire. The grain and resources of Egypt flowed to the new capital, and its public spaces were decorated with ancient Egyptian monuments just as those of Rome had been in previous centuries (Curran et al. 2009).

Imperial statues continued to be erected in Egypt (Jones and McFadden 2015), and jewellery worked in imperial workshops was sent there, probably as imperial gifts (Fig. 22; Stolz 2006). Alexandria continued to be a great hub of learning for both Christians and non-Christians, sometimes studying under the same teachers (Cribiore 2001). Despite the moral concerns of emperors and Church leaders regarding public spectacles, people continued to flock to the races and the theatre. The city also continued to be famous for the riots and violence that flared up between and within its communities (Haas 1997; Papaconstantinou 2003; Watts 2010). These were often stirred up by the city’s archbishops, supported by monks, and targeting the followers of the old gods, philosophers and Jews. In the sixth and seventh centuries, the heart of the city was still occupied by lecture halls where philosophers, now Christian, taught (Derda, Markiewicz and Wipszycka 2007; Kominko,

Fig. 22: Necklace and earrings from the Asyut hoard, c. AD 600, said to be from the Lykopolis/Asyut area. A large hoard of gold items, probably produced for the imperial court, was discovered in the early twentieth century and is today distributed between international collections. The hoard contains a combination of pieces containing imperial portraits and Christian motifs (BM 1916,0704.2–4).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

21

Fig. 23: Limonite amulet with suspension loop, late fifth century, said to be from Karnak. On one side of this Christian amulet, Abraham holds Isaac on an altar, restrained by an angel holding a cross. Behind them a ram is tied to a tree. On the other side is a large handled cross, around which is inscribed in Greek ‘One God in Heaven’ (BM OA.9940).

this volume). Its great temple complexes, the Caesareum and Serapeum, were occupied by churches (McKenzie 2007). Heis theos, or ‘one God’, now referred to the God in the Bible (Fig. 23). Modern scholars debate the rate of conversion to Christianity, but a study of names in the thousands of papyrus documents found in Egypt suggests that by the fifth century AD most people regarded themselves as Christian (Depauw and Clarysse 2013; 2015; cf. Frankfurter 2014). While Christian parents increasingly gave their children names that were taken from the Bible, or explicitly Christian, such as Christodoros (Gift of Christ), they also continued to use traditional Greek, Roman and Egyptian names (Fig. 24). Greek continued to be the official language of Egypt, but people also increasingly wrote Coptic (Fournet 2020). This was the last stage of the Egyptian language, written in a modified Greek script. At first it was used to translate literature in Greek such as the Bible, and for everyday writing. From the end of the fourth century it was also used to write original Christian literature. Both Jews and Christians drew upon classical sources for their visual culture (Fine 2005; Spier and Charles-Murray 2007). Christians adopted the Greek version of the Hebrew Bible as their Old Testament and understood many of its stories in the context of the

death and resurrection of Jesus. Jewish narratives in which faith in God was rewarded with salvation were often shown on Christian objects. For example, an ivory box, now in two halves, shows Daniel, who was thrown into a lions’ den by the Babylonian king Nebuchadnezzar but saved by his faith (Fig. 25). Some of the most famous stories from the Bible take place in Egypt. Joseph, the favourite son of Jacob, sold into slavery by his jealous brothers, was taken to Egypt and rose to power in the pharaoh’s palace (Fig. 26). The number of textiles that show scenes from the life of Joseph suggest that the story may have been popular among Egyptians precisely because of its location. Stories from the Christian Old Testament and newer narratives of the martyrdom of Christian saints could combine. For example, a c. AD 600 wall-painting shows the story of the Three Hebrews, captive in Babylon, and sentenced to burn in a fiery furnace but rescued by divine intervention, represented by an angel (Fig. 27). Flanking this scene are the twin physician saints Kosmas and Damian, carrying their bags of medical instruments, and, below, their three brothers, all said to have been martyred under the emperor Diocletian and popularized around the empire under the emperor Justinian. A Coptic text commemorates the otherwise unattested sixty Christian martyrs of Samalut

22

E. R. O’CONNELL

Fig. 24: Limestone gravestone for Besa, c. fifth–eighth century, Egypt. Originally a temple relief showing the falcon-god Horus, centuries later it was re-carved on the back as a Christian gravestone for a man named Besa, a Greek version of the name of the ancient Egyptian god Bes (BM EA 937).

Fig. 25: Ivory Daniel pyxis, fifth–sixth century, Syria or Egypt? Daniel, with arms raised in prayer, is flanked by two lions. In a separate scene, an angel rushes forward behind a ram, referencing the story of Abraham, in which a ram is substituted at the last minute as a sacrifice to God in place of Isaac (BM 1877,0706.3).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

23

Fig. 26: Tapestry panel depicting Joseph’s dream, said to be from Panopolis/Akhmim, seventh–eighth century. The panel formed the decoration from a tunic sleeve and depicts Joseph’s dream foretelling the future. He is shown lying surrounded by the sun, moon and stars, which, in his dream, represent his parents and brothers (BM EA 17175).

and probably names the painting’s donors. The wallpainting thus combines proto-martyrs of the ancient, biblical past with more recent international martyrs together with local, Egyptian martyrs. As members of the elite adopted Christianity, the Church became well funded through donations and gifts. From the fourth century onwards, Christians began to build churches, saints’ shrines and monasteries, redefining the sacred landscape of Egypt (McKenzie 2007). Some buildings were newly constructed; others were built on earlier sacred sites. Some occupied earlier temple complexes or shrines; others were built of reused stone. Such reuse was mainly practical, but it

could also be perceived as a triumph of Christianity over traditional religion. The Temple of Isis on the island of Philae was the last functioning temple in Egypt (Fig. 28). While most temples were closed by the fourth century, the temple on Philae remained open so that people from Nubia could worship there (Dijkstra 2008; 2011, 398). The latest known hieroglyphic and demotic inscriptions, dating to AD 394 and AD 452, are both found on Philae. The temple went out of use shortly afterwards. The story of how the people of Philae were converted to Christianity, as told in the Life of Aaron, is dramatic. In the temple, there was a caged falcon that people believed embodied the god Horus.

24

E. R. O’CONNELL

Fig. 27: Wall-painting, c. AD 600, excavated from a mud-brick complex near Lykopolis/Asyut. The central panel depicts the Three Hebrews in the Fiery Furnace, flanked by the Syrian physician saints, brothers Kosmas and Damian and, below, their three brothers, who were said to have been martyred under Diocletian (BM EA 73139).

Macedonius, Philae’s newly appointed bishop, went to the temple asking ‘to make a sacrifice to God’. As the altar was prepared for the sacrifice, he went up to the cage, cut off the falcon’s head and threw it on the burning altar. According to the Life, he later baptized the city’s entire population (Dijkstra and van der Vliet 2020). The archaeology of Philae, however, tells a rather different story. Conversion did not occur as a single event in the fourth century. Instead, Christians lived side by side with followers of traditional religion until around the middle of the fifth century. In the sixth century a small church dedicated to St Stephen was installed within the Temple of Isis. Unlike ancient temples, where access was restricted, churches were based on Roman basilicae (public buildings housing law courts); they were, and are, congregational spaces inside which worshippers gather together (Fig. 29). At that time, their interiors were decorated with textile hangings and wall-paintings, lit with lamps, filled with the scent of incense and crowded with congregations led by priests and other Church officials, such as deacons and readers.

Across the Christian world people visited the places they knew from the Bible, the tombs of early martyr saints and ‘living saints’ or monks (Markus 1994). Sometimes the monuments of ancient Egypt were interpreted through the lens of the Bible. In the story of Joseph, for instance, he advises pharaoh to store up grain in order to survive a famine. Some thought that the Great Pyramids at Giza were his grain stores (History of the monks in Egypt XVIII.3; Russell 1981, 102). In Egypt, as elsewhere, many towns had their own local martyr (Papaconstantinou 2001). In time, their tombs were marked by shrines and churches, and whole new settlements grew up around them. A web of pilgrimage routes connected these sites and attracted local and international travellers (Frankfurter 1998). One of Egypt’s most popular destinations was the shrine of Menas, a saint who is said to have been a soldier in the Roman army, martyred in about AD 300 during the Great Persecution of Diocletian (Fig. 30). Christian pilgrims flocking to the shrine of St Menas, just southwest of Alexandria, took away with them the saint’s blessings in the form of earth, oil or water in

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

25

Fig. 28: Vue des temples de l’isle de Philae, by Baron Dominique Vivant Denon, c. 1802 (BM 1836,0109.110).

little flasks (Fig. 31). Today these flasks have been discovered at sites from Britain to Uzbekistan (W. Anderson 2007). The shrine was very popular. From the fifth century onwards the church over St Menas’ tomb had to be rebuilt and extended several times to accommodate visitors, and a town, with guest houses and other facilities, grew up around it. Monasticism as it was adopted in the West had its origins in Egypt. Antony was an ideal model for

Fig. 29: Sandstone niche head, sixth–seventh century, Philae, said to be from the East Church, the main church of the island and probably its cathedral church (BM EA 1422).

hermits, while Pachomius established a communitybased monasticism, which was exported across the empire (Harmless 2004). Monks were called ‘living saints’ or ‘bloodless martyrs’ and were visited by pilgrims from near and far seeking advice and blessing (Frank 2000). Even the emperor Theodosius had occasion to consult Apa John of Lykopolis/Asyut at the end of the fourth century (Wipszycka 2009, 83–85; Choat 2017, 37–40). Although early Christian writings suggest that monks lived in the ‘desert’, far from cities and towns, archaeological remains show that most lived within sight of settlements and were actively engaged in people’s daily lives (O’Connell 2007). Shenoute (d. AD 465) is one of the most famous Egyptian saints, although unknown outside Egypt. As abbot of the White Monastery and other monasteries on the desert’s edge near Atripe, he governed separate communities of male and female monks according to a series of strict rules (Layton 2014). He was an outspoken critic of both other Christians and non-Christians. Locally, he attacked temples and private shrines, reportedly destroying their statues. The walls of the monastery’s church were built to imitate the profile of ancient Egyptian temples and were partially constructed of limestone blocks from a temple at nearby Atripe (Fig. 32; Blanke 2019). Shenoute was a great speaker and writer, trained in the Greek art of rhetoric, but writing in Coptic (inter alia Emmel 2004). In doing so he helped to establish Coptic as a literary language, his sermons and correspondence copied and circulated for centuries

26

E. R. O’CONNELL

Fig. 30: Ivory Menas pyxis, sixth century AD, said to have been found in a chapel dedicated to St Menas, located in the church of St Paul’s-outside-the-Walls, Rome. Menas, standing in his martyr’s shrine, is flanked by two camels and approached by two women from the left and two men from the right, probably the donors of this luxury object. A second scene narrates his martyrdom with a prefect ordering the execution, a soldier holding a diptych, while the executioner stands with sword raised grasping the hair of Menas, and an angel sweeps in to receive his soul (BM 1879,1220.1).

Fig. 31: Pilgrim flask from Abu Mina, sixth–seventh century AD, excavated at Burgate, Kent, UK. The saint, his arms raised in prayer, is wearing a soldier’s costume (short, belted tunic, cloak and boots) and is flanked by camels, which kneel at his feet (BM 1929,0108.1).

(Fig. 33). Other monasteries were built in and around earlier monuments also at the edge of the desert. The Monastery of the Holy Martyr Phoibammon was built in and around the famous temple of Hatshepsut at Deir el-Bahri in the Theban Necropolis (Godłewski 1986). Archaeologists have excavated thousands of papyri, pottery fragments and limestone flakes with writing, mainly in Coptic and some in Greek. Even the papyrus wills of the founder, the Bishop Abraham of Hermonthis (P.Lond. I 77, c. 600), and many of his successors bequeathing the monastery have survived. Abraham’s will was written in Greek. His successors’ wills, written after the Arab conquest of Egypt, are written in Coptic (Garel 2020). The evidence from Egyptian monasteries thus complicates many of the received traditions about monks: that they lived in the ‘desert’, did not own property, were illiterate and existed separately from the church hierarchy (Fig. 34). Across the empire people wore similar tunics and mantles (Fluck, this volume), and drank wine and ate using similar tablewares (Bagnall 1993). Christians inherited a rich classical repertoire which was usually not thought incompatible with their faith (Török 2005). Some writers in Egypt, such as Dioscorus of Aphrodito, wrote in Greek and Coptic and probably knew Latin, and read the Iliad and Isocrates alongside books from the Bible (MacCoull 1988; Fournet 1999). Curtains and wall-hangings that furnished homes and

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

27

Fig. 32: Déir beyâdh le Couvent Blanc by Baron Dominique Vivant Denon, c. 1800.

Fig. 33: Papyrus codex leaf, seventh–eighth century AD, Egypt, probably from Western Thebes, Egypt. This codex leaf contains accounts of miracles that were experienced in Shenoute’s monastery (BM EA 71005/7).

public spaces combined Christian motifs with the earlier classical visual culture (Fig. 35). Elements from ancient Egyptian culture were reinterpreted, such as when the ankh, the hieroglyph meaning ‘life’, became a handled cross (crux ansata) (Fig. 36). Although mummification by evisceration probably ended by the fourth century AD, shrouded mummiform burials wrapped in purpose-made bands continued to resemble earlier mummies (Gessler-Löhr 2010; Fluck, this volume). The grave stelae that marked burials could combine earlier classical framing devices with Christian iconography (Fig. 37). People continued to turn to ritual specialists using traditional methods to negotiate this world and the next. Some handbooks and curses instructed people to place spells in tombs or graves, directly on corpses or in their mouths, even writing spells directly on to the corpse (Ritner 1993, 178–83). One long Greek spell was found in a pot with two wax figures held in an embrace. It calls on the dead in the cemetery to cause Euphemia, daughter of Dorothea, to burn with desire for Theon, son of Proechia. In exchange, Theon promises to call upon the Egyptian god Osiris to quench their thirst and give them rest (PGM CI AD 400–500). Another set of four illustrated spells written by or for a man named Severus, son of Anna (or Joanna) was written in Coptic. One contains an illustration of Jesus’ death by crucifixion alongside a prayer to cast out an

28

E. R. O’CONNELL

Fig. 34: Palimpsest ivory diptych, c. AD 400, with main visible text naming Bishop Apa Abraham of Hermonthis, written between AD 623 and 662, Hermonthis/Armant or Theban region. The text lists the archbishops of Alexandria, beginning with St Mark’s legendary successor Ananias and the historical bishops of Hermonthis, including the bishop Abraham, who was also abbot of the Monastery of Phoibammon (BM 1920,1214.1).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

29

Fig. 35: Pair of linen and wool curtains or hangings, sixth–seventh century, Panopolis/Akhmim. At the top, two Victory figures hold a wreath containing a jewelled cross with a Greek inscription. Above them are erotes, chubby babies personifying love, standing among baskets of produce and holding flower garlands. The curtains have survived nearly intact because they were reused as a burial shroud (BM EA 29771).

unclean spirit. Another spell asks for success in fishing, calling on successful figures from the Bible including a man with a fishing pole and fish labelled as Jesus. A spell for a good singing voice shows the biblical figure King David with an instrument, and the last contains a prayer made by Mary, the mother of Jesus (Meyer and Smith 1994, nos 129–32, c. AD 475–525). During the early centuries of Christianity, the new religion was still in flux, with long and bitter disputes between leading bishops as to which beliefs and practices were accepted and which were not. Among the

diversity of Christian groups are those known today as Manichaeans, Arians, Melitians, Nestorians, Chalcedonians, anti-Chalcedonians, Julianists and Gaianites. Such partisans at the time simply considered themselves Christians. Only occasionally are communities visible in the material remains. For example, followers of Mani (AD 216–274) can be found in texts excavated in Kellis in the Dakhla Oasis. Mani was a self-proclaimed last prophet in a line that included Buddha, Zoroaster and Jesus. Manichaeans were persecuted in the Roman empire, but sacred texts, as well as letters

30

E. R. O’CONNELL

Fig. 36: Tapestry tabula depicting a bird and cross, fifth–seventh centuries AD, said to be from Panopolis/Akhmim. Once used to decorate a garment or soft furnishing, this tapestry panel belongs to a set depicting different birds and crosses. Here, a victory wreath encloses a handled cross adapted from the Egyptian hieroglyph ankh, meaning ‘life’ (BM EA 22868).

and other documents found in the actual houses they inhabited, provide evidence that they were living in Egypt (P.Kellis V; Mazza, this volume). Many modern scholars assumed that Jewish communities did not survive in Egypt after the Diaspora Revolt of AD 115 to 117. In fact, papyrus documents show they continued to live in cities and towns, marry, pay taxes and transact business. In one c. AD 400 rental agreement, ‘Aurelius Jose son of Judas, Jew’, leases part of a house from two

female monks, Aurelia Theodora and Aurelia Tayris, daughters of Silvanus. Their titles suggest that they probably lived as virgins in their town, rather than as hermits or in a monastic community. While the two sisters have Greek and Egyptian names (Theodora means ‘gift of God’ in Greek and Tayris means ‘she of Horus’ in Egyptian), their father’s name, Silvanus, is Latin (P.Oxy. XLIV 3203).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

31

Fig. 37: Sandstone gravestone of ‘Abraham, the perfect monk’, seventh century, Egypt. A classical pediment decorated with vine leaves supported by columns provides the architectural frame for the Greek tau and rho forming a cross at centre. In the top left and right quadrants, alpha and omega refer to Jesus Christ. Below are handled crosses, and above, the Coptic inscription (EA 1257).

Arrival of Islam In AD 639 Muslim armies entered Egypt under the command of ῾Amr b. al-῾As (Petry 1998; Booth 2013). The Byzantine empire at the time was weak and vulnerable. By AD 642 Alexandria had fallen and all of Egypt soon came under Muslim rule. Since Alexandria was open to attack from the Byzantine navy, ῾Amr chose a site at the tip of the Nile Delta for his new capital Fustat (Map 2). Built just outside the walls of the Roman fortress called Babylon, it became a bustling city and the commercial centre of Egypt (Sheehan 2010). The Muslim state recognized Jews and

Christians as ‘People of the Book’ and did not interfere with their religious practices. Jews and Christians did, however, pay a special poll tax in exchange for their protected status. In the first centuries of Muslim rule, Egypt was governed by empires based in faraway capitals. At first, the Umayyad dynasty used the existing system of Byzantine coins. They also made imitations, modifying the Christian cross by removing the crossbar (Figs 38–39). In AD 696 the caliph ꜥAbd al-Malik introduced the first coins replacing images with verses from the Qur’an on both sides, making each coin a messenger of the faith and bringing Islam into the service of the state just as

E. R. O’CONNELL

32

Mediterranean Sea D Damietta

Marina el-Alamein

al-Arish h

Tell el-Farama m

Nikiu N ik k DE E LTA L A

WA D I A N TRUN

Ismailia ail ailia WAD ADI TUMILAT

Siwa Oasis

Saqqar Saqqara q

B ah

Petra

Su S uez

Aqaba

Atfih Naqlun Beni Suef

SINAI

e

r Yu

suf

Medinet Me et al-Fayu a ayu ayum yum yu

F Fustat/Cairo Ba Babylon

Nil

FAY U M

Bahariya Oasis

Gaza za

Port Said

Alexandria

Bahnasa B h Minya Ashmunein n

Sheikh Ibada She

Manfalut Manfa Asyut sy Farafra

EASTE R N D E S E RT

Qaw el-Kebir

Red

Kom Ishqaw I aw

Oasis

Sea

Akhmim Girga Ba Balayana a Dende Dendera De n Abydos os Qena Qe Ho Hou Qift Q ift fftt Armantt Luxor Lu Kharga K h Esna a Sohag

W E STE R N D E S E RT Dakhla Oas Oasis

Mutt Mu K harga O asis

Douch

Quseir el-Qadim

E Edfu Kom Ombo Om Phila ilae

Aswan First Cataract

Baranis is

Lak ake Nasser ak

Abu Simbel

Qasr srr Ib IIbrim Principal deser t r oa ds

0

Map 2: Medieval Egypt (Illustration by Claire Thorne).

200 km

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 38: Byzantine gold solidus of Constantine IV, minted in Constantinople, AD 654–85, found in Hampshire, UK (BM 1998,0430.4).

Fig. 39: Umayyad gold dinar of ꜥAbd al-Malik in imitation of Byzantine solidus, crossbar removed, minted in Syria, AD 690 (BM 1954,1011.1).

Fig. 40: Umayyad gold dinar of ꜥAbd al-Malik, minted in about AD 698 (BM 1849,1121.88).

33

34

E. R. O’CONNELL

Constantine had promoted Christianity (Fig. 40). The visual language of ꜥAbd al-Malik’s coins became the standard for hundreds of years. In AD 641 work began on the first mosque to be built in Egypt, the Mosque of ῾Amr b. al-῾As. It was completed the following year and stood in the centre of the new city of Fustat. From then on, Muslim rulers established mosques in Alexandria and other administrative centres across the country. Like churches, mosques are congregational spaces crowded with people led by an imam and other personnel. The mihrab or prayer niche, containing the qibla, marks the direction of Mecca, to which Muslims turn at prayer. Next to it is the minbar, a raised platform used by the imam to deliver the Friday sermon. At the time, interiors were often lit by glass lamps filled with oil floating on water. In the first centuries of Islam, pilgrims undertaking the Hajj would gather in the capital cities of Egypt and elsewhere, and travel in groups of tens of thousands. The Muslim rulers were responsible for safeguarding the route to Mecca and for protecting pilgrims. The first mention of the Hajj from Egypt is found on a papyrus letter dated between AD 705 and 709 (Sijpesteijn 2014). Under the Fatimids, pilgrims from across the empire began to travel via Cairo. There were two main routes from Egypt to Mecca. One ran from Cairo across the Sinai Peninsula; the other followed the Nile south and then crossed the desert to the Red Sea port of ꜥAidhab. The arrival of the Muslim armies in the seventh century brought to Egypt not just a new religion, but also a new calendar, language and script. They brought the Hijri calendar as a new form of reckoning time, starting from AD 622, when the Prophet Muhammad emigrated from Medina to Mecca. The highest officials were now Arab Muslims, but administrative documents were often written in Greek or Coptic. Arabic, the new state language, was increasingly used until it replaced Greek by the end of the eighth century. It took another 400 years for Arabic to replace Coptic as the most common language, used not just by Muslim officials but also by non-Muslim populations (Papaconstantinou, this volume). Governors ruled Egypt on behalf of the Umayyads (658–750) and Abbasids (750–969). In AD 868 Ahmad ibn Tulun established himself as a governor of Egypt and thereafter ran it as a semi-autonomous state, founding the Tulunid dynasty (Kennedy, this volume). With the collapse of the Abbasid empire, the Fatimids, previously based elsewhere in North Africa, established their

power base in Egypt. Their new capital was named alQahira, meaning ‘the Victorious’ and known in English as Cairo (Sheehan 2010). Absorbing Fustat, it quickly became a flourishing city (Fig. 41). Alexandria with its routes west became less important, while Cairo, facing east, became the new centre of scholarship, learning and the arts. The madrasa of the mosque of al-Azhar developed into one of the greatest universities in the world. The Fatimid rulers belonged to the Ismaili Shiꜥi branch of Islam, tracing its line back to Fatima, the daughter of the Prophet Muhammad, and her husband, the Prophet’s cousin Ali. The Fatimids emphasized their Shiꜥi identity on their monuments and coins, which were extremely fine quality with a high gold content, and became the most common trade coins used around the Mediterranean. Large-scale inscriptions became the focus of decoration, further promoting the status of Arabic script (Fig. 42; Ali and Wood 2017, 150). The Muslim people they governed belonged mainly to the Sunni branch of Islam, and Jews and Christians gained prominence in their administration. The Fatimid period was one of great wealth for the elite. At its height, the empire extended across North Africa to the Atlantic and into the Levant and the Hijaz (in present-day Saudi Arabia). Mediterranean, Red Sea and Indian Ocean trade was revitalized, and prized materials such as ivory and spices were imported from the far reaches of the empire and beyond (Power 2012). Fustat, now part of Cairo, became a major centre for the production of pottery, glass and metalwork, as well as carved rock crystal, ivory and woodcarving (Fig. 43). Although images of people and animals were not used in religious art, they often decorated luxury objects (Fig. 44). For 200 years, from 1095, Europe’s Christians carried out a series of military campaigns, known as the Crusades, fighting, inter alia, the Fatimids over control of the Holy Land which was considered sacred to Jews, Christians and Muslims. The first campaign was waged against the Fatimids, who controlled Jerusalem at the time. The Crusaders took the city and established the Kingdom of Jerusalem, which ruled the Eastern Mediterranean from Antioch to Ascalon (Fig. 45). Many pilgrims on Hajj travelled via Egypt to avoid the fighting. Medieval Muslim scholars were fascinated by Egypt’s ancient history and by the great monuments still dominating the landscape (el-Daly 2005; Van Bladel 2009). They travelled the country, visiting temples and tombs, including the Great Pyramids of Giza. They recorded monuments and their decoration,

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 41: ‘Gate of the Metwáleys – Cairo’, etching and engraving after David Roberts, 1855. The gate in the distance is the last remaining southern gate from the walls of Fatimid Cairo (BM 1872,1012.2224).

35

36

E. R. O’CONNELL

Fig. 42: Marble panel from a monumental inscription, ninth–tenth century, reused for a gravestone AH 356 Jumada II / AD 967, probably Cairo. The original panel was part of a much larger monumental inscription carved in an angular Arabic script known as Kufic and containing part of the basmala. Not long afterwards the relief was carved on the back and reused as a gravestone for a man named Muhammad bin Fatik Ashmuli, inscribed with a verse from the Qur’an (BM 1975,0415.1).

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

37

Fig. 43: Gold filigree and enamel pendant, c. eleventh century, Egypt. The gold filigree work has been decorated with coloured enamels, including images of birds. The style is typical of the Byzantine period, but it continued to be popular and produced in the Islamic period (BM 1981,0707.2).

studying hieroglyphs, which some recognized to be an ancient script. Thus, hundreds of years before Western Egyptologists began to study ancient Egyptian culture, Medieval travellers and scholars described and interpreted ancient monuments. For example, in the ninth century the famous Sufi mystic Dhu al-Nun al-Misri lived inside the Ptolemaic-period Temple of Min at Akhmim and was said to be fluent in the ‘language of the walls’, i.e., hieroglyphs. The accounts of Muslim scholars are the best evidence for the size and decoration of the temple because it was destroyed in about 1350, when it was quarried for stone (el-Daly 2005). Like the Romans and Christians before them, Muslims often adopted and adapted existing architecture or quarried it for building material (Behrens-Abouseif 2014). At Luxor, the temple complex begun in the

1300s BC under the pharaoh Amenhotep III was fortified by the Roman army in about AD 300, at which time they inserted a chapel dedicated to the imperial cult into one of the ancient Egyptian courtyards (McKenzie 2007). From the sixth century, Christians built at least four churches within the temple complex, and centuries later Muslims built a mosque dedicated to the Egyptian saint Abu el-Haggag on the probable ruins of one of them (Boraik 2014). In Alexandria, the sarcophagus of Nectanebo II (r. 360–342 BC) was put to a new purpose when it was used for washing before prayer at al-‘Attarin Mosque in Alexandria (Fig. 46; McKenzie 2007). In everyday life there was continuity with the past and a layering of the new and the old. While the Hijri calendar was adopted immediately in the highest

38

E. R. O’CONNELL

Fig. 44: Lustreware dish fragment depicting faces, twelfth century, Syria or Egypt. Ceramics known as lustreware used a metal glaze technique to give a shimmering effect. The technique was first developed in Iraq, but Fustat was famed for its production (BM 1986,0415.1).

echelons of the administration (Sijpesteijn 2013; see also P.Paramone 18; Gonis 2010), other aspects of timekeeping remained distinctively local (Cooper and Barak, this volume). Egyptians had long used nilometers to measure the water level of the Nile during its annual flooding from July to November. A flood that was too high would mean devastation, while too low an inundation would cause famine. From AD 715 the ancient nilometer on Roda Island, opposite Fustat, was the official measure of the Nile flood. The nilometer that stands today was built in the ninth century and restored under the Fatimid dynasty (Fig. 47). Roman-style tunics decorated with clavi continued to be worn for centuries, while elements of Eastern horse-riding costume, such as tunics with flared sides and trousers, came into fashion along with embellished fabrics (Fig. 48; Fluck and Thomas, this volume). Tiraz fabrics in particular provide some of the earliest surviving evidence of a specifically Arab Muslim

material culture, often bearing verses from the Qur’an or the name of a caliph (Fig. 49). As in earlier periods, some aspects of death and burial changed, and others remain the same. Muslims were ideally buried on their sides facing Mecca (Gayraud 1999; Zych 2003). Elite graves were marked by monuments or gravestones. Unlike Christian gravestones, which were often decorated with images, Muslim grave markers bore Arabic script, which often expanded to fill the frame. Such texts provide a wealth of information on the social status of the person commemorated. Among the hundreds of gravestones from the Muslim cemetery of Aswan (Björnesjö 2013; Speiser et al. 2013), one records in Kufic script the death of Fatima, daughter of Ja’far, son of Muhammad, the dyer, in the year AH 412/AD 1021 (Fig. 50). Methods for cursing remained the same as in earlier centuries. One Coptic curse, written on a piece of parchment, invokes the dead at ‘the door of this tomb of the Greeks’. It also calls on Apollo, understood to be a demon, to cause the victim Sipa, son of Siheu, to be separated from Ouarteihla, daughter of Cauhare, probably his wife. The parchment was deliberately cut in the shape of a blade to strengthen its aggressive intent. The words draw on names from the Bible: ‘I am the sheet that separated pharaoh from his nation … raised Judas against … Jesus’. The spell was ‘to be placed under the head of the victim while he slept’ (Meyer and Smith 1994, no. 109; c. tenth century AD). On another parchment sheet, another curse depicts a figure labelled Ebbael who is shown at the centre of this sheet crowded with Coptic text. In it, a man with the Arabic name Abdallah, which means ‘servant of God’, curses his rival, seeking to confuse and render him speechless ‘like the dead’. He invokes the biblical story of Moses and pharaoh—the pharaoh here is ‘king of Egypt, the one who has power over all the magicians of Satan, lord of all the tribes of the earth’ (Meyer and Smith 1994, no. 95, c. ninth century AD). Under Islamic rule Christians remained the majority population for centuries (Sahner 2019). Muslims called Egypt’s residents qibti in Arabic, from the Greek Aigyptioi, meaning ‘Egyptians’. As most Egyptians at the time were Christian, qibti came to refer only to Christian Egyptians, today known as Copts (Booth, this volume). Although there were periods of repression and violence, Egypt’s Christian community also thrived with international links across the Medieval world (Fig. 51; Hunt 1989). Under the Fatimids, Egypt’s economy boomed, and Cairo became an international

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

39

Fig. 45: Painting on paper depicting a battle between Arabs and Franks, twelfth–thirteenth century, found in Fustat. The fragment shows the siege of a town, possibly Ascalon (BM 1938,0312.1).

centre of learning. This boosted both the prestige and usefulness of Arabic (Papaconstantinou, this volume). The Church benefited from economic prosperity under Fatimid rule, shown in the quality and quantity of manuscripts surviving in monastery libraries. By the tenth century, Coptic literary production had largely ceased and scholars focused instead on reworking and compiling texts for use in church. Early in the eleventh century the Coptic Church adopted Arabic as its sole language for correspondence, and gradually Coptic was abandoned (Papaconstantinou, this volume). At about the same time, Muslims began to outnumber Christians (Rapoport, this volume). Jews also largely thrived under the Fatimids. By that time, there were three main communities of Jews living in Cairo: those following the Babylonian and Palestinian rabbinic traditions, and the Karaites (Rustow 2014). The Karaites accepted only the authority of the Bible and not the teachings compiled in the Babylonian and Palestinian Talmuds. In Egypt all three branches spoke

and often wrote in Arabic. Since Jewish men were taught to write Hebrew from a young age, they also used the Hebrew script to write Arabic, known as Judeo-Arabic. Much of what we know about Medieval Egypt today comes from a vast store of documents known as the Cairo Genizah––hundreds of thousands of literary texts, letters and other communications that lay discarded for centuries in Ben Ezra Synagogue in Fustat (Hoffman and Cole 2011; Soskice 2011). This synagogue and several Medieval Christian churches were built within the walls of the earlier Roman fortress of Babylon (Sheehan 2010). For hundreds of years, the Jewish community of Fustat placed their worn-out books and documents in a storeroom, or genizah, of the Ben Ezra Synagogue. According to custom, anything that might have the name of God written on it cannot be destroyed, but instead must be stored away and buried. Every synagogue had its genizah, but for some reason the documents in the Ben Ezra Synagogue were never disposed of. The Cairo

40

E. R. O’CONNELL

Figs. 46: Agglomerate granite sarcophagus of Nectanebo II, 345 BC, removed from the al-‘Attarin Mosque in Alexandria, where it was housed in an octagonal building and used for washing before prayer. Contemporaries understood the sarcophagus of Nectanebo II to be the tomb of Alexander the Great. Conceded by the French to the British under the terms of the Treaty of Alexandria in 1801 and donated to the British Museum by King George III (EA 10).

Genizah contained everything from Bibles and prayer books to shopping lists, from marriage contracts to works of philosophy and medicine, from business contracts to pages from Indian animal fables retold in Arabic (Goitein 1967–92). The texts, written in an astonishing range of languages and scripts, reveal a thriving Jewish community with international links ranging from the Atlantic coast to India (Goldberg 2012). Under the Fatimid caliphs, Cairo was a centre of Jewish intellectual debate. In 1166 Maimonides, one of the most influential Jewish scholars of all time, moved to the city from Cordoba in al-Andalus, and became head of the Jewish community in Cairo and physician to the caliph. The Cairo Genizah contains his papers, including autograph drafts of his Mishnah Torah. At the other end of the social spectrum, the Cairo Genizah witnesses the voices of the poor: debtors, widows, refugees, captives, and the sick and disabled (Cohen 2005). The world illuminated by the Cairo Genizah gives insights

not only into the daily life of Jews under Islamic rule, but more generally into those of Jews, Christians and Muslims in Egypt and beyond. Since paper was a valuable material, it was recycled and many of the documents in the Cairo Genizah were actually reused scrap paper from the caliphal court, thus giving astonishing access into the Fatimid empire’s record-keeping apparatus (Rustow 2020). Our sources for Egypt are uneven. First Millennium AD papyri have been found in a limited number of sites in the Fayum, Nile Valley, Eastern Desert and oases, while the wet or damp conditions of the Nile Delta, Egypt’s economic engine in all periods, preclude their survival. The contents of the Cairo Genizah provide an uneven portrait of Egypt in the Medieval period. Yet, together with other archaeological and archival sources found in Egypt, they provide unparalleled access to otherwise invisible communities and, equally important, inform the kinds of questions we

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Fig. 47: Le Miqyàs, by Baron Dominique Vivant Denon, c. 1802 (BM 1836,0109.20.c).

Fig. 48: Child’s striped tunic, seventh–eighth century AD, Egypt (BM 1990,0612.103.a–b).

41

42

E. R. O’CONNELL

Fig. 49: Fragment of silk and linen pseudo-script tiraz, c. eleventh–twelfth century, Panopolis/Akhmim, Egypt. Tiraz cloth often bears text from the Qur’an or the name of a caliph. The workshops were a caliphal monopoly, but owing to their high status the cloths were widely imitated (BM 1893,0514.188).

might ask elsewhere in their contemporary worlds. This volume demonstrates the breadth and depth of the results. Empire and religious identity: Case studies from Egypt The following chapters explore the dialectical relationship between empire and religious identity through

a series of case studies from Egypt. The volume is divided into seven parts. After establishing a range of definitions, approaches and sources (I), the next four sections are organized around cultural markers, different combinations of which have come to be understood as constituting aspects of religious identity (e.g., Durkheim 1917/2001), here, conceptions of time and space (II–III); customs and appearance (IV); and languages and scripts (V).3 The final two sections seek to identify

3

Any number of markers could have been included; I regret in particular not including cooking and consumption practices.

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

43

Fig. 50: Sandstone gravestone of a woman named Fatima, AH 412/AD 1021, found in the Muslim cemetery at Aswan. The epitaph records that Fatima, daughter of Jaf’ar, son of Muhammad the dyer, died on 13 August 1021 (BM 1887,0402.1437).

and locate religious minority populations in Egypt (VI) and address the construction of religious identities in the writing of both local and universal histories (VII). Within these admittedly artificial and intersecting spheres, most of the contributors are explicit in their choice of theoretical models, employing the tools of the overlapping disciplines of sociology, anthropology, philosophy and literary criticism. A combination of historians, archaeologists and art historians, they bring a range of perspectives which usefully challenge received narratives concerning relationships between adherents of traditional Mediterranean religions, Jews, Christians and Muslims in an Egypt ruled by successive empires.

I. Empire and religious identity: Definitions, sources and approaches The first two chapters grapple with historiographical aspects of the study of empire and religion in Egypt from the perspective of documents and material culture, respectively. The bibliographies on ‘empire’ and ‘religion’ are vast, and these chapters serve to define some of the terms of engagement for the ancient and modern periods, as well as suggesting ways forward for future study. Roger S. Bagnall argues that it was precisely in the context of the Roman empire that religious identity as

44

E. R. O’CONNELL

Fig. 51: Panels from the Church of the Virgin, c. AD 1300, Old Cairo, Egypt. Known as al-Mu’allaqa or the Hanging Church, it is one of several Medieval churches, as well as a synagogue, built within the walls of the former Roman fortress and now part of Old Cairo. Probably once gilded, the ten panels decorated doors inside the church. Four are carved with crosses and decorative designs, and six show scenes from the life of Jesus, which demonstrate an international style suggesting strong contacts with the wider Mediterranean world (BM 1878,1203.1 and 8).

a discursive category was formed when, in the middle of the third century, Roman emperors sought to maintain the pax deorum (‘peace of the gods’) by compelling the empire’s inhabitants to sacrifice. He finds that ‘it is precisely the needs of empire that call into existence the recognition of such a thing as religious identity’. Whereas previous attestations of Christians

concerned individuals, it is only in the period after the Decian persecutions, in 256, that we find ‘Christian’ used in the papyrological record, and thereafter very infrequently. Since everyday life was not structured by religious difference, such categories are almost absent from public business or private legal matters, which concern most written documents.

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Darlene L. Brooks Hedstrom uses other forms of material culture to explore the modern construction of the discipline of Late Antique archaeology developed within the modern rubric of empire, foremost the British imperial project in Egypt. Just like the textual world of the documents examined by Bagnall, the material world of objects rarely displays religious affiliation. Brooks Hestrom both sketches the history of scholarship—with its many implicit and explicit biases—and makes useful suggestions for the application of modern archaeological theory in the service of more self-critical approaches to material religion in Egypt (inter alia, Meyer and Houtman 2012). She explores three case studies to illuminate the ways in which religious identity has been assumed or prescribed by scholars in the past, and to apply new methodologies in order to revise the ways in which objects, burials and landscape can be studied to investigate identity. II. Calendar and cosmology In the ancient world, communities and regions had their own agricultural calendars, marked by festivals, which we would now term ‘religious’. In Egypt, the solar year had long been structured by the rise and fall of the Nile, and this only ended with the construction of successive dams near the First Cataract culminating in the Aswan High Dam completed in 1970. Following the adoption of Christianity, Egypt’s year was in time reorganized around the Christian liturgy and the feasts of saints (Papaconstantinou 2001) and, later, around Muslim feasts and mawlids. For reckoning years, regnal dating continued to be used under the Romans (Clarysse, Van der Veken and Vleeming 1983; Bagnall 2009b, 180–83). Consular years took precedence in Egypt only from the mid-fourth to mid-sixth centuries, until Justinian prescribed the combination of regnal year, consular year, indiction year and month and day in legal documents (Bagnall and Worp 2004, 45). The indiction cycle was introduced beginning from AD 312 as a fifteen-year cycle for tax calculation. Under Sasanian Persian rule, and following the briefly restored Byzantine rule, emperors do not appear in dating formulae, but dating by indiction continued (Bagnall and Worp 2004, 5). The Hijri lunar year was introduced in the upper echelons of the Arab Muslim administration immediately following the conquest (e.g., SB 9576, AH 22/AD 643). The system appears in Greek and Coptic documents as ‘the year of the Saracens’ or ‘according to the Saracens’ (Bagnall and Worp 2004,

45

300). Centuries later, the Era of Diocletian was (re-) introduced as a means by which to calculate Christian time, now understood as the ‘Era of the Martyrs’ (Bagnall and Worp 2004, 5). Dating from the first year of the reign of Diocletian had complex origins. It was first used for the purpose of casting horoscopes in documents dating from 304/5 to 507/8 and in private inscriptions (i.e., graffiti and gravestones) from as early as 316/17 until 456/57 in non-Christian contexts (cf. Fig. 6, above). Such dating was also used extensively in Christian epigraphic contexts (e.g., gravestones) and, following the conquest, from as early as 657/58, in Greek and Coptic documents (Bagnall and Worp 2004, 63–87). From as early as 785/86 (in a Greek gravestone from Old Dongola, Nubia), the system of reckoning was recast as the ‘Era of the Martyrs’, emphasizing its now exclusively Christian character and serving the needs of this community (Bagnall and Worp 2004, 67–68). Each system of dating held its own authority. Thus, when Anna, daughter of daughter of John and Taham, donated her property to the Monastery of Apa Paul in the eighth century (P.KRU 106, 31 May 734), her notary used multiple methods of dating the Coptic document: the indiction year; the year of Diocletian; year of the Saracens; and by stating the names of the amir of Hermonthis and the kastron Memnonion under whom it was drawn up. John P. Cooper examines the Medieval perception of the annual Nile flood as the measure of the legitimacy of Egypt’s rulers and the authenticity of their faith. The Egyptian calendar year was divided according to its agricultural seasons (the inundation, growing and harvest season), with the new year beginning upon the appearance of Sirius (the Dog star) upon the eastern horizon coinciding with the beginning of the annual Nile flood. Until the construction of successive Aswan dams in the twentieth century, the success of the Nile flood governed all lives in Egypt. Food security was threatened if it was too high or too low, with implications for tax revenue, navigability and commerce, and the very stability of society (Cooper 2014). The explanation, prediction and, ultimately, the control of Nile flood waters via irrigation systems (and dependent upon conscripted labour) was the responsibility of the state from the earliest times. Once the river reached its optimal level around mid-August, the major seasonal canals were opened so that the water could spread over the cultivated land. In the Medieval period, political legitimacy was no less contingent upon the successful management of the inundation. Cooper examines a

46

E. R. O’CONNELL

narrative recorded in the Christian History of the patriarchs of Alexandria detailing events in late summer 753, a few years after the Abbasids came to power in 750. The narrative describes how God stopped the rising Nile flood waters to punish Egypt’s Muslim rulers, and Christian officials and their congregations gathered at Giza so that God allowed the water to rise again. When the amazed Muslim governor gathered his faithful to pray at the Muqattam hills in order to establish ‘which is the true religion’, the level of the Nile sank again. After a third attempt in which Muslims were joined by Jews and Samaritans, the Nile was unchanged, so that the Muslim governor was forced to concede and ask the Christians to pray again, with the result that the Nile rose to its optimal level. In recognition of God’s support, the governor endowed church buildings, providing the reconciliation required for the coexistence of the two faiths. Just as the custodianship of the measure used to gauge the Nile had previously been reported to have been transferred from the temple of Serapis to a Christian church at Alexandria (Rufinus, Church history 11.30), control of the rituals attending the inundation were eventually transferred to Muslims. About 220 years after the Arab conquest, in the middle of the ninth century (c. 861), control of the ideology of the nilometer shifted from Christian officials to a Muslim man whose job became hereditary. The nilometer established at the Roda in Cairo was a Muslim monument, inscribed with the name of the caliph and a programme of inscriptions including verses from the Qur’an, which represent the integration of Islamic cosmology and the Egyptian sacred landscape. On Barak considers the British imposition of ‘clock-time’ and the Gregorian calendar upon Egypt as an instrument of imperial control and Egyptian resistance. He finds that new technologies such as the printing press, steamers, railways and telegraph lines all contributed to the swift relegation of the Hijri calendar to mark only religious events such as festivals and holidays. The Hijri calendar established by ‘Umar to mark the Prophet Muhammad’s emigration from Mecca to Medina in AD 622 organizes time according to the lunar year, with each day beginning at sunset and the length of hours varying by season. While the traditional Egyptian or Coptic solar year continued after the seventh-century conquest to have practical application to the agricultural cycle and thus financial affairs (foremost taxation), the lunar cycle was used for state administration and the educated public. Following the arrival of the first Arabic printing press

with Napoleon’s army in 1798, harmonization of lunar and solar calendars became more readily feasible. From 1870 the newspaper Wadi al-Nil, which was modelled upon European newspapers, began to systematically render the coordination of the Hijri, the Coptic and Gregorian calendars as well as other calendars serving other Christians in Ottoman lands. But Barak shows that the results were often mismatched, demonstrating just how labour-intensive synchronization of time was (and is). New technologies required the negotiation of authority within the Muslim world; for example, should the day or a festival begin when the moon was sighted locally or when it was sighted in Cairo (or Mecca, or Bombay) and telegraphed to other locations? As part of the British imperial system, Egypt’s train and steamer schedules, primarily for the transport of agricultural products, were determined, within a few short decades, beginning in the 1870s, by the Gregorian calendar and the ‘Frankish’ division of the day into twelve equal hours. Economic concerns required temporal standardization imposed by colonial powers. At the same time, Egyptian recognition of Western mechanical time as a cultural form allowed for its critique and forms of resistance, the invention and celebration of ‘Egyptian time’. III. Topography and cosmology Visual representations of space, such as maps and itineraries, are as much about social constructions as getting from one place to another (Black 1997; B. Anderson 2006, 170–78). In the second century AD, the Alexandrian mathematician and scientist Claudius Ptolemy composed his Geography as an examination of the ‘inhabited world’ as known to the Romans, and stretching from the Atlantic coast of Africa to China and from the most northerly British Isles to the sources of the Nile in East Africa (A. Jones 2020). The work provides instructions for calculating latitude and longitude and for representing the world as a sphere on a two-dimensional surface. Composed in Greek, it was later reproduced in Latin and Arabic and in this way Ptolemy’s authoritative account was passed on through the centuries. The anonymous author of another Greek treatise was a sixth-century Alexandrian merchant known from as early as the eleventh century as Kosmas Indikopleustes (Kominko 2013 and this volume). His world map comes from a work in which he sets out to disprove the theory that the universe is a sphere, with the globe of the earth in the centre. According to his

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

world view, the earth was flat with the heavens arching over it like the curved lid of a box. He presents time and space collapsed in the service of God’s plan for salvation. Under the Fatimids, an anonymous Muslim author working in Egypt composed an Arabic treatise, the Book of curiosities (Rapoport and Savage-Smith 2014). It compiles earlier scholarship including otherwise lost works by Muslim astronomers, geographers and travellers from the tenth and eleventh centuries. Among its many maps is a world map with a graduated scale, which extends from the Atlantic coast of Europe and West Africa to China. Egypt’s Nile Delta is nearly in the centre. The author cites Ptolemy’s Geography as a source for the map, as well as works by scholars commissioned in about AD 830 by the caliph al-Ma’mun. Katherine Blouin uses literary and visual sources to explore the imagined geographies of the Delta in the Roman and Byzantine periods. Since sites in Egypt’s Delta are poorly preserved compared with those in the Nile Valley, the region has received less attention from modern scholars; however, the Delta was arguably always the economic engine of Egypt, providing its high agricultural yield as well as being highly urbanized (Blouin 2014). Blouin examines three sets of sources produced by religious elites: priests of Isis composing an invocation to the goddess; Christian pilgrims; and officials who commissioned a mosaic pavement for a church. Each source situates the Delta within a given frame of reference. She argues that the invocation to Isis, the best-preserved witness to which survives on a long papyrus roll dating to the second century (P.Oxy. XI 1380), was a Ptolemaic-period composition written in Egypt. Its Eastern Mediterranean and Near Eastern orientation is consistent with the Hellenistic world beyond Egypt’s boundaries, with named locations extending as far as Persia and India. By contrast, she finds the Western Mediterranean locations to be treated superficially and thus likely to have been added in an augmented and extended composition in the Roman period. The Christian pilgrims John Cassian, originally from the Danube delta region of the Black Sea, and Egeria, from Spain, both travelled to Egypt as part of their pilgrimages to the Holy Land in the 380s; they were able to undertake their journeys safely in the context of well-supported itineraries protected by Roman imperial interests. As a result of their subsequent travels (for Egeria, home to Spain; for Cassian, to Marseilles via Constantinople and Rome), they carried back geographies of Egypt defined by stories from the Bible and monastic centres. Cassian, indeed, is

47

especially famous for importing Egyptian monasticism to the West. Blouin’s last source, the late sixth- or early seventh-century mosaic pavement of a church today in Jordan, the Madaba Mosaic Map, was based on a combination of Christian and secular sources; the portrayal of Egypt’s Delta, however, relied upon Herodotus or a source based on Herodotus, testifying to the enduring authority of classical literature. While she concludes that religious movements can succeed and their works can be produced without imperial patronage, she argues that imperial support extends the quantity, quality, reach and longevity of our sources. It was precisely within the context of a territory united by imperial shipping lanes and road networks that each set of religious sources is possible. Maja Kominko sets the cosmography of the anonymous Alexandrian trader known by at least the eleventh century as Kosmas Indikopleustes, ‘Kosmas who sailed to India’, firmly in his contemporary context showing that, for him, trade, empire and Christianity are all part of God’s plan to unify humanity. Written in Alexandria in the 540s, the ten books of the Christian topography aim to disprove the assertion that earth and heaven are spherical, promoting instead a cosmos comprised of a flat, rectangular earth surmounted by a firmament. Kominko argues that no matter how tangential, all the details in these ten books are marshalled to support his cosmography. Previous scholarship had not sufficiently appreciated that two other books, probably written prior to the Christian topography, are not part of the same treatise, but were appended to Books I–X in later copies of the manuscript. Book XI is on the plants and animals of Ethiopia and India and contains a description of Sri Lanka, and Book XII is a compilation of excerpts from ancient authors, which affirm the antiquity and accuracy of the Bible. Whereas earlier scholars have inferred that Kosmas’ commercial pursuits were facilitated by networks of ‘Nestorian Christians’, Kominko argues that these identifications are anachronistic (with the East Syrian Church only adopting a dual-nature Christology from the 610s), and also exaggerate his doctrinal allegiance. She aims to clarify the author’s intellectual and religious context, positing informal networks of exchange that allowed the author to engage with contemporary debates taking place in Alexandria (in particular in opposition to John Philoponus), and the biblical exegesis of the Nisibis school of the East Syrian Church (as represented by Patrikios, identified as Mar Aba, who travelled to and taught in Alexandria where Kosmas became his student).

48

E. R. O’CONNELL

Kominko’s close reading of Book XI and survey of the latest research on Red Sea and Indian Ocean trade allows her further to debunk wider claims concerning Sasanian Persian dominance of the Indian Ocean trade at the expense of Rome from the sixth century. Rather, she finds a network of contacts between Rome, Aksum, South Arabia, North Somalia, Persia, India and Sri Lanka that were neither dependent upon the imperial powers of Rome and Persia nor upon the Church of the East, but largely functioned according to convenience. Thus while empires might facilitate and shape religious identity, networks of communication of course existed beyond their boundaries. Yossef Rapoport compares two sets of sources which capture a village in the Fayum at two moments in time, exploring the mechanics of Islamization with profound consequences for our understanding of the process. Around c. AD 1000 the village is largely Christian; 200 years later, it is entirely Muslim. The earlier family archive of Banu Bifam comprises around fifty parchment and paper documents dating between c. 992–1029 and yielding around 100 Coptic names. Only in 1025 is a mosque mentioned at the edge of town with about a dozen Muslims living nearby. By 1245, when al-Nabulusi undertook a fiscal survey of the Fayum, his work shows an overwhelming Muslim majority. Whereas previous scholars have adopted a nomadic settlement model, with Arab Bedouin tribes displacing Christians in Egypt and elsewhere, Rapoport argues instead that the Christian families converted, adopting not only a religious identity, but also a tribal affiliation. Thus, like timekeeping, cosmology, dress, language and food, so too common ancestry was constructed to mark boundaries, and here ancestry was tethered to religious identity. By mapping the tribes and clans reported in al-Nabulusi’s survey, Rapoport further recognizes patterns, finding that groups of Fayum villages belonging to the same tribe followed sections of the canal systems, suggesting that the construction of shared ancestry arose from shared interests, such as water rights and the ability to settle disputes. IV. Customs and appearance Elements of dress and appearance signal inclusion and exclusion, and are sometimes policed. Certain hairstyles were associated with religious adherents; for example, the shaved head of priests for traditional ancient Egyptian cults. In Rome, tunics with clavi,

once reserved for Roman citizens, were a widely adopted fashion (Walker 2000). Later, tiraz workshops were a caliphal monopoly, but the cloths were widely imitated, sometimes bearing pseudo-Arabic script (Helmecke 2015). The so-called ‘Pact of ῾Umar’ or ‘Laws of differentiation’, perhaps compiled in the ninth century and enumerating rights and restrictions of the People of the Book (dhimmis), specified the hairstyles and clothing that could and could not be worn by Christians, Jews, Zoroastrians, Samaritans and other groups recognized by Muslims, with periodic enforcement in Egypt (Swanson 2010, 103). Cäcilia Fluck explores continuity and change in the use of textiles in burials in the transition from Roman and Late Antique Egypt. After describing the mummification process in ancient Egypt, she argues that Roman-period funerary workers concentrated upon the outer form of mummies at the expense of long-held practices treating the body (foremost through evisceration). In Late Antiquity people were buried in or with their clothes and were often further wrapped in shrouds tied with woven bands, in displays that remind modern viewers of the elaborate Roman-period wrappings. In the adoption of the practice of dressing the dead in their finest clothes, Fluck finds an extension of the same individuation of the deceased that resulted in portraits painted on plaster masks, wooden panels and shrouds. In Late Antique wrapping practices, she finds continuity with earlier Egyptian funerary ideals. Rather than a shift based primarily on religious identity, she sees the third-century economic crisis as the likely point of rupture between Roman and Late Antique funerary practices. Thelma K. Thomas traces the visual language of dress deployed to signal the authority of Christian monks in representations produced from Late Antiquity and into the Medieval period. Forms of dress had long been associated with different kinds of secular authority in the Roman world. As the outermost garment, the mantle most visibly characterized the wearer. By Late Antiquity, the toga or its updated version, the loros, represented civic authority, whereas the chlamys belonged to high-ranking military and imperial officials and the himation (in Greek) or pallium (in Latin) to philosophers and intellectuals. While the last two were used in the depictions of angels and military saints, it was the himation-pallium which was used to depict Christ and his followers, including monks (cf. Tertullian, De pallio). These options could be modified, for example, decorated with crosses or reduced in scale to

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

represent humility, as shown by the small mantle worn by the abbot Shenoute and others (and probably to be identified with the maphorion and mafort known from Greek and Coptic sources, respectively). In addition to these mantles of secular authority, the mantle of scriptural authority, the melote, defined in the biblical description of the prophet Elijah, who gifts it to his disciple, Elisha, was another, biblical option for conveying status. Depictions of monks thus showed their status both in the secular sphere, adopting the visual vocabulary of philosophers, and within the monastic sphere, with variations in group portraits demonstrating (spiritual) familial relationships and relative status between individuals. In the Medieval period, Christians again adopted contemporary dress to signal status so that Shenoute is shown in striped tunics and highly decorated mantles, consistent with the elite tastes in the ninth–eleventh centuries, and demonstrating not only his Christian, but also his worldly, authority in the context of Abbasid and Fatimid Egypt. V. Languages and scripts Each successive empire ruling Egypt (the Achaemenid Persians, the Greeks, the Romans, Sasanian Persians and Arabs) brought its own language of power, each with different degrees of penetration (Adams, Janse and Swain 2002; Adams 2003; Papaconstantinou 2012, 4; 2016). Egypt’s dry climate preserves an astonishing range of documents and literature, foremost papyri (Bagnall 1995; 2009), and later the Cairo Genizah documents, mainly on paper and parchment (e.g., Goitein 1967–93). The statistical possibilities of using these enormous datasets have long been recognized and exploited to characterize the chronology and genre of texts, their languages, and material (e.g., Trismegistos. org). As elsewhere in the Roman empire, Latin was used in some top domains (e.g., to convey law, to register citizen births and in the army) and, as in the rest of the East, Greek was used in other registers. Once the last stage of the Egyptian script, demotic, had ceased to be used in everyday writing, it was almost two centuries before the Egyptian language was commonly written again. Known today as Coptic, it was written in a modified Greek script with about 30% of its vocabulary borrowed from Greek. While Greek-demotic language contact, interference and social aspects of language choice (e.g., code-switching) are well-developed fields of study, the relationships between Greek, Coptic and Arabic are now also being analysed in the same

49

depth (Papaconstantinou 2012). As well as characterizing the broad sweep of the adoption and abandonment of languages and, through naming practices especially, faiths, manuscripts from Egypt also allow us access to communities that may be otherwise largely invisible. Although they are rarely writers themselves, women, notoriously obscure to modern historians despite comprising half the population at any given moment, are somewhat better represented in documents from Egypt than in, say, literary sources (Bagnall and Cribiore 2006). Religious minorities are represented by documents from Egypt on an unparalleled scale. The Aramaic papyri from Elephantine Island, opposite Aswan, give a unique glimpse into the world of a largely Jewish garrison on the Achaemenid Persian frontier in the fifth century BC (Nirenberg, this volume). It has often been observed that the field of Syriac literature would not exist without the survival of manuscripts in Egypt. For the Medieval period the c. 400,000 manuscripts of the Cairo Genizah (largely in Hebrew, Aramaic, Judeo-Arabic and Arabic, but also witnessing Coptic and other languages and scripts) are by far the best source for the Mediterranean and Indian Ocean (Goitein 1967–93, and below). Sofía Torallas Tovar maintains that there was no prescriptive linguistic policy in the ancient world. Supplementing legal and historical sources with documentary evidence from Ptolemaic and Roman Egypt, she argues that these states promoted the use of a language in a particular domain and tolerated languages other than Greek and, later, Latin in additional domains. Thus, in Ptolemaic Egypt, revenue laws were proclaimed in both Greek and Egyptian (259 BC). About a century later, Egyptian documents registered in the local notary’s office (grapheion) had to be registered in Greek with a Greek abstract (by 145 BC), thus promoting Greek in this sphere of public life. A decree of 118 BC stated that the language of a contested contract would govern which court, Egyptian or Greek, the parties attended, so that Egyptian language and law continued to be sanctioned. Greek was incentivized, for example, with tax exemptions given to Greek schoolteachers. The Roman state also did not impose a linguistic policy, but demotic Egyptian nevertheless quickly disappeared from the economic and administrative spheres, surviving only in the realm of religious and/or literary texts. Roman legal sources refer only to Latin and Greek and rarely to vernaculars in general, and never to Egyptian in particular. This is not surprising given the size of the Roman empire and the many

50

E. R. O’CONNELL

different languages and scripts used by its populations. Common to both the Ptolemaic and Roman states was their flexibility, which, Torallas Tovar argues, was guided by intelligibility in legal and other public domains. Whereas Egyptian was used for a millennium alongside Greek, Arietta Papaconstantinou details the process whereby Greek gave way to Arabic within half that time. Just like the Ptolemies or Romans, the Arab rulers of Egypt did not impose a linguistic policy in Egypt. Arabic was used immediately after the conquest at the highest levels of the administration, but Greek continued to be used until it gave way to Coptic in local legal contexts in the eighth century. As Egypt’s official language, Arabic bestowed advantages upon its speakers, foremost social mobility, improving career paths and access to legal and administrative resources. Especially when, under the Shiꜥi Fatimids based in Cairo, Christians and Jews achieved better integration and participation in civil society (albeit with periodic conflicts arising), Arabic was the common tongue of elites, regardless of their religious affiliation. In this period, when the Red Sea trade flourished again, Arabic, not Coptic, provided access to commerce. It also conferred access to the latest science and learning beyond the borders of Egypt. Papaconstantinou recognizes resistance in some monastic circles, which equated Coptic with Christian heritage, but once the Church adopted Arabic in the early eleventh century, and moved the patriarchate from Alexandria to Cairo, the Arabization of Egyptian Christians was well on its way. VI. Minorities and majorities This volume is mainly concerned with the religions associated with successive empires in Egypt, i.e., traditional cult practice, Christianity and Islam. When exactly the shift occurred from minority to majority populations has been hotly contested by scholars, but the scholarly consensus largely agrees that Egypt’s population had become a Christian majority by the fifth century (inter alia, Depauw and Clarysse 2013), and Muslim in the eleventh century (Rapoport, Papaconstantinou and Kennedy, this volume). Significant minorities were present within these populations, either adherents to other faiths or groups within a given faith. These have variously been written out of history or indeed used as the mirror of self-examination, to serve as the ‘other’ by which one may define one’s self (inter

alia, Nirenberg 2013). One of the extraordinary features of Egypt’s rich papyrological record is that it allows, sometimes, for members of communities ‘othered’ in the historical record to present themselves. Jewish populations which were supposed to have been completely eradicated or expelled from Egypt appear in documents (see above P.Oxy. IX 1205, AD 291; P.Oxy. XLIV 3203, AD 400). When they appear, it is in remarkably ordinary circumstances. The first Christians to become visible in the documentary record appear not as a hidden, secretive sect fleeing persecution, but in the midst of everyday correspondence and identifiable by their distinctive greetings, titles and conventions such as the use of nomina sacra (Choat 2006; Luijendijk 2009). Later, among Christians, the so-called Melitians who were so vilified in the writings of Athanasius, among others, appear in the course of everyday business transactions (P.Lond. VI). So too, Manichaeans lived side by side with (other) Christians in Egypt’s Dakhla Oasis (P.Kellis). While first finding favour at the court of Shapur I in Sasanian Persia, Mani (AD 216–274) was later condemned to death. His teaching spread in Syriac-speaking lands from Persia to Syria, and from the Greek-speaking East to the Latin West, with Augustine himself self-identifying as a Manichaean for about a decade (Brown 2000, 35–49). Roman imperial rescripts under Diocletian parallel those against Christians, but whereas Christianity was later adopted by the imperial house, later emperors continued to issue legislation against Manichaeans. Our main sources, apart from legal codes and the polemical literature, are documents preserved foremost on papyri in Egypt (Römer 2009) and from ‘Silk Road’ oases in what is now Western China. Whereas the main powers in Late Antiquity, Rome/Byzantium and Persia, maintained their own state religions (Christianity and Zoroastrianism respectively), only the Turkic Uighurs of the mid-eighth to mid-ninth century maintained Manichaeanism as their state religion, which scholars have argued was at least partly to distinguish themselves from their Buddhist Tang Dynasty neighbour. Building on his 2013 book, David Nirenberg explores how the idea of Judaism was used by nonJews in Egypt for over a millennium to define themselves against the ‘other’ precisely in the context of successive empires: Persian, Greek, Roman and Byzantine. Only occasionally does the papyrological record show Jews themselves. The extraordinary preservation of fifth-century BC documents from Elephantine Island on Egypt’s traditional southern border at the First

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Cataract records a garrison of Jews and others employed by the Persians (see now Van der Toorn 2020). Following the Egyptians’ destruction of their ‘Temple of YHW’, King Darius himself writes to his governor to warn Egyptians to keep away from the garrison during the coming Passover and recounts the process whereby the garrison was allowed to rebuild the temple, but without permission to sacrifice animals, presumably the prerogative of Jerusalem alone by this period (Porten et al. 1996). Nirenberg can only speculate concerning the Egyptian perspective to explain the hostility, e.g., that the Jews were the agents of a foreign empire, or that the Passover festival (with its ritual re-enactment of the defeat of Egypt and the slaughter of the paschal lamb) offended the priests of the ram-god Khnum. Subsequent Hellenistic and Egyptian histories of Egypt show how Jews were perceived. The Greek historian, Hecataeus of Abdera, recounts the earliest non-biblical witness to the Exodus story, in which a plague in Egypt was blamed on foreigners who practised different rites. They were led by a man called Moses (‘outstanding in his wisdom and courage’) who banned the worship of images and whose followers were distinguished by their unsocial and intolerant way of life. The Egyptian historian, Manetho, among others, ‘cast Jews as degenerate descendants of defeated invaders from the distant past’. In the Acts of the Alexandrians, the citizens were to have defended their city against Rome and its agents, specified as Jews, an antagonism repeated in the first and second centuries and witnessed by Philo of Alexandria and recounted by Josephus (see above, P.Lond. VI 1912). Nirenberg argues that the Jews in Egypt were eliminated by 117, but that even while there were no Jews in Egypt (i.e., until the ninth century, when Jews reappear in the Genizah corpus), the prolific abbot Shenoute (d. 465) could accuse his opponent or indeed his own congregation of impieties he casts as Jewish. Where Nirenberg finds second-century eradication until the ninth century, Petra Sijpesteijn traces the papyrus evidence found at sites in the Fayum and Edfu. She finds Jewish names, as well as institutions such as a communal synagogue on one of the Appion estates in the Fayum in the middle of the sixth century (P.Oxy. LV 3805). Only a handful of administrative texts in Greek and Arabic name ‘Jews’ (hebre(a)ios, Ioudaios; al-yahūdī) in the first centuries of Muslim rule and Sijpesteijn argues this does not signify an absence of Jews, but rather that their identification in texts was not necessary until the mid-eighth-century changes. Prior

51

to this time, ‘people of the land’ (ahl al-arḍ) referred to Egyptians as distinct from the occupying army (jund). Only in the mid-eighth century did Arab Muslims begin to settle outside of the garrisons of Fustat and Alexandria, and the term qibṭ (al-aqbāṭ) first appears at this time in order to distinguish non-Muslims with respect to an agricultural tax. It simply identifies non-Muslims (i.e., Egyptians), without regard for whether they were Christians or Jews. The late eighthcentury and early ninth-century migration of PersianTurkish soldiers replaced the Arab Muslim families who had held offices and land from the initial conquest in the seventh century (Kennedy, this volume). In this context, an Arab Muslim identity was promoted as distinct from the new arrivals, and names attributing origin (nisba) or other ethnic or religious affiliations were employed at the local level where these Arab Muslim families still enjoyed some status. As a result, terms to distinguish Christians and Jews were employed and so we see an explosion of ‘Jews’ in the record where they became more systematically identified as al-yahūdī or al-isrā’īlī. Thus, Sijpesteijn argues that Jews were not necessarily more numerous than they had been in earlier centuries, but they are simply named in the documents from this time owing to the complex socio-political dynamics of the eighth and ninth century. Roberta Mazza re-evaluates a polemical letter (P.Ryl. III 469) accusing Manichaeans of a litany of wrongs. Missing the beginning and therefore the address, it accused them of: 1) being against marriage; 2) idolatry; 3) carrying out deviant rituals involving the menstrual blood of Elect women. Mazza re-dates a papyrus letter, probably circulated by a bishop from the third century (and thus contemporary with Mani) to the fourth century, situating its rhetoric and polemics within a Constantinian milieu. Previously dated by its palaeography to the second half of the third century, the earlier range of which has been reproduced in subsequent scholarship, she suggests that the contents are more in keeping with the imperial Christian Church, which was better able to focus its efforts on competitors, rather than defending itself. Instead of a Festal letter circulated on the occasion of Easter, the letter aims to halt the proselytizing of Manichaeans and claims first-hand knowledge of their rites. Accusations of being anti-marriage find correspondence in the writings of Didymus the Blind and others in Athanasius’ circle. Improper use of bodily fluids such as semen and menstrual blood are a common heresiological accusation, applied to Manichaeans by no less a theologian

E. R. O’CONNELL

52

than Augustine himself. Mazza situates the letter in the fourth century as part of a movement to normatize Christianity following the challenge of the Council of Nicaea, called by Constantine in 325. Mazza thus identifies this letter as part of a movement to resist pluralism and diversity and establish a single universal Church. VII. Universal and local histories While histories written at the end of the First Millennium BC, e.g., by Polybius (d. 118 BC) and Diodorus of Sicily (d. after 36 BC), established a view of the fulfilment of human history through the establishment of the Roman empire, political narrative was more strongly intertwined with religious developments following the Chronicle and Church History of Eusebius (d. 339) and his successors writing in Greek, Latin and Syriac (Fowden 2014, 68–82). Finding coincidence in the time of Augustus and Christ, Eusebius twinned the empire of the Romans and the pious teaching of Christ, articulated most forcefully in his In praise of Constantine (Drake 1976, 16.4–7). Muslim historians too, e.g., Muhammad ibn Ishaq (d. 767) and Tabari (d. 923), wrote universal histories providing a narrative for the early history of the caliphate (Fowden 2014, 79). But there are frequently tensions between universal and local histories. As a rule, each history made significant use of earlier sources, which were then copied, edited, modified and extended to suit the writer and his audience (for the origins of the chronicle and its subgenres, see Burgess and Kulikowski 2013). Local histories often gained popularity among readers, continuing to be copied and read, sometimes to the detriment of universal histories (Fowden 2014, 79; Booth, this volume).4 Phil Booth compares the Chronicle of the bishop John of Nikiu to the competing and more successful narratives of his near contemporaries, which were later compiled in the History of the patriarchs of Alexandria. Modern scholars have largely adopted the perspective given in the latter, a regional, Egyptian

4

The historiography of Egypt is underdeveloped in comparison to contemporary traditions (e.g., for the Church of the East: Debié 2015; Wood 2013).

history, wherein an anti-Chalcedonian Severan Church was to be distinguished from that of the pro-Chalcedonian Roman imperial oppressors, with the result that Egypt was indifferent to or even welcoming of Arab Muslim rule. Thus the patriarch of Alexandra, Benjamin, is portrayed as mediating the transfer of power from Roman to Arab rule, characterized as the ‘philosopher-cum-prophet’ at the court of ῾Amr b. al-῾As (see Cooper, this volume). In this context, the History of the patriarchs of Alexandria employs ‘Copts’ (al-aqbāṭ) to describe a regional ethnic Church. By comparing the History of patriarchs with the universal Chronicle of John of Nikiu, Booth finds a multiplicity of voices even within the anti-Chalcedonian tradition. By contrast to the regional and ecclesiastical horizons of the History of the patriarchs, the Chronicle is a universal history, and indeed the only universal chronicle of the seventh century. Composed in the late seventh century, the Chronicle combines several Roman historiographic and hagiographic texts. Booth disputes the charges of Egyptocentrism that have been levelled by modern scholars at the Chronicle, arguing that John’s aim is always to produce a universal chronicle, even when his main source, John Malalas, runs out in the reign of Justinian (r. 527–65), narrating politics in Constantinople for the period until 641–42. Like the History of the patriarchs, the Chronicle is anti-Chalcedonian, expressing the view that God punished the Roman empire for its persecution of the orthodox. Booth points out that John lived under both Sasanian Persian and Arab rule, and there is no suggestion that he expected Egypt to return to Roman rule. Nevertheless, unlike the History of the patriarchs, John’s world view is fundamentally Roman, in genre and in chronological and geographical scope. Originally written in Greek or Coptic, the text was translated into Arabic, then, at the Ethiopian court in 1602, into Ge’ez. All four surviving Ge’ez manuscripts, or parts thereof, are copies of this single manuscript. It has not survived in Greek, Coptic or Arabic as it was not perhaps as relevant to Egyptians in the succeeding centuries as it was to its more regionally and religiously specific rivals.

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Hugh Kennedy situates the first centuries of Muslim rule in the context of regional politics. While historians of the universal caliphate tell us very little about Egypt, historians living in Egypt, Ibn ῾Abd al-Hakam (d. 871) and Muhammad bin Yusuf al-Kindi (d. 961), provide a much fuller picture. Both belonged to the leading Arab Muslim families who maintained political and military power in Egypt until the beginning of the ninth century. While, in the first centuries, the appointed governors of Egypt appear to have had absolute authority, Kennedy finds that their negotiation between two powers with fundamentally different priorities made their roles much more precarious. Umayyad (658–750) and Abbasid (750–969) caliphs sought to maximize revenue for their capitals but also to maintain military control over other provinces. By contrast, the local Arab Muslim elite, led by the dozen families who dominated politics and the army (jund) in Egypt, sought to retain revenue at home and discourage the frequent taxrevolts of the Coptic majority (starting from 725/26). These Arab Muslim elites had their origins not among nomadic Bedouin, as often imagined by modern scholars, but among well-established tribes previously settled in the villages and towns of South Arabia. At first, these families and the vast majority of Arab Muslims lived in Fustat and, in time, Alexandria. While under the Umayyads, the chief of police, a member of one of these families, mediated between the governor and the Egyptian army (jund), their power was increasingly reduced under the Abbasids. Kennedy explains that the Egyptian elite families were further diminished by the arrival of Andalusi Arabs expelled from Cordoba in the early ninth century, who took control of Alexandria. Once order was restored in 826, the Abbasids sent Persian-Turkish administrators to Egypt, who effectively dismantled the financial basis for maintaining the Arab Muslim families. The histories as told by al-Hakam and al-Kindi thus favour the Umayyads at the expense of the Abbasids, in accordance with their own families’ experience. It was precisely in the context of such Muslims becoming ‘Egyptian’, alongside Christians and Jews, that names attributing origin (nisba) or other ethnic or religious affiliation became common (Sijpesteijn, this volume).

53

In Egypt the shift from a belief in many gods to a belief in one God mirrors what took place across much of Eurasia and North Africa. It reflects the transformation of the ancient to the Medieval world, a transition that shaped the world we live in today. In defining identity, this volume supports the position that religion is only part of a wider world view that can encompass different combinations of cultural markers, only some of which are addressed herein: diet and consumption practices; art and architecture; timekeeping; geography; cosmology; customs and appearance; languages and scripts; and political and economic ideologies. At the same time these cases show how religious identity remains central to the politics of similarity and difference, inclusion and exclusion, continuity and change. Acknowledgements Some of my early thinking about the subject of Egypt and empire was developed at the ‘Second workshop of the ancient and modern imperialisms network’ at Stanford University in 2007, and I thank Phiroze Vasunia and Susan Stephens for the invitation. The chronological narrative presented above was refined in the course of many presentations to the general public, and I am extremely grateful for their questions, which forced me to distil many complex subjects. I thank Susanna Elm, Arietta Papaconstantinou, Jitse Dijkstra and Neal Spencer for their comments and suggestions on this chapter. All errors in judgement and fact remain my own. All images in this chapter are courtesy of the Trustees of the British Museum. Throughout this chapter, I have Anglicized many Egyptian, Greek and Arabic names, thus, e.g., Fustat, not Fusṭāṭ. While I have given the Greek place names for Roman and Byzantine Egypt, according to the administrative usage, I have endeavoured also to give the modern place name in the text. The limited space available on maps, however, has required that only the former appear (see Map 1).

54

E. R. O’CONNELL

Bibliography All papyrological abbreviations are cited throughout the chapter in accordance with the conventions established in J. F. Oates et al., Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed Sept. 2018), and P. M. Sijpesteijn et al. Checklist of Arabic documents, http://www.naher-osten.lmu. de/isapchecklist (last accessed Sept. 2018). Adams, J. N. 2003. Bilingualism and the Latin language. Cambridge. Adams, J. N., M. Janse and S. Swain. 2002. Bilingualism in ancient society: Language contact and the written word. New York; Oxford. Alcock, S. E., T. N. D’Altroy, K. D. Morrison and C. M. Sinopoli (eds). 2001. Empires: Perspectives from archaeology and history. Cambridge. Ali, N. and R. Wood. 2017. The emergence of Islamic art. In Imagining the divine: Art and the rise of world religions, J. Elsner and S. Lenk (eds), 134–59. Oxford. Anderson, B. 2006. Imagined communities: Reflections on the origin and spread of nationalism. Revised ed. London; New York. Anderson, W. 2007. Menas flasks in the West: Pilgrimage and trade at the end of antiquity. Ancient West & East 6: 221–43. Bagnall, R. S. 1993. Egypt in Late Antiquity. Princeton, NJ. ———. 1995. Reading papyri, writing ancient history. New York. ———. 2009a. Early Christian books in Egypt. Princeton, NJ. ———. 2009b. Practical help. In Oxford handbook of papyrology, R. S. Bagnall (ed.), 179–96. Oxford. ———. 2011. Everyday writing in the Graeco-Roman East. Berkeley, CA. Bagnall, R. S. and R. Cribiore. 2006. Women’s letters from ancient Egypt, 300 BC–AD 800. Ann Arbor, MI. Bagnall, R. S. and D. Rathbone (eds). 2016. Egypt from Alexander to the Copts: An archaeological and historical guide. 2nd revised ed. Cairo; New York. Bagnall, R. S. and K. A. Worp. 2004. Chronological systems of Byzantine Egypt. 2nd ed. Leiden; Boston, MA. Bang, P. F. and C. A. Bayly (eds). 2011. Tributary empires in global history. Cambridge Imperial and Post-Colonial Studies. New York, NY. Behrens-Abouseif, D. 2014. Between quarry and magic: The selective approach to spolia in the Islamic monuments of Egypt. In Dalmatia and the Mediterranean: Portable archaeology and the poetics of influence, A. Payne (ed.), 402–25. Leiden; Boston, MA. Björnesjö, S. 2013. The history of Aswan and its cemetery in the Middle Ages. In The First Cataract of the Nile: One region – Diverse perspectives, D. Raue, S. J. Seidl-

mayer and P. Speiser (eds), 9–14. Sonderschriften des Deutschen Archäologischen Instituts, Kairo 36. Berlin. Black, J. 1997. Maps and history: Constructing images of the past. New Haven, CT. Blanke, L. 2019. An archaeology of Egyptian monasticism: Settlement, economy and daily life at the White Monastery Federation. New Haven, CT. Blouin, K. 2014. Triangular landscapes: Environment, society and the state in the Nile Delta under Roman rule. Oxford. Blue, L. K. and E. Khalil. 2011. The Lake Mareotis Research Project: A multidisciplinary approach to Alexandria’s economic past. BAR International Series 2285. Oxford. Booth, P. 2013. The Muslim conquest of Egypt reconsidered. In Constructing the seventh century, C. Zuckerman (ed.), 639–70. Travaux et mémoires 17. Paris. Boraik, M. 2014. SCA excavations at Luxor: New discoveries from the First Millennium AD. In Egypt in the First Millennium AD: Perspectives from new fieldwork, E. R. O’Connell (ed.), 207–13. British Museum Publications on Egypt and Sudan 2. Leuven. Boyarin, D. and C. A. Barton. 2016. Imagine no religion: How modern abstractions hide ancient realities. New York. Brophy, E. 2015. Royal statues in Egypt 300 BC–AD 220: Context and function. Oxford. Brown, P. 2000. Augustine of Hippo: A biography. Berkeley, CA. Burgess, R. W. and M. Kulikowski. 2013. Mosaics of time 1: A historical introduction to the chronicle genre from its origins to the high Middle Ages. Studies in the Early Middle Ages 33. Turnhout. Carruthers, W. (ed.). 2015. Histories of Egyptology: Interdisciplinary measures. New York; London. Choat, M. 2006. Belief and cult in fourth-century papyri. Studia Antiqua Australiensia 1. Turnhout. ———. 2010. Coptic. In Oxford handbook of Roman Egypt, C. Riggs (ed.), 581–93. Oxford. ———. 2017. Writing and communication in early Egyptian monasticism. In Writing and communication in early Egyptian monasticism, M. Choat and M. C. Giorda (eds), 17–72. Texts and Studies in Eastern Christianity. Leiden. Clark, E. A. 1992. The Origenist controversy: The cultural construction of an early Christian debate. Princeton, NJ. Clarysse, W. 1993. Egyptian scribes writing Greek. Chronique d’Égypte 68: 186–201. Clarysse, W., G. van der Veken and S. P. Vleeming. 1983. The eponymous priests of Ptolemaic Egypt: Chronological lists of the priests of Alexandria and Ptolemais with a study of the demotic transcriptions of their names. Papyrologica Lugduno-Batava 24. Leiden. Cohen, M. R. 2005. Poverty and charity in the Jewish community of Medieval Egypt. Princeton, NJ; Oxford. Cooper, J. P. 2014. The Medieval Nile: Route, navigation and landscape in Islamic Egypt. Cairo.

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Cribiore, R. 1996. Writing, teachers and students in GraecoRoman Egypt. American Studies in Papyrology 36. Atlanta, GA. ———. 2001. Gymnastics of the mind: Greek education in Hellenistic and Roman Egypt. Princeton, NJ. Curran, B. A., A. Grafton, P. O. Long and B. Weiss. 2009. Obelisk: A history. Cambridge, MA. el-Daly, O. 2005. Egyptology: The missing millennium: Ancient Egypt in Medieval Arabic writings. London. Davis, S. 2004. The early Coptic papacy. New York. Debié, M. 2015. L’Écriture de l’histoire en Syriaque: Transmissions interculturelles et constructions identitaires entre hellénisme et islam. Leuven. Depauw, M. and W. Clarysse. 2013. How Christian was fourth century Egypt? Onomastic perspectives on conversion. Vigiliae Christianae 67: 407–35. ———. 2015. Christian onomastics: A response to Frankfurter. Vigiliae Christianae 69: 327–29. Derda, T., T. Markiewicz and E. Wipszycka. 2007. Alexandria: Auditoria of Kom el-Dikka and Late Antique education. Journal of Juristic Papyrology Supplement 8. Warsaw. Dieleman, J. 2005. Priests, tongues and rites: The LondonLeiden magical manuscripts and translation in Egyptian ritual (100–300 CE). Leiden. Dietler, M. 2010. Archaeologies of colonialism: Consumption, entanglement and violence in ancient Mediterranean France. Berkeley, CA. Dijkstra, J. H. F. 2008. Philae and the end of ancient Egyptian religion: A regional study of religious transformation (298–642 CE). Orientalia Lovaniensia Analecta 173. Leuven. ———. 2011. The fate of the temples in Late Antique Egypt. In The archaeology of Late Antique ‘paganism’, L. Lavan and M. Mulryan (eds), 389–486. Late Antique Archaeology 7. Leiden; Boston. ———. 2020. Crowd behaviour and the destruction of the Serapeum at Alexandria in 391/392 CE. In Religious violence in the ancient world: From classical Athens to Late Antiquity, J. H. F. Dijkstra and C. R. Raschle (eds), 286–305. Cambridge. Dijkstra, J. H. F. and van der Vliet, J. 2020. The Coptic Life of Aaron: Critical edition, translation and commentary. Vigiliae Christianae, Supplements 155. Leiden; Boston. Durkheim, E. 1917/2001. The elementary forms of religious life. Translated by C. Cosman. Oxford; New York. Elm, S. 1996. ‘Pierced by bronze needles’: Anti-montanist charges of ritual stigmatization in their fourth-century context. Journal of Early Christian Studies 4: 409–39. Elsner, J. (ed.). 2020. Empires of faith in Late Antiquity: Histories of art and religion from India to Ireland. Cambridge. Emmel, S. 2004. Shenoute’s literary corpus. 2 vols. Corpus Scriptorum Christianorum Orientalium 599–600. Leuven.

55

Fine, S. 2005. Art and Judaism in the Greco-Roman world: Toward a new Jewish archaeology. Cambridge; New York. Fournet, J.-L. 1999. Hellénisme dans l’Égypte du VIe siècle: La bibliothèque et l’œuvre de Dioscore d’Aphrodité. 2 vols. Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 115. Cairo. Fowden, G. 2014. Before and after Muhammad: The First Millennium refocused. Princeton, NJ. ———. 2020. The rise of Coptic: Egyptian versus Greek in Late Antiquity. Princeton, NJ. Frank, G. 2000. The memory of the eyes: Pilgrims to living saints in Christian Late Antiquity. Transformation of the Classical Heritage 30. Berkeley, CA. Frankfurter, D. (ed.). 1998. Pilgrimage and holy space in Late Antique Egypt. Religions in the Graeco-Roman World 134. Leiden; Boston. ———. 2014. Onomastic statistics and the Christianization of Egypt: A response to Depauw and Clarysse. Vigiliae Christianae 68: 284–89. ———. 2015. Review of B. Nongbri, Before religion: A history of a modern concept (New Haven, CT, 2013). In Journal of Early Christian Studies 23 (4): 632–34. ———. 2017. Christianizing Egypt: Syncretism and local worlds in Late Antiquity. Princeton, NJ. Garel, E. 2020. Héritage et transmission dans le monachisme égyptien: Les Testaments des supérieurs du topos de Saint-Phoibammôn à Thèbes (P. Mon. Phoib. Test.). Bibliothèque d’études coptes 27. Cairo. Gayraud R.-P. 1999. Le Qarafa al-Kabra, derniére demeure des Fatimides. In L’Égypte fatimide – son art et son histoire: Actes du colloque organisé à Paris les 28, 29 et 30 mai 1998, M. Barrucand (ed.), 443–64. Paris. Gessler-Löhr, B. 2010. Two child mummies and some grave goods of the Byzantine period from the Egyptian collection at Heidelberg University, Germany. In Mummies of the world, A. Wieczorek and W. Rosendahl (eds), 310– 15. Munich. Godłewski, W. 1986. Deir el-Bahari V: Le monastère de St Phoibammon. Warsaw. Goitein, S. D. 1967–93. A Mediterranean society: The Jewish communities of the Arab world as portrayed in the documents of the Cairo Geniza. 6 vols. Berkeley, CA. Goldberg, J. 2012. Trade and institutions in the Medieval Mediterranean: The Geniza merchants and their business world. Cambridge. Gonis, N. 2010. P. Paramone: Emperors, conquerors and vassals. Zeitschrift für Papyrologie und Epigraphik 173: 133–35. Grenier, J. C. 1983. La stèle funéraire du dernier taureau Bouchis (Caire JE 31901=Stèle Bucheum 20): Ermant – 4 novembre 340. Bulletin de l’Institut français d’archéologie orientale 83: 197–208.

56

E. R. O’CONNELL

Haas, C. 1997. Alexandria in Late Antiquity: Topography and social conflict. Ancient Society and History. Baltimore, MA. Harmless, W. 2004. Desert Christians: An introduction to the literature of early monasticism. Oxford; New York. Hoffman, A. and P. Cole. 2011. Sacred trash: The lost and found world of the Cairo Geniza. New York. Hunt, L.-A. 1989. The Al-Mu’allaqa doors reconstructed: An early 14th-century sanctuary screen from Old Cairo. Gesta 28 (1): 61–77. Jones, A. 2020. The ancient Ptolemy. In Ptolemy’s science of the stars in the Middle Ages, D. Juste, B. van Dalen, C. Burnett and D. N. Hasse (eds), 13–34. Turnhout. Jones, C. 1987. Stigma: Tattooing and branding in GraecoRoman antiquity. Journal of Roman Studies 77: 139–55. Jones, M. and S. McFadden. 2015. Art of empire: The Roman frescoes and imperial cult chamber in Luxor temple. New Haven, CT. Keulen, W. H. et al. (eds). 2015. Apuleius Madaurensis Metamorphoses, Book XI: The Isis Book. Leiden. Kominko, M. 2013. The world of Kosmas: Illustrated Byzantine codices of the Christian topography. Cambridge. Kristensen, T. M. 2012. Miraculous bodies: Christian viewers and the transformation of ‘pagan’ sculpture in Late Antiquity. In Patrons and viewers in Late Antiquity, S. Birk and B. Poulsen (eds), 31–66. Aarhus Studies in Mediterranean Antiquity 10. Aarhus. ———. 2013. Making and breaking the gods: Christian responses to pagan sculpture in Late Antiquity. Aarhus. Layton, B. 2014. The canons of our fathers: Monastic rules of Shenoute. Oxford. Luijendijk, A. M. 2009. Greetings in the Lord: Early Christians and the Oxyrhynchus papyri. Cambridge, MA. MacCoull, L. S. B. 1988. Dioscorus of Aphrodito: His work and his world. Berkeley, CA. McKenzie, J. 2007. The architecture of Alexandria and Egypt 300 B.C.–A.D. 700. New Haven, CT; London. Markus, R. 1994. How on earth could places become holy? Origins of the Christian idea of holy places. Journal of Early Christian Studies 2: 257–73. Meyer, B. and D. Houtman. 2012. Introduction: Material religion—How things matter. In Things: Religion and the question of materiality, D. Houtman and B. Meyer (eds), 1–23. New York. Meyer, M. W. and J. Smith. 1994. Ancient Christian magic: Coptic texts of ritual power. San Francisco, CA. Michel, S. 2001. Die Magischen Gemmen im Britischen Museum. London. Mitchell, S. and P. Van Nuffelen. 2010. One God: Pagan monotheism in the Roman empire. New York. Modrzejewski, J. and R. Cornman. 1995. The Jews of Egypt: From Rameses II to Emperor Hadrian. Edinburgh. Nirenberg, D. 2013. Anti-Judaism: The history of a way of thinking. New York.

Nongbri, B. 2013. Before religion: A history of a modern concept. New Haven, CT. O’Connell, E. R. 2007. Transforming monumental landscapes in Late Antique Egypt. Journal of Early Christian Studies 15: 239–74. Ogden, D. 1999. Binding spells: Curse tablets and voodoo dolls. In Witchcraft and magic in Europe: Ancient Greece and Rome, B. Ankarloo and S. Clark (eds), 1–90. London. Opper, T. 2014. The Meroë head of Augustus. London. Papaconstantinou, A. 2001. Le culte des saints en Égypte: Des Byzantins aux Abbassides. L’apport des inscriptions et des papyrus grecs et coptes. Paris. ———. 2003. Review of C. Haas, Alexandria in Late Antiquity (Baltimore, MA, 1997). The Journal of Egyptian Archaeology 89: 295–97. ———. 2012. Why did Coptic fail where Aramaic succeeded? Linguistic developments in Egypt and the Near East after the Arab conquest. In Multilingualism in the Graeco-Roman worlds, A. Mullen and P. James (eds), 58–76. Cambridge. ———. (ed.). 2016. The multilingual experience in Egypt, from the Ptolemies to the Abbasids. New York, NY. Petry, C. F. 1998. Cambridge history of Egypt 1: Islamic Egypt, 640–1517. Cambridge. Porten, B., J. J. Farber, S. Hopkins and R. Katzoff (ed.). 1996. The Elephantine papyri in English: Three millennia of cross-cultural continuity and change. 2nd revised ed. Leiden. Powell, E. Troutt. 2003. A different shade of colonialism: Egypt, Great Britain and the mastery of the Sudan. Berkeley, CA. Power, T. 2012. The Red Sea from Byzantium to the caliphate: AD 500–1000. Cairo; New York. Preiser-Kapeller, J., L. Reinfandt and Y. Stouraitis (eds). 2020. Migration histories of the Medieval Afroeurasian transition zone: Aspects of mobility between Africa, Asia and Europe, 300–1500 C.E. Leiden. Rapoport, Y. and E. Savage-Smith. 2014. An eleventh-century Egyptian guide to the universe: The Book of Curiosities. Leiden. Rebillard, É. 2012. Christians and their many identities in Late Antiquity, North Africa, 200–450 CE. Ithaca, NY. Rebillard, É. and J. Rüpke (eds). 2015. Group identity and religious individuality in Late Antiquity. Washington DC. Reid, D. M. 2002. Whose pharaohs? Archeology, museums and Egyptian national identity from Napoleon to World War I. Berkeley, CA. ———. 2015. Contesting antiquity in Egypt: Archaeologies, museums and the struggle for identities from World War I to Nasser. Cairo. Riggs, C. 2005. The beautiful burial in Roman Egypt: Art, identity and funerary religion. Oxford.

ONE GOD: EGYPT AND EMPIRE IN THE FIRST MILLENNIUM AD

Ritner, R. K. 1993. The mechanics of ancient Egyptian magic. Studies in Ancient Oriental Civilization 54. Chicago, IL. Rogers, J. M. 2017. Didymus the Blind and the Alexandrian Christian reception of Philo. Atlanta, GA. Römer, C. 2009. Manichaeism and gnosticism in the papyri. In Oxford handbook of papyrology, R. S. Bagnall (ed.), 622–43. Oxford; New York. Rondot, V. 2013. Derniers visages des dieux d’Égypte: Iconographies, panthéons et cultes dans le Fayoum hellénisé des IIe–IIIe siècles de notre ère. Paris. Russell, N. 1981. The Lives of the Desert Fathers. Kalamazoo, MI. Rustow, M. 2014. Heresy and the politics of community: The Jews of the Fatimid caliphate. Ithaca, NY. ———. 2020. The lost archive: Traces of a caliphate in a Cairo synagogue. Princeton, NJ. Sahner, C. C. 2019. Christian martyrs under Islam: Religious violence and the making of the Muslim world. Princeton, NJ. Sharkey, H. J. 2003. Living with colonialism: Nationalism and culture in the Anglo-Egyptian Sudan. Berkeley, CA. Sheehan, P. 2010. Babylon of Egypt: The archaeology of old Cairo and the origins of the city. Cairo. Sijpesteijn, P. M. 2013. Shaping a Muslim state: The world of a mid-eighth-century Egyptian official. Oxford Studies in Byzantium. Oxford. ———. 2014. Early Umayyad papyrus invitation for the Ḥajj. Journal of Near Eastern Studies 73 (2): 179–90. Smith, M. 2017. Following Osiris: Perspectives on the Osirian afterlife from four millennia. Oxford. Smith, R. R. R. and B. Ward-Perkins (eds). 2016. The last statues of antiquity. Oxford. Soskice, J. 2011. Sisters of Sinai: How two lady adventurers found the hidden gospels. London. Speiser P. et al. 2013. Umayyad, Tulunid and Fatimid tombs at Aswan. In The First Cataract of the Nile: One region – Diverse perspectives, D. Raue, S. J. Seidlmayer and P. Speiser (eds), 211–22. Sonderschriften des Deutschen Archäologischen Instituts, Kairo 36. Berlin. Spier, J. 2007. Late Antique and early Christian gems. Spätantike, Frühes Christentum, Byzanz, Reihe B: Studien und Perspektiven 20. Wiesbaden. Spier, J. and M. Charles-Murray (eds). 2007. Picturing the Bible: The earliest Christian art (exhibition held at the

57

Kimbell Art Museum, Fort Worth, November 18, 2007– March 30, 2008). New Haven, CT. Stolz, Y. 2006. Eine kaiserliche Insignie? Der Juwelenkragen aus dem sog. Schatzfund von Assiut. Jahrbuch des romisch-germanischen Zentralmuseums Mainz 53: 521–603. Swanson, M. N. 2010. The Coptic papacy in Islamic Egypt (641–1517). New York. Tkaczow, B. 1993. Topography of ancient Alexandria: An archaeological map. Travaux du Centre d’archéologie méditerranéenne de l’Académie polonaise des sciences 32. Warsaw. Török, L. 2005. Transfigurations of Hellenism: Aspects of Late Antique art in Egypt, AD 250–700. Probleme der Ägyptologie 23. Leiden; Boston, MA. Tyson Smith, S. 2014. Desert and river: Consumption and colonial entanglements in Roman and Late Antique Nubia. In Inside and out: Interactions between Rome and the peoples on the Arabian and Egyptian frontiers in Late Antiquity, J. H. F. Dijkstra and G. Fisher (eds), 89–107. Leuven. Van Bladel, K. T. 2009. The Arabic Hermes: From pagan sage to prophet of science. Oxford; New York. Van der Toorn, K. 2020. Becoming diaspora Jews: Behind the story of Elephantine. New Haven, CT. Walker, S. (ed.). 2000. Ancient faces: Mummy portraits from Roman Egypt. London; New York. Walter, C. 1994. Some unpublished intaglios of Solomon in the British Museum. In Thymiama Stē Mnēmē Tēs Laskarinas Boura, 1: 365–68, 2: pls 205–6. 2 vols. Athens. Watts, E. J. 2010. Riot in Alexandria: Tradition and group dynamics in Late Antique pagan and Christian communities. Berkeley, CA. Wilburn, A. 2012. Materia magica: The archaeology of magic in Roman Egypt, Cyprus and Spain. Ann Arbor, MI. Wipszycka, E. 2009. Moines et communautés monastiques en Égypte (IVe–VIIIe siècles). Journal of Juristic Papyrology Supplement 4. Warsaw. Wood, P. 2013. The Chronicle of Seert: Christian historical imagination in Late Antique Iraq. Oxford. Zanker, P. 1988. The power of images in the age of Augustus. Ann Arbor, MI. Zych, I. 2003. Wooden coffins from the Moslem cemetery at Kom el-Dikka. Polish Archaeology in the Mediterranean 14: 32–37.

I EMPIRE AND RELIGIOUS IDENTITY: DEFINITIONS, SOURCES AND APPROACHES

EGYPTIAN RELIGIOUS IDENTITIES UNDER ROMAN IMPERIAL RULE: CRITICAL REFLECTIONS Roger S. BAGNALL

The themes of the colloquium in which this volume originated were as vast and sprawling as any ancient or medieval empire. When I was asked to use this address to reflect in general terms on the connections of empire and religion, surely a daunting assignment, I first turned to the academic poltroon’s last refuge, definition. It would have been easily possible to avoid the topic entirely by spending my space trying to define empire and religion. Theoreticians and comparatists of empire have been active, perhaps hyperactive, in the last three decades, and most of the issues about religion that I shall be discussing have their parallels in the debate about empire. But I am going to leave that aspect aside, because I shall in fact be discussing almost entirely a single empire, the Roman, which, moreover, tends to be at the root of many generalizations about empire and definitions of the phenomenon. The ‘almost’ in that statement refers to the Arab empire, which will come in at the end. Religion, however, is another matter. As we shall see, and as many of the other papers in this volume show, wrestling with the question of what religion is turns out to be anything but peripheral to our theme, and it leads to that theme rather than helping me to escape it. I shall be arguing, in fact, that the Roman empire played a central role in causing concepts of religion and religious identity to come into existence. In a certain fashion, there was no such thing as religion without the Roman empire and its evolving needs. Egypt is thus part of a larger developmental process; but the richness of its evidence allows us to look at some aspects of the subject in a way not possible elsewhere. To see how and why this is the case, we must confront the thorny question of what we mean by ‘religion’.

This matter of definition has been a major preoccupation of the field of religious studies.1 The nature of the problem and one of the main directions of argument are encapsulated in the title of Brent Nongbri’s recent book (2013) Before religion: A history of a modern concept. As the wording of his title indicates, Nongbri does not believe that there was a concept of religion in antiquity, and he argues that the ancient terms that we translate with the English ‘religion’ and its equivalents in other modern languages did not mean anything like what we mean by religion. The phrase ‘a common universal religion’, which Katherine Blouin (this volume) has quoted from Jan Assmann, could thus not have had any meaning for the Egyptians or any other ancient people. The second proposition, that ancient terms did not mean what religion does for us, is not new, but Nongbri argues it well. A recent book by Carlin Barton and Daniel Boyarin (Barton and Boyarin 2016)2 develops this argument in much greater detail, and I do not think that anyone who reads their acute discussions of religio and threskeia will be able to come away tempted to translate either word as ‘religion’. (The reality is in fact much more interesting than that.) Like many others, all three of these scholars make forcefully the point that the ancients had no abstract category under which they would have listed all of the things we call religion; this abstract category of analysis is a product of modern Western, for the most part Christian, thought. If we write about ancient religion, then, we must recognize that we are not pursuing a line of inquiry that the Greeks, Romans or Jews would have understood at all as we do. But that is not the end of the story. Nongbri in the end follows Jonathan Z. Smith (2004) in arguing that religion remains a second-order, redescriptive category

1

2

The vast territory covered (superficially) here inevitably means that bibliographical references have been limited to a few works that have helped shape my thinking and several helpful suggestions from the referees; they will, however, lead the reader to much more of the relevant literature.

I am deeply grateful to the authors for access to the proofs of this stimulating book in advance of publication. Asad (2001) is useful as a consideration of the difficulties of ‘religion’ as a noun, from a more theoretical and less historical perspective than Nongbri.

62

R. S. BAGNALL

of analysis useful for scholarly inquiry. As he points out (Nongbri 2013, 158), ‘All of our concepts are modern and hence anachronistic when applied to the ancient world’. The problem is not anachronism in itself but lack of awareness of anachronism on the part of scholars. Barton and Boyarin (2016), in contrast, stake out a position hostile to this kind of redescriptive treatment, regarding the term ‘religion’ as ‘misleading and distracting’. This is by no means an academic conceit, recently invented in our most fortunate and self-aware era. In a review of two books by Alfred North Whitehead, written in 1927 but unpublished until 2015, T. S. Eliot remarked ‘that there is such a thing as “religion” above the various particular religions seems to me very doubtful’. I come out much closer to Nongbri in this regard. Religion is, after all, not unique in the conceptual problems it poses. Economic historians had this debate a generation ago, and, fortunately, study of that other anachronism, the economies of the ancient world, continues, even if the ancients also had no real concept of ‘the economy’.3 Our choices are not simply between using ‘religion’ as an unproblematic category of analysis and abandoning it as too historically contingent a construct (Barton and Boyarin 2016, 8–9). And, to be fair, in their detailed discussion Barton and Boyarin are much more nuanced, arguing, for example—in part following Mary Beard (1986)—that a notion of religion as proper cult to be regulated by the Roman state and even as an object of discourse goes back to Cicero, although they stress the originality and tendentiousness of his attempt to redefine religio (2016, 18–19, 39ff.). And they suggest that awareness of difference in cult practices, which is crucial to the question of the existence of religious identity, goes back to the secondcentury Christian apologists. The conceptualization of a Jewish cult distinct from those of the surrounding society can be found already in the first century AD in

Philo (Barton and Boyarin 2016, 201). They note, moreover, that Tertullian ‘pressed the word [religio] to serve new functions with new companions along several trajectories that will lead, in the third, and particularly the fourth century, to resemble (some of) our notions of “religion”’ (Barton and Boyarin 2016, 55). They speak (2016, 100) of the empire moving ‘toward an increasingly sacralized universality, toward legal uniformity and common citizenship’. These observations, which I think are sound, point us in an important direction, to which we shall return. But for our present purposes the most interesting questions actually lie elsewhere. It is not so much whether the ancients had a concept of ‘religion’ in the abstract, or of ‘religions’ as discrete entities, that concerns me here. It is whether they recognized religious identity in people.4 It is a commonplace that the cults of the Greeks and Romans were generally linked to particular places; they were local rather than general or imperial, mostly connected to cities, villages, or sometimes natural features in the landscape. This remains broadly true after classical times, even though the Hellenistic monarchies developed royal cults that were spread throughout their empires, including the cities.5 Individual cults were not conceived of as excluding others; being a devotee of one god did not lead to the exclusion of worship of others or a sense that those gods were less real. Exclusive religious identity in the modern sense can hardly have existed in such an environment. The Jews have often been discussed as the earliest counter-example in the Graeco-Roman world. Shaye Cohen’s book The beginnings of Jewishness (1999) tackled this difficult question. For our purposes, the fundamental problem is the one he confronts in his third chapter: when does Ioudaios cease to have an exclusively ethnic-geographic meaning, ‘Judaean’, and become the designation of those who, whatever their

3

4

The classic formulation of the ancient lack of the concept of ‘economy’ appears in Finley 1973, 18–34, esp. 21: ‘In stressing this [the inability to translate ‘economics’ into Greek or Latin] I am not suggesting that the ancients were like Molière’s M. Jourdain, who spoke prose without knowing it, but that they in fact lacked the concept of an “economy”, and a fortiori, that they lacked the conceptual elements which together constitute what we call “the economy”.’ For a concise summary of the evolution of the debate on the ancient economy since Finley, see Scheidel, Morris and Saller 2007, 1–12.

5

This subject too has a vast bibliography, with many approaches. See recently, for example, Rebillard and Rüpke 2015. Identity is, of course, a complicated concept, involving both attributed identity, as when the Roman state classified the residents of Egypt as Romans, Greeks, or Egyptians, and internalized or subjectively constructed identities (as with Hellenic identity in the Roman empire; cf., e.g., Whitmarsh 2001, esp. 90–130). In this article I am primarily concerned with the first. See Ma 1999, 219–26, for a brief treatment with references to the bibliography, especially the classic Habicht 1970.

EGYPTIAN RELIGIOUS IDENTITIES UNDER ROMAN IMPERIAL RULE

63

origin, worship the god whose temple was in Jerusalem? His thesis (Cohen 1999, 70) is that ‘all occurrences of the term Ioudaios before the middle or end of the second century B.C.E. should be translated not as “Jew,” a religious term, but as “Judaean,” an ethnic-geographic term’. He points to the latter part of that century as the time when the semantic shift occurs, through which it becomes possible to use Ioudaios for individuals who have become worshippers of the god of the Judaeans or else the political partners of the Judaeans. This shift implies also the introduction of the concept of conversion, i.e., to turning to share in the worship of the god of the Judaeans, although this move never in antiquity results in a clear and coherent definition of what practices were required in a convert. Despite continued ethnic-geographic use of Ioudaios, in some late Hellenistic and Roman cases it is clearly possible to translate the word as ‘Jew’. That some cases are doubtful and that it is only gradually that the ‘religious’ sense of the term becomes dominant is less significant for our purposes than that a designation comes into play marking adherence to a cult as a conceptual marker distinguishing individuals from those who do not adhere to this cult. The telos ioudaikon, ‘Jewish tax’, levied by the Roman authorities after AD 70, was explicitly a tax on those who observed Jewish practices, not on Judaeans, and thus recognized an identity that we would call religious, or that at least was based on what we regard as religious practices, laid down in Jewish law (Cohen 1999, 58–59). Although one might be tempted to call this ethnicity, that is hardly better in terms of avoiding anachronistic modern conceptualization. There has indeed been much discussion of exactly what the Judaism covered by the tax was, and how far it may have differed from one emperor to another.6 At all events, it is a term regarding identity, based largely in cultic practices, which meant above all exclusivity in worship of their one god.

Rives sees Decius’ edict in the middle of the third century, requiring sacrifice from all inhabitants of the empire, as a step, however ineffectual, towards creating a ‘collective religious identity’, trying ‘to treat the empire as though it were a city’.8 In a subsequent article he called the decree ‘a highly innovative and important step towards a radical restructuring of religious organization in the Roman world’ (Rives 1999, 135).9 It for the first time created a ‘religious obligation between the individual and the Empire’. But, although in a bureaucratic sense the sacrifice and certificate procedure linked to it were perfectly within the

6

9

7

8

See Schwartz 2001, 187–88, discussing the views of S. Cohen and M. Goodman, among others. Barton and Boyarin (2016) indeed conclude with a clear sense that it was response to the fact of empire and the emperor that drove much of the development they describe. Rives 1995, 260; the decree is discussed in greater detail in Rives 1999.

But that takes us only so far. As James Rives (1995, 241) pointed out, ‘The Jews were first and foremost a people, an ethnos; in so far as there was such a thing as “Judaism” it was the traditional form of worship peculiar to that people’. Rives argues that Roman policies concerning Judaism in the first two centuries AD are not to be seen as imperial religious policy, but as policies concerning peoples and their customs. In this respect they are in line with a general absence of positive religious policy for the provinces and indeed very little presence of negative policy either; that is, the authorities prohibited practices they thought threatening to the well-being of the state, but these prohibitions did not form part of any larger conceptual framework or rest on a basis of formal legislation about religion, there being no such concept. In this sense, his argument fits into the broad lines of the resistance to abstract notions of religion with which I began. But it has further implications. Rives says (1995, 245–46): The religious pluralism, not to say anarchy, of the empire reflected the absence of any organized system of official religion. The civic model [sc. of religion] was in the last analysis fundamentally incompatible with the empire, and the attempts to maintain it led ultimately to a radical separation of religious identity and political authority.7

In this sense, Rome was a city like any other; its civic cults were specific to it and did not scale to the empire: see Ando 2008, esp. 95–119. Although critical of Rives’ analysis, he remarks (p. 119), ‘The failure of any Roman cult to become a religion for and of the empire should therefore not surprise us’.

64

R. S. BAGNALL

capacities of the imperial administration, the nature of civic religion was so fundamentally at odds with any imperial unifying goals that sacrifice alone made a poor basis for this attempt, and no succeeding emperor found a better approach within the traditional cultic framework of the Graeco-Roman world, not even the imperial cult. Still, the middle of the third century is the moment at which we can first define an imperial concept of empire-wide religious observance, and the persecution of non-conformant Christians thus points up the existence of a group that had an identity different from that of the majority of the empire’s population. That is, it is precisely the needs of empire that call into existence the recognition of such a thing as religious identity. The process may, indeed, have its beginnings visible already in the Antonine Constitution of Caracalla, which Claudia Moatti (2016) has argued shows signs of the same conception of the empire as Decius’ edict. A similar view also emerges from Jörg Rüpke’s analysis of the relevant imperial legislation of the Tetrarchic period and of the Edict of Milan (313) issued by Constantine and Licinius (Rüpke 2016, esp. 65–76). It is notable that Rüpke does not hesitate to render religio as ‘religion’ in these texts, marking a more abstract understanding of the term in the third and fourth centuries than one might be willing to accept in earlier times. My avoidance of a focus on Christianity until now has perhaps not escaped notice. In part that is a reflection of our twin focus on Egypt and on the relationship of religion and empire. Despite the abundance of legend preserved in Eusebius, the wishful thinking of some papyrologists, and the blatant misrepresentation of the evidence by those with a need to project backward a fully formed Christian church that can be considered ancestral to later ecclesial bodies, there is almost no real evidence for Christianity in Egypt before the Severan period (Bagnall 2009, 3–5; Wipszycka 2015), and it is in the period after Decius that the record begins to become more visible, as AnneMarie Luijendijk’s (2008) account of the Oxyrhynchus material has shown. It is not until 256 that the first mention of the word ‘Christian’ in a documentary context occurs (P.Oxy. 42.3035). This particular gap could be an accident of preservation, but the coincidence of this attestation with both the general growth in evidence for Christians in the papyri and the movement towards greater interest in cultic practice on the part of the imperial government is telling. The Decian persecution seems to be the first time that the government took on

an identifiable church structure and hierarchy, rather than dealing reactively with individual Christians when the case arose, as we see happening in Pliny’s famous correspondence with Trajan. There is another inescapable issue here, however, and that is the validity of the category ‘Christianity’. A telling marker of the questioning of the category is the appearance in modern discourse of the plural ‘Christianities’. A 1990 book by Jonathan Z. Smith used the term in its title, and uses have multiplied since then. Bart Ehrman’s book (2003) Lost Christianities: The battle for scripture and the faiths we never knew helps to show the polemical force of the plural. The gathering recovery during the past seventy years of texts that did not become part of the orthodox canon of scripture, a recovery owed in large part to the discovery and editing of papyri, has made it possible to get a far fuller and clearer view of something long known: that Nicene Christianity represents only one strand in a profusion of views found among those in the Roman world who counted themselves as followers of Christ. That Christian belief and practice were not monolithic is not news, of course. No more are they today, and even at the height of Catholic dominance of the West and Orthodox rule of the East there were many church bodies, especially in Asia, that owed no obedience to either Rome or Constantinople and differed from them on many points of doctrine and practice. What is new in the last quarter-century, rather, is the insistence on giving those who did not prevail in the power struggles of Late Antiquity an equal voice, to put them on the same footing as the organized strain that emerged as imperially sanctioned orthodoxy. Even that latter phrase is full of problems, as the continued battles at church councils from Ephesos on down to the iconoclastic controversies show. It takes only the slightest imagination to realize that Judaism is subject to the same pluralizing approach; my late colleague Alan Segal already used the plural in The other Judaisms of Late Antiquity in 1987, and in the same year appeared a collective volume (Frerichs, Green and Neusner 1987) titled Judaisms and their messiahs at the turn of the Christian era. One of the editors of that volume was Jacob Neusner, who is often credited with bringing the plural into scholarly discourse (see Schwartz 2001, 9). But not everyone has been persuaded. Seth Schwartz instead proceeds from the assumption ‘that ancient Judaism was complex, capacious, and rather frayed at the edges’, concluding ‘I reject the characterization of Judaism as multiple,

EGYPTIAN RELIGIOUS IDENTITIES UNDER ROMAN IMPERIAL RULE

as well as the atomistic reading of the sources that justifies it’ (Schwartz 2001, 9). In the case of Christianity, similarly, I am going to avoid this byway, partly because it would take too long to explore properly the ideological underpinnings of this movement, but more because I think it is even more problematic than the whole category of ‘religion’ in imposing on antiquity a point of view that derives from a modern pluralistic world. I do not think that the ancient Christians believed they were adherents of one out of many equally valid strains. Each group thought they possessed the true Christianity and that those who disagreed were wrong, or worse than wrong. The striving for unity and conformity in Christianity is just as deep and ancient as its inherent tendency to fissure. So, like Schwartz, I shall stick with the singular Christianity, despite its limitations in conveying the reality. But neither Christianity nor Judaism can claim the title of most debated categorization; that crown belongs to ‘paganism’ without a doubt. That the use of ‘pagan’ to characterize those who still accepted the traditional civic religion of the Roman world is a Christian adaptation of an existing word is not in doubt. Alan Cameron (2011, ch. 1) has argued, in my view convincingly, that it was for the most part intended descriptively as an omnibus term for non-Christians rather than polemically. This usage in itself, of course, shows the existence of a willingness to abstract religious identity as an individual characteristic. Paganism was an invention of Christianity. That no one has found a better term, as Cameron argues, also seems to me patent; ‘polytheist’ is equally a Christian invention, meaningless except from a monotheistic perspective. But in response to Christian categorization, clearly those who in various ways still participated in the older cults developed a sense of themselves as part of a distinct tradition, which they might well have refused to see as representing ‘religion’ as opposed to a more comprehensive culture.10 There is, as must be obvious by now, no such thing as innocent categorization. All forms of aggregation and disaggregation serve someone’s interests. And yet categorization seems both a natural human drive, which

10

Fundamental works for this subject are Bowersock (1990) and Frankfurter (1998).

65

is not likely to go away, and a necessary element in historical analysis. We will not escape it. All we can do is to avoid assuming that our own use of categories is innocent. In looking at the Roman situation, we have to keep asking cui bono? What interests were served by categorization, in this case by describing individuals as a defined group? This is surely one nexus of empire and religion. If Decius was a pioneer in seeing the possibility of using religion as a force for cohesion in the Roman empire, beset as it was by external threats, economic problems and internal political fission, he unleashed a process of which he can hardly have foreseen the consequences. For the first time as a matter of imperial policy, cultic practice became a matter of distinguishing us and them: us, the majority whose practices supported the well-being of a unified empire; them, the minority who refused to participate in these practices. Categories in this context are rhetorical moves aimed at the exercise of power. Of course empires were old hands at categorization; that was not in itself new. But when the Assyrians, Persians or Ptolemies categorized, it was by official ethnicity, tax status, occupation and similar descriptors of obvious relevance to administration. They could not have conceived of ‘religion’ as such an administratively relevant descriptor. It is, at all events, hardly to be wondered at if, when Christians came to power, they found the Decian strategy readily reversible; not only what they came to call pagans, and Jews, but also Christians not conforming to the views adopted by the emperor of the moment could be categorized and disadvantaged in various ways. Given my topic, it is hardly surprising that the problem of religion has occupied such a central place up to this point. But I think it is essential to question that place. Such questioning comes naturally to a papyrologist centrally interested in social and economic matters, of course. But the central role of religion in the discussion is not an inevitability in any case. It is the product of a long historiography, going back certainly to Gibbon, who gave the rise of Christianity a central (and largely negative) place in his history. It

66

R. S. BAGNALL

has been reinforced in the last few decades by the development of the field of Late Antiquity, which has been heavily oriented towards religious questions and texts. Those directions reflect an abundant source material largely ignored by historians until recent decades, being previously left to the care of patristics scholars. They also reflect the interests of a small group of highly influential and distinguished scholars who have trained and influenced younger generations. One could, however, imagine an alternative Late Antiquity, in which angels, the soul and asceticism got less attention, and rents, administration, land ownership and taxation got more.11 That is, of course, the Late Antiquity of the documentary papyri, revealed above all (but not only) in an Egypt in which people buy and lease land, borrow money or wheat, pay rent and taxes, petition the governor and so on. It is striking just how little mention of religion there is in the papyri and how difficult it is to say anything about the religious commitments of individuals and families. A few soundings will illustrate what I mean. A search in the Duke Databank of Documentary Papyri for the string χριστιαν- in texts before AD 500 produces just seven occurrences,12 of which two belong to the well-known archive of correspondence, now in the British Library, originating in a Melitian community. That leaves five from all other contexts taken together. Out of some 9,200 papyri and ostraca from the period 250–500, that is not very impressive. Other terms, like Arian or Melitian, produce still fewer (αριαν- in fact yields only the name Arianos). These words do not figure in the daily discourse of people’s written lives, not even very much in letters. This is a superficial observation, but it points to a deeper problem, that of identifying the religious affiliations of individuals in the documents. This problematic dominates the issue of inclusion in the Corpus Papyrorum Judaicarum, which, I have argued elsewhere (Bagnall 1982, 110), got it largely wrong for Late Antiquity by virtue of identifying most people

11

12

That alternative Late Antiquity, of course, is that already represented a half-century ago by A. H. M. Jones (1964). At www.papyri.info, specifying ‘strict’ for the date; ‘loose’ adds only two more. Search carried out 3 June 2016. Adding χρηστιαν- (a common enough spelling) brings only four more

with biblical names as Jews. It has been enormously thorny for Christians, as one can see, for instance, in the polemics over the inclusion or exclusion of letters in Mario Naldini’s volume of Lettere cristiane (see especially Wipszycka 1974). Malcolm Choat has devoted an entire volume to the range of criteria deployed by scholars to identify Christians in the papyri, trying with great learning, prudence, and precision to distinguish what he considers fairly certain from what is only speculative.13 One need not agree with all of the positions he takes in matters of detail to be grateful for the care of the analysis and the elicitation of areas of doubt. Choat also points out the similar problems facing us in distinguishing Manichaeans from (other) Christians: ‘No less than any other group, Manichaeans are unlikely to have always marked themselves unquestionably as adherents of their sect through the use of unique vocabulary’ (Choat 2006a, 137). Much of the difficulty stems from the atomized nature of the papyrological corpus. An individual letter or lease gives us little explanatory context. As Choat has insisted, reference to a cult title even in a private letter does not of itself indicate that the writer shared the religious commitment of the person referred to. A title such as ‘priest’ or ‘deacon’ may occur in an official document purely for description or identification, and the same can be true in private contexts. Approval or disapproval is not implied. When Aurelius Isidoros tells us in a petition of AD 324 (P.Col. 7.171 = P.Coll.Youtie 2.177) that he was rescued from an altercation by a passing monk and deacon, we are not entitled to assume that Isidoros was a Christian, only that he knew that the titles would adequately identify the individuals to a magistrate. Papyrus archives offer more promising ground, as they give us multiple points of observation on individuals and families. But even there, certainty may be frustratingly elusive. A defining case is the archive of Theophanes of Hermopolis. It is clear that his circle included both Christians and pagans, and he himself

13

passages, two of them from the Melitian archive and one only vaguely dated. Choat 2006a. See also Luijendijk 2008, 25–55 on the difficulties of identifying Christians in the third- and early fourth-century documentation.

EGYPTIAN RELIGIOUS IDENTITIES UNDER ROMAN IMPERIAL RULE

has often been identified by modern commentators as an adherent of traditional religion. But this is far from certain, as a detailed study by Choat (Choat 2006b; see also Choat 2006a, 90–92) has shown. As he says, ‘there is evidence to suggest that Theophanes’s family were not, in fact, worshippers of Hermes at all, and with this must go the certainty that he himself was’ (2006b, 42). The elite circles of Hermopolis in the 310s and 320s appear to have included a mix, and there is no sign that the internal strife visible within the small circle of the city councillors of the day had anything to do with religious affiliation. Rather, economic interests appear at the fore, to the point that religious identity is obscured. To quote Choat again, ‘common cultural values, religious differences notwithstanding, bound elites together’ and ‘the majority of the educated classes that filled the municipal offices and imperial administration did not attach as much importance to religion as do modern discussions’ (Choat 2006b, 42 and 73 respectively). One could argue that economic interests are inevitably the core concerns of documents drawn up for the needs of material life. Moreover, such documents in general have no need to provide any indication of religious affiliation; neither in public business nor in private legal matters was this something to be recorded. The names of parents and grandparents, place of origin, occupation, age, and distinguishing marks all serve the purposes of precise identification; religion does not. But we might expect more reflection of religious difference in two categories of text, private letters and petitions. Surely these would offer opportunities to complain about mistreatment on religious grounds by those of a different persuasion. But such references hardly exist. Even in Theophanes’ archive, as Choat points out, it is the coexistence within his papers of devotees of Hermes and Christians that is striking. The letters and petitions more generally are no less centred on daily economic and familial concerns than the rest of the documents. This statement remains as true when we come to the controversies arising from the councils of the fifth century, particularly Chalcedon, and the resulting fissures developing inside Egyptian Christianity in the sixth century. These are virtually impossible to identify in the papyri. This is in a sense hardly surprising. It does not much matter in drawing up a lease or a will whether one believes that Christ had one nature or two natures. Once again, we find that looking at an archive is revealing. That of Dioskoros of Aphrodito is one of the two

67

great archives of the sixth century, and it is rich in discursive petitions drawn up by him as well as poetry composed by him. And yet, as the late Leslie MacCoull observed in 1988, ‘One cannot even say with the hard-and-fast certainty of labeling by hindsight which side of the confessional fence, Chalcedonian or nonChalcedonian, he came down on’ (MacCoull 1988, 151). His family had an ancestral tie with a Pachomian monastery in Tesmine, but despite the predominantly Chalcedonian character of the Pachomian congregation in the sixth century, some Pachomian establishments were non-Chalcedonian or at least housed nonChalcedonian monks (Gascou 2008; Wipszycka 2015, 434–38). This is not to say that religious difference was unimportant, and certainly not to claim that the papyri present a complete picture, any more than the ecclesiastical and literary sources do. The documents had functions to fulfil, and these functions had little to do with religion, so it is natural that they tend to ignore it. But they do suggest a society in which the functioning of everyday life was not structured by religious difference. There is a less quotidian level at which it has been claimed that religious difference was of importance, and precisely in the relationship of Egypt to the later Roman empire. There is a longstanding, if often contested, narrative according to which the miaphysite branch of Egyptian Christianity, with its opposition to the doctrines promulgated at Chalcedon, was inextricably tied together with Egyptian nationalism and hostility to Roman rule. This view was most definitively demolished by Ewa Wipszycka in a classic article (1992). As she notes, it has resisted previous demolitions and is hard to destroy. But at least in serious scholarly circles, it now seems to have been replaced by a recognition of the fact that both miaphysite and Chalcedonian Christians in Egypt spoke both Greek and Egyptian and belonged to various social levels. What I have said so far runs a serious risk of seeming to offer a static picture of Egypt at a level of experienced reality. But ‘eternal Egypt’ is not actually a concept with which I have any sympathy. Egypt underwent many changes throughout the period of Roman rule (see generally Bagnall 1993; 2007) and under Arab rule. Among these were the development of the nome capitals into self-administering cities and the concomitant rise of the class of local notables that ran these cities; the privatization of the public land, turning tenants into owners; the integration of Egypt into the production and trading economy of the Roman empire;

68

R. S. BAGNALL

the transformation of the tax system from one centred on wheat into one largely based on money; a shift from a monetary regime based on silver and bronze to one organized around gold and bronze; the development of a more professionalized bureaucracy from Diocletian on; and a greater concentration of wealth. The degree of this last development is controversial, as is any shift in the legal forms though which labour was employed. Two things are evident from this list: a close connection to developments in the larger empire and the absence of any link to religious change. Of course the second generalization is somewhat artificial, as I omitted from the list anything that would contradict it. This period, to be sure, also saw the decline and disappearance of the institutional structures of the Egyptian temples and the abandonment or, often after a considerable period of time, reuse of their physical manifestations. In its place arose a vast array of churches and monasteries throughout the country, with their own extensive personnel. By the later fifth century, along with these changes in physical infrastructure, sacred time had been reorganized around the Christian liturgical year and the calendar of the commemoration of saints (Papaconstantinou 2001). Demotic writing went out of use, and Coptic was created to express the Egyptian language, incorporating many Greek loanwords, appearing first (as it seems) in Christian use and then in everyday contexts. Burial customs were transformed, although only slowly. Bishops and priests sought, in detail if not always with success, to regulate the private lives of the faithful. Little of this, and the list could be extended, is unique to Egypt. As in other areas of life, Egypt was by Late Antiquity well integrated into the common practices of the eastern part of the empire. This integration itself represents a significant development since the start of Roman domination. Although Egypt in the Late Period and Ptolemaic period was much less self-contained than has sometimes been thought, and Augustus already started the process of making Egypt’s administration conform to imperial norms (Bowman and Rathbone 1992), there is no doubt that Egypt under Augustus was far less Roman than Egypt under Diocletian or Theodosius. It is impossible here to give a proper assessment of how far this change represented imperial purposes and agency and how far it reflects the interests and actions of the local elites of Roman Egypt, particularly after the Severan reform that gave the nome capitals their full civic institutions. But either way, the imprint of empire was profound.

Is this then true with religion as well? The most obvious instance where one can see that imprint is the disaster that befell the Jewish population of Egypt in the wake of the suppression, at the start of Hadrian’s reign, of the great revolt that had broken out in Cyrenaica and Egypt under Trajan. We know very little in detail about what happened, yet again a sobering reminder of the silences of the papyrological documentation. From the near-absence of identifiable Jews in the documents of the period from Hadrian to Diocletian it has been surmised that much of the Jewish population was killed, although we have no means of estimating how many might have been enslaved instead. We cannot suppose that many simply hid from view; Egyptian society did not offer that kind of anonymity, and the authorities had lists of who had paid the Jewish tax in the past. Certainly much property was confiscated, and the demographic catastrophe must have been profound (Mélèze Modrzejewski 1991). Tracing even a partial recovery is very difficult, but I do not think that Jews ever again in antiquity represented the proportion of Egypt’s population that they did before Trajan. David Nirenberg (this volume) argues that the rebuilding of the Jewish community in Egypt is in fact largely the product of Arab rule, and the silences of the documents from Byzantine Egypt support that view. The process by which Egypt became a Christian land has been the subject of much controversy, but it is in many ways a process analogous to the way in which Egypt became Roman, in that a combination of policy and local initiative was necessarily in play. How to assess their relative roles is hard to say; certainly Egypt’s Christianization traces its origins and much of its development to a period when imperial policy was anti-Christian, even if not actively so during the ‘Little Peace’ of the church in the second half of the third century. On the other hand, it is hard to imagine that the proliferation of churches in Egypt in the second quarter of the fourth century, built on a limited repertory of plans, owes nothing to central direction, even if that came from ecclesiastical authorities now supported by officialdom rather than persecuted by it. There is, however, a less direct but perhaps more interesting, even unintended, possible point of intersection. This takes us back to the actions of Decius in beginning to forge a common denominator religious basis at the level of the entire empire. What interests me is less a specifically religious aspect in his order of sacrifice than the perception that the empire as a whole needed to become a more unified state. It was not only

EGYPTIAN RELIGIOUS IDENTITIES UNDER ROMAN IMPERIAL RULE

69

religion, after all, that was largely local; indeed, the localism is closely tied to the difficulty of talking about religion as something separate from the rest of life. The spate of recent studies of Diocletian and the Tetrarchy have brought out the tendency of his reforms and of his ideological work to create a Roman nation: to make an empire into a nation, in effect.14 Diocletian’s efforts to make the imperial cult a centrepiece of this nation did not succeed, but the larger effort to make the empire a more unified and uniform state cannot be so readily dismissed. It provided a basis on which Constantine could imagine constructing a unified Christian church, using councils and bishops to create the ecclesiastical counterpart of the civil and military administrations and presiding himself over the bishops as much as over the provincial governors and praetorian prefects (Barnes 1981; 1982). In this sense one may plausibly speak of empire and religion feeding one another’s tendencies to centralization and uniformity. That neither fully succeeded in this project is self-evident, but I would be more inclined to dwell on their relative success in a world used to localism and held together with difficulty, at a time before modern communications and transportation; not least, their success in framing the way we tend to think about religion today as a result of many hundreds of years of the accumulation of power by Rome and Constantinople. Neither time nor competence will permit me to say as much about post-Roman Egypt. That the Arab conquest did not lead to major change in administrative practices until well into the eighth century is well established. Petra Sijpesteijn (2013) has shown that this is not a reflection of a lack of experience and expertise in either writing or administration on the part of the conquerors, who had their own documentary traditions and

show signs of bureaucratic competence. Moreover, she has identified a number of administrative and legal innovations that can be found in Greek and Coptic texts starting from the earliest days of Arab rule. The most striking was the introduction of a money poll-tax, a capitation levy not based on property-holdings, something not seen in Roman practice since the third century. It is striking to observe the tension in Sijpesteijn’s account between the terms ‘Arab’ and ‘Muslim’. Robert Hoyland (2015) has recently argued in favour of privileging ‘Arab’ as a description of the conquerors and their administration, because it is far from clear just how far we can realistically speak of Islam as a common characteristic of the conquerors at this early date. ‘Arab’ itself, to be sure, is not free of difficulties, as the military forces that overthrew Roman rule certainly included allies from parts of the Near East outside the Arabian peninsula who might not best be described as Arabs. And Sijpesteijn has pointed out that the hijra system of dating, which is usually regarded as a hallmark of Islam, is found in documents from shortly after the conquest of Egypt. Indeed, Hoyland’s subtitle speaks of the ‘creation of an Islamic empire’, a process to be seen as unfolding over time. As with the Roman empire, that process implied in some degree an attempt to use religion to unify and define a political entity. It was pursued with less rigour than Decius or Theodosius displayed as they went about their work, but in the long run religious unity fared better than political. With the Roman empire, it was essentially the attempt to change the nature of empire, from an assemblage of states into a more unitary state, that demanded something more than a sprawling patchwork of cults. In that sense, with the third century we may speak of being present at the invention of religion.

Bibliography

Asad, T. 2001. Reading a modern classic: W. C. Smith’s The meaning and end of religion. History of Religions 40: 205–22. Bagnall, R. S. 1982. Religious conversion and onomastic change in early Byzantine Egypt. Bulletin of the American Society of Papyrologists 19: 105–24; reprinted in R. S. Bagnall, Later Roman Egypt: Society, religion, economy and administration (Aldershot 2003, ch. VIII).

All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http:// papyri.info/docs/checklist (last accessed 8 February 2019). Ando, C. 2008. The matter of the gods: Religion and the Roman empire. Berkeley, CA.

14

See particularly Corcoran 2000 and the dissertation of Monica Hellström (2013), applying theories of Charles Tilly (1990) on the building of national states.

70

R. S. BAGNALL

———. 1993. Egypt in Late Antiquity. Princeton, NJ. ———. (ed.). 2007. Egypt in the Byzantine world 300–700. Cambridge. ———. 2009. Early Christian books in Egypt. Princeton, NJ. Barnes, T. D. 1981. Constantine and Eusebius. Cambridge, MA. ———. 1982. The new empire of Diocletian and Constantine. Cambridge, MA. Barton C. and D. Boyarin. 2016. Imagine no religion: How modern abstractions hide ancient realities. New York. Beard, M. 1986. Cicero and divination: The formulation of a Latin discourse. Journal of Roman Studies 76: 33–46. Bowersock, G. W. 1990. Hellenism in Late Antiquity. Ann Arbor, MI. Bowman, A. K. and D. W. Rathbone. 1992. Cities and administration in Roman Egypt. Journal of Roman Studies 82: 107–27. Cameron, A. D. E. 2011. The last pagans of Rome. Oxford. Choat, M. 2006a. Belief and cult in fourth-century papyri. Turnhout. ———. 2006b. The public and private worlds of Theophanes of Hermopolis Magna. Bulletin of the John Rylands Library 2006: 41–75. Cohen, S. J. D. 1999. The beginnings of Jewishness. Berkeley, CA. Corcoran, S. 2000. The empire of the Tetrarchs: Imperial pronouncements and government, AD 284–324. Oxford. Ehrman, B. 2003. Lost Christianities: The battle for Scripture and the faiths we never knew. New York. Eliot, T. S. 2015. The return of Foxy Grandpa. New York Review of Books 8 October 2015: 31. Finley, M. I. 1973. The ancient economy. Berkeley, CA. Frankfurter, D. 1998. Religion in Roman Egypt: Assimilation and resistance. Princeton, NJ. Frerichs, E. S., W. S. Green and J. Neusner. 1987. Judaisms and their messiahs at the turn of the Christian era. Cambridge. Gascou, J. 2008. Les Pachômiens à Aphrodité. In Les archives de Dioscore d’Aphrodité cent ans après leur découverte, J.-L. Fournet (ed.), 275–82. Paris. Habicht, C. 1970. Gottmenschentum und griechische Städte. 2nd ed. Munich. Hellström, M. 2013. Public construction under Diocletian: A study of state involvement in construction in Roman era towns in present day Tunisia and eastern Algeria. PhD diss. Columbia University. Hoyland, R. G. 2015. In God’s path: The Arab conquests and the creation of an Islamic empire. Oxford. Jones, A. H. M. 1964. The later Roman empire: A social, economic, and administrative survey. Oxford. Luijendijk, A. 2008. Greetings in the Lord: Early Christians and the Oxyrhynchus papyri. Cambridge, MA.

Ma, J. 1999. Antiochos III and the cities of western Asia Minor. Oxford. MacCoull, L. S. B. 1988. Dioscorus of Aphrodito: His work and his world. Berkeley, CA. Mélèze Modrzejewski, J. 1991. Les Juifs d’Égypte: De Ramsès II à Hadrien. Paris. Moatti, C. 2016. The notion of res publica in the age of Caracalla. In Citizenship and empire in Europe 200–1900: The Antonine Constitution after 1800 years, C. Ando (ed.), 63–98. Stuttgart. Nongbri, B. 2013. Before religion: A history of a modern concept. New Haven, CT. Papaconstantinou, A. 2001. Le culte des saints en Égypte: Des Byzantins aux Abbassides. L’apport des inscriptions et des papyrus grecs et coptes. Paris. Rebillard, E. and J. Rüpke (eds). 2015. Group identity and religious individuality in Late Antiquity. Washington DC. Rives, J. B. 1995. Religion and authority in Roman Carthage from Augustus to Constantine. Oxford. ———. 1999. The Decree of Decius and the religion of empire. Journal of Roman Studies 89: 135–54. Rüpke, J. 2016. Religious deviance in the Roman world: Superstition or individuality? Cambridge. Scheidel, W., I. Morris and R. P. Saller. 2007. The Cambridge economic history of the Greco-Roman world. Cambridge. Schwartz, S. R. 2001. Imperialism and Jewish society, 200 B.C.E. to 640 C.E. Princeton, NJ. Segal, A. F. 1987. The other Judaisms of Late Antiquity. Atlanta, GA. Sijpesteijn, P. M. 2013. Shaping a Muslim state: The world of a mid-eighth-century Egyptian official. Oxford. Smith, J. Z. 1990. Drudgery divine: On the comparison of early Christianities and the religions of Late Antiquity. Chicago, IL. ———. 2004. Relating religion: Essays in the study of religion. Chicago, IL. Tilly, C. 1990. Coercion, capital and European states. Oxford. Whitmarsh, T. 2001. Greek literature and the Roman empire: The politics of imitation. Oxford. Wipszycka, E. 1974. Remarques sur les lettres privées chrétiennes des IIe–IVe siècles (à propos d’un livre de M. Naldini). Journal of Juristic Papyrology 18: 203–21. ———. 1992. Le nationalisme a-t-il existé dans l’Égypte byzantine? Journal of Juristic Papyrology 22: 83–128; reprinted in E. Wipszycka, Études sur le christianisme dans l’Égypte de l’antiquité tardive (Rome 1996, 9–61). ———. 2015. The Alexandrian church: People and institutions. Warsaw.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST: LATE ANTIQUE CHRISTIAN MATERIAL CULTURE AND THE HISTORY OF ARCHAEOLOGY IN EGYPT Darlene L. BROOKS HEDSTROM

In the nineteenth and early twentieth century, archaeologists working in Egypt were often disappointed when they encountered Late Antique materials. The artefacts looked different from their Egyptian, Hellenistic and Roman precursors, and their strangeness was judged as poorer quality. Mud-brick housing, colourful textiles and Coptic ostraca were not as highly prized as items made of stone, or in the styles of previous eras. The artefacts of Late Antique communities were read as inferior to those of early civilizations and, therefore, scholars’ attitudes towards them were often dismissive. They found that the simplest solution was to remove and discard any eroded, mud-brick settlements that cluttered pharaonic sites. Occasionally, archaeologists extracted the most interesting elements—a sculpture or a textile—for shipment back to museums, but after arrival the objects were often misclassified, divided up or never put on display because they could not be easily categorized to suit existing collections.1 Late Antique objects were often allocated to certain parts of museum collections based upon the presence of recognizable religious iconographic elements, such as a Christian cross or figures from classical mythology. Pharaonic Egypt and its material remains generated far greater interest for scholars, and (ultimately) far greater financial rewards for antiquities dealers serving their wealthy clients. Some archaeologists collected colourful wool textiles, carved lintels and painted pottery, but the findspots for many of the objects were simply unknown.2

In the end, archaeologists were handmaids for their museums, universities and patrons (Jenkins 1995; Stobiecka 2016; Stevenson 2019; Noor et al. 2019). As primarily white scholars, they perceived their actions in Egypt to be part of a larger effort to acquire and protect a past that they considered neglected and endangered. One of the only ways to ‘rescue’ the past from lack of interest or neglect, they believed, was to take objects out of Egypt back to home institutions and museums (Challis 2013; Moser 2015 and 2018). Excavators were pleased to uncover the monuments of ancient Egypt, but, when material from Christian Egypt and later periods was found, they often spoke of ‘clearing the rubble’ to get to the more important finds.3 Late Antique Egypt, built in and around ancient sites, was perplexing and often excised from the material record. The story of how archaeologists and their colleagues reacted to and presented Egypt’s Late Antique material is a complex one. The period known now as Late Antiquity did not exist as a concept for early scholars to examine; rather, they classified times according to assumed religious identity (such as ‘pagan’, ‘Coptic Christian’4 or Muslim), or imperial identity (such as late Roman, Byzantine or, later, ‘Islamic’).5 Less specific terms, such as post-classical or medieval, helped address uncertainty about the date of objects and which communities produced them. Now, we speak of Late Antiquity as a temporal approach that is not tied to a particular religion or imperial entity

1

4

2

3

For an example of the division of a body of material, see Dospěl Williams 2014. As an illustration, the Rijksmuseum recently published its Late Antique textiles from Egypt for the first time: see Janssen 2013. In Naville’s description of the Christian material at Deir elBahri, we find this negative reading of the reuse of abandoned, pharaonic monuments by (in this case) a monastic community, which ‘took material for building the wretched walls which divided their convent into different rooms’ (Naville 1894, 6).

5

‘Coptic’ was often used as a descriptive label to mean Egyptian Christian. Discussion of the term as used in earlier scholarship is found in Curating collections: Dalton at the British Museum below. Islamic dynasties such as the Umayyads or Abbasids, were not usually distinguished, and thus ‘Islamic’ covered all material from after the Arab conquest, unless an object had clearly recognized motifs of other religious communities.

72

D. L. BROOKS HEDSTROM

Map. General map of Late Antique Egypt with sites mentioned in the chapter (Map by D. L. Brooks Hedstrom).

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

(P. Brown 1978 and 1982; Clark 2015; Cameron 2016). Scholars, however, still debate the beginning and end dates of Late Antiquity and whether this periodization captures the many changes that were taking place between the third and eighth centuries AD (E. James 2008; Marcone 2008; Whittow 2009; Kadellis 2015; Rousseau 2015). The problem is compounded when we acknowledge that the artefacts from Christian Egypt lack reliable archaeological contexts and, even when dated, are generally assigned a three- to four-hundredyear span. This chapter explores the intersection of religion, empire and artefacts in the Late Antique world. I will also explore how scholars since the nineteenth century have approached the Christian artefacts of Late Antique Egypt. In order to illustrate the ways in which inherited knowledge about Late Antique objects was constructed by scholars, and how we can use new approaches to better understand what objects are and how they reflect past identities, I will look at three examples: objects; burials; and landscape. In exploring these topics as case studies, I propose two central questions to survey how material culture was once viewed, and how it is viewed now by archaeologists wielding an array of new methodological theories. First, how did interpretations of past religious identity and modern concepts of empire influence the development of a history of Christianity in Egypt? This question helps us examine the frameworks that overlaid how scholars viewed religious identity and to what degree the structures impinged upon their reading, classification and interpretation of artefacts. I sketch the story of a few early scholars—not always archaeologists—to demonstrate how they used religious objects to try to reconstruct Christian Egypt. In surveying these thinkers, I will highlight how they were rooted within their own nationalist and imperial structures, which then shaped their actions and interpretations of Egypt’s Christian material culture (Meskell 2005; Stobiecka 2016; Stevenson 2019). The second question is centred on the practical application of new methodologies and analytical techniques related to new materialism as a methodology for analysing objects, both small and monumental (Domanska 2006; Webmoor and Witmore 2008; Olsen 2010; Nativ 2014). How can the adoption of theoretical hermeneutics in archaeology offer revisions and new insights to enhance our interpretation of Late Antique Egypt and the religious identities of Christian objects? I will use illustrations from the material turn (Bräunlein

73

2016; Versluys 2017), the archaeology of death and body theory (Parker Pearson 1999; Borić and Robb 2008) and landscape archaeology (Ashmore 2002; Brooks Hedstrom 2017) to demonstrate the richness of material cultural studies in archaeology and how this approach allows us to revise earlier encounters with Christian Egypt (Hicks 2010). With poststructuralism and postprocessual archaeology, scholars have asked what forces or factors shaped earlier narratives of religious empires and, by extension, what conclusions were then presented in museum collections to educate the general public about the sequence of empires in Egypt (Moser 2006). In order to understand how archaeology facilitated modern definitions of civilizations, I will illustrate the application of archaeological theories to revise the ways in which objects, burials and landscape can be studied to examine identity. The chapter begins with a brief survey of the intellectual, cultural and social background of late eighteenth- and nineteenth-century interest in Egypt and its material remains, especially those produced by Christian communities. I then outline the challenges in ascertaining religious identity in Late Antique materials and how identity may or may not be attributed to things (Barsch and van Norden 2020). To illustrate the two themes of early approaches to Christian materials and the efficacy of new theories, I offer in-depth discussions on three case studies of Late Antique material: 1) objects and the material turn; 2) cemeteries and the archaeology of the dead; and 3) Late Antique landscapes and the archaeology of space. All three cases begin with a consideration of the work of early scholars in each field, followed by a demonstration of more recent archaeological theories that allow us to reread material, in order to illustrate what archaeology can and cannot tell us about Christian identity. Archaeologists today consider the role of makers as individuals, and also the role of the living in constructing burials; see objects as artefacts with many meanings; and use caution in ascribing religious identity to material remains. Ian Hodder and his fellow postprocessual archaeologists urge a ‘turn to things’ in which one ‘explore[s] how the objectness of things contributes to the ways things assemble us’ and not vice versa (Hodder 2012, 14). This reversal of focus underlies the following discussion in illustrating how the field has moved from a view that material culture ‘reveals’ cultural identity, to an attempt to trace how things, places and people interact and create personhood in Late Antique Egypt (Webmoor 2007; Fowler 2010, 376–78).

74

D. L. BROOKS HEDSTROM

Modern empires, archaeology and Late Antique Egypt The world of ancient Egypt, in particular, became a fetish of the nineteenth and early twentieth centuries. From climbing the pyramids to collecting Egyptian things, the fascination reflected how Americans and Europeans, in general, created what Scott Trafton has called the ‘overlapping iconographies of Orientalism and Egyptomania’ in order to sell products, such as music, cigarettes and furniture; to create a sense of ownership of the Egyptian past by building mausoleums, churches, and even prisons in styles akin to Egyptian temples; and to create a racialization that diminished the presence of modern Egyptians (or removed them entirely) from stewardship of the past (Trafton 2004, 175) (Fig. 1). The Egyptian Revival movement, as an illustration, became important in the decorative arts from the mid-nineteenth century until the 1940s, with its second period coinciding with the Art Deco movement (Curl 2005). David Gange traces the fascinating ways in which Egyptological discoveries and the archaeologists behind them, coupled with the visual pleasure of Egyptian styles, helped fuel growing secularism and new forms of spiritualism that ran counter to traditional Christian theology (Gange 2006; 2013). Archaeology, therefore, functioned as a popular catalyst bringing scholars, collectors and tourists together in a way that popularized the study of material culture and blurred the lines between scholarship and popular history (Matic 2018).6 Initially, the archaeology of Late Antique Egypt was overshadowed by greater interest in the world of ancient temples, tombs and mummies. There was less curiosity, comparatively, about the archaeology of Egypt’s other religious traditions and their material remains: Judaism, Christianity and Islam.7

6

7

For an engaging discussion of how nineteenth- and twentiethcentury archaeologists altered their publications and produced material that would cater to their financial supporters’ preferences and extend interest in archaeology to a reading public, see Thornton 2018. The history of archaeology, Egyptology, and museum collecting is complex and shaped by: legal developments with the advent of antiquities laws; the complex network of patrons and excavators; the evolution of scientific knowledge, with developments in anthropology and eugenics; and an increasing interest in critical study of the past. The early study of the Islamic archaeology of Egypt shared a similar trajectory (see Reid 2015 and Goode

The collections of foreign government officials, missionaries and other travellers and collectors, both men and women, fostered interest in Egyptian artefacts as things to own and display (Fagan 2004; Donker van Heel 2012; Evans 2013; Kalloniatis 2013; Hagen and Ryholt 2016). Dynamics of power and ownership of the past shifted dramatically during the nineteenth century as Egypt and the world experienced radical changes with new approaches to knowledge, science and concepts of empire (Wood 1998; Reid 2002; Goode 2007; Reid 2015; Zena 2018).8 Questioning how objects of antiquity related to the developing narratives of nation states was an important topic (Wood 1998; Zena 2018). Alongside these forces, Western empires drew upon a desire to be heirs to the past and early civilizations, as they simultaneously helped reinforce their own, new, narratives of identity (Barringer and Flynn 1997; Díaz-Andreu García 2007, 99–105; Picchi 2015; Raven and Insinger 2018). The prominence of Christianity in the West also strengthened a desire to be closer to the biblical past and its material remains (Cormack and Jeffreys 2000). Such ideas are keenly evident in Austin Layard’s discovery of the remains of Nineveh, where the physical evidence of the Old Testament world helped bolster interest in the British Museum. The Museum held the monumental artefacts from the 1848 excavations and published lavishly illustrated books on the archaeological discoveries (Malley 1996; Holloway 2006; Malley 2008). Antiquarian collecting and historical study would give way to the first archaeological excavations, with their new emphasis on artefacts, typologies, stratigraphy and mapping (Thompson 1992; 2015a; 2015b; 2018). ‘Seeing archaeologically’ while touring Egypt and visiting sites was a significant form of history tourism that flourished in the second half of the nineteenth century (Willis 2016, 115–42). The consumption of

8

2007). Similar challenges existed in other regions, where contemporary empires devalued religious artefacts or monuments as they challenged modern notions of empire. The bibliography examining the primary relationships between Britain, France and Egypt during the long nineteenth century is extensive. For the development and history of the national museums in Egypt, see Reid 2002 and 2015; for the exploitative nature of antiquities dealers, see Fagan 2004 and Greenhalgh 2019; and for the development of Egyptology and its association with collecting and excavation see T. G. H. James 1982; Jenkins 1995; Stevenson 2014 and 2019; and Thompson 2015a, 2015b and 2018.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Fig. 1: ‘Ascension d’une pyramide’ (Climbing a pyramid) (General Research Division, The New York Public Library 1870–1875).

75

76

D. L. BROOKS HEDSTROM

Egypt’s history and its material remains was further complicated by how foreign visitors and scholars interacted with the people of Egypt. The racialization and othering of Egyptians and Turks emerged against the backdrop of the demise of the Ottoman empire in Egypt and the rise of Muhammad Ali and his successors (Reid 2002, 54–58). As these major changes altered existing networks of purchase, sale and gifting of artefacts, we can sometimes lose track of the complex paths that artefacts followed through and out of Egypt (Colla 2008, 11–14).9 The practice of visiting Egypt and taking trophies home is well attested in nineteenth-century travelogues, but many antiquities were purchased legally and later bequeathed to national museums by private donors or their heirs (Jørgensen 2015; Stevenson 2019, 259–60). Some individuals made it their business to travel to Egypt to acquire artefacts and manuscripts from dealers, both foreign and Egyptian, intending either to sell their treasures to museums or to enrich their own collections. Prior to the early twentieth century, Christian objects were often separated from their original, archaeological contexts.10 While this fate was not necessarily unique to Christian objects, and many materials lacked a reliable provenance, it does illuminate why some early scholars could not conceptualize a multi-religious landscape of Late Antique Egypt. The establishment of the Egypt Exploration Fund (EEF) in 1883 unequivocally linked Egypt to its biblical past ‘with a view to the elucidation of the History and Arts of Ancient Egypt and the illustration of the Old Testament narrative … also to explore sites connected with Greek history or with the antiquities of the Coptic Church’ (Egypt Exploration Fund 1895, 17; T. G. H. James 1982; Spencer 2007). A clear agenda emerged: a search for religious identity and for linkages to a sense of Christian and Greek history. The mission of the EEF was

9

10

11

For an examination of the founding of the Coptic Museum in Cairo and the role of Egyptian Copts in its creation, see Henein and Simaika 2017. See Kozelsky 2008 for the archaeology of Christian churches in the Crimea and the role of the Orthodox Church as a point of comparison to Egypt. Early excavation work with Christian remains will be discussed in Case study 1: Late Antique objects and the material turn, below. Numerous studies now outline the complex history of colonial archaeology in service to imperialism and conquest, such as in

similar to that of the Palestinian Exploration Fund, but the EEF had a more explicit interest in Christian artefacts and history (Drower 1982; Bárta 2010, 100). Museums, universities and private organizations, such as the Byzantine Research Fund and the Byzantine Institute of America, provided scholars with the necessary resources to undertake excavations, pursue collections and publish definitive records of monuments and artefacts, which in turn helped to establish accepted classifications and reception of material culture (Kakissis 2009; Stevenson 2014).11 As objects moved from a variety of owners and dealers into the hands of museums, it is understandable that early interpretors missed the crucial context to situate Christianity not as a hegemonic force, but as one religion within a diverse setting. Difficulties in ascertaining the religious identity of Late Antique materials The Late Antique population of Egypt did not consider themselves Late Antique persons as we might conceive of them now (Riggs and Baines 2012). We might ask what identity or identities they did hold. Did they consider themselves Jews, Christians, Manichaeans, pagans, Egyptians, Romans or Greeks, or several of these at once (BeDuhn 2015)? Or was their identity defined by their spoken or written language? How did people living in Egypt conceive of their relationship to Rome and later to Constantinople or to empire more broadly (Palme 2012)? Such questions point to the complexity of classifying Egypt from the third to eighth centuries AD, a time when dating is difficult to pinpoint even with the objects we do have. When we look at material remains, we are disadvantaged because we often do not know who made the objects, who owned them originally, or if they were

the Ottoman and Greek empires. The studies highlight the role of archaeology, foreign and indigenous archaeologists within each country in helping to use the past to construct new ideas of empire for the present: see Tanyeri-Erdemir 2006; Kourelis 2007; Hamilakis 2011; Bahrani, Çelik and Eldem 2011; Çelik 2016; Greenhalgh 2019. The introduction of antiquities laws in various nations also helped facilitate the role of archaeology in defining new conceptions of empire status along religious and ethnic lines: see Majd 2003; Shaw 2003; Voudouri 2010; Corbett 2014, 1–19.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

reused and repurposed. We can no longer embrace the concept that mute stones speak, and instead need to embrace a humility of knowledge when we begin to read objects. Richard Alston offered such an illustration of the limits of what we can know and ask of material when discussing houses at the Late Antique Christian pilgrimage centre at Abu Mina. He addresses the categories of identity and representation that we cannot ascertain, such as: the ethnic identity of the inhabitants, the population of the house, its temporal usage, whether the pattern of occupation was gendered, the interior social relationships of the house (male-dominated/slave-master household, patriarchal family, extended family, nuclear family, diverse household, multi-family occupation), the religion of the occupants, its ownership status, its emotional hold over the occupants, its symbolic value as representation of its inhabitants, its function in social and economic production (Alston 2007, 373–74).

Alston’s list is not intended to discourage study or limit the possibilities of what we might ask about material culture as it relates to religious identity, but rather to highlight the underlying assumptions that were made continuously in the past by earlier scholars in the field. The ascription of meaning and identity stemmed from the person or community to objects but did not begin with objects. While textual sources may appear to offer fewer complications, determining how individual writers and readers conceived of their own identity and the world they inhabited can be equally challenging. Arietta Papaconstantinou explains that Late Antique ‘administrators, religious leaders, family groups, urban elites [and] monastic and ecclesiastical authors’ employed multi-lingualism ‘for the construction of social and religious models and in the negotiation of socio-economic or political status’ (Papaconstantinou 2010, 7–8). Her nuanced reading of the strategic use of demotic, Greek and Coptic, for example, points to the complex networks shaped by the nature of Egypt’s Late Antique society. An individual may activate different identities depending upon circumstances. Just as a person’s choice of language may reflect specific networks of connectivity, Late Antique objects reflect networks of the material and visual cultures of individuals and their communities. Once we take stock of the inherited knowledge of the past through the written reports of archaeologists, their excavations, objects and texts at our disposal, we

77

can then consider the value of adopting new theoretical approaches to revisit and revise earlier interpretations. Acknowledging the limitations of earlier archaeological work, which preceded the advent of significant changes in the field in the twentieth century, allows us to draw upon the insights that were not possible in previous generations (Brooks Hedstrom 2017, 60–62; Doyon 2018). For the purposes of this discussion, I will focus on the current approach to materials and objects as a corrective to postprocessual archaeology, and on the recent impetus to move the field beyond the desire for data and hyper-taxonomies of things. Reading Christian objects Humans are easily considered agents of change and this is where archaeology began: humans make things. They build structures, they craft objects, they paint, they write (after a certain point), and they create sites— all activities that attract the attention of archaeologists. The ‘material turn’ and ‘thing theory’ grant agency to objects themselves as materials that can act upon humans and shape their activities and even religious identity (Gell 1998; Brown 2001 and 2015; Latour 2005, 16). The material turn is a current effort by humanists and social scientists to dismiss the notion that objects are merely things representative of humans and their actions (Malafouris 2004; Insoll 2009). For archaeologists, an approach that places things, rather than humans, at the centre is a welcome change enabling us to appreciate how material things affected communities and their identities (Latour 2005; Bynum 2011; Olsen et al. 2012; Morgan 2020). Recent studies illustrating the effective use of materiality and the non-human turn include: identifying Persian imperial things (Khatchadourian 2016); assessing Romanness (Van Oyen and Pitts 2015); and looking at communities of makers (Langlands 2017; Wendrich 2012). All are illustrations of the effectiveness of an objectcentred approach. Miguel John Versluys calls for archaeologists to adopt an ‘object-scape’ perspective in which objects are the main characters (Versluys 2017, 197). He recommends a move away from the traditional approach that considers things as reflections of daily life, for just as in the case of religious identity, objects do not necessarily tell us how they were used. In addition, Versluys goes beyond the postcolonial readings of objects, for in many ways these readings are more anticolonial

78

D. L. BROOKS HEDSTROM

and more concerned with the humans around the objects than the objects themselves.12 In the case of Christian Egypt, the material turn asserts that objects do not convey religious identity as if they are labels that one can simply read. Objects can be interrogated with new ontological and phenomenological questions, which require us to ask what objects did in the past and how they expressed multiple meanings and functions (Harman 2002).13 The shift in focus is an important one, for it reveals current modes of reading materials that situate things, and not people, at the centre (Prown 1982; Yonan 2011). The material turn, therefore, regards objects, small and large, as actors in their own right. There is a sense of instability between what objects mean, their function, their use, and how they are later interpreted (in the present). In many ways the material turn considers how objects shape human behaviour as ‘structuring structures’ along the line of Bourdieu (Bräunlein 2016). The idea is that objects, or anthropogenic things such as buildings and settlements, influence humans and that this inversion of focus moves us firmly away from reading artefacts as labels or mirrors of identity. The history and historiography of how Christian objects were interpreted and classified in the past reflects the traditional reading of objects as culturalhistorical markers. They were not seen as materials that could be appreciated as entities in and of themselves, but rather as by-products of people and, therefore, representatives of identity. Fabrics bearing Christian images on decorative panels were identified as the clothes of a Christian; but panels with images that were too mysterious or not clearly Christian were assigned to the belongings of a ‘Gnostic’ believer.14 Similarly trouble in determining date, style and religious identity surfaced with objects from the end of Late Antiquity

that might have been owned by a Muslim, a Jew or a Christian (Ball 2019). Other examples of complexity surround non-canonical texts that were assumed to belong to the library of a Gnostic and not a Christian. And even landscapes, like objects, could be labelled without substantial evidence. For example, the ruins of a church would be marked as signs of a monastic site, without any further investigation. Or a reused quarry cave with Coptic graffiti mentioning Christ was deemed the residence of a hermit monk, without considering whether Christian soldiers or traders used the space as shelter during their travels. In all of these cases, the object, whether small or monumental, was read in biographical terms, meaning it had a single, humanfocused, story to tell about the past. An object-centred approach might reveal other stories. Building a taxonomy of objects was an essential part of the work done by earlier archaeologists, art historians and museum curators, who looked for marks of religious identity and the shifting nature of imperial power. The cultural-historical reading of portable objects or landscapes as Christian things enabled scholars to write histories of Egypt as part of the Roman and Byzantine empire before the arrival of the Arabs in the seventh century. But what happened to the Christian objects, buildings and sites with the arrival of new people, new ideas and new things? The following three case studies examine the past and the present as they relate to Christian objects from Late Antique Egypt. First, I will examine how interest in objects of Christian Egypt developed among collectors and museum curators, before considering how materiality studies illuminate what we can and cannot know about identity and objects. Second, I turn to the role of cemeteries, the dead and the materials within graves to review how early excavators and their

12

13

A recent example of this type of object-scape is Ellen Swift’s exploration of objects in Roman artefacts and society: Design, behaviour and experience (Swift 2017). Her starting point is with things, such as dice, rings and reed pens, which appear in museum collections, but are rarely studied for their life histories. Similar work is being done for artefacts at Pompeii, putting things back into rooms and houses to ascertain what the objects, both big and small, can reveal. In many cases the objects contradict the normative Roman accounts of domestic space and thus the objects take on an important role in creating an objectcentred space (Allison 2004).

14

Object-oriented ontology, or OOO, is sometimes referred to as ‘speculative realism’. For critiques of this approach, see Cole 2013 and Boysen 2018. Tim Ingold presents a medium pathway through the tensions between objects, agency, and materials in landscape: see Ingold 2007. See Interpreting material culture: Crum on Strzygowski and Gayet below for discussion of earlier usage of the label ‘Gnostic’ for Late Antique material.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

publications identified what was, or was not, considered a reflection of Christian identity. By adopting reading strategies from the archaeology of death and burial, I demonstrate the limitations of what may be deduced from graves about the religious identity of the dead and those who designed the burial. Finally, the third case study takes up the topic of Egypt’s Christian landscape, from the first surveys and excavations to current interpretations of defacement of ‘pagan’ monuments and monastic settlements. By using theories from landscape archaeology and the spatial turn, I will demonstrate the effective use of new materialism to put things back into the landscape—the recluttering of the landscape (Brooks Hedstrom 2017 and forthcoming a). Case study 1: Late Antique objects and the material turn The first case study considers the discovery, collection and identification of Christian things from Late Antique Egypt. Early scholars of Christian Egypt struggled to understand the religious complexities of Late Antiquity, and their view could not accommodate the diverse identities that helped define the period of the third to eighth centuries AD. Objects without clear Christian motifs could not be easily classified and finds from Egypt were not regarded as reflecting the same quality of craftsmanship as Christian material from other regions of the Mediterranean. The material turn, which illuminates the quest to locate identities of objects and the humans who are associated with them, demonstrates the benefit of new approaches to Late Antique objects. Christian artefacts: Manuscripts European and American travellers, some with academic credentials, have left numerous stories, narratives and letters describing their visits to Egypt and their acquisition of manuscripts to contribute to the history of the New Testament and early Christianity (O’Connell 2014a). Their work was not archaeology, of course,

79

but collecting. In the end we are left with what Brent Nongbri calls an ‘imaginary archaeology’ (or, less charitably, a ‘fake archaeology’) of Christian manuscripts (Nongbri 2018, 14).15 The volumes made their way out of Egypt and into various locations, permanently separated from their archaeological contexts or the monastic libraries where they were originally found. The initial interest in Christian Egypt was tied, almost from the beginning, to the collection of manuscripts, which contained the earliest copies of the New Testament. There was little interest in the copyists [or writers] themselves. And there certainly was little understanding of the religious diversity of Jewish, Egyptian and Christian communities in the late Roman world. The central goal of both collecting and excavation was the hunt for biblical manuscripts. For some, this objective was motivated by the impetus to shore up the historicity and antiquity of the New Testament, whereas for others it was a desire to acquire new objects that would create legacies for the collector or museum. Even when biblical scholars sought artefacts of Egypt’s Christian history specifically for academic purposes, the perception that the objects were neglected by modern Egyptians fuelled the paternalism of the curatorial vein of collecting and removing material from Egypt (Muhs and Vorderstrasse 2008). For example, English dealer and biblical scholar Rev. Henry Tattam (1788–1868), who travelled to Egypt explicitly to acquire Christian manuscripts, encouraged Lord Algernon Prudhoe (1792–1865) to go to Egyptian monasteries in search of additional documents in 1828 (Littell’s Living Age 1846). While in Wadi Natrun, home to four famous Coptic monasteries, Prudhoe found numerous manuscripts to take back to England. The Christian Register explained to its readers the importance of obtaining the Coptic, Ethiopic, Syriac and Arabic books: ‘The monks were wholly ignorant of their value and had no critical knowledge of the languages in which most of them were written, so that it cannot be regretted that they have been purchased and transferred to places where they may be consulted and studied’ (Christian Register 1846).

15

The PAThs project https://atlas.paths-erc.eu/ is a multi-disciplinary and collaborative effort by scholars to bring together palaeography, codicology, provenance studies and archaeology.

80

D. L. BROOKS HEDSTROM

The imperative to rescue, protect and study manuscripts from Egypt lay behind the recovery of Codex Sinaiticus, the earliest, most complete Greek Bible, which dates to the fourth century. The German biblical scholar Constantine Tischendorf (1815–1874) identified the Codex Sinaiticus at the Monastery of St Catherine’s in the Sinai in 1859. Tischendorf negotiated for a loan of the codex, which he then transferred to St Petersburg as a gift to Tsar Alexander II, as the tsar was recognized by the Ottomans as the Protector of the Orthodox Christian communities.16 Then, by 1933, the Russians sold the codex to the British and today it resides in the British Library (Parker 2010). The story of Codex Sinaiticus reflects how objects became important artefacts of national and communal identities outside of Egypt, while simultaneously being heralded as reflections of Christianity’s dominance in the region. Despite the great desire for manuscripts, the hunt for biblical texts left much of the history and material of Egypt’s Christian culture in the shadows (Hurtado 2006, 26; Nongbri 2018, 214).17 With exciting new discoveries being made at Oxyrhynchos in 1903, there was little time to consider whether or not one could determine who exactly read the texts being found, what doctrinal position a community held, and what relationship Christians had with other religious communities in Egypt (van Minnen 2010) (Fig. 2).

Ormonde M. Dalton (1866–1945), Assistant Keeper for the Department of British and Medieval Antiquities and Ethnography at the British Museum, was interested in Christian Egypt (Fig. 3). He wanted the museum to increase its holdings of materials that would not be found in either the classical or the Egyptian collections (Stevenson 2014). He faced opposition from his colleagues, especially from E. A. Wallis Budge, and was in competition with other museums seeking new material for their museums (Brooks Hedstrom 2017, 33–35;

O’Connell 2014b, 125–26). Despite these obstacles, his persistence was rewarded and he gained the support of the Byzantine Research Fund and the British Museum, sending R. Campbell Thompson to excavate at the monastic settlement of Wadi Sarga in Middle Egypt in 1913/14 (O’Connell 2014b). Even before the arrival of the new finds at the British Museum, Dalton was paving a new path for his vision for Byzantine objects with his publications. In 1901 Dalton published the British Museum’s first catalogue of Byzantine material associated with the Eastern Mediterranean (Dalton 1901; Entwistle 2000). His catalogue provided an interesting assessment of the cultural and intellectual landscape of the early twentieth century and the difficulties he and others had in finding a place for Late Antique and Byzantine material culture in their museums. In assembling the catalogue, he acknowledged that he needed to decide ‘whether all Byzantine antiquities were to be admitted, or only those which bore some obvious mark of Christian ownership’ (Dalton 1901, xvi–xvii). Like the Bulaq Museum in Cairo, the British Museum placed Late Antique remains in different departments, such as Egyptian and Assyrian Antiquities or British and Medieval Antiquities, thereby fragmenting a viewer’s perception of what was happening in Late Antique Egypt. Objects without evident Christian iconography could be found in the Department of Greek and Roman Antiquities, where the assumption was that the objects were linked to older religious traditions classified simply as ‘pagan’ (Dalton 1901, xvi–xvii). Dalton admitted that there were limits to what his colleagues in Greek and Roman Antiquities would accept—if an object was dated after the establishment of Constantinople, they considered it too late for their collection (Dalton 1901, xvi–xvii). In 1911 he wrote his seminal work on Byzantine art and archaeology, in which he told a story of the Egypt that existed outside of Alexandria. Despite his evident commitment to elevate imperial Byzantine art and the

16

17

Curating collections: Dalton at the British Museum

The Codex Sinaiticus Project is a joint initiative between the British Museum, the National Library of Russia, Leipzig University Library and St Catherine’s Monastery to create digital access to the Codex Sinaiticus and promote study of the manuscript: https://codexsinaiticus.org/en/. See ‘Modern histories of Codex Sinaiticus’ in McKendrick et al. 2015.

The site of Oxyrhynchos, whose trash mounds became a treasure trove of early Christian (along with other Greek, Latin and Jewish) texts, provides an excellent example of the type of laser focus scholars could have in collecting for their institutions. The directors of the recovery project were Bernard Grenfell (1869– 1926) and Arthur Hunt (1871–1934). See Luijendijk 2010.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

81

Fig. 2: Egyptian men and boys gathering fragments of papyri in baskets at the excavation at Oxyrhynchos in 1903 (Courtesy of Egypt Exploration Society).

material culture of Constantinople as a topic of scholarly appreciation, he was less open to regard Egypt’s Christian materials as reflections of the empire of Byzantium; rather, he told his readers how Egypt was a land in ‘silent revolt against the invading [Greek] culture’ and far removed from a ‘great intellectual or artistic centre’ (Dalton 1911, 71). Dalton’s difficulties in curating Egyptian material revealed the scholarly uncertainties about who made or owned Late Antique objects. He explained that finds without ‘outward evidence of Christianity … introduced an unavoidable complication’ (Dalton 1901, xviii). He also avoided using ‘Byzantine’ as a descriptor, for the word ‘implies too close an association with the metropolis of the Empire’; thus suggesting to his readers that Egyptian material could not be a mirror to Constantinople or to a Christian empire (Dalton 1901, xix–xx). Dalton wanted to see and display Christian Egypt, but he did classify Christian Egyptian art and architecture of the third through seventh centuries as the worst expression of ‘sensuous oriental taste’ reflecting a ‘naïve realism’ fighting ‘against the finer sentiment of the Greek’ (Dalton 1911, 152). For Dalton the appellation ‘Coptic’, as a descriptor, allowed him to separate

out what he considered truly ‘Byzantine’ (that is, reflecting the refinement of Greek Christianity) from Egyptian expressions of Christian identity. ‘Coptic’ would also be adopted by later archaeologists and museums in classifying and ordering material that reflected neither Roman nor ‘pagan’ Egyptian identities. It would also be used to identify Christian Egyptian material, as indicated by obvious Christian motifs. Interpreting material culture: Crum on Strzygowski and Gayet At the time that Dalton was beginning to build the Byzantine collection and the Coptic collection, the emerging field of Coptic studies was linked more to texts and the language than to the study of the material remains of Christian Egypt. Because language, textual analysis and philology shaped early patristic studies of the Eastern Mediterranean world, the investigation of Christian Egypt required linguists with familiarity in Coptic, Greek, Arabic and Syriac. Walter E. Crum (1865–1944) became Britain’s foremost Coptologist, teaching Coptic at the University College London between 1893 and 1910, and publishing an annual

82

D. L. BROOKS HEDSTROM

Fig. 3: Ormonde Maddock Dalton (1866–1945) (© National Portrait Gallery).

Fig. 4: Walter E. Crum (1865–1944) (© National Portrait Gallery).

account of scholarship associated with ‘Coptic’ studies in the newly established Archaeological report for the EEF in 1893 (Griffith 1892–93, vii; Cromwell 2014) (Fig. 4). Crum’s initial reports were entitled ‘Coptic studies’ until 1900, when the chapter title was changed to ‘Christian Egypt’, which indicated clearly that Egypt’s history was now divided by periods of dominant religious affiliation. For the first four years, Crum’s summaries did not contain any reference to archaeological work; rather, the focus of his contribution was first and foremost on literature, philology and history. His review of the literature included the categories closest to his own expertise: Biblical; Apocryphal, Gnostic; Liturgical; Church Literature; History; Non-literary texts; Inscriptions; and Philological. At the end of each annual summary, he included ‘Art, archaeology, excavations’ and when necessary ‘Miscellaneous’. The placement of art and archaeology at the end of the exhaustive treatments of literature in a myriad of categories reveals the lesser value placed on material culture for the field. Despite treating

it as a minor topic, Crum did provide commentary on new archaeological discoveries beginning in 1902. In addressing the field of art history, Crum drew attention to the work of Josef Strzygowski, who offered new arguments for Egypt’s greater connections within a wider Mediterranean and West Asian world (Elsner 2002). His 1901 Orient oder Rom? posed the question of whether Christian art really was linked to Rome or to roots found further to the East (Crum 1901, 233; Zäh 2012). Strzygowski’s reputation earned him the opportunity to create a catalogue of the Christian artefacts in the Cairo museum (Koptische Kunst—Musée du Caire) in 1904, but his overall thesis of style affinities with the East was met with strong opposition (Marchand 1994). Specifically, Crum highlighted Strzygowski’s claim that ‘Coptic’ art had a ‘primitive clumsiness’ and that ultimately ‘Coptic art [wa]s merely “depraved hellenism”’ (Crum 1903–1904, 84). In addition to his low opinion of Christian Egyptian art, Strzygowski also adopted ‘Gnostic’ as a descriptor for any object that could not be easily classified.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Crum faithfully reported on the wider response to Strzygowski’s ideas, but he saved his most critical assessment for Albert Gayet (1856–1916), the French archaeologist whose work shaped early perceptions of Late Antique Egypt. Crum was utterly unconvinced by Gayet’s work and considered it highly flawed. Gayet published interpretations of the material he excavated from the Christian site of Antinoopolis (Gayet 1902a; 1902b; 1903a). Crum drew readers’ attention to how Gayet’s classified late antique Egyptian material as ‘ungainly Coptic productions’ and that the artefacts of the Christian period were ‘manifestations of the native spirit, significant even under Christian disguises’ (Crum 1901–1902, 55). Crum was disappointed by Gayet’s loose treatment of ‘chronological sequence and distinctive characteristics of objects’ for ‘it is impossible to grasp [Gayet’s] grounds for assigning his various finds to the different periods’ (Crum 1900– 1901, 77). An illustration of Gayet’s methods of excavation, published in 1902, shows a series of bodies, stripped of their textiles and burial garments, piled in a haphazard manner (Fig. 5). No wonder Crum and others put so little stock in Gayet’s early interpretations.18

83

The first step in adopting an object-centred reading is to ask what it means to identify a thing as Christian, or indeed any other religious identity (Morgan 2020; Meyer and Houtman 2012; Plate 2014). For early twentieth-century scholars, identifying objects as Christian was a simple matter of looking for Christian symbols or texts. When objects lacked these elements, they were placed in the ambiguous category of ‘Gnostic’ as a way to explain the missing Christian components. Now, a series of questions are utilized to frame how we know what we know about Christian things and highlight why Dalton and Crum were not able to. Is an object Christian because of motifs it bears, such as a chi-rho, or a haloed individual or an inscribed biblical

passage? Is the application of these signs of Christian belief and religious identity the act of Christianizing an otherwise generic item, whether it be a ring or a column or a pair of shoes? How does the act of marking Christian elements transform an object? Does it make the physical thing act in new ways that it could not without such signs of identity?19 Essentially, we must ask whether objects are imbued with religious identity and upon what basis scholars in the past ascertain the Christianness of any object (Rebillard 2015). Equally important to the identity of an object is the relationship between a Christian object and the humans who interact with it. A maker (or people in a workshop) crafts things. Sometimes artisans and craftsmen work in communities and those communities of practice adopt methods and techniques of work that reflect the makers’ identity (Wendrich 2012 and 2013). Does the object need or require to be made by a person who identifies as Christian? For example, a lamp that bears an inverted chi-rho reveals that the craftsman may not understand how to read the anagram; does this alter the Christian identity of the object (Fig. 6)? Or is the owner or user of the thing the one who creates the identity of an object as Christian? Is it the function of the object as a reflection of imagined communities? Many objects outlive their makers and users. If objects have generational lives of use, can we speak of the many identities of an object? Buildings, for example, have extended lives whose periods of use and occupation are problems for archaeologists to puzzle out. Objects may be gifts that do not reflect the religion of the receiver, or they may be owned by a community or family for several generations. All these elements shape the lives and identities of an object. Two illustrations will help demonstrate how to think in material ways about the identity of the object, the maker and the users. The site of Philae is an example of a big object. The island was home to a thriving Ptolemaic and Roman cult to Isis. The importance of the place for religious activity was continued by a Christian community; two new churches were built in the nearby

18

19

New approaches to religious objects and Christian object-scapes

I describe Gayet’s work in greater detail on cemeteries as part of Case Study 2 below.

Similar questions are asked by Smith (2018) in his exploration of Jewish and Christian material things as ways to map communities.

84

D. L. BROOKS HEDSTROM

Fig. 5: ‘Les corps après le depouillement’, Albert Gayet’s illustration of bodies stripped of textiles during his excavations (Gayet 1902a, 19).

settlement and three churches were placed within the abandoned temples (Dijkstra 2008, 323–24). Built around these religious centres was a small town constructed in mud brick, and epigraphic evidence demonstrates the older local community and the Christian community lived side-by-side in Late Antiquity (Lyons 1896, Plan 1; Dijkstra 2008, 125–29) (Fig. 7). Signs of defacement on some of the Ptolemaic temple walls at Philae, discussed below, were once viewed as the hallmark of a religious conquest or conversion of the site, and reflective of the broader conversion of all of Egypt. But archaeological and epigraphic evidence from the site reveals that the island cannot be neatly divided into ‘pagan’ and ‘Christian’ periods during Late Antiquity, but rather is a site where boundaries between religions are blurred (Dijkstra 2015). This is most evident when we look at the mud-brick housing around the Egyptian temple buildings and the later churches (Fig. 8). In earlier generations the site of Philae might be identified as a Gnostic town; now we regard Philae as highlighting the complexity of religious plurality where Egyptians embraced Christianity and traditional Egyptian beliefs within the same community.

Fig. 6: Moulded clay lamp from Egypt with chi-rho in reverse, fifth–sixth century (17.194.1818, Courtesy of the Metropolitan Museum of Art).

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

85

Fig. 7: Plan of the island of Philae according to H. G. Lyons in 1896 illustrating the Late Antique domestic quarters built around the older temples and new churches (Reproduced from Dijkstra 2008, fig. 13, which is based on Lyons 1896, Plan 1).

Similarly, smaller things can illuminate the same complexities. Censers, or incense burners, are frequently identified as religious items used in creating a purified or holy space through the distribution of aromatics (Caseau 2015) (Fig. 9). Made of copper alloys, or sometimes clay, the censers are designed for a particular function: transforming the sensorial nature of space (Caseau 2007). When found in Late Antique contexts, the incense burners are understood to exist within a particular non-domestic religious space, despite the fact that many of them do not come from archaeological contexts in churches or other religious spaces. At the Monastery of Epiphanius, a small monastic community in Thebes, a Metropolitan Museum of Art excavation team found a remarkable bronze censer in 1912

20

The discovery coincided with the arrival of American collector and Christian enthusiast, P. Morgan, who reportedly was allowed to remove the item from the excavation, take it to his boat, and

(Winlock and Crum 1926, I.95; Ross 1942) (Fig. 10). The censer does not bear any Christian motifs but rather a composition of a lioness atop a boar, who is pinned to the lid of the container, which is elevated by four paws. Below this scene of triumph an elephant appears in relief on the front of the box. The smoke and fragrance escaped the censer through the nostrils, mouth and ears of the boar and from the nose and mouth of the lioness. The visual appearance of the censer is not accidental; the maker made choices regarding the animals and the movement of air into and out of the vessel. The censer was found on a shelf in a rock wall by one of the monastery’s ovens in the West Court. It still contained ashes of incense.20 Is the censer a Christian object? It was found within a monastery and near

hold on to it until it was eventually returned to the MMA in 1944. See Humphrey 2015 for details on this story.

86

D. L. BROOKS HEDSTROM

Fig. 8: Mud-brick houses from the island of Philae in 1896 as photographed by H. G. Lyons (Lyons 1896, pl. 50) (© Universität Heidelberg).

an oven so the incense could be lit. Does the lack of obvious Christian motifs render this a non-religious object? Did the object come to the monastery from outside and was its identity therefore changed, or did it carry multiple identities, much like the island of Philae? Another Late Antique censer found in Egypt and now in the Brooklyn Museum is identified as having been transformed from a generic vessel into a Jewish one at an unknown date (Fig. 11). Unlike the Epiphanius censer, the Brooklyn artefact has a Greek

inscription and a menorah added to the exterior of the bowl.21 The inscription does not follow the original inscribed lines of the bowl and therefore the censer has moved from a vague to a specified identity (Ness 2015, 139–40). The inscribing of the object does not necessarily mean that it moved physically from one community to another. The act of marking could be part of a moment within the community that required Auxanon, the man named in the dedication, to act upon the vessel, and vice versa.

21

The inscription reads: In fulfilment of a vow of Auxanon: Blessed are you, O Lord, who separates the holy from the profane.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Fig. 9: Cast bronze censer with three crosses added to three corners of the hexagonal body, from Egypt, sixth to seventh century (BM 1923,0512.378, Courtesy of the Trustees of the British Museum).

87

Fig. 11: Bronze censer with a Greek inscription, including a menorah, from Egypt, fifth century (41.684, Courtesy of the Brooklyn Museum of Art).

Fig. 10: Bronze censer from the Monastery of Epiphanius in Thebes, sixth–seventh centuries (MMA 44.20a, b, Courtesy of the Metropolitan Museum of Art).

88

D. L. BROOKS HEDSTROM

An object-centred approach highlights the new and expanding questions that we can ask about a Christian thing and the object-scape. Where should the Epiphanius censer be placed in a museum gallery? Gayet would have dubbed it a Gnostic artefact as a way of explaining the confusing nature of motifs not in alignment with his conception of the visual culture of Late Antique Egypt. A discussion about makers, identity and the blurring of religious lines also appears in discussions of Late Antique sculpture, in which ‘pagan’ and Christian designs appear together, as if in conversation. Thelma Thomas has demonstrated how the historiographical assumptions of early generations hampered our ability to see the textured meanings of Late Antique objects (Thomas 2000 and 2010). New materialism proves to be the path forward in considering how to approach objects anew and consider the networks of objects and how they may redefine what we know of the people who used them (Witmor 2014). Case study 2: Late Antique cemeteries and the archaeology of the dead The second case study relates to the use and interpretations of burials, the dead and their cemeteries. Excavations at late Roman and Late Antique settlements produced a wealth of new material from sites such as Antinoopolis. More recent excavations at Naqlun, Kellis and Deir el-Bakhit illustrate the continued challenges in dating and interpreting the burials of the dead. In reviewing these sites, we see the power of the excavator as the translator, historian and authoritative voice on how to read, or misread, the significance of Christian material. In many cases, objects have been overwritten by the assumptions of the archaeologists: on how Christianity emerged; on the relationship between ethnicity, religion, gender and authority; and on the quality of Christian art as a direct indicator of the intellectual capacity of the makers (Becker 2014; Rebillard 2015). The archaeology of death and burial presents new modes of seeing and reading archaeological material.

Digging cemeteries: Gayet at Antinoopolis The late nineteenth and early twentieth centuries witnessed an exciting period in field archaeology with greater interest in the unusual materials being recovered from excavations. Additionally, Christianity’s uniqueness and the narrative of its superiority were called into question with the greater acceptance of the German ‘History of Religions’ school, which presented Christianity not as exceptional, but rather as a religion emerging as part of the natural progression of ideas along with the emergence of teachers of Gnosticism (Brakke 2011).22 Placing Christianity in dialogue with earlier religions made the classifying of material remains yet more problematic, as seen in the catalogues and introductions to Christian, Byzantine and Coptic art from this period. Gayet excavated the Roman and Byzantine cemeteries at Antinoopolis in Middle Egypt from 1896 until 1912. He was well placed to offer a substantial introduction to a new body of material and to demonstrate the cultural and religious diversity of the post-Roman period (Gayet 1902b). Gayet’s work at the cemeteries prompted Gaston Maspero (1846–1916), the director of the Antiquities Service in Egypt in 1899, to ask him to write the first catalogue of the Christian remains that became part of the first, permanent display of Coptic material in Cairo at the Bulaq Museum (Gayet 1889). At the prompting of Maspero, a new room was opened to display material that was not considered appropriate for the GraecoRoman rooms or was ‘too crude’ to have been displayed previously (Gayet 1889, 1). In Gayet’s introduction to the catalogue he pointed to the great decline in art and architecture at the advent of the Byzantine period in Egypt. He lamented the waning in artistic skills, evident in what he described as a primitive style that reflected ignorance of the art of craftsmanship (Gayet 1889, 6–7). For Gayet, Christian Egypt was marked by the reuse of pharaonic monuments by monks and the translating of pharaonic theology directly into Christian iconography. In his discussion of ‘Coptic’ art

22

For a discussion of whether Gnosticism should be a category for identity, see Van Den Broek 2013, 9–11.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

and architecture, he posited the thesis that Egyptians readily embraced Christianity because it was so similar in nature to ancient beliefs: Jesus was regarded as the new Ra, the Trinity was a new Theban triad of deities, and Mary was a new incarnation of Isis. These ideas resonated with others who embraced the new historical-critical methods from the History of Religions school. The full development of Gayet’s ideas would come with his later excavation and publication of the material from Antinoopolis (Gayet 1903). In 1900, in order to introduce his discoveries to the world, the remarkable Late Antique burial assemblages were showcased at two exhibitions in France (Gayet 1900). Sponsored by the Société du Palais du Costume and the Musée Guimet, Gayet displayed materials he recovered from the additional cemeteries at Panopolis, Dayr al-Dik, Asyut and Damietta (Calament 2005) (Fig. 12). The exhibition at the Musée national des arts asiatiques Guimet presented the finds exclusively from Antinoopolis and highlighted the burials Gayet dated to the sixth century, complete with wax models of people wearing reconstructions of clothing of the period to help viewers envision the past (Gayet 1902a). Gayet’s publications and exhibits provided a physical context for viewing Christian Egypt, but there was little of substance to enable an exploration of religious plurality in Egypt in the first few centuries AD. Gayet’s work shaped the reception of Christian Egypt for popular consumption. For example, Anatole France’s 1890 novel Thaïs, about an Alexandrian prostitute who converts to Christianity, and Jules Massenet’s later adaptation of France’s work as an opera, highlighted Gayet’s influence on presentations of early Christianity in Egypt.23 Later, when Thaïs was translated into English, G. Monkshood described in his preference how the story offered a ‘new sensation’ to ‘even the sex-worn brain of Paris’ (Monkshood 1911, ix). The publication highlighted Gayet as the archaeologist who ‘discovered the body of Thaïs, at Antinöe’, and whose work truly offered ‘a great re-conquest of antiquity’ (Monkshood 1911, xii). The popular novel brought Gayet’s archaeological discoveries to life in prose, offering readers a vision of Late Antique Egypt

23

The opera was first performed on 16 March 1894 with Sibyl Sanderson, the American soprano, in the title role of Thaïs. The

89

Fig. 12: Cover of Albert Gayet’s publication of finds from the Roman and Byzantine cemeteries for Palais du Costume (Gayet 1900).

with ‘cities, palaces, philosophers, actors, courtezans [sic] and naked vine-wreathed corybantes dancing, full of life, to drones and shrieks of the syrinx, sullen clank of brass, and the murmur of harps beneath a golden plectrum’ (Monkshood 1911, xi). Monkshood’s summary captured the clash between the opulence of paganism and the emergence of Christianity as expressed in the displays of Gayet’s discoveries.

opera gained popularity after 1898 and was performed as recently as 2017 by the Metropolitan Opera in New York.

D. L. BROOKS HEDSTROM

90

Gayet’s popular L’Art copte (1902b), written to educate the general public following the success of the earlier exhibitions, presented the realia of Egyptian Christianity with samples of colourful textiles, grave goods and modern mannequins. In his descriptions of ‘Coptic’ art, he created two broad categories: Christian and Gnostic. The broad umbrella of Gnostic material included all non-liturgical items, such as ivory furniture inlays, bronze pitchers and even spindle whorls. Despite the fact that these items from domestic contexts could be assigned to any household, regardless of religion, Gayet took the opportunity to differentiate between orthodox Christian communities and supposed Gnostic adherents—although the simple fact was that the objects revealed no obvious religious clues whatsoever (Gayet 1902b, 42–61). In fact, Gayet’s contributions were far more damaging. He asserted interpretations based upon specious claims, offered false identification of motifs, and applied a vivid imagination to the question of how Egyptians responded to the emergence of Christianity. By stating that an object or site was Coptic in nature, Gayet signalled to readers that the materials were not ancient Egyptian; nor Byzantine, meaning Greek Christian; nor Roman, meaning belonging to the cultural protectors of Hellenistic Greek ideals. And although Coptic was a religious and ethnic identity adopted, with pride, by Egyptian Christians of the Coptic Orthodox Church, the name carried a slightly different connotation when used by foreign scholars in assessing the value of Late Antique material culture (Mondal 2003, 150–56; Brooks Hedstrom 2012). Seeking religious identity among the dead: Scott-Moncrieff Despite the promise of new discoveries at the turn of the twentieth century, there were few archaeologists working in Egypt who truly wanted to study exclusively Christian Egypt.24 This was due in large part to the practical interests of museums and the development of the first departments in Egyptology outside of Germany. One scholar who had the potential to set the

24

When Dalton needed an archaeologist to work at Wadi Sarga, for example, he employed a specialist in Assyriology, R. Campbell Thompson.

stage for a more careful treatment of the archaeological material and its insights into religious life was Philip David Scott-Moncrieff (1882–1911). Unfortunately, his premature death at age twenty-nine extinguished the spark of a young scholar who developed a keen interest in Egypt and the religious complexities in Late Antique society. As an Assistant Keeper in the Department of Egyptian and Assyrian Antiquities at the British Museum, Scott-Moncrieff was able to travel to Egypt and to work with a wide array of material remains. His most important work was posthumously published in 1913 as Paganism and Christianity in Egypt. The book was Scott-Moncrieff’s effort to explore the development of a new religion and how the ‘old beliefs fell into decay, and the effect of the new faith on a stubborn and conservative people’ (Scott-Moncrieff 1913, 3). He reveals his prejudiced view of practitioners of traditional Egyptian religions in the light of the advent of Christianity. Despite this, he was very interested in the work of B. P. Grenfell and A. S. Hunt at Oxyrhynchos, which had begun in 1897 with the discovery of fragments of the Gospel of Thomas and other variant Christian texts. Because the texts were written in Greek, ScottMoncrieff asserted: As far as we know at present there was no attempt to translate any of the gospels or ‘Sayings’ at that period into the native language so that they might be understood by the fellahîn population, who still worshipped the ancient gods of Egypt unaffected by Hellenic influences (ScottMoncrieff 1913, 79).

The underlying belief in Scott-Moncrieff’s reconstruction of Roman Egypt was a clear division between Hellenized Egyptians (embracing Greek as an expression of Christianity in urban settings), and the more illiterate countryside marked by ‘disintegration and decay’ (Scott-Moncrieff 1913, 23). Even the priesthood attending to the temples had evolved—in his mind—to be ‘like people, practically conversant only with the garbled and distorted ideas which were all that remained of the great religious system of the past’ (ScottMoncrieff 1913, 23).

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Based upon the papyrological evidence of Christian and Gnostic literature and personal correspondence, Scott-Moncrieff was convinced Greek was the language of the new religion and thus impacted only a specific stratum of society, which was ‘entirely Hellenic and Alexandrine’ in birth and education. For ‘although there may have been many non-Greek speaking, native converts, Greek ideas and Greek education, tinged perhaps with native Egyptian elements, were the guiding influence of the early Church’ (ScottMoncrieff 1913, 98). His view of the social and intellectual gulf between Greek and Egyptian communities (the latter still using demotic) mirrored the hierarchical view of the past as described by early twentieth-century Coptic intellectuals and foreign scholars seeking to distance Christianity from the remnants of pagan beliefs (Fournet 2020, 6–18). Egyptians in the countryside were cast as an ignorant and destructive force that watered down Greek philosophical values and resisted new Christian theology. Ignorance and limited education in religious matters, in Scott-Moncrieff’s view, helped explain why semi-paganism crept into early Christianity in Egypt, causing it to be disorganized and making it difficult to discern when individuals converted when analysing material remains. It is perhaps not surprising that while Scott-Moncrieff was seeking to build a more ethnically diverse Egypt, he was still shaped by classism and its ties to education and intellectual achievements. The discovery and exploration of the Roman period cemeteries at Antinoopolis and Panopolis prompted Scott-Moncrieff to analyse how Christian burials compared to those of non-Christians. He was particularly interested in Gayet’s earlier excavations, which provided the world with a stunning array of ‘semi-pagan objects’ from the Christian cemetery, ‘such as seals impressed with various symbols, gnostic figures, scorpions, whips, figures of pagan gods, the cross and imperial portraits’ (Scott-Moncrieff 1913, 119; Calament 2005). Scott-Moncrieff, like Crum, was critical of Gayet’s methodology and in particular his ‘much abused term “Byzantine”’, when dating the remains (Scott-Moncrieff 1913, 107; see also Gayet 1903). Scott-Moncrieff viewed Antinoopolis as ‘a centre chiefly of Graeco-Roman life or at any rate of Egyptian life with a strong veneer of Hellenism, and this doubtless accounts for the extraordinary jumble of Hellenic and Egyptian elements in both pagan and Christian burials’ (Scott-Moncrieff 1913, 121). The persistence

91

of Hellenism as a reflection of high culture and taste permeated Scott-Moncrieff’s scholarly analysis. He acknowledged that ascertaining religious identity was difficult, ‘because the earliest Christians freely employed pagan designs and were at the same time very restrained in their use of Christian symbolism’ because early Christian designs were ‘furtive symbols of the faith’ such as ‘the fish, the dove, and the hare’ (Scott-Moncrieff 1913, 123). When looking at burial remains at Panopolis, Scott-Moncrieff admitted it was also difficult to articulate the religious beliefs of the deceased or the community responsible for the burials. In his attempts to explain the syncretism evident in the archaeological materials he wrote: ‘[T]he conservative and non-philosophical character of the Egyptians themselves might easily lead to the perpetuation of the old native beliefs, especially with regard to funerary customs and the deeply rooted ideas concerning a future life, under a thin disguise of Christianity’ (ScottMoncrieff 1913, 131). Scott-Moncrieff conceded that it was nearly impossible to frame an accurate story of ownership, use and craftsmanship for many of the objects produced in the first centuries AD, when religious change and conversion were impossible to track. New approaches to the archaeology of the dead Burials, on the surface, appear to offer clear indications of religious identity owing to the design of symbols on clothing, the presence of religious artefacts or the positioning of the body (see Fluck, this volume). If a funerary marker was found in the vicinity of the body, archaeologists could confidently assert that the individual was linked to the inscribed object and thereby assign religious identity based upon the presence of a Christian invocation. If the deceased was draped in textiles with easily recognizable Christian emblems with inscriptions, the person was classified as a Christian. If the textiles bore images from mythology, such as images of nudes or scenes without any text, the person was categorized as a pagan or even a Gnostic. But what about objects that do not bear any motifs or indications of identity, such as woven hats or hoods (Fig. 13)? The objects recovered from the bodies were frequently separated, placed into different collections and not regarded as coming from the same community. Classifications of dead bodies into Christian, ‘pagan’ or ‘other’ led to a reductionist system that reinforced bias in interpreting identity in Late Antique Egypt.

92

D. L. BROOKS HEDSTROM

Fig. 13: Red woven hood from a woman’s burial from Deir el-Bahri, Egypt, second to early fourth century (25.3.217a, Courtesy of the Metropolitan Museum of Art).

Body theory and the archaeology of death are important for the field of archaeology as significant correctives to earlier methods, sharpening our focus and turning it away from assertions of identity towards asking questions about the design of burials and cemeteries as non-human things (Yasin 2005; Borić and Robb 2008; Fahlander and Oestigaard 2008; Nilsson Stutz 2008). The archaeology of death and burial is a relatively new approach to interpreting living communities who care for and prepare the dead for burial (Parker Pearson 1999, 28). The distinction is an important one, for it allows for a societal history of the mortuary practices of a community to be viewed as distinct from the bioarchaeological history of the person and their life as seen in their human remains. Mummification, for example, continued to be used in later periods along with the practice of wrapping bodies, dressed in clothes, in linens that were secured with a series of fabric tapes in interlocking patterns (Dunand 2007). Thus, mummification was not, itself, a reflection of a specific religious identity as it continued into Late Antiquity (Gessler-Löhr 2012). Many of the burials from post-pharaonic periods do not contain grave goods, and thus it is even more difficult to ascribe religious identity to any one person or to those within a particular necropolis. Part of the challenge lies in the fact that burial practices did begin to change in Late Antiquity, but the reasons are not so easily linked to religious motives or Christian identity. Historians and archaeologists often looked to burial customs as a way to differentiate

between ‘pagans’ and Christian converts purely according to the presence of grave items with iconographic clues. But, in Late Antique Egypt, the appropriation and syncretism of various motifs and styles could not provide easy answers for scholars (O’Connell 2008). Thus, it seemed easier to early scholars to separate material according to whether it was Egyptian, meaning poorly executed, or Hellenistic, meaning produced by a skilled artisan, and thereby not truly Egyptian. For example, when looking at Late Antique textiles, William Lowrie viewed the products as works commissioned by those outside of Egypt for it ‘explains the uniformity of textile work’ and the lack of anything ‘purely Egyptian’ in character (Lowrie 1901, 377). Clothing and adornments found within the graves, for example, could point to identity, but whose identity, exactly, was reflected (Swift 2004; Harlow 2012; Swift 2017)? Do the clothes, textiles, and grave goods reflect the identity of the deceased or that of the living persons who placed him or her in the grave? How can we reconstruct or replace objects within a network of connections, which would help us better understand objects as items that act upon the living (Lavan, Swift and Putzeys 2007)? Such questions are used now to revisit earlier assessments of what we can learn from burials and the content of graves. Digging cemeteries: Kellis, el-Deir and Deir el-Bakhit Examinations of burial practices at Kharga and Dakhla oases help reveal the longevity of traditional

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Egyptian religious practices, even in communities where Christians began to change their approaches to burial. Some components of religious identity can be inferred from alterations in the disposition and orientation of the bodies within the actual grave. The dead are oriented toward the east and their arms are not crossed over the chest in what is often referred to as the Osiris position, and which suggests alignment with the traditional Egyptian cults (Dunand and Letellier-Willemin 2019, 246). At the Kellis 2 Cemetery, the majority of the bodies lacked any funerary material, in a departure from earlier burials that had included traditional Egyptian amulets and the like. The omission of funerary markers was interpreted as a shift in community beliefs, but not necessarily linked to the specific individuals (Bowen 2003). At the West Cemetery at el-Deir, two males had their index and middle fingers raised as if to form the gesture of blessing seen in Christian art; this unusual position of the hands suggests to the excavators that the men might be Christian priests (Dunand and Coudert 2014). Clothing found within the graves does not indicate specific religious identity, although the quality of the textiles used may suggest variations in socio-economic status. Textiles at the cemetery of el-Deir, for example, can suggest the clothing styles in use, but may not reveal religious affiliation unless they bear expressly iconographic components. Fleur Letellier-Willemin used the discovery of woven headscarves from a cemetery tentatively identified as Christian to demonstrate that women’s head coverings existed in both ‘pagan’ and Christian contexts in the Nile valley (LetellierWillemin 2011). Scarves as (required) headdresses are found to be in use for Jewish, Manichaean and Christian women—and these communities all coexisted at Kharga. Therefore, while burials certainly can be seen as important indicators of religious beliefs, the evidence is very fragmentary and often lies in the eyes of the excavators, who may see hand gestures, a headcover or a linen shroud as pointing toward religious identity. At Deir el-Bakhit, a site identified as the Monastery of Apa Paulos in Western Thebes, the remains of twenty-two individuals were found in a necropolis dating between the sixth and tenth centuries (Lösch et al. 2013). Despite the cemetery’s close proximity to the monastery, the burials lacked any expressly religious indicators of identity. Excavators and physical anthropologists deduced that those individuals who were mummified with salt and juniper berries were

93

likely to be associated with the monastery, reflecting the community’s greater care for the bodies of monks. The other bodies, whose remains were not mummified, were regarded as individuals who worked within the monastery, and whose burials reflected their lack of status as non-monastic persons (Lösch, Hower-Tilmann and Zink 2013, 36). Yet, such a reading of bodies as objects of identity and social class within a community masks the reality of what we can genuinely infer from dead bodies and graves. Closer scrutiny is required of the variations in treatment of the dead by family or community, or even the specific time and place. While the graves are considered to be concurrent, the range of dates still spans several generations, and without firmer links to specific periods it is not possible to conclude with certainty when the dead were wrapped, clothed and buried. To infer the importance of one set of individuals relative to another may not be a true reflection of what is extant in a cemetery where even 14C dating offers a span of ± 50–100 years (Shortland et al. 2013). The placement of the cemetery is certainly important for spatial context, but when we examine the physical remains themselves and the graves, none of the material points to any individual expressly as Christian, Muslim or Jewish. Within a four-hundred-year window, it is necessary to consider who is burying the dead and when the cemetery was actually created. Case study 3: Late Antique landscapes and the archaeology of space The last case study centres on the topic of buildings, sites and the landscape. As discussed above, identity is ascribed to a wide variety of things, and the built environment can be considered as a ‘big thing’ that may have multiple identities at play at any given time. It is important to remember that an archaeological site does not usually preserve a single moment, unless it has experienced a natural disaster such as those at Herculaneum, Pompeii or the island of Thera. Otherwise, sites, just like other things, can survive generations of communities, and the task of dating levels and building phases and areas of use requires flexibility and latitude. Early efforts to survey and document the surviving architecture of Christian Egypt provided a foundation for tracking new settlements. Then, the excavation of previously unknown Christian sites, which contained extensive physical remains, further enriched the corpus of physical evidence of Christian communities. Despite this work and a plethora of material that

D. L. BROOKS HEDSTROM

94

was identified as clearly Christian, there were many factors that limited the impact of the discoveries in illustrating the diversity of Late Antique Egypt and the importance of the material remains of Christian Egypt. The advent of landscape archaeology has enabled scholars to re-examine older interpretations of Christian sites and assess to what degree identity may be ascribed to sites. Surveying architecture: Butler and Clarke The first consideration of Christian material as a subject of study, in English, was published in 1884 by historian Alfred J. Butler (1850–1936) (Fig. 14). Trained in England in the history of Roman Egypt, Butler departed for Egypt in 1880 to work as a tutor in Classical Studies for the royal children of Khedive Tawfiq (1852–1892). While in Egypt, Butler was amazed to see extensive Christian architectural remains that were not catalogued nor described in any English publication to date (Butler 1887). Butler took it upon himself to rectify this lacuna by writing a two-volume work, The ancient Coptic churches of Egypt. As the first author to treat the material culture of Egyptian church architecture, Butler acknowledged how every traveller to Egypt noticed Christian remains which ‘lie buried in the gloom of fortress walls, or encircled and masked by almost impassable deserts’, but lacked reliable information to help them appreciate the value of these Christian remains (Butler 1884, vii). He wrote for a diverse audience of churchmen, historians and antiquarians, as well as some private collectors. With this audience in mind, Butler alerted his readers to the shortcomings of his own work by stating: the very incompleteness of this work proves how much is still lacking … for the history of Christian Egypt is still unwritten, or at least that part of it about which the most romantic interest gathers, the period which witnessed the passing away of the ancient cults and the change of the pagan world. We have yet to learn how . . . faces like those sculptured on the monuments of the pharaohs became the faces of anchorites, saints, and martyrs (Butler 1884, ix–x).

Butler acknowledged that the study of Christian Egypt, in particular, needed more advocates like himself; he knew that foreigners were attracted to the mystery and allure of the pharaonic world and not Christian Egypt. What concerned him most was the lack of awareness of and attention to the still existing monuments of Christian Egypt. Recognizing the

Fig. 14: Cover to Alfred J. Butler’s The ancient Coptic churches of Egypt 1 (1884).

groundbreaking nature of his two-volume work, he entreated his readers to appreciate the Christian antiquities of Egypt, for they were ‘so vast in extent, so venerable in years, so unique in character, so rich in known and unknown possibilities of interest’ (Butler 1884, 371). He also worried about the fate of the sites owing to a lack of conservation and interest: ‘[D]ay by day they are perishing, unknown to western travellers, and little regarded by the Copts themselves; and nothing, absolutely nothing, has been done or is doing to rescue them from oblivion, or to save them from destruction’ (Butler 1884, 371). The importance of Butler’s work was immediately recognized by late nineteenth-century scholars, who shared his interest in Christian antiquities. The archaeologist and art historian of Christian Rome, Arthur L. Frothingham (1859–1923), wrote a review of Butler’s work for the newly established American Journal of

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Archaeology. He acknowledged ‘Christian Egypt was still a terra incognita’ and asked his readers to consider why this period had not been explored ‘long ago’ and ‘diligently investigated by the historian and the archaeologist’ (Frothingham 1886, 443–44). Frothingham’s recognition that the study of Christian Egypt required both excavation and interpretation illustrated how convincing a case Butler had made that a wealth of physical remains still existed and needed attention. Despite his recognition of Butler’s contribution, Frothingham was pained by Butler’s lack of art historical training. Himself a professor of Christian architecture at Princeton, he pointed out where Butler misunderstood the designs of churches, or the evolution of mosaics, thereby diminishing the scholarly value of his work. While Butler’s work was foundational for the study of the landscape of Christian Egypt and existing church architecture, he did not have the skills to contextualize the material culture of Christian Egypt or the makers behind the art and architecture he sought to celebrate. Frothingham summarized Butler’s view of Coptic art as: ‘never . . . tied and bound by rigid laws of tradition in the same manner as the art of the Greek Church’ (Frothingham 1886, 451). Frothingham agreed with Butler’s overall conclusions and his hopes that scholarly attention might be given to Egypt: ‘Remains so vast in extent, so venerable in years, so unique in character, so rich in known and unknown possibilities of interest, are surely as well worthy of research and exploration as the colossal monuments of pagan Egypt’ (Frothingham 1886, 451). Interest in Christian architecture continued with Somers Clarke (1841–1926), an English architect who deeply appreciated the craft of Egyptian brickmaking. As a trained architect Clarke first started surveying sites in the 1890s and then turned his attention to Christian sites in 1901 (Warner 2011 and 2014). He published a detailed volume in 1912 in which he admitted that the material condition of specifically Christian antiquities was dire: The remains of ancient Egyptian buildings have, not unnaturally, absorbed the attention of travellers and antiquaries … [but t]he buildings which grew up under Christian influence have as yet been but little studied, and it must be admitted that as we see them to-day, decrepit, neglected, and ruined, they are not very attractive (Clarke 1912, 7).

In documenting the Christian antiquities of Egypt, Clarke knew that he could not make a case for their

95

beauty; mud bricks palled in comparison to the stone architecture of the ancient pharaohs. The fate of portable artefacts was slightly more positive, but such material was often mischaracterized by archaeologists who were not attuned to the religious complexity that emerged in the first millennium AD. Butler and Clarke’s efforts to document surviving Christian monuments were an essential first step in assessing what still existed in the landscape. For both men, documenting these structures was a conscious act of preservation. They feared the monuments would not survive, owing to a lack of attention (or interest) from a variety of communities. Today, their work allows us to rewrite and reconstruct landscapes that have changed over the last 100 years, and is still an essential resource for archaeologists in conceptualizing the presence of monumental objects of Christianity in Late Antique Egypt. Excavation of Christian sites: Bawit, Saqqara and Abu Mina To complement the survey work of Butler and Clarke, in the early twentieth century archaeologists from the French Institute, the EEF, the Metropolitan Museum of Art and the British Museum started large excavation projects of Christian settlements that would significantly challenge older theories and provide a wealth of new material to examine. The excavations, while not necessarily models of systematic excavation techniques, were an improvement over earlier projects. Three excavations at expressly Byzantine period sites (all specifically Christian in nature), offered great potential to present the world with a window into Late Antique society. The French excavations at the Monastery of Apa Apollo at Bawit, led by Jean Clédat (1871–1943); the British excavations at the Monastery of Apa Jeremias at Saqqara, led by James Quibell (1867–1935); and the German excavations at the Christian pilgrimage site of Abu Mina, led by Carl Kaufmann (1872–1951) revealed the foundation for large-scale Christian settlements with skilfully carved stone features; complex architectural plans of monastic and domestic quarters; and a plethora of monastic documents in the form of graffiti, ostraca and papyri. Clédat carried out four campaigns at Bawit in Middle Egypt between 1901 and 1905 (Clédat 1904; Rutschowscaya 1995). He cleared several rooms, which he believed were Christian chapels similar to the tombs found earlier at Bagawat, although the results of this excavation have not been fully published (Lythgoe 1908a and

96

D. L. BROOKS HEDSTROM

1908b; Hauser 1932).25 The first substantial publication of the first two seasons appeared in 1904 in Mémoires publiés par les membres de l’Institut français d’archéologie orientale and provided eight pages of introduction to the site followed by a partial catalogue of twenty-eight buildings with a description of the architecture; subsequent publications focused more upon reproductions of wall paintings than on other aspects of the settlement and its buildings (Clédat 1904– 1906; Chassinat 1911; Clédat 1916 and 1999; Maspero 1931–43). Clédat’s report included discussions of mudbrick rooms only when they contained wall paintings, graffiti or any objects of significant interest to him (Clédat 1904–1906, viii). Evidently, during both excavation and publication, Clédat winnowed the material to focus on the areas considered of greatest interest. Crum questioned Clédat’s approach and his initial interpretation of ‘[f]ifteen little “chapels” form[ing] a single group of buildings’ and wondered if the structures instead ‘represented the main body of monastic buildings’ (Crum 1903–1904, 84). Crum’s hunch was later confirmed when large-scale excavations resumed in 1913 under the direction of Jean Maspero (1885–1915), who identified forty more monastic structures (Maspero 1913). At least seventy structures across a 40ha area yielded a wealth of new material, enabling study of the development of Christian Egypt and its material remains within a specifically monastic context (Bénazeth 1994). Just after Clédat’s last season ended at Bawit, Quibell, a British Egyptologist working on behalf of the EEF at Saqqara in 1906, was alerted to the discovery of colourful wall paintings that he recognized as similar to those found at Bawit. Aware of Clédat’s work further south, Quibell decided to explore more of the settlement. In the following seasons from 1907 to 1910 he cleared and mapped a large-scale monastery, making it the second most important discovery in the history of monastic archaeology in Egypt. Although Bawit was larger in area, at Saqqara Quibell recorded more than 200 residential rooms, three churches, a refectory, a

hospital and several storage facilities (Quibell 1908– 1912). Quibell recognized the hand of various painters and craftsmen involved in the construction of the monastery over several periods, but he offered only a rather crude assessment that most paintings were ‘barbarous’, while a few were more carefully executed (Fig. 15). The construction methods at the Monastery of Apa Jeremias were poor, in Quibell’s opinion. The buildings were the ‘rudest’ he had seen and the paintings were ‘debased’ (Quibell 1908, 64–65).26 Like Clarke and Dalton, Quibell applied an aesthetic assessment to the cultural value of the materials he excavated at Saqqara. He was little interested in how the materials reflected the multi-religious nature of Egyptian identity, except to equate the poor quality with a form of monasticism he considered ignorant of true Christianity. The best pieces from the excavations conducted at Bawit and Saqqara were divided up between the newly established Coptic Museum and museums in Europe and the USA, such as the Louvre, the British Museum and the Metropolitan Museum of Art.27 Founded in 1908, the Coptic Museum was the first museum in Egypt to be entirely controlled by Egyptians and also dedicated to Christian material (Reid 2002, 258–86; Gabra 2006; O’Connell 2014b, 172; Henein and Simaika 2017).28 The initiative was led by Murqus Simaika (1864–1944), a Copt who was able to negotiate with the Coptic Patriarchate to help preserve manuscripts in Cairo and to assert Egyptian ownership of excavated remains. While Bawit and Saqqara offered new evidence for Egypt’s monastic history, it was not clear how the monastic remains related to earlier Christian history in Egypt—especially since the two sites were not part of the monastic narratives found in Greek and Latin patristic accounts. The archaeological work at Abu Mina, a large settlement south of Alexandria, provided further evidence that Egypt contained more Christian material than ever expected. The site proved to be the largest Christian pilgrimage centre in the Eastern Mediterranean. Under the

25

27

26

Access to the archives of the work at al-Bagawat will provide unique opportunities for scholars to revisit the landscape of a Christian necropolis with complex chapels and numerous burials: https://libmma.contentdm.oclc.org/digital/collection/ p16028coll10/id/1161. Quibell’s most opinionated and unguarded statements assessing the history of the site are actually reserved for the unidentified Arabs who, in his opinion, destroyed the monastery.

28

Several noteworthy pieces also found their way into the hands of antiquities dealers and into private collections. Not until 1947 was the Coptic material held in the Egyptian Museum permanently relocated from storage to the Coptic Museum.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

97

Fig. 15: Excavated room at the Monastery of Apa Jeremias at Saqqara, sixth–eighth century (Quibell 1908, pl. XLV).

direction of German archaeologist Carl Kaufmann, who began excavations in 1905, the story of St Menas became a reality rather than myth (Kaufmann 1906–1908; Kaufmann 1910a; Ewald Falls 1911). Kaufmann discovered the source for the production of numerous clay ampullae showing St Menas flanked by two camels. The Menas pilgrim flasks found their way into various museums and private collections around the world and are perhaps the most recognized artefact of Christian Egypt (Kaufmann 1910b) (Fig. 16). Kaufmann’s first report on the site described the clearing of a large transept church, a series of domestic quarters and sculpture pieces that he

transported back to Frankfurt (Kaufmann 1910a). Crum accepted Kaufmann’s interpretation that the site had monastic buildings and the shrine of Menas, whereas others, such as Ugo Monneret de Villard, considered Kaufmann’s work ‘badly excavated and worse published’ (Crum 1906–1907, 74; Ward-Perkins 1949, 29). Despite the limitations of his methodology, Kaufmann introduced the world to the largest Christian pilgrimage centre in Egypt, which is still providing insights into religious practice, the intersection between the episcopate and the laity, and a Late Antique city made of stone and mud brick (Grossmann 1998).29

29

Grossmann has worked for decades at the site. For bibliography, see Grossmann 2002.

98

D. L. BROOKS HEDSTROM

Fig. 16: Menas flask, c. 480–650 AD (BM 1875,1012.16, Courtesy of the Trustees of the British Museum).

Read together, Bawit, Saqqara and Abu Mina offered new material that could help shift the field away from a focus on manuscripts to the study of Christian Egypt through more diverse materials. However, the field still lacked the methodological skills to question how Egypt was Christianized and how material remains could be used to effectively illustrate that Christianity existed alongside traditional cults, Jewish communities and various expressions of Christian theology such as Manichaeanism (Elm and Ocker 2020). Material culture and landscape archaeology: Defacing older buildings Just as burials were seen as a straightforward way to define religious identity, so the defacement of pharaonic monuments was often very readily assumed to be the work of Christians in the changing religious landscape of Roman Egypt. Many studies have looked at the sociology of religious conversion and acts of aggression

against the past; new materialism considers defacement in the light of material religion and as an action that underscores, rather than removes, the element of sacredness. The temple to Isis on the island of Philae is a monument marked with religious texts and covered with wall reliefs of the Egyptian pantheon. Constructed in the Ptolemaic era, the island became a centre for Egyptian traditional religion. The impressive remains standing today reflect the significant importance of Isis into the late Roman period. The carving of Christian crosses over older Ptolemaic images was often interpreted as an act of deliberate defacement and of conquest over earlier religious communities. Even when Christian iconography was not included in the damage to pharaonic images, the chiselling out of faces was considered the work of Christian vandals. Fortunately, Jitse H. F. Dijkstra’s work on the epigraphic record within the context of the site has demonstrated that the transformation of the area into a Christian community was a complex one, in which Christians were operating

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

99

alongside the older ‘pagan’ community (Dijkstra 2008). The swift ‘temple conversion’, rather, is a myth created by ancient Christian authors who were persuasive enough to convince later scholars (Hahn 2008). The Christianizing acts of defacement warrant closer scrutiny and illustrate the value of considering the construction of the Christian landscape in Late Antique Egypt (Dijkstra 2015). In order to frame the rereading of the landscape of Philae, some questions need to be raised. When did the acts of inscribing on monuments happen? Who actually carried out these activities (anonymously)? Additionally, how do we ascribe religious identity to the individuals altering the monuments? These may appear to be simple questions, but they belie the importance of phenomenological theory in considering how we know what we know about the past and the ways humans have interacted with the built environment.30 Early scholarship started by seeing all changes to pharaonic temples as an act of religious triumphalism and violation by Christians, but this says more about how scholars viewed Christianity in their own time, and Egyptian Christians, than it does about the history of Christian conversion in Late Antiquity (Kristensen 2013, 155– 58; Mayer 2013).31 The literary turn in Late Antique studies has forced historians of early Christianity to read hagiographical texts through a more critical lens and not accept accounts simply at face value. For example, stories of monastic expulsion of demons from temples or from individuals are now read, not as celebrations of individual acts of defacement or destruction, but rather as exaggerated accounts of Christian power to reassure and encourage readers. Many times, it was the literary accounts that created the foundation for interpreting the defaced monuments and blaming Christians. But, if we need to set aside the hagiographical accounts of Christian power, where can we look in order to understand the archaeological record of defacement? Repurposing of existing monuments is an area of landscape archaeology that will help us reconsider how we read space and its ties to religious identity and

change. While the earlier schools of landscape archaeology created spaces without people, critical refinement of landscape theory has led to a consideration of the complexity of the anthropogenic landscape; buildings and space, like texts, can be stories of changing power dynamics within communities (Fleming 2006; Brooks Hedstrom forthcoming b). For some archaeologists, consideration of the landscape and those who live within it and act upon it is a reflection of the spatial turn, which requires us to think about how space is understood both in physical and imagined realities. Myrto Veikou emphasizes the need for the spatial turn for ‘[b]oth texts and archaeology reveal specific . . . strategies for the construction of settled spaces, ranging from the selection of location based on practical or symbolic concerns to the architecture and decoration of the buildings’ (Veikou 2016, 146). The focus upon the physical and textual creation of space is important when we seek to assign motivations to those who altered Egyptian temples. David Frankfurter, drawing upon theories of the sociology of memory and modern approaches to archaeological space, explores acts of Christian identity formation within existing religious landscapes (Frankfurter 1998 and 2017, 233–37). He draws attention to the importance of the spaces themselves in drawing Christians to want to lay claim to them—not because there was aggression towards the spaces, but because they held an inherent sacrality that the new religious community wished to share. He explains that defacement is an act that does not reflect ‘methodical iconoclasm’, but rather is a sign of Christian devotional acts of laying claim to a site (Frankfurter 2017, 235). He admits that some acts may reflect ‘fearful and apotropaic’, ‘hostile’ or even ‘destructive’ objectives, but that by failing to consider the acts of inscription and inscribing as performative activities of devotion, we may attribute emotions to Christians that they did not possess. Earlier scholars regarded Christianity’s primacy as a swift erasure of traditional cults and spaces, but in fact the process of Egypt’s Christianization was a

30

31

Delattre (2018) explores what he calls the ‘social competition’ found in graffiti, in which the monument is not really the focus of the writing, but rather the interaction between the two inscribers.

Sauer (2003) does not define what ‘religious hatred’ means in Late Antique Egypt and relies upon interpreting Egyptian monuments without grounding his reading of monuments in the complexity of the variegated Christian responses to paganism. For an example of more nuanced readings, see Westerfeld 2019.

100

D. L. BROOKS HEDSTROM

multi-generational process. For buildings, as for cemeteries, trying to associate a moment in time to a community’s religious identity is difficult even now with meticulous archaeological methods and documentation. The participation in construction, reuse and redesign of existing buildings, which may or may not be in use, reflects a purposeful engagement with the religious landscapes of the past and not their erasure. Material culture and landscape archaeology: Monasticism and the myth of the desert Inscribing graffiti on the walls of abandoned buildings was one way of marking a person’s or a community’s identity in a visual manner. Another mode of establishing identity was through the construction of buildings used by new communities, such as those built by monks when they designed areas for ascetic living in Egypt. Nineteenth- and twentieth-century historical accounts of the monastic movement cast monks as fleeing from the city and running to the seclusion of the desert as they abandoned familial ties to find space to live alone. The historians relied upon the writings of Christian writers such as Athanasius, Jerome, John Cassian, Shenoute and others to write a story of Egyptian monasticism (Goehring 1993 and 2003). Most of the earlier histories accepted at face value the narrative of the early Christian authors and regarded saints’ lives, hagiographies, as reliable, historical accounts of the period (van der Vliet 2006). The literary turn in early Christian studies and in the field of Late Antique history has revealed the importance of reading the Christian fathers as men with their own agendas and desires in creating a history of monasticism that reinforced their worldview (Clark 2015). The use of papyrology and epigraphy to offer correctives to purely religious literary sources has enriched our view of how monks viewed the Egyptian landscape and their place within in it (O’Connell 2007; Wipszycka 2009 and 2018). Archaeology has also provided new tools for reexamining earlier excavated material and for considering landscape as a space rather than simply focusing on individual buildings at a site. These developments

32

Similar investigations and approaches are happening in the study of other homosocial communities of Western monasticism and Buddhism in south central Asia. For comparative material from

reflect the spatial turn by which the study of the larger environment helps to situate Christian built environments in a wider context.32 As discussed above, the first excavations of monastic sites in Egypt did not garner the type of attention that would rewrite monastic history (Brooks Hedstrom 2012). Instead, the material was classified as ‘Coptic’ and thus was seen as ‘other’ in comparison to Christian material from elsewhere in the Mediterranean. Additionally, the monastic material, which dated from the fifth to the eighth century, was not early enough to reflect the stories found in Christian literature. The material, therefore, presented a perplexing problem; it did not help elucidate early Christian Egypt, but rather represented a world that was lesser known, as it was not described in as much detail as the beginnings of the monastic movement (Brooks Hedstrom 2017). The larger monastic sites of Bawit, Saqqara and Western Thebes included numerous monastic residences, offering scholars the first opportunities to study domestic settlements within a religious landscape. Landscape archaeology considers both the social and the physical forms of connectivity between communities and households to illuminate the history of Late Antique Egypt (Brooks Hedstrom 2020). At Western Thebes there was a wide array of smaller settlements clustered in and around the long-abandoned pharaonic tombs. Monks adapted the anthropogenic landscape to suit their new purposes by cutting loom pits for weaving textiles, erecting a few chapels, writing mnemonic devices on walls for prayer, and hosting numerous visitors. Rather than being apart from settled civilization, the monks in Western Thebes were very much part of the world around them (O’Connell 2007; Brooks Hedstrom 2017). Deir el-Bakhit was the largest of the monastic sites in Western Thebes, with interconnections to the other monks living in more modest communities or in independent residences. In the case of Western Thebes, we can see the footpaths that link the various terraced monastic sites and connect them to the settlement of Jeme. Papyrological evidence from the area reveals that the large town of Jeme was built in and around the temple of Medinet Habu (Wilfong

other monastic communities outside of the Byzantine world, see Fogelin 2010; Bowes 2011; Dailey 2013.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

2002; Cromwell 2013; O’Connell and Ruffini 2017). Many of the residents interacted with monks in the adjacent areas of Western Thebes and even bequeathed some of their wealth to local monasteries. The raw materials of the landscape also help inform us as to why monks occupied abandoned caves and used mud bricks to create additional spaces for workshops and storage on the terraces. Such an examination of the wider setting of a monastic site can create a more nuanced understanding of Late Antique Christian communities and how different identities were activated within a monastic/non-monastic setting. The earliest studies of Egypt’s Christian landscape considered monuments only as objects in and of themselves without consideration of the wider context of their locations. The first excavations of monastic settlements afforded a unique body of evidence for examining Christian domestic settlements, but a broader view of Christian Egypt did not seem to include the unusual physical remains of monastic communities. Landscape archaeology has helped archaeologists revisit sites to consider the dynamics of imagined communities and has challenged the presence of borders and boundaries that did not exist in the past as we once thought. Conclusion Writing in 1944, John D. Cooney presented the problem of the study of Late Antique Egypt as a time caught between chronological fields of specialization. He hoped that progress would come in the second half of the twentieth century: The old attitude that Coptic art, along with Roman art, was little more than a bastard descendant of late antiquity has yielded to the obvious truth that the Copts developed an expression of their own. To some extent this new art was part of an international style common throughout the Mediterranean areas from Spain to Syria. … Falling, as it does, however, between the fields of the Egyptologist, the classical archaeologist and the medievalist, the period has not been so thoroughly studied as might be supposed, and much work remains to be done (Cooney 1944, 39).

These words remained true for much of the twentieth century. The field of Late Antique archaeology and a greater interest in the complexity of religious identity have helped the study of Egypt in the first millennium move beyond the ideas of Dalton, Gayet and Quibell. The shift to materiality, identity and the complexities of the spatial turn have enriched how scholars approach

101

artefacts whose identities and owners are not immediately apparent. The interface between place, thing, persons and their communities is essential for considering the shifts in religious identities as people in Egypt adopted Christianity in the first five centuries AD. We can also better appreciate the challenges faced by early archaeologists, art historians and historians when confronting a world that involved greater connectivity and transfer of ideas than they could have possibly imagined—it was too complex, and in need of a field of its own, as Cooney suggests (Pohl 2010, 11). What archaeology can do is to observe the ‘negotiation of identity’ and to realize that material remains cannot easily reflect a single identity, but rather the nature of the fluctuating identities that a person or community might have at any single moment in time (Casella and Fowler 2005, 8). The advent of new materialism and the object turn have provided unique ways to reassess the relationship between things, both small and large, and humans. Landscape archaeology, similarly, poses new questions about the larger context of the physical environment, the built environment, and humans within a network of activity and action. Archaeology can help us see the value in reformulating questions, in considering the methods for religious conversion outside of literary texts, and the importance of the less pretty things of the past to rewrite Egypt’s religious landscape in Late Antiquity. With greater attention to things of Christian Egypt, we are in a better position to examine the complexity of identity construction and thereby see the interconnected nature of religious views in Late Antiquity. Acknowledgements I would like to thank Elisabeth R. O’Connell for the invitation to participate in this volume. I am grateful for her suggestions and questions, which helped improve this text. Marica Cassis and Christian Raffensperger offered constructive feedback. A portion of this chapter was presented as ‘Misplacing Byzantine Egypt’, for Constructing and deconstructing Byzantine elements: Perceptions of a Medieval world in the late 19th and early 20th centuries, as a panel at the International Medieval Congress in 2018 at the University of Leeds. Figure 8, courtesy of Univ. of Heidelberg, is here available at https://digi.ub.uni-heidelberg.de/ diglit/lyons1896/0119.

102

D. L. BROOKS HEDSTROM

Bibliography Allison, P. M. 2004. Pompeian households: An analysis of material culture. Los Angeles, CA. Alston, R. 2007. Some theoretical considerations and a late antique house from Roman Egypt. British School at Athens Studies 15: 373–78. Ashmore, W. 2002. ‘Decisions and dispositions’: Socializing spatial archaeology. American Anthropologist 104 (4): 1172–83. Bahrani, Z., Z. Çelik and E. Eldem (eds). 2011. Scramble for the past: A story of archaeology in the Ottoman empire, 1753–1914. Istanbul. Ball, J. 2019. Rich interiors: The remnant of a hanging from Late Antique Egypt in the collection of Dumbarton Oaks. In Catalogue of the textiles in the Dumbarton Oaks Byzantine Collection, G. Bühl and E. Dospěl Williams (eds), Washington DC. Online: https://www. doaks.org/resources/textiles/essays/ball. Barringer, T. and T. Flynn (eds). 1997. Colonialism and the object: Empire, material culture and the museum. London. Barsch, S. and J. van Norden. 2020. Kann Materie historisch sein? Eine Einleitung. In Historisches Lernen und Materielle Kultur: Von Dingen und Objekten in der Geschichtsdidaktik, S. Barsch and J. van Norden (eds), 9–13. Bielefeld. Bárta, M. 2010. ‘Biblical archaeology’ and Egyptology. In Historical biblical archaeology and the future: A new pragmatism, T. E. Levy (ed.), 99–122. London. Becker, A. 2014. Ethnicity, ethnic identity: Some remarks for the Late Antiquity. Gerión-Revista de historia antigua 32: 289–305. BeDuhn, J. D. 2015. Am I a Christian?: The individual at the Manichaean-Christian interface. In Group identity and religious individuality in Late Antiquity, É. Rebillard and J. Rüpke (eds), 31–53. Washington, DC. Bénazeth, D. 1995. Histoire des fouilles de Baouît. In Études coptes IV: Quatrième journée d’études, Strasbourg, 26–27 mai 1988, D. Bénazeth (ed.), 53–62. Cahiers de la Bibliothèque copte 8. Louvain. Borić, D. and J. Robb. 2008. Body theory in archaeology. In Past bodies: Body-centered research in archaeology, D. Borić and J. Robb (eds), 1–8. Oxford. Bowen, G. 2003. Early Christian burial practices at Kellis, Dakhleh Oasis, Egypt. Artefact: The Journal of the Archaeological and Anthropological Society of Victoria 26: 77–88. Bowes, M. 2011. Inventing ascetic space: Houses, monasteries and the ‘archaeology of monasticism’. In Western monasticism ante litteram: The spaces of monastic observance in Late Antiquity and the Early Middle Ages, H. Dey and E. Fentress (eds), 315–51. Turnhout.

Boysen, B. 2018. The embarrassment of being human: A critique of new materialism and object-oriented ontology. Orbis Litterarum 73 (3): 225–42. Brakke, D. 2011. The Gnostics: Myth, ritual and diversity in early Christianity. Cambridge. Bräunlein, P. J. 2016. Thinking religion through things: Reflections on the material turn in the scientific study of religions. Method & Theory in the Study of Religion 28 (4–5): 365–99. Brooks Hedstrom, D. L. 2012. Treading on antiquity: AngloAmerican missionaries and the religious landscape of nineteenth-century Coptic Egypt. Material Religion 8 (2): 128–53. ———. 2017. The monastic landscape of Late Antique Egypt: An archaeological reconstruction. Cambridge. ———. 2020. Monastic settlements as monastic domestic households. In The Oxford handbook of Christian monasticism, B. Kaczynski (ed.), 185–205. Oxford. ———. Forthcoming a. ‘The humility of the desert’: The symbolic and cultural landscapes of Egyptian monasticism. In From the human body to the universe: Spatialities of Byzantine culture, I. Nilsson and M. Veikou (eds). Leiden. ———. Forthcoming b. Three monks and a house: The archaeology of monastic householding in Late Antique Egypt. In Households in context: Dwelling in Ptolemaic and Roman Egypt, C. E. Barrett (ed.). Ithaca, NY. Brown, B. 2001. Thing theory. Critical Inquiry 28 (1): 1–22. ———. 2015. Other things. Chicago, IL. Brown, P. 1978. The making of Late Antiquity. Cambridge. ———. 1982. Society and the holy in Late Antiquity. Berkeley, CA. Butler, A. J. 1884. The ancient Coptic churches of Egypt. 2 vols. Oxford. ———. 1887. Court life in Egypt. London. Bynum, C. 2011. Christian materiality: An essay on religion in late Medieval Europe. New York. Calament, F. 2005. La révélation d’Antinoé par Albert Gayet: Histoire, archéologie, muséographie. 2 vols. Bibliothèque d’études coptes 18. Cairo. Cameron, A. 2016. Late antiquity and Byzantium: An identity problem. Byzantine and Modern Greek Studies 40 (1): 27–37. Caseau, B. 2007. Incense and fragrances: From house to church. A study of the introduction of incense in the early Byzantine Christian churches. In Material culture and well-being in Byzantium (400–1453), M. Grünbart, E. Kislinger, A. Muthesius and D. Stathakopoulos (eds), 75–92. Vienna. ———. 2015. Nourritures terrestres, nourritures célestes: La culture alimentaire à Byzance. Paris. Casella, E. C. and C. Fowler. 2005. Beyond identification: An introduction. In The archaeology of plural and

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

changing identities beyond identification, E. C. Casella and C. Fowler (eds), 1–8. New York. Çelik, Z. 2016. About antiquities: Politics of archaeology in the Ottoman empire. Austin, TX. Challis, D. 2013. The archaeology of race: The eugenic ideas of Francis Galton and Flinders Petrie. London. Chassinat, É. 1911. Fouilles à Baouît. Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 13. Cairo. Christian Register. 1846. Valuable manuscripts. 21 February: 25, 8. Clark, E. A. 2015. From patristics to Late Antiquity at the Catholic Historical Review. Catholic Historical Review 101 (2): 27–71. Clarke, S. 1912. Christian antiquities in the Nile Valley: A contribution towards the study of the ancient churches. Oxford. Clédat, J. 1904. Nouvelles recherches à Baouît (Haute Égypte): Campagnes 1903–04. Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres 48 (5): 517–26. ———. 1904–1906. Le monastère et la nécropole de Baouît. Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 12.1–2. Cairo. ———. 1916. Le monastère et la nécropole de Baouît. Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 39. Cairo. ———. 1999. Le monastère et la nécropole de Baouît, D. Bénazeth and M.-H. Rutschowscaya (eds). Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 111. Cairo. Cole, A. 2013. The call of things: A critique of object-oriented ontologies. The Minnesota Review NS 80: 106–18. Colla, E. 2008. Conflicted antiquities. Durham, NC. Cooney, J. D. 1944. Problems of Coptic art. In Coptic Egypt: Papers read at a symposium held under the joint auspices of New York University and the Brooklyn Museum, February 15, 1941, 35–41. New York. Corbett, E. D. 2014. Competitive archaeology in Jordan: Narrating identity from the Ottomans to the Hashemites. Austin, TX. Cormack, R. and E. Jeffreys (eds). 2000. Through the looking glass: Byzantium through British eyes. Aldershot. Cromwell, J. 2013. Keeping it in the family: Property concerns in eighth century A.D. Thebes. Journal of Near Eastern Studies 72 (2): 213–32. ———. 2014. Walter Ewing Crum (1865–1944). A Coptic scholar ‘sui generis’. In Christlicher Orient im Porträt. Wissenschaftsgeschichte des christlichen Orients, P. Bukovec (ed.), 407–22. Hamburg. Crum, W. E. 1900–1901. Christian Egypt. Archaeological report (Egypt Exploration Fund), 64–81. ———. 1901. Communiqué on Strzygowski’s Orient oder Rom. The Classical Review 15 (4): 232–34.

103

———. 1901–1902. Christian Egypt. Archaeological report (Egypt Exploration Fund): 48–59. ———. 1903–1904. Christian Egypt. Archaeological report (Egypt Exploration Fund): 74–88. ———. 1906–1907. Christian Egypt. Archaeological report (Egypt Exploration Fund): 74. Curl, J. S. 2005. The Egyptian revival: Ancient Egypt as an inspiration for design motifs in the west. London. Dailey, E. T. 2013. Introducing monastic space: The early years, 250–750. In Monastic space through time, E. T. Dailey and S. Werronen (eds), 5–15. Leeds. Dalton, O. M. 1901. Catalogue of early Christian antiquities and objects from the Christian East in the Department of British and Mediaeval Antiquities and Ethnography of the British Museum. London. ———. 1911. Byzantine art and architecture. Oxford. Delattre, A. 2018. Christian graffiti in Egypt: Case studies on the Theban mountain. In Scribbling through history: Graffiti, places and people from antiquity to modernity, C. Ragazzoli, Ö. Harmanşah, C. Salvador and E. Frood (eds), 37–47. London. Díaz-Andreu García, M. 2007. A world history of nineteenthcentury archaeology nationalism, colonialism and the past. Oxford. Dijkstra, J. H. F. 2008. Philae and the end of ancient Egyptian religion: A regional study of religious transformation (298–642 CE). Leuven. ———. 2015. Religious violence in Late Antique Egypt reconsidered: The cases of Alexandria, Panopolis and Philae. Journal of Early Christian History 5 (2): 24–48. Domanska, E. 2006. The return to things. Archaeologia Polona 44: 171–85. Donker van Heel, K. 2012. Djekhy & son: Doing business in ancient Egypt. Cairo. Dospěl Williams, E. 2014. ‘Into the hands of a well-known antiquary of Cairo’: The Assiut Treasure and the making of an archaeological hoard. West 86th: A Journal of Decorative Arts, Design History and Material Culture 21 (2): 251–72. Doyon, W. 2018. The history of archaeology through the eyes of Egyptians. In Unmasking ideology in imperial and colonial archaeology: Vocabulary, symbols and legacy, B. Effros and G. Lai (eds), 173–200. Los Angeles, CA. Drower, M. 1982. Gaston Maspero and the birth of the Egypt Exploration Fund (1881–3). Journal of Egyptian Archaeology 68: 299–317. Dunand, F. 2007. Between tradition and innovation: Egyptian funerary practices in late antiquity. In Egypt in the Byzantine world 300–700, R. Bagnall (ed.), 163–84. Cambridge. Dunand, F. and M. Coudert. 2014. Le passage du paganisme au christianisme: Le cas de l’oasis de Kharga. In

104

D. L. BROOKS HEDSTROM

Alexandrie la divine, Ch. Méla and F. Möri (eds), 796– 801. Geneva. Dunand, F. and F. Letellier-Willemin. 2019. Funerary practices in the Great Oasis during antiquity. In The Great Oasis of Egypt: The Kharga and Dakhla Oases in antiquity, R. S. Bagnall (ed.), 237–68. Cambridge. Egypt Exploration Fund. 1895. President’s report [John Fowler]. In Report of the ninth ordinary general meeting, 16–18. London. Elm, S. and C. Ocker. 2020. Christianity and the material, Medieval to Modern. In Material Christianity: Western religion and the agency of things, C. Ocker and S. Elm (eds), 1–28. New York. Elsner, J. 2002. The birth of Late Antiquity: Riegl and Strzygowski in 1901. Art History 25: 358–79. Entwistle, C. 2000. O. M. Dalton: Ploughing the Byzantine furrow. In Through the looking glass: Byzantium through British eyes, R. Cormack and E. Jeffreys (eds), 177–83. Aldershot. Evans, E. A. 2013. Edward Libbey: An American glass magnate collects in Egypt. In Souvenirs and new ideas, D. Fortenberry (ed.), 24–38. Oxford. Ewald Falls, J. C. 1911. Drei Jahre in der Libyschen Wüste: Reisen, Entdeckungen und Ausgrabungen der Frankfurter Menasexpedition (Kaufmannsche Expedition). Freiburg im Breisgau. Fagan, B. M. 2004. The rape of the Nile: Tomb robbers, tourists and archaeologists in Egypt. Boulder, CO. Fahlander, F. and T. Oestigaard (eds). 2008. The materiality of death: Bodies, burials, beliefs. Oxford. Fleming, A. 2006. Post-processual landscape archaeology: A critique. Cambridge Archaeological Journal 16 (3): 267–80. Fogelin, L. 2010. Material practice and the metamorphosis of a sign: Early Buddhist stupas and the origin of Mahayana Buddhism. Asian Perspectives 51 (2): 278–310. Fournet, J.-L. 2020. The rise of Coptic: Egyptian versus Greek in Late Antiquity. Princeton, NJ. Fowler, C. 2010. From identity and material culture to personhood and materiality. In The Oxford handbook of material culture studies, D. Hicks and M. C. Beaudry (eds), 352–85. Oxford. Frankfurter, D. 1998. Religion in Roman Egypt: Assimilation and resistance. Princeton, NJ. ———. 2017. Christianizing Egypt: Syncretism and local worlds in Late Antiquity. Princeton, NJ. Frothingham, A. L. 1886. Reviewed work: The ancient Coptic churches of Egypt by A. J. Butler. American Journal of Archaeology 2 (4): 443–51. Gabra, G. 2006. Treasures from the Coptic Museum. Cairo. Gange, D. 2006. Religion and science in late nineteenthcentury British Egyptology. Historical Journal 49: 1083–1103.

———. 2013. Dialogues with the dead: Egyptology in British culture and religion 1822–1922. Oxford. Gayet, A. 1889. Les monuments coptes du Musée de Boulaq. Paris. ———. 1900. Exposition universelle de 1900. Le palais du costume. Le costume en Égypte du IIIe au XIIIe siècle. Paris. ———. 1902a. Antinoë et les sépultures de Thaïs et Sérapion. Paris. ———. 1902b. L’Art copte: École d’Alexandrie-Architecture monastique. Sculpture-Peinture. Art Somptuaire. Paris. ———. 1903. L’exploration des nécropoles de la montagne d’Antinoë: Fouilles exécutées en 1901–1902. Annales du Musée Guimet 30: 115–40. Gell, A. 1998. Art and agency: An anthropological theory. Oxford. General Research Division, The New York Public Library. 1870–1875. Ascension d’une pyramide. Retrieved from https://digitalcollections.nypl.org/items/510d47d94b34-a3d9-e040-e00a18064a99. Gessler-Löhr, B. 2012. Mummies and mummification. In The Oxford handbook of Roman Egypt, C. Riggs (ed.), 664–83. Oxford. Goehring, J. 1993. The encroaching desert: Literary production and ascetic space in early Christian Egypt. Journal of Early Christian Studies 1 (3): 281–96. ———. 2003. The dark side of landscape: Ideology and power in the Christian myth of the Desert. Journal of Medieval and Early Modern Studies 33 (3): 437–51. Goode, J. F. 2007. Negotiating for the past: Archaeology, nationalism and diplomacy in the Middle East, 1919– 1941. Austin, TX. Greenhalgh, M. 2019. Plundered empire: Acquiring antiquities from Ottoman lands. Leiden. Griffith, F. Ll. 1892–93. Preface. Archaeological report (Egypt Exploration Fund), vii–viii. Grossmann, P. 1998. The pilgrimage center of Abu Mina. In Pilgrimage and holy space in Late Antique Egypt, D. Frankfurter (ed.), 281–302. Leiden. ———. 2002. Christliche architektur in Ägypten. Leiden. Hagen, F. and K. S. B. Ryholt. 2016. The antiquities trade in Egypt 1880–1930: The H. O. Lange papers. Copenhagen. Hahn, J. 2008. The conversion of the cult statues: The destruction of the Serapeum 392 A.D. and the transformation of Alexandria into the ‘Christ-Loving’ City. In From temple to church: Destruction and renewal of local cultic topography in Late Antiquity, J. Hahn, S. Emmel and U. Gotter (eds), 336–66. Boston, MA. Hamilakis, Y. 2011. Indigenous archaeologies in Ottoman Greece. In Scramble for the past: The story of archaeology in the Ottoman empire 1733–1914, Z. Bahrani, Z. Celik and E. Eldem (eds), 49–69. Istanbul.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Harlow, M. 2012. Dress and identity: An introduction. In Dress and identity, M. Harlow (ed.), 2–10. Oxford. Harman, G. 2002. Tool-being: Heidegger and the metaphysics of objects. Chicago, IL. Hauser, W. 1932. The Egyptian Expedition 1930–1931. Bulletin of the Metropolitan Museum of Art 27: 38–40. Henein, N. and S. Simaika. 2017. Marcus Simaika: Father of Coptic archaeology. Cairo. Hicks, D. 2010. The material-cultural turn: Event and effect. In The Oxford handbook of material culture studies, D. Hicks and M. C. Beaudry (eds), 25–98. Oxford. Hodder, I. 2012. Entangled: An archaeology of the relationships between humans and things. Oxford. Holloway, S. W. 2006. Orientalism, Assyriology and the Bible. Sheffield. Humphrey, L. 2015. Collecting Christianity on the Nile: J. Pierpont Morgan and the Metropolitan Museum of Art. In Age of transition: Byzantine culture in the Islamic world, H. Evans (ed.), 2–19. New York. Hurtado, L. W. 2006. The earliest Christian artifacts. Grand Rapids, MI. Ingold, T. 2007. Materials against materiality. Archaeological Dialogues 14 (1): 1–16. Insoll, T. 2009. Materiality, belief, ritual-archaeology and material religion: An introduction. Material Religion 5 (3): 260–64. James, E. 2008. The rise and function of the concept ‘Late Antiquity’. Journal of Late Antiquity 1 (1): 20–30. James, T. G. H. 1982. Excavating in Egypt: The Egypt Exploration Society 1882–1982. London. Janssen, E. 2013. Coptic textiles in the Rijksmuseum. The Rijksmuseum Bulletin 61 (3): 226–49. Jenkins, I. 1995. Archaeologists and aesthetes in the sculpture galleries of the British Museum 1800–1939. London. Jørgensen, M. 2015. How it all began: The story of Carl Jacobsen’s Egyptian collection, 1884–1925. Copenhagen. Kadellis, A. 2015. Late Antiquity dissolves. Marginalia: Los Angeles Review of Books. https://marginalia.lareviewofbooks.org/late-antiquity-dissolves-by-anthony-kaldellis/ Kakissis, A. G. 2009. The Byzantine Research Fund Archive: Encounters of Arts and Crafts architects in Byzantium. In Scholars, travels, archives: Greek history and culture through the British School at Athens, M. LlewellynSmith, P. M. Kitromilides and E. Calligas (eds), 143–61. Athens. Kalloniatis, F. 2013. Jeremiah Colman: A Norfolk traveller in Egypt. In Souvenirs and new ideas, D. Fortenberry (ed.), 68–79. Oxford. Kaufmann, C. M. 1906–1908. Die Ausgrabung der Menasheiligtümer in der Mareotiswüste, 3 vols. Cairo. ———. 1910a. Die Menasstadt und das Nationalheiligtum der altchristlichen Ägypter in der westalexandrinischen Wüste. Leipzig.

105

———. 1910b. Zur Ikonographie der Menas-Ampullen mit besonderer Berücksichtigung der Funde in der Menasstadt nebst einem einführenden Kapitel über die neuentdeckten nubischen und aethiopischen Menastexte. Cairo. Khatchadourian, L. 2016. Imperial matter: Ancient Persia and the archaeology of empires. Berkeley, CA. Kourelis, K. 2007. Byzantium and the avant-garde: Excavations at Corinth, 1920s–1930s. Hesperia 76 (2): 391– 442. Kozelsky, M. 2008. The challenges of church archaeology in post-Soviet Crimea. In Selective remembrances: Archaeology in the construction, commemoration and consecration of national pasts, P. Kohl, M. Kozelsky and N. Ben-Yehuda (eds), 71–98. Chicago, IL. Kristensen, T. M. 2013. Making and breaking the gods: Christian responses to pagan sculpture in Late Antiquity. Aarhus. Langlands, A. 2017. Craeft: An inquiry into the origins and true meaning of traditional crafts. New York. Latour, B. 2005. Reassembling the social: An introduction to actor-network-theory. Oxford. Lavan, L., E. Swift and T. Putzeys (eds). 2007. Objects in context objects in use: Material spirituality in Late Antiquity. Leiden. Letellier-Willemin, F. 2011. Accessories from the Christian cemetery of el-Deir, in the Egyptian oasis of Kharga, Western Desert. In Dress accessories of the 1st millennium AD from Egypt: Proceedings of the 6th conference of the research group ‘Textiles from the Nile valley’ Antwerp, 2–3 October 2009, A. De Moor and C. Fluck (eds), 80–93. Tielt. Littell’s Living Age. 1846. Mss. from the Egyptian monasteries. 8: 393–405. Lösch, S., E. Hower-Tilmann and A. Zink. 2013. Mummies and skeleton from the Coptic monastery in ThebesWest, Egypt. Anthropologischer Anzeiger 70 (1): 27–41. Lowrie, W. 1901. Monuments of the early Church. London. Luijendijk, A. M. 2010. Sacred scriptures as trash: Biblical papyri from Oxyrhynchus. Vigiliae Christianae 64: 217–54. Lyons, H. G. 1896. A report on the island and temples of Philae. London. Lythgoe, A. 1908a. The Egyptian expedition. Bulletin of the Metropolitan Museum of Art 3 (5): 84–86. ———. 1908b. The oasis of Kharga. Bulletin of the Metropolitan Museum of Art 3 (11): 203–208. Majd, Mohammad Gholi. 2003. The great American plunder of Persia’s antiquities, 1925–1941. Lanham, MD. Malafouris, L. 2008. At the potter’s wheel: An argument for material agency. In Material agency: Towards a non-anthropocentric approach, C. Knappett and L. Malafouris (eds), 19–36. New York.

106

D. L. BROOKS HEDSTROM

Malley, S. 1996. Austen Henry Layard and the periodical press: Middle Eastern archaeology and the excavation of cultural identity in mid-nineteenth century Britain. Victorian Review 22 (2): 152–70. ———. 2008. Layard enterprise: Victorian archaeology and informal imperialism in Mesopotamia. International Journal of Middle East Studies 40 (4): 623–64. Marcone, A. 2008. A long Late Antiquity?: Considerations on a controversial periodization. Journal of Late Antiquity 1 (1): 4–19. Marchand, S. L. 1994. The rhetoric of artifacts and the decline of classical humanism: The case of Josef Strzygowski. History and Theory 33: 106–30. Maspero, J. 1913. Rapport sur les fouilles entreprises à Baouît. Comptes rendus des séances de l’Académie des Inscriptions et Belles–Lettres 43: 287–301. ———. 1931–43. Fouilles exécutées à Baouît, E. Drioton (ed.). Cairo. Matic, U. 2018. De-colonizing the historiography and archaeology of ancient Egypt and Nubia Part 1: Scientific racism. Journal of Egyptian History 11 (1–2): 19–44. Mayer, W. 2013. Religious conflict: Definitions, problems and theoretical approaches. In Religious conflict from early Christianity to the rise of Islam, W. Mayer and B. Neil (eds), 1–20. Berlin. McKendrick, S., D. C. Parker, A. Myshrall and C. O’Hogan (eds). 2015. Codex Sinaiticus: New perspectives on the ancient biblical manuscript. London. Meskell, L. 2005. Pharaonic legacies: Postcolonialism, heritage and hyperreality. In The politics of archaeology and identity in a global context, S. Kane (ed.), 149–71. Los Angeles, CA. Meyer, B. and D. Houtman. 2012. Introduction: Material religion—How things matter. In Things: Religion and the question of materiality, D. Houtman and B. Meyer (eds), 1–23. New York. Mondal, A. A. 2003. Nationalism and post-colonial identity: Culture and ideology in India and Egypt. London. Monkshood, G. F. 1911. A note upon ‘Thaïs’. In Thaïs, Anatole France, trans. by Ernest Tristan (ix–xii). New York. Morgan, D. 2020. Assembling inferences in material analysis. In The Wiley Blackwell companion to religion and materiality, V. Narayanan (ed.), 293–315. Hoboken, NJ. Moser, S. 2006. Wondrous curiosities: Ancient Egypt at the British Museum. Chicago, IL. ———. 2015. Reconstructing ancient worlds: Reception studies, archaeological representation and the interpretation of ancient Egypt. Journal of Archaeological Method and Theory 22 (4): 1263–1308. ———. 2018. The antiquities trade in Egypt 1880–1930. The H. O. Lange papers. Journal of the History of Collections 30 (3): 533–34.

Muhs, B. and T. Vorderstrasse. 2008. Collecting Egyptian antiquities in the year 1838: Reverend William Hodge Mill and Robert Curzon, Baron Zouche. Journal of Egyptian Archaeology 94: 229–40. Nativ, A. 2014. Anthropocentricity and the archaeological record: Towards a sociology of things. Norwegian Archaeological Review 47 (2): 180–95. Naville, E. 1894. The temple of Deir el Bahri: Its plan, its founders and its first explorers. London. Ness, L. 2015. Perspectives on early Islamic art in Jerusalem. Leiden. Nilsson Stutz, L. 2008. More than metaphor: Approaching the human cadaver in archaeology. In The materiality of death: Bodies, burials, beliefs, F. Fahlander and T. Oestigaard (eds), 19–28. Oxford. Nongbri, B. 2018. God’s library: The archaeology of the earliest Christian manuscripts. New Haven, CT. Noor, F. A. et al. 2019. Raffles in southeast Asia: Revisiting the scholar and statesman, S. A. Murphy, N. Wang, and A. Green (eds). Singapore. O’Connell, E. R. 2007. Transforming monumental landscapes in Late Antique Egypt: Monastic dwellings in legal documents from Western Thebes. Journal of Early Christian Studies 15 (2): 239–73. ———. 2008. Representation and self-presentation in Late Antique Egypt: ‘Coptic’ textiles in the British Museum. In Textile Society of America Symposium Proceedings. Paper 121. http://digitalcommons.unl.edu/tsaconf/121. ———. 2014a. The discovery of Christian Egypt: From manuscript hunters toward an archaeology of Late Antique Egypt. In Coptic civilization: Two thousand years of Christianity in Egypt, G. Gabra (ed.), 163–76. Cairo. ———. 2014b. R. Campbell Thompson’s 1913/14 Excavation of Wadi Sarga and other sites. British Museum Studies in Ancient Egypt and Sudan 21: 121–92. O’Connell, E. R. and G. Ruffini. 2017. The social networks of Late Antique Western Thebes. In Sinews of empire: Networks in the Roman Near East and beyond, H. Teigen and E. Seland (eds), 167–84. Oxford. Olsen, B. 2010. In defense of things: Archaeology and the ontology of objects. Lanham, MD. Olsen, B., M. Shanks, T. Webmoor and C. Witmore. 2012. Archaeology: The discipline of things. Berkeley, CA. Palme, B. 2012. Political identity versus religious distinction? The case of Egypt in the Later Roman empire. In Visions of community in the post-Roman world: The west, Byzantium and the Islamic world, 300– 1100, W. Pohl, C. Gantner and R. Payne (eds), 81–98. New York. Papaconstantinou, A. 2010. Introduction. In The multilingual experience in Egypt, from the Ptolemies to the ‘Abbāsids, A. Papaconstantinou (ed.), 1–16. Farnham. Parker, D. C. 2010. Codex Sinaiticus: The story of the world’s oldest Bible. London.

ARTEFACTS, ARCHAEOLOGY AND THE ARCHAEOLOGIST

Parker Pearson, M. 1999. The archaeology of death and burial. College Station, TX. Picchi, D. 2015. Amalia Sola Nizzoli (1805–45/49), Archaeologist Ante Litteram in Egypt and the origins of the Third Nizzoli Collection. In Every traveller needs a compass: Travel and collecting in Egypt and the Near East, N. Cooke and V. Daubney (eds), 131–42. Oxford. Plate, S. B. 2014. A history of religion in 5½ objects: Bringing the spiritual to its senses. Boston, MA. Pohl, W. 2010. Archaeology of identity: Introduction. In Archaeology of identity—Archäologie der identität, W. Pohl and M. Mehofer (eds), 1–13. Vienna. Prown, J. D. 1982. Mind in matter: An introduction to material culture theory and method. Winterthur Portfolio 17 (1): 1–19. Quibell, J. E. 1908–1912. Excavations at Saqqara, II–IV. Cairo. Raven, M. J. and J. H. Insinger. 2018. The most prominent Dutchman in Egypt: Jan Herman Insinger and the Egyptian collection in Leiden. Leiden. Rebillard, É. 2015. Late Antique limits of Christianness: North Africa in the age of Augustine. In Group identity and religious individuality in Late Antiquity, É. Rebillard and J. Rüpke (eds), 293–318. Washington, DC. Reid, D. 2002. Whose pharaohs? Archaeology, museums, and Egyptian national identity from Napoleon to World War I. Cairo. ———. 2015. Contesting antiquity in Egypt: Archaeologies, museums and the struggle for identities from World War I to Nasser. Cairo. Riggs, C. and J. Baines. 2012. Ethnicity. In UCLA encyclopedia of Egyptology, https://escholarship.org/uc/ item/32r9x0jr. Ross, M. C. 1942. A group of Coptic incense burners. American journal of archaeology 46 (1): 10–12. Rousseau, P. 2015. Can ‘Late Antiquity’ be saved? Marginalia: Los Angeles Review of Books. https://marginalia. lareviewofbooks.org/can-late-antiquity-be-saved-byphilip-rousseau/. Rutschowscaya, M.-H. 1995. Le monastère de Baouît – État des publications. In Divitiae Aegypti: Koptologische und verwandte Studien zu Ehren von Martin Krause, C. Fluck, L. Langener, S. Richter, S. Schaten and G. Wurst (eds), 279–88. Wiesbaden. Sauer, E. W. 2003. The archaeology of religious hatred: In the Roman and early Medieval world. Stroud. Scott-Moncrieff, P. D. 1913. Paganism and Christianity in Egypt. Cambridge. Shaw, W. M. K. 2003. Possessors and possessed: Museums, archaeology and the visualization of history in the late Ottoman empire. Berkeley, CA. Shortland, A. J., C. B. Ramsey, M. Dee and F. Brock (eds). 2013. Radiocarbon and the chronologies of ancient Egypt. Oxford.

107

Spencer, P. (ed.). 2007. The Egypt Exploration Society: The early years. London. Stevenson, A. 2014. Artefacts of excavation: The British collection and distribution of Egyptian finds to museums, 1880–1915. Journal of the History of Collections 26 (1): 89–102. ———. 2019. Scattered finds: Archaeology, Egyptology and museums. London. Stobiecka, M. 2016. Entangled objects: The place of archaeological artifacts in past collections and its influence on modern archaeological museums. Brukentalia. Romanian Cultural History Review. Supplement of Brukenthal. Acta Musei 6: 675–93. Swift, E. 2004. Dress accessories, culture and identity in the late Roman period: An introduction. Antiquité Tardive 12: 217–22. ———. 2017. Roman artefacts and society: Design, behaviour and experience. Oxford. Tanyeri-Erdemir, T. 2006. Archaeology as a source of national pride in the early years of the Turkish Republic. Journal of Field Archaeology 31 (4): 381–93. Thomas, T. K. 2000. Late Antique Egyptian funerary sculpture: Images for this world and the next. Princeton, NJ. ———. 2010. Egyptian art of Late Antiquity. In A companion to ancient Egypt II, A. B. Lloyd (ed.), 1032–63. London. Thompson, J. 1992. Sir Gardner Wilkinson and his circle. Austin, TX. ———. 2015a. Wonderful things: A history of Egyptology 1: From antiquity to 1881. Cairo. ———. 2015b. Wonderful things: A history of Egyptology 2: The golden age: 1881–1914. Cairo. ———. 2018. Wonderful things: A history of Egyptology 3: From 1914 to the twenty-first century. Cairo. Thornton, A. 2018. Archaeologists in print: Publishing for the people. London. Trafton, S. 2004. Egypt land: Race and nineteenth-century American Egyptomania. Durham, NC. Van Den Broek, R. 2013. Gnostic religion in antiquity. Cambridge. van der Vliet, J. 2006. Bringing home the homeless: Landscape and history in Egyptian hagiography. Church History and Religious Culture 86 (1–4): 39–55. van Minnen, P. 2010. Archaeology and papyrology: Digging and filling holes? In Tradition and transformation: Egypt under Roman rule. Proceedings of the international conference, Hildesheim, Roemer- and PelizaeusMuseum, 3–6 July 2008, K. Lembke, M. Minas-Nerpel and S. Pfeiffer (eds), 437–74. Leiden. Van Oyen, A. and M. Pitts (eds). 2017. Materialising Roman histories. Oxford. Veikou, M. 2016. Space in texts and space as text: A new approach to Byzantine spatial notions. Scandinavian Journal of Byzantine and Modern Greek Studies 2: 143–75.

108

D. L. BROOKS HEDSTROM

Versluys, M. J. 2017. Discussion. Object-scapes. Towards a material constitution of Romanness? In Materialising Roman histories, A. Van Oyen and M. Pitts (eds), 191– 200. Oxford. Voudouri, D. 2010. Law and the politics of the past: Legal protection of cultural heritage in Greece. Journal of Cultural Property 17: 547–68. Ward-Perkins, J. B. 1949. The shrine of St. Menas in the Maryût. Papers of the British School at Rome 17: 26–71. Warner, N. 2011. Somers Clarke: A pioneer of mud brick architecture in Egypt. In Hassan Fathy: Une ambition égyptienne, S. Santelli (ed.), 1–11. Cairo. ———. 2014. An architect abroad: The life and work of Somers Clarke in Egypt. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 68: 237–61. Webmoor, T. 2007. What about ‘one more turn after the social’ in archaeological reasoning? Taking things seriously. World Archaeology 39: 563–78. Webmoor, T. and C. L. Witmore. 2008. Things are us! A commentary on human/things relations under the banner of a ‘social’ archaeology. Norwegian Archaeological Review 41 (1): 53–70. Wendrich, W. 2012. Recognizing knowledge transfer in the archaeological record. In Archaeology and apprenticeship: Body knowledge, identity and communities of practice, W. Wendrich (ed.), 255–62. Tucson, AZ. Westerfeld, J. T. 2019. Egyptian hieroglyphs in the Late Antique imagination. Philadelphia, PA. Whittow, M. 2009. Early Medieval Byzantium and the end of the ancient world. Journal of Agrarian Change 9 (1): 134–53.

Wilfong, T. G. 2002. Women of Jeme: Lives in a Coptic town in Late Antique Egypt. Ann Arbor, MI. Willis, M. 2016. Vision, science and literature, 1870–1920: Ocular horizons. Pittsburgh, PA. Winlock, H. E. and W. E. Crum. 1926. The Monastery of Epiphanius at Thebes. 2 vols. New York. Wipszycka, E. 2009. Moines et communautés monastiques en Égypte (IVe–VIIIe siècles). Warsaw. ———. 2018. The second gift of the Nile, monks and monasteries in Late Antique Egypt. Warsaw. Witmoor, C. 2014. Archaeology and the new materialisms. Journal of Contemporary Archaeology 1 (2): 1–44. Wood, M. 1998. The use of the pharaonic past in modern Egyptian nationalism. Journal of the American Research Center in Egypt 35: 179–96. Yasin, A. M. 2005. Funerary monuments and collective identity: From Roman family to Christian community. Art Bulletin 87 (3): 433–57. Yonan, M. 2011. Toward a fusion of art history and material culture studies. West 86th: A Journal of Decorative Arts, Design History and Material Culture 18 (2): 232– 48. Zäh, A. 2012. Josef Strzygowski als Initiator der christlichkunsthistorischen Orientforschung. Römische Quartalschrift 107: 251–69. Zena, A. G. 2018. Archaeology, politics and nationalism in nineteenth- and early twentieth-century Ethiopia: the use of archaeology to consolidate monarchical power. Azania 53 (3): 398–416.

II CALENDAR AND COSMOLOGY

‘WE DESIRE TO KNOW WHICH IS THE TRUE RELIGION’: INTER-COMMUNAL RIVALRY AND THE VERDICT OF THE NILE IN AN EPISODE FROM THE HISTORY OF THE PATRIARCHS OF ALEXANDRIA John P. COOPER

That pre-modern Egyptian society depended for its prosperity and very survival on the success of the annual cycle of the Nile flood is a truism of published scholarship that is no less salient for that. Egyptians lived or died, prospered or were poor, experienced peace or upheaval, on the annual turn of the card that was the flood. It is, therefore, not surprising that this reliance on a phenomenon that was both mysterious in its origin and inconstant in its outcome fostered cosmological narratives that sought to explain, foretell and ultimately control the rise of the river and its consequences. It follows also that the political legitimacy of social elites was contingent on their ability to demonstrate that they enjoyed the favour of the divine architect—or architects—of the cosmos through both their custodianship of the rituals surrounding the Nile, and the delivery of the life-giving floods that those rituals were designed to ensure. This contingency was no less the case in the monotheistic Medieval period—based on non-native Abrahamic traditions—than it had been in polytheistic antiquity, when several indigenous deities had been associated with particular characteristics and mythologies surrounding the life of the river. This study focuses on a particular Medieval Egyptian narrative, of Christian authorship, that portrays a high-stakes and very public competition for religious legitimacy and communal authority between the Christian community and the Muslim elite, with Jewish and Samaritan minorities implicated for good measure. The narrative is found in what is commonly known in English as the History of the patriarchs of Alexandria (HPA): the underlying Arabic manuscripts bear several titles, including the Siyar al-bī῾a al-muqaddasa (Biographies of the Holy Church). This Arabic text is a composite work, presented in the form of a series of biographies of the successive heads of the Coptic Orthodox church, to which a number of authors, editors and scribes have contributed over the centuries. Several editions exist, but all have been shown to be methodologically unsound (see den Heijer and Pilette 2013, 109–11; Pilette 2013, 420–23; den Heijer, Lev and Swanson 2015, 343–44; den Heijer 2015, 454–67). The

first part, which includes the narrative with which this article is concerned, is based on earlier sources written in Coptic, and is often referred to as the ‘primitive recension’. The excerpts reproduced here are taken from Christian F. Seybold’s 1912 edition. Much more widely known, however, are the many later versions commonly referred to as the ‘Vulgate’ of the HPA, which—for the sections that concern us here—exist in editions by Seybold (1904–1910) and Evetts (1910). The latter of these is accompanied by an English translation: it is cited here alongside the English translations of excerpts from Seybold’s edition in order to facilitate comparison between the two. Our narrative is set in the year AD 753, in the wake of the tumultuous Abbasid conquest of Egypt. Its central event is the annual inundation of the Nile, and its theme the role that the inundation plays in communicating the disposition of God—whether favourable or otherwise—towards the beliefs, authority and legitimacy of Egypt’s principal communities. That the communal loyalties and religious affiliations of the author are ultimately vindicated in our narrative is, of course, unsurprising. What makes this story illuminating is that it offers us an insight into contemporary perceptions of the Nile as a vehicle of divine agency that is understood to be active in the political and communal affairs of the land. It should also be considered within the wider context of the Christian–Muslim theological encounter across Arabia, Mesopotamia, the Levant and Egypt in the first centuries after the Islamic conquest, during which these mutually encumbered religious communities learned more about each other and sought to differentiate, identify and justify themselves with respect to the beliefs and practices of the others (Thomas 2015). The HPA and its authorship Recent research has demonstrated that the HPA as an Arabic text was compiled from the earlier Coptic material towards the end of the eleventh century AD by the Alexandrian layman Mawhūb ibn Manṣūr ibn Mufarrij

112

J. P. COOPER

and his collaborators (den Heijer 1984, 346–47; 1989, 81–116; 1991b). Mawhūb was a member of the Coptic urban elite, and was close to both the Coptic church hierarchy and the Fatimid authorities (den Heijer 2015, 456, 483). His intervention, a radical departure in Coptic ecclesiastical historiography, is reflective of underlying socio-political and linguistic transformations under way in the upper echelons of Egyptian society at that time (Graf 1947, 2: 300; Farag 1977, 66; den Heijer 1991b; Rubenson 1996; Swanson 2010; Mikhail 2014, 6–8). From the mid-thirteenth century, the primitive recension translated and edited by Mawhūb (and later completed, in Arabic, by his successors) came to be systematically rewritten in order to address the cultural climate of a late-Ayyubid and early-Mamluk readership. It was during this process that it came to be attributed—spuriously—to a famous author of the tenth century, Sawīrūs (Severus) Ibn al-Muqaffa῾, bishop of Ashmūnayn. This anachronistic attribution was repeated in many of the manuscripts produced in subsequent centuries that together constitute the ‘Vulgate’—a term now rightly contested (Pilette 2013). This attribution persisted both in Egyptian and international scholarship for several centuries. In contrast, the attribution of the primitive version to Mawhūb remains unchallenged (e.g., Swanson 2010, 7 and passim; Mikhail 2014, 6–8 and passim; Mikhail 2016, 10–21 and passim). The authors of the Coptic sources informing it have been identified incrementally (Johnson 1977; den Heijer 1989, 1–13, 117–56). The author of the Coptic source containing the particular narrative with which this paper is concerned is believed to be a certain John (in the Arabic version, Yūḥannā), a Coptic Christian born in Giza who was a deacon at the time of the narrative, and who went on to become a bishop. With the assistance of two men, both called Maqārah, he wrote the biographies covering the period AD 705–68. John was a close companion of the patriarch Khael (Michael I) (r. AD 744–68), who is the main Christian protagonist in the narrative. Indeed, the text is written— at least with respect to the central events unfolding in Giza and al-Fusṭāṭ—from an eyewitness perspective, as if of John himself (Fig. 1). The palimpsestic nature of the HPA should not be forgotten: from the above we can see that the analysis here is based on a late-eleventh-century Arabic translation and heavy redaction of a lost Coptic original. The cultural frame of reference, terminology and style of Mawhūb’s writing are undeniably that of the Coptic

elite of Cairo and Alexandria. However, while we cannot regard the narrative under consideration here as a direct and unmediated window onto eighth-century life in al-Fusṭāṭ, we can nevertheless find in it evidence of Medieval communities fixated upon the Nile flood and its implications for both communal and food security. The narrative of the Nile flood Our narrative is from the forty-sixth biography of the HPA, which deals with the reign of the already mentioned patriarch, Khael. The events revolve around the Nile flood of the 471st year of the Coptic calendar, beginning in late summer AD 753, and in particular its delayed and erratic rise in the approach to the customary plenitude celebrations in the month of Thouth (11 September–10 October in the Gregorian calendar). Its political context is the aftermath of the Abbasid revolution, which swept through most of the Islamic provinces in 750. The text condemns the tyranny of the Umayyads in the dying days of their rule (Seybold 1912, 153–54, 157–58, 181, 184–85; cf. Evetts 1910, 92–95 [346–49], 101–3 [355–57], 153–54 [393–94], 162–64 [416–18]). His account of the overthrow of that dynasty immediately precedes our flood narrative, and contains glimpses of the themes to come. In this, the last Umayyad governor of Egypt, ῾Abd al-Malik, imprisons Khael in order to pressure him into surrendering wealth from the church to support the Umayyad campaign (Seybold 1912, 173; cf. Evetts 1910, 134–36 [388–90]). In response, God sends an earthquake, which causes widespread damage but touches no Coptic churches—a manifestation that prompts a change of heart from ῾Abd al-Malik, who releases the patriarch (Seybold 1912, 175–76; cf. Evetts 1910, 139–40 [393– 94]). This motif of a large geophysical event presented as a divine response to Muslim molestation of the church goes on to be repeated in our central narrative. Following the Battle of the Zāb in 750, the final Umayyad caliph, Marwān II, fled to Egypt; the text says he pillaged town and country on his way, and ‘wrecked and robbed’ monasteries. On his arrival in Egypt, Marwān finds Copts in rebellion at Shubrā and in the Delta marshlands, refusing to pay taxes. Unable to suppress the uprisings, Marwān plunders Alexandria and again imprisons the patriarch, once more demanding that he hand over wealth from the church (Seybold 1912, 184–81; cf. Evetts 1910, 154–60 [408–10]). As the invading Abbasid forces reach the east bank of the Nile,

‘WE DESIRE TO KNOW WHICH

IS THE TRUE RELIGION’

Fig. 1: Map of the Egyptian Nile showing locations mentioned in the text (Map: J. P. Cooper).

113

114

J. P. COOPER

Marwān keeps to the west side, all the while holding the patriarch, the bishop Moses—and, it seems, the author of the original narrative—as hostages. The clergy are freed only after the Abbasid rout of Marwān’s forces (Seybold 1912, 188; cf. Evetts 1910, 186 [440]). Here, the text relates the rapid response of the Nile to the news of the Abbasid defeat of the Umayyad oppressors: And the river rose from the first day of Misri, and rose every day about a cubit until it reached eighteen cubits that year, and for that reason people were saying: ‘the hand of God is with those people [i.e., the Abbasids]’. And because of the crosses that were with them, they used to announce [lit. to spread] in all districts that they would lighten the land tax for them (Seybolt 1912, 195–96; cf. Evetts 1910, 184 [438]; this author’s emphasis).

Clearly the narrative makes connections between the behaviour of the inundation and the disposition of God towards the combatants. Our narrative proper is set just three years later, by which time the Abbasids had consolidated their control and, in the condemnatory eyes of the HPA, ‘brought the country back into the state in which it had been under Marwān’. According to the text, in the second year of their rule, the Abbasids applied new imposts, and in the third ‘… they doubled the money and completed the land tax [imposed] on the Christians, and did not fulfil for them what they had promised to them’. The text reminds us, moreover, that the Abbasids ‘… forgot God who had given them the kingdom, and the Holy Cross which had given them the victory’. Perhaps worse, from the narrator’s perspective, is his observation that, as a result, ‘many of the poor and the rich denied Christ’ (Seybolt 1912, 196; cf. Evetts 1910, 188–89). According to John, God’s displeasure with the Abbasids is manifest in their failure in the Coptic year 470 (AD 752–753) to conquer North Africa, and in the infighting that ensued. In Egypt itself, our narrator detects further evidence of God’s displeasure with Abbasid rule in the form of a halt in the annual rise of the Nile when the Abbasid conqueror and first governor of Egypt, Abū ῾Awn Ibn ῾Abd al-Malik Ibn Yazīd (called Baūn or Ba᾿ūn in the HPA, possibly from the Coptic Paōn), returns from the stalled African campaign to Miṣr-al-Fusṭāṭ and its satellite settlements. It is here that our central narrative begins. John says that: And God hindered the water from rising while Ba᾿ūn was in Egypt; for its highest level [lit. the total of what it reached] was fourteen cubits, and [there] it stopped, whereas the measure required by [lit. of] the government

was sixteen cubits. And this was because of the deeds of the two comrades of the Antichrist. And the hindering of the water took place [lit. was] by God’s will, that he might show his wonders which he manifests at all times, and the truth of the religion of Christianity, and [of that of] the Muslims and others than them (Seybold 1910, 193–94; cf. Evetts 1910, 193–94 [447–48]).

The puzzling final words of the last sentence are as they appear in the primitive recension: later redactions simply omitted them. The measurements of the river were taken on the column of the nilometer at al-Rawḍa (Roda) Island, in the Nile alongside al-Fusṭāṭ (Fig. 2). The river was in a typical year expected to reach the crucial sixteen cubits level in mid-August (Popper 1951, 192). But in this year, it remained at fourteen cubits for several weeks, even until the Festival of the Cross, which took place on the seventeenth of the Coptic month of Thout, or 27 September in the Gregorian calendar. As late as the tenth century AD, this was the date on which the major seasonal canals of Egypt were opened, an event that started to dissipate the flood waters as they spread across the cultivated landscape (al-Muqaddasī; de Goeje 1906, 206; al-Mas῾ūdī; Barbier de Meynard and Pavet de Courteilles 1863, 2: 312, 364, 367; Cooper 2014, 117–23). According to the HPA, it was on this critical day in AD 753 that the church hierarchy gathered with the laity at the church of St Peter at Giza, on the west bank of the Nile, having cancelled their annual synod in Alexandria for the purpose. Significantly, we are told of this church that ‘its fundament was in the river’, and the crowd was so great that people ‘… found themselves [lit. were] in the fields and the places [around]’ (Seybold 1912, 199; cf. Evetts 1910, 194 [448], transl. 194–95 [448–49]). The narrative has Michael and his attendants carry crosses and Gospels to the riverbank before sunrise: And my father prayed, and Anbā Mennas [Mīnā] [as well]; and the crowd did not cease to shout Kyrie eleison until the third hour of the day, so that all the crowds marvelled at the cry of the crowd to God, the Exalted. And He heard [it], may His noble name [lit. mention] be elated, and the water rose by one cubit; and everyone glorified God and thanked Him (Seybold 1912, 199; cf. Evetts 1910, 195 [449]).

The text is explicit in portraying this act as a challenge to Abbasid authority: And when the news reached Ba᾿ūn, he marvelled and feared, he and his entire army. And God inspired him

‘WE DESIRE TO KNOW WHICH

IS THE TRUE RELIGION’

115

Fig. 2: The column and well of the nilometer at Roda in 2006 (Photo: J. P. Cooper).

[lit. cast in his heart] to say to his army and to the people of Miṣr: ‘We desire to know which is the true religion’ (Seybold 1912, 199; cf. Evetts 1910, 195 [449]).

Abū ῾Awn’s response to this challenge is to assemble the Muslims of Miṣr on the Muqattam Hills, to the east of al-Fusṭāṭ, and to pray thus: We desire to see your miracles today, which you work, that we may know and be convinced that there is no religion like our religion, which we inherited from our fathers; and we ask you to work for us a miracle, as you did for the Christians yesterday, who are the adversaries of our creed, who place with You another [god]—while you [exist] since the beginning—and [who] say that Christ, the son of Mary, that he is Your son, and the holy ghost, that who both are from You while You are the third of them. And much other talk they were saying, asking [God] to show them a miracle [lit. sign] in the water (Seybold 1912, 199; cf. Evetts 1910, 196 [450]; this author’s emphasis).

However, as the Muslims pray, ‘the water measurer’ runs to them and announces that: ‘As much as the water rose yesterday, it has sunk [lit. is lacking] today’. As a result ‘great sadness came upon them, and they did not know what to say’ (Seybold 1912, 199; cf. Evetts 1910, 196 [450]). The next day, the governor orders a second Muslim prayer attempt. For good measure, the HPA tells us that ‘the Jews and the Samaritans’ also went out to pray; the result, inevitably, was Nilotic indifference: ‘… the water neither rose nor sank’, prompting the chastened Abū ῾Awn to reflect that: ‘By the prayer of the Christians the water rose, and at our prayer it sank’ (Seybold 1912, 199–200; cf. Evetts 1910, 196–97 [450–51], transl. 196 [450]). On the third day, the exasperated governor orders that no one at all should pray. Inevitably, the river remains implacable, and the water level unchanged. Finally, with all alternatives exhausted, Abū ῾Awn has

116

J. P. COOPER

no choice but to accept the verdict of the river, and capitulate to those he had declared his enemy: After that he gave orders to bring forward the Christians, and a great multitude gathered towards him […] and he commanded my father Anba Moses to pray, him and his people, that they would recite the prayers, and they prayed and were giving thanks to God till the sixth hour of the day. And they went down and walked round Miṣr, and we came to the bank of the river, and prayed for the rest of the day. And during [lit. in] that night the water rose three cubits, until it reached [lit. became], eventually, seventeen cubits. Then all the people rejoiced greatly, and thanked the Lord Christ and glorified his name (Seybold 1912, 200; cf. Evetts 1910, 197 [451]).

Seventeen cubits was the optimal result—even better than the sixteen cubits level at which ‘plenitude’ was declared, irrigation made possible, and land tax applied to the harvest. The tenth-century AD historian and geographer al-Mas῾ūdī (Barbier de Meynard and Pavet de Courteilles 1863, 2: 363) reports that it enabled irrigation even of higher ground on the periphery of the normally irrigated area. The narrative concludes with a humbled Abū ῾Awn not only reversing his unjust imposts, but also seeking actively to support the Christian community: As for Baūn, because of that, he once more acted favourably towards [lit. did good with] the Christians and their churches, and lightened the land tax for them. And from that day the Father Patriarch and the bishops, and the Church and the Sons of Baptism, lived in [lit. were under] security, joy and gladness, in the land of Egypt and the Pentapolis [lit. the Five Cities] and all [uncertain reading] the places that are in the see of the Father, the evangelist, Mark, because of what the governor had seen in terms of miracles of the Church and the force of its deeds (Seybold 1912, 200; cf. Evetts 1910, 197 [451]).

Interpretation, context and impact Clearly, we should not be reading this narrative of an evidentiary miracle—absurd as it appears to the modern reader—as an objective account of events of the late summer of AD 753. But nor should we reject it out of hand for its legendary nature and improbable events, as positivist-minded historians such as Amélineau (1914) have done. Its symbolism is rich, and offers an insight into the preoccupations of the church authorities at the time of its writing. The HPA’s portrayal of a causal relationship between unjust and illegitimate rule and poor Nile floods is clear. The alleged

oppression of the Christian community by the Abbasid elite—no doubt regarded by John as presaging subsequent Abbasid abuse (Swanson 2010, 31–35)—not only provokes the river into not rising, but also into vindicating through its fluctuations the truth of Christian belief, the inefficacy of Islamic prayer, and the consequent authority of the Christian clergy as arbiters of the inundation. Abū ῾Awn’s putative lack of cosmological understanding is demonstrated through his selection of location for the Muslim communal prayer. His choice of the mountains may be a reference to the first Islamic revelation to the Prophet Muḥammad in a mountain cave, or to the sacred status that both Muslims and Christians ascribed to the al-Muqaṭṭam (Makhoul 2015, 287–90; Du Roy et al. 2018). It may also reflect an association of Arabs with the highground quarters of al-Fusṭāṭ (Gayraud 1998, 437–39). In any case, it is misdirected, away from the Nile. In contrast, the Christians pray in a church with its very foundations in the river and, later, on the river bank in Miṣr—thereby demonstrating their connectedness to and understanding of the flood cycle, and with the sanctity of that relationship embodied in a consecrated structure. The text emphasizes both the success and credibility of Christian ritual through the motif of the voluble crowds, which spill out of the church into its surroundings. Meanwhile, Muslims are seen to fail while praying alone, but also when joined in their endeavours by Jews and Samaritans. Crucially, Abū ῾Awn is explicitly made hostage to the notion that the river’s imminent responses will reveal ‘which of the religions is the true one’. In the end, he finds he has left himself no choice but to accept the verdict of the river—which turns out to be that God is on the side of the Christian Copts, and that their religion is demonstrably ‘the true one’. Abū ῾Awn’s endowment of church buildings is acknowledgement not only of this, but also of the indispensability of the Coptic Orthodox church in securing the future prosperity of Egypt. Moreover, it signifies a moment of reconciliation between the church and Abbasid rule after a difficult episode, in which the church was compelled to take action entailing confrontation with the governor by both circumstance and the duty entailed by its custodianship of the true cosmology. John regards such episodes as dangerous to the survival of the Christian community: Swanson (2010, 28–31) highlights his repeated insistence on the church’s obedience to the secular power.

‘WE DESIRE TO KNOW WHICH

IS THE TRUE RELIGION’

117

Although John does not make the parallel explicit, there are also resonances in this narrative with the biblical account of the confrontation between Moses and Aaron on the one hand, and pharaoh and his priests on the other, over the release of the Israelites from servitude in Egypt (Exodus 4–13). The community of the righteous—the biblical Jews, the Medieval Christians—are able to overcome the injustices of Egypt’s ruler—pharaoh, the Abbasid governor—by demonstrating through natural phenomena that the favour of God lies with them, in the end forcing the ruler to relent.1 Such a compelling tale of vindication retained traction in the Coptic historical tradition for some centuries. The History of the monasteries and churches of Egypt, a multi-layered text partly attributed to the twelfth-century Coptic priest Abū al-Mukārim—and formerly to the obscure Abū Ṣāliḥ, ‘the Armenian’— repeats the narrative in abbreviated form (Evetts and Butler 1895, 75–76 [transl. 175–76]; den Heijer 1993). It adds the extra details that the Christians prayed night and day, and fasted for a week, and puts into the mouth of Abū ῾Awn the final exclamation: ‘How does God receive the prayers of the Christians!’ The flood narrative finds itself replicated in several respects in the more famous evidentiary miracle of the moving of the al-Muqaṭṭam mountain, related by Bishop Michael of Tinnīs, who wrote the lives of the patriarchs from Kha’il II to Shenoute II (880–1046) in either AD 1051 or 1058 (den Heijer 1991a, 1241a; Swanson 2010, 49–52). In this case, enemies of the patriarch Abraham (r. 975–978) in the caliphal court call upon him to demonstrate the validity of the Christian faith, citing the Gospel of Matthew (17:20): ‘If you have faith as a grain of mustard seed, you will say to this mountain, “Move from here to there”, and it will move.’ In response, Abraham gathers the clergy and laity in three days of united fasting and prayer. Inspired by a saintly tanner, they then process together to the al-Muqaṭṭam mountain, calling ‘Kyrie eleison’. Watched by the caliph and his entourage, they prostrate themselves three times: the mountain lifts in the air and returns with each prostration. The caliph capitulates,

declaring ‘I have recognized, indeed, the correctness of thy faith’, and as a result allows the restoration of several churches, including that of St Mercurius (Swanson 2010, 49–52). The mode of presentation, in the Arabic text, of the flood narrative may very well have been informed by this narrative (cf. den Heijer 1994; Boutros 2013; Makhoul 2015). The translation from Coptic to Arabic, which has been demonstrated to be heavily edited and adapted, was done by the same scribes and redactors of all Coptic sources available to them in the late eleventh century AD.2

1

2

I am grateful to the Revd James Monger and Dr Alex Mallett for drawing my attention to this parallel.

Explaining the flood Completion of the immense Aswan High Dam in 1970 has done such a comprehensive job of erasing the annual phenomenon of the Nile flood from the Egyptian experience that it is difficult today to appreciate the immanence of its annual cycle in the past for the entire population. Before that intervention, Egyptians would have typically first noticed an incipient rise in river level in the Nile valley in late June. The surge would accelerate exponentially in July and August, and peak in early September. It would recede only slightly more slowly than it had risen, and normally would have dissipated by the end of December. At the height of the inundation, the flow of water entering Egypt at Aswan might typically have been twelve times greater than between March and June, when the river was at its annual low (Hurst 1952, 241). Modern perceptions of the Nile flood are couched in the methods and insights of the earth sciences. The cause of this great annual oscillation we understand today to be early summer snow melt and heavy rains falling in the distant mountains of Ethiopia and feeding into the Atbara and Blue Nile tributaries; and, to a lesser extent, of rains in the African Great Lakes region feeding the White Nile (Hurst 1952, 6–7; Shaheen 1985, 105–8; Springuel and Ali 2005, 349). In the past, however, the great distance of these generative events from the phenomenon experienced by Egyptians magnified its mystery and left ample scope for speculation

On this concept of ‘internal intertextuality’ in the HPA, with its use of specific expressions, formulae, themes, and topoi, see den Heijer 2015, 480–81.

118

J. P. COOPER

as to its cause. Adding to the curiosity of the flood is the fact that annual rainfall in the south of Egypt is almost zero; moreover, the flood gathered momentum just as the country entered its hottest, and otherwise driest, season. The Nile flood in cosmological context Naturally, ancient and Medieval Egyptians sought to make sense of this phenomenon. The polytheistic cosmology of ancient Egypt held that the inundation was an annual manifestation of Nun, the primaeval ocean, which emerged from underground caverns below the island of Biggeh, near Elephantine, at the southern limit of Egypt proper (Černý 1952, 50, 100; Meeks 1996, 17, 30; Zivie-Coche 2004, 46; Sheehan 2010, 30). This source was guarded by Khnum, god of the First Cataract (Hornung 1982, 79; Favard-Meeks 1996, 129). Meanwhile, texts from the Book of the Dead and the walls of the Chamber of the Nile at Edfu temple identify a corresponding ‘hidden source’ of the flood for Lower Egypt. Its ‘Great Chamber’ was inhabited by the Nile god, Hapi—androgynous, corpulent, and with pendulous breasts—who was associated with the inundation and the fertility it brought (Petrie 1924, 36; Quirke 1992, 50; Zivie-Coche 2004, 26, 156; Sheehan 2010, 30). Ancient Egyptians held that the immediate cause of the upwelling was the tears of the goddess Isis, lamenting the loss of her husband Osiris, who had been murdered by his brother Seth. Philae, on the first cataract, was the cultic centre of Isis in Upper Egypt: it was here Osiris was said to have been buried, and Egyptians believed that the flood welled up from beneath Isis’ feet as she wept (Witt 1971, 14–15, 23, 61, 64). Ancient understandings of the flood were not all mystical. Herodotus in the fifth century BC relates three ‘Greek’ hypotheses for the inundation that sought cause in physical phenomena. These were: that the flood was caused by the northwesterly etesian winds that blow across the Mediterranean in summer, pushing the river’s waters upstream; that the flood surged from the ocean ‘which flows round all the world’; or that it was caused by melting snow in Ethiopia. Herodotus’ rejection of these hypotheses is followed by his own rather opaque explanation that attributes the flood to the sun’s capacity to ‘draw’ water towards it, arguing in doing so that the Nile receives the water of other rivers, which, unlike the Nile, run low in the summer heat (Herodotus, Histories 2.19–26; Godley 1966, 297–305). This explanation allowed him, by the by, to

reconcile his essentially meteorological view with the contemporary religious explanation that the flood emerged through springs located near Elephantine (Herodotus, Histories 2.28; Godley 1966, 305–7). A similar concept—where the world’s rivers are connected by underground springs, and the Nile flood draws down other rivers—continued to be related in the ninth century AD by Ibn ῾Abd al-Ḥakam (Torrey 1922, 149), and in the twelfth or thirteenth in one of the layers of The history of the churches and monasteries of Egypt (Evetts and Butler 1895, 26–27 [transl. 67–68]). By the second century AD, Ptolemy Claudius was able to offer an explanation that echoes our modern meteorological interpretation: citing his predecessor Marinos of Tyre, he relates that a Greek trader, Diogenes, had travelled to the African interior from East Africa around AD 50 and found a snowy mountain range and two lakes which he concluded were the source of the river (Berggren and Jones 2000, 161–62). However, the emergence of such an earth-scientific understanding of the flood did not detract from its mythic power or pragmatic importance. The fate of Egypt remained in the balance each year, and so people still turned to the river in midsummer in anticipation of the first signs of increase, and thereafter to monitor the progress of that increase to the level needed to ensure full irrigation of agricultural lands. Since failure of the flood augured hunger, social disorder, lost taxation, and the increased vulnerability of the state to external threats, Egypt’s ruler was deeply implicated in the annual cycle. The connection between the success of the flood—a divine mechanism—and the legitimacy of the regime was made early on. In ancient Egypt, the pharaoh had been considered the ultimate officiant of religious ritual, responsible ‘… for the correct functioning of the temple and the cosmos’ (Zivie-Coche 2004, 99). As Zivie-Coche observes, it is the pharaoh, rather than the clergy, that is portrayed performing rites in temple reliefs. The temple priests were merely his or her vicars, who performed the role in his or her stead. The duty of pharaoh was further to ensure that Maat—order and justice—held sway in both society and the cosmos (Zivie-Coche 2004, 179– 83). In later, Graeco-Roman times, the living ruler became the direct object of cult; from the third century BC statues of rulers were placed in temples (Dunand 2004, 200, 203). Soon after the beginning of Roman dominion, the emperor Augustus was proclaimed in Philae as nothing less than ‘the Nile of Egypt, who inundates the land with food’ (Dunand 2004, 203),

‘WE DESIRE TO KNOW WHICH reflecting a more general Roman dominance of Egyptian religious practice (Dunand 2004, 211–22). Despite its formally subordinate position, the ancient priesthood played a vital role in generating and sustaining the legitimacy of the ruler. ‘Those who were in a position to testify to this legitimacy, which was both political and religious, were those in charge of the sacred traditions, that is to say, the priests’, says Dunand, who calls this a ‘well established tradition’, not least at times when kingship was being contested or a foreign dynasty installed, such as that of Alexander and his Ptolemaic successors (Dunand 2004, 199–200). The same is of course also true of the time of our AD 753 narrative. The nilometer The centrality of the inundation rituals in the ‘management’ of the cosmic order is reflected in the manner by which its progress was measured. Since the earliest times, this was achieved via a nilometer. These instruments were often—particularly in the pharaonic and Ptolemaic periods—associated with a temple precinct (Daressy 1915; Wild 1981, 25–26). The structure usually comprised: a well that was connected to the river; a scale, on which the river level was measured; and straight or spiral stairs leading down into it. Pharaonicera examples are attested archaeologically on Elephantine Island at Aswan, at the temples of Philae, Kom Ombo, Edfu, Esna, Karnak and Luxor, and beside the Serapeum in Alexandria (Borchardt 1906; Wild 1981, 26–34). In the early Roman era, Strabo (Geography; Jones 1917–32, XVII: 1.48) identifies nilometers at Elephantine Island and Memphis. As he and later authors make clear, its further function was to foretell the size of the harvest, and hence the revenues accruing from it (Popper 1951, 69–82). One clear religious trend in the Ptolemaic and Roman periods was the growth in the cult of the ‘new’ god Serapis, alongside Isis, and with it the Nile itself as an object of ritual. The cult spread across the GraecoRoman world (Wild 1981, 34–60). Scholars including Lindsay (1968, 268–69) and Wild (1981, 49–53) argue that crypts associated with temples of Isis and Serapis found in Italy and Greece would have held Nile water —or, perhaps, water emulating it—thus enabling even distant believers to partake in the annual miracle of the inundation. Texts from the Roman era underline the continued significance of Isis in the cosmology of the Nile flood:

IS THE TRUE RELIGION’

119

one papyrus calls the inundation ‘the rising of the freshet of Isis’ (P.Lond. Lit. 239, in Lewis 1983, 95). A hymn in her praise says: ‘By your power all the channels of the Nile are filled’, and ‘You induce golden-streamed Nile and lead him in due season over the land of Egypt’ (SEG VIII, nos 549 and 549, in Lewis 1983, 95). Serapis, meanwhile, was god of fertility and guarantor of the state. His pairing with Isis connected him with the bounty brought about by the inundation, and he was closely associated with the imperial supply of Egyptian grain to Rome (Lewis 1983, 86; Frankfurter 1998, 216–18; Dunand 2004, 42–48). Theological innovations in the Roman era appear to have been accompanied by a change in the form of the nilometer. Iconography of the Roman era—on coins, textiles and mosaics—depicts nilometers as comprising a column bearing a scale (Friedman 2008; Kupelian 2015, 183, 185). Only one nilometer conforming to this columnar type is attested archaeologically from the pre-Islamic period: this was found at the turn of the twentieth century at Kom el-Giza near Karyūn in the northwest Delta (Daressy 1900). It comprised a column inscribed with four scales that had Greek letters marking whole cubits: it was found in situ within the remains of a square well. This particular nilometer would have been vital to the management of the seasonal Schedia canal, which connected the Canopic branch of the Nile to Alexandria (Cooper 2014, 49, 52, 256, 295). Its column is today on display in the Egyptian Museum in Cairo. The overall form of the Kom el-Giza structure—a column set within a well—was to persist in the Islamic era in the form of the nilometer on Roda (al-Rawḍa) Island. The triumph of Christianity The close association and mutual contingency between the Nile inundation, the divine, and the ruler found in ancient, polytheistic Egypt succeeded in making the incremental transition to monotheism that took place during Late Antiquity. While the physical geographical understanding of the source of the Nile remained essentially that of Ptolemy (Evetts and Butler 1895, 34 [transl. 93–94], 127–28 [transl. 276]), the mystical genesis of the inundation required a new explanation to satisfy the new religion, with its focus on the one almighty God. In the late fourth/early fifth century, the abbot of the White Monastery, Shenoute, describes the annual flood as God’s ‘yearly mercy’. Meanwhile, his disciple and biographer Besa relates

120

J. P. COOPER

the account of a year in which the Nile flood did not appear as expected. Investing Shenoute with a clear intimacy with the divine, Besa says that his abbot ‘knew from God the hidden reason for it’. Shenoute orders his disciples to pray, and takes himself into the desert for a week to do the same, at the end of which the Nile rises. He tells his disciples: You know what I said to you, that God has commanded that there should be no flooding of the land this year. Behold, then, I prayed to him, and he, as the good and merciful God, promised me that this year again he would cause the waters to come and cover the face of the land (quoted in Oestigaard and Firew 2013, 34).

Again, the Nile is moved by the power of Christian prayer. Meanwhile, a more mundane report from an estate official among the Oxyrhynchus papyri captures the new association of the Nile with Christian belief. It reads: ‘I again bring the good news to your honor that the blessed fertilizing river of Egypt has risen by the power of Christ’ (P. Oxy. 1830, in Hardy 1952, 176). Meanwhile, the late-fourth to early-fifth-century AD monk and historian Rufinus gives an impression of the exhilaration—and sense of vindication—that Christians felt in wresting control of the Nile rituals from pagans. Writing of Alexandria, he says: Now it was the custom in Egypt to bring the gauge of the rising Nile River to the Temple of Serapis, as being the one who caused the increase of water and the flooding; so when his statue was overthrown and burned, everyone of course unanimously declared that Serapis, mindful of this injury, would never again bestow the waters in their usual abundance. But so that God could show that it was he who ordered the waters of the river to rise in season, and not Serapis, who after all was much younger than the Nile, there began then such a succession of floods as never before recorded. And thus the practice began of bringing the very measuring rod, or water gauge, which they call a pêchys, to the Lord of the waters in the church. When these events were reported to the pious sovereign, he is said to have stretched out his arms to heaven and exclaimed with great joy, ‘I thank you, Christ, that this age-old error has been demolished without harm to that great city’ (Rufinus 11.30; Amidon 1997, 87).

A broadly contemporary narrative by Eusebius of Caeserea tells of a similar vanquishing of the pagan clergy, deriding their masculinity for good measure. In his Life of Constantine, he writes: To those in Egypt and especially Alexandria, who had a custom of worshipping their river through the offices of

effeminate men, another law was sent out, declaring that the whole class of homosexuals should be abolished as a thing depraved, and that it was unlawful for those infected with this gross indecency to be seen anywhere. Whereas the superstitious supposed that the river would no longer flow for them in its customary way, God co-operated with the Emperor’s law by achieving quite the opposite of what they expected. For although those who defiled the cities by their abomination were no more, the river, as though the land had been cleared for it, flowed as never before, and rose in abundant flood to overflow all the arable land, by its action teaching the senseless that one should reject polluted men and attribute the cause of prosperity to the sole giver of all good (Eusebius 4.25 [I]; Cameron and Hall 1999, 161).

Despite the great transformation in the religious cosmology of the flood, the unchanged physical fundamentals of the Egyptian landscape—not least the cycles of the Nile—meant that much established praxis went unaltered into the new era. Christian Egypt retained the ancient Egyptian ‘civil’ calendar—albeit with its Roman-era ‘Alexandrian’ correction—which had as its new year an idealized date for the onset of the Nile flood. The extent to which the Egyptian year and its beginning was based mathematically on astronomical observation and/or average timings of the start of the Nile inundation has been debated (Winlock 1940; Neugebauer 1942). But there is no doubt that it was in essence structured to coincide with the Nile cycle, and the patterns of harvest and taxation that accompanied it. To this extent the ancient calendar remained useful, even though the polytheistic cosmology that accompanied it was clearly no longer acceptable. For example, the old religion had held that the year, and the flood, began when the star of Isis, Sirius, appeared over the horizon; this was the signal for the priests of the Philae temple to carry out the rituals of the inundation (Witt 1971, 14–23). Christians could do nothing about the timing or solar basis of the Nilotic cycle, but they could supplant—or at least co-opt and adapt—the polytheistic festivals that punctuated the calendar. The annual commemoration of the tears of Isis was displaced by the feast of St Michael, the archangel, celebrated on 12 Paone of the Coptic calendar, or 6 June in the Julian reckoning. Copts believed that, on that day, St Michael ‘asks the Lord’ to cause the river to rise: the introduction of this celebration, replacing temple ceremonies, is attributed to the emperor Constantine (r. AD 324–37) (Amélineau 1888, 1: 17, 19, 34). Other continuities with the pagan past persited. In the Medieval period at least, the night of the festival

‘WE DESIRE TO KNOW WHICH was known in Arabic as laylat al-nuqṭa, or ‘the night of the drop’, because a drop of water was believed to fall on that day from heaven into the river, thus initiating its rise. Parallels with the tears of Isis are clear. The day was also known as the day of ‘weighing the mud’, since a portion of mud was wetted with Nile water and weighed before being left overnight. It was weighed again in the morning, with the proportional change in weight being held to predict the size of the coming flood (Popper 1951, 68). Other annual ceremonies were also Christianized. Medieval Arab authors report that the day of the ceremony of the opening of the main irrigation and navigation canals of Egypt was in Christian times held on the day of the Festival of the Cross—17 Thout in the Coptic calendar, or 14–15 September in the Julian (al-Muqaddasī; de Goeje 1906, 206; al-Mas῾ūdī; Barbier de Meynard and Pavet de Courteilles 1863, 2: 312, 364, 367; Halm 2003, 64–68). Ancient Egyptians had formerly celebrated this entry of the Nile waters into the seasonal canals as the union of Isis and Osiris (Popper 1951, 85). Finally, in mid-January, on 11 Tobi of the Coptic calendar, Christian Egyptians celebrated the feast of Epiphany with communal submersions in the Nile. It was on this day that they believed Christ’s divinity was revealed through his baptism in the River Jordan. In the Medieval period, the festival became known as al-ghiṭās, which in Arabic means both ‘immersion’ in the mundane sense and ‘baptism’ in the spiritual (Basilios 1991; Kupelian 2015, 185–86; al-Maqrīzī; Sayyid 2002, 1: 718). The syncretic urge is apparent here both in the substitution of the River Nile for the River Jordan and in the implicit association of submergence in or by the Nile with Christian baptism. Another Christian (and indeed Jewish) association was between the Nile and the River Gehon, one of four rivers that the Bible says issue directly from Paradise (1. Gen. 2.13; Eccli., 24.27; Maguire 1999). This sacred connection is also made in a gloss that was inserted in later, probably thirteenth-century, versions of the HPA, and indeed in the very biography—that of Khael—from which our central narrative is taken (Evetts 1910, 169 [423]). The corresponding passage in the primitive recension mentions only ‘the river that is Gehon’ (Seybold 1912, 188). The Epiphany festival became a major public spectacle—as it would have been at the time of our narrative—until the Fatimid caliph al-Ḥākim banned the public revelry associated with it, perhaps because of

IS THE TRUE RELIGION’

121

the degree of participation by Muslims (al-Maqrīzī; Sayyid 2002, 1: 718, 2: 603–4; Walker 2009). One consequence of this prohibition was that Christians, especially in churches in Cairo, created immersion cisterns that are known as Epiphany tanks, so that the rituals could be continued out of public view (Grossman 1991): their similarity with the crypts found alongside the ancient temples of Isis and Serapis is striking. An example of an Epiphany tank survives at the church of St Shenoute in Old Cairo’s Abū Sayfayn monastery (Butler 1884, 1: 81). Just as Christians gradually took over the temples of Egypt in Late Antiquity, either through their demolition or conversion into churches (Frankfurter 1998, 265– 84), so too they wrested the function of the nilometers that were the ritual fulcrum of flood, faith and secular power. We have seen the report of Rufinus that custodianship of the ‘measuring rod’ used to assess the flood level was transferred from the Serapeum to a church, and its offices adapted to Christian practice. By the time of the Islamic conquest, it appears that Egypt’s principal nilometer was that at Memphis, the ancient city where previous pagan gauges had stood: it too took the form of a portable measure that was kept in a Christian church, and was carried to the river to take the measurements (Popper 1951, 5–7). Mamluk authors cite the lost work of al-Quḍā῾ī, who was writing before AD 1062, in saying that the device was a raṣāṣah, or ‘lead’, and therefore possibly some sort of sounding device (al-Qalqashandī; Anon. 1913–1922, 2: 298; see also al-Maqrīzī; Sayyid 2002, 1: 169; Ibn Taghrī Birdī; Anon. 1929–1972, 2: 309). At some point, even the foundation narrative of the nilometer as a measuring implement became entwined with the Abrahamic tradition: the Muslim Ibn ῾Abd al-Ḥakam in the ninth century AD asserts that the first person to establish a nilometer was the prophet Yūsuf (Joseph), this being at Memphis (Ibn ῾Abd al-Ḥakam: Torrey 1922, 16). The belief is repeated in the twelfth and thirteenth centuries in the Christian History and the churches and monasteries of Egypt (Evetts and Butler 1895, 42 [transl. 50]). The Arab conquest and ‘Coptic’ resistance When Arab forces led by ῾Amr ibn al-῾Ās conquered Egypt between AD 639 and 642, they encountered a society that had been overwhelmingly Christianized for more than two centuries, albeit with a notable Jewish minority. Whatever Christianity meant for the

122

J. P. COOPER

world-views of individuals in rural communities, institutionally it meant that public worship occurred in churches, and public ceremonial was Christian in ideology. The conquest installed a relatively small military elite in al-Fusṭāṭ that followed an emergent religion, Islam, with a quite different perspective on its monotheism, yet sharing many of the narratives of the JudeoChristian tradition. The indigenous population operated a complex irrigated agrarian system that was quite alien both to the experience and religious calendar of the new elite. Realizing the tight integration of Christianized cosmology, management of the agricultural cycle, and political legitimacy, the new elite was shrewd enough to leave well alone. The conquest was, after all, a military invasion by a small but highly effective army. It brought in its immediate wake neither a large settler-colonial population nor a cadre of bureaucrats. The elite was, therefore, dependent for civil functions on Egyptians, writing and communicating in Coptic or Greek, a pattern repeated throughout the empire (Donner 2005, 40–41). We have little contemporary historical material from the immediate aftermath of the conquest to inform us of the rituals, practices and monumental structures associated with the inundation. The central nilometer of ritual and taxation appears to have remained that of Memphis (al-Qalqashandī; Anon. 1913–1922, 2: 298), which continued as a prominent Christian centre and episcopal seat (Popper 1951, 5–6; Evetts and Butler 1895, 86 [transl. 200]; Yāqūt; Wüstenfeld 1866–1873, 1: VI.668). Ibn ῾Abd al-Ḥakam (Torrey 1922, 16) cites a report of Yaḥya Ibn Bukayr (AD 771/2–846) that, even in his lifetime, the Nile had been measured there daily by a ‘measurer’ (Ar. qayyās; cf. the quote from the HPA given above (p. 116) and reported to the governor in al-Fusṭāṭ. As we shall see, this was apparently after the Umayyad establishment of the ‘Tanūkhī’ nilometer on Roda island, close to the capital, but before Christian officials were stripped of their custodianship of the nilometer and its rituals. How a putative Memphis nilometer at this time fitted with the functions of the Roda nilometer is not at all clear. The Memphis nilometer was probably not the only gauge active at the time of the conquest, however: Ibn ῾Abd al-Ḥakam (Torrey 1922, 16) says there were others at Ikhmīm and Anṣinā in Middle Egypt, and al-Quḍā῾ī (cited in Yāqūt; Wüstenfeld 1866, 6: 611) says there had been one at the site of al-Fusṭāṭ’s ‘Clothes Bazaar’ until it was overtaken by the growth

of the town. Meanwhile, the Mamluk historian al-Qalqashandī appears to be citing al-Quḍā῾ī when he claims that the conquering ῾Amr ‘built’ (Ar. banā) nilometers at Aswan, Dandara and Anṣinā. As Popper (1951, 1) points out, this probably signifies some sort of refurbishment of existing structures rather than a new construction programme. In any case, we are as yet far from any evidence of the Arab Muslim elite expropriating any of the ritual functions of the nilometer. Inroads of Islamic government into the operation of the nilometer were to prove incremental. Al-Maqrīzī (Sayyid 2002, 1: 172) also cites al-Quḍā῾ī in claiming that the caliph ῾Umar ordered a recalibration of the nilometer scale no more than three years after the conquest in response to apparently unnecessary panic buying of food. If this anecdote is anything more than a reform motif, the effect in terms of taxation would rather have been a tightening of the screw: since the scale was adjusted upwards by two cubits, this would have shifted Nile levels that were previously at fourteen–sixteen cubits to sixteen– eighteen cubits, and so above the sixteen-cubit plenitude level that triggered country-wide taxation. Clearly, given the revenue implications of the nilometer’s function, Egypt’s Arab rulers were keen to establish an element of control over the structure, at least with respect to taxation and the maintenance of social stability: the recalibration attributed to ῾Umar is indicative of that. But that desire for control soon also acquired a spatial dimension. In 80 AH (AD 699/700), apparently in response to an epidemic, the Umayyad governor of Egypt, ῾Abd al-῾Azīz ibn Marwān (r. AD 685–705), relocated the Egyptian capital from al-Fusṭāṭ to Ḥulwān, some 18km upstream (Yāqūt; Wüstenfeld 1866, 1: 321; al-Kindī; Naṣṣār 1959, 71). According to the historian and biographer al-Kindī (Naṣṣār 1959, 71), ‘He built houses and mosques and such like of the strongest construction’. Ibn ῾Abd al-Ḥakam (Torrey 1922, 36) reports that one of Ḥulwan’s public buildings was a nilometer: hence a gauge had, for the first time in Islamic Egypt, been established within the precincts of Islamic political power. The close spatial relationship between the gauge and the seat of Islamic government within Egypt was never to be broken again. Seventeen years later, in AD 715—the final year of the reign of the caliph al-Walīd ibn ῾Abd alMalik (r. 705–15), and the first of his brother Sulaymān—after the Islamic capital had returned to al-Fusṭāṭ, the Roda nilometer was established by Egypt’s head of kharāj, or land taxation, Usāmah bin

‘WE DESIRE TO KNOW WHICH Zāyid al-Tanūkhī. The Mamluk chronicler Ibn Taghrī Birdī (Anon. 1929–1972, 2: 130) claims that other nilometers across Egypt were abandoned once this Roda gauge was established—thereby concentrating the administration of the flood at the centre of government. While this may be formally true, it is unlikely to have been the case in practice. The incorporation of the nilometer into the Islamic capital did not happen in isolation. Al-Kindī (Naṣṣār 1959, 80) relates that ῾Abd al-῾Azīz bin Marwān’s nephew and successor ῾Abd Allāh bin ῾Abd al-Malik (r. AD 705–9) ordered that the language of Egypt’s bureaucracy—formerly Greek and Coptic—should henceforth be Arabic (Papaconstantinou 2010, 196). The immediate reason was a desire to reintegrate a province into Umayyad government that ῾Abd al-῾Azīz bin Marwān had run as something of a personal fiefdom for two decades. But it also reflects a certain confidence among the elite in the security of its position, and its growing integration into Egyptian society. That said, papyrus evidence from the final year of ῾Abd al-῾Azīz’s governorship, as well as from the era of his successor, Qurra bin Sharīk (r. AD 709–14), demonstrates that the reality was an unresolved situation of ongoing multilingualism—with texts written in Arabic, Coptic and Greek—rather than reflecting the orchestrated switch to Arabic that al-Kindī would have us believe (Papaconstantinou 2010, 196). In the meantime, the incorporation of the nilometer into the spatial nexus of Islamic political power in Egypt was not yet a ‘conversion’ of its offices and rituals to reflect Islamic cosmology and scriptural belief: operation of its structure had come under the scrutiny of the Arab elite, but it remained in the hands of Christians. Again, we know little of the practices and ceremonies surrounding the nilometer in the Umayyad period. But given that its offices were conducted by Christians, we can infer that it was not yet positively Islamic in its ritual or material character. With conversion to Islam yet to gain momentum within the Egyptian population, the public performance of the rites of the Roda nilometer—and no doubt others around Egypt—had perforce to maintain a Christian character in order to secure the confidence of the populace, and through it the legitimacy of the Muslim elite. To meddle with the rituals of the nilometer would have been to tamper with the fundamentals of popular Egyptian belief: as Egyptians knew, that way lay the failure of the inundation, and the terrible consequences that attended it.

IS THE TRUE RELIGION’

123

Political context The flood narrative presented in this chapter is set—and in its lost original form was composed—in the middle of the eighth century AD, a period of ongoing tension between Egypt’s Muslim rulers and its still-majority Christian population, during which the first major political and institutional reforms and transformations of the Islamic era were implemented. These began at the end of the seventh century AD with the changes arising from the centralizing agenda of the Marwānid caliphs, including ῾Abd al-Malik’s monetary reform of AD 694; standardizations of weights and measures; and the edict, already mentioned, that established Arabic as the official language of administration. Maged Mikhail (2014, 106) claims at this time that ‘… the seeds of permanent cultural change were sewn within the span of a single extraordinary century’, although the process was sporadic, and it would take a number of centuries yet before that cultural change achieved primacy. Nevertheless, the linguistic switch to Arabic began to exclude non-Muslims from administrative positions, and the proportion of Arab Muslims ‘at all levels of the bureaucracy’ grew significantly (Mikhail 2014, 116). Another aspect of this change was a re-evaluation of the land taxation conditions agreed or imposed in Egypt by the new Arab elite at the time of the conquests (Morimoto 1981, 54–144; Mikhail 2014, 108–13). For Egyptians, this meant that the country as a whole was identified as having been conquered, rather than according to local traditions that categorized only some areas as such, with still others acknowledged as reduced by treaty. The net result was a mounting tax burden for many, including Christians, Muslims and others. What emerges from this period was, as Mikhail puts it, ‘… a hegemonic program that effected every aspect of Egyptian culture along with the whole of the population: Muslims, Christians and Jews. Arab Muslim identity began to permeate the public sphere on nearly every facet…’ (2014, 118). Perhaps not surprisingly, this programme, accompanied by a mounting tax burden, provoked a series of broad-based ‘Coptic’ rebellions, in Upper Egypt in AD 697 and 712, and in the eastern Delta in 726, and in 739–40. As Abbasid rule took hold, rebellions took place in AD 750, including the Bashmūr uprising in the eastern Delta, and in the central Delta, around Samannūd, in the year before our narrative is set, encountered by a fleeting Marwān.

124

J. P. COOPER

Such rebellions, though by no means exclusively Christian in their particiation, continued well into the ninth century (Mikhail 2014, 118–27). The constitution of these rebellions in many ways reflected the increasing mixture of Egyptian society, with Egyptians of all confessions joining forces against crippling taxation. However, from the point of view of the Christian establishment—whence our narrative emerges—they were part of a wider ideological struggle with the Arab regime in which the clergy was engaged on cosmological and ritual fronts for control over the ritual landscape, in which the ‘blessed’ Nile played the central role. If the Nile was the ritual focus of Egypt’s annual existential drama, then al-Fusṭāṭ was the stage on which the country’s political narratives were played out: it is hence a significant backdrop to John’s account. The city was founded in the wake of the Arab-Islamic conquest as Egypt’s new capital, alongside the Roman fort of Babylon. Under Roman rule, the latter had been ‘primarily a strategic military post’ (Sheehan 2010, 88), with Alexandria, some 180km away, constituting Egypt’s political capital. Its fall to the forces of ῾Amr ibn al-῾Ās in AD 640 proved to be a blow from which the Byzantines did not recover. The subsequent foundation of al-Fusṭāṭ, on the model of military garrisons (amṣār) established throughout the newly conquered Islamic lands, established a new political and sectarian geography around the nucleus of the old fort. The elite established itself around the mosque of ῾Amr, to the north of the fort (Kubiak 1987, 96); Sheehan (2010, 87–88) interprets the disappearance of the northern wall of the fort as an act of deliberate post-siege demolition intended to join this new area with the former imperial quarters of the Roman fort. However, the remaking of Babylon was not all Arab-Islamic in character: Sheehan’s archaeological investigations at the site have revealed that the Copts took the opportunity of the end of Byzantine religious persecution to establish a number of new churches in the southern part of the fort, on the opposite side of the Via Principalis to the precincts of Arab power and administration (Sheehan 2010, 85–92). It was to these churches—first that of Abu Serga and later the Hanging Church—that the election and enthronement of the Coptic patriarch had shifted by the time of our narrative. Hence, at al-Fusṭāṭ, the centres of Arab and Coptic power in Egypt sat cheek by jowl.

Islamic counter-narratives Protestations of legitimacy such as the HPA’s telling of the events of AD 753 were, of course, not only the preserve of Christians. A century later than John the Deacon, the lost Coptic source of the narrative, was active, the Muslim Egyptian historian Ibn ῾Abd al-Ḥakam was compiling and editing narratives from three centuries earlier than his time—in his case, those of the Islamic conquest of Egypt (AD 639–42) and North Africa. Just as the HPA narrative seeks to demonstrate the legitimacy and authenticity of Christian stewardship of the Nile, so Ibn ῾Abd al-Ḥakam seeks to demonstrate the need for a corrective Islamic intervention against alleged wayward Christian practices— echoing, by the by, the purifying sentiment of Rufinius and Eusebius, four-and-a-half centuries earlier. A narrative in Ibn ῾Abd al-Ḥakam’s Futūh Miṣr relates that, soon after the conquest, the people of Miṣr approach ῾Amr Ibn al-῾Ās to report to him that the Nile has failed to rise at the appointed time. They tell him that the customary solution is to take a virgin girl from her parents, dress her in fine clothing, and throw her into the river (Torrey 1922, 150–51). ῾Amr responds firmly that: ‘This is not in Islam, and Islam vanquishes what came before’. The people seek clarification of what should be done, and ῾Amr writes to Medina for advice from the Caliph ῾Umar, who responds by sending a note (Ar. biṭāqah) that he instructs ῾Amr to throw into the Nile. On the note is written: From the Servant of God ῾Umar, Commander of the Faithful, to the Nile of the People of Egypt: You flowed before, and you do not now. It was the victorious One God who made you flow before, so we ask the victorious One God to make you flow [again].

In public view, ῾Amr throws the note into the river on the day before the Festival of the Cross. Overnight, the river reaches the plenitude mark, and all is well: ‘[God has] cut ill off from the people of Egypt this year’. Ibn ῾Abd al-Ḥakam goes on to deploy a parallel narrative in which the prophet Moses (Mūsā) is the protagonist and harks back to the Exodus narrative already discussed. In it, the pharaonic family are unwilling to heed the preaching of Moses, so God in response holds back the waters of the Nile. The pharaonic family ask Moses to pray to God, which he does, and next day the river flows at sixteen cubits. The narrative concludes: ‘And, after all that time, God responded to ῾Umar Ibn al-Khaṭṭāb just as he responded to his

‘WE DESIRE TO KNOW WHICH prophet, Moses (Peace Be Upon Him)’ (Torrey 1922, 151). Again, echoing Rufinus and Eusebius, Ibn ῾Abd al-Ḥakam presents the response of the Nile as a vindication of the truth of the new religion of Islam, and the waywardness of pre-existing Christian belief. Moreover, Ibn ῾Abd al-Ḥakam uses the narrative of Moses in Egypt to lay claim both to a key figure within the Christian tradition, and to the river as part of a Muslim sacred geography. Whether the casting into the Nile of young girls related in Ibn ῾Abd al-Ḥakam’s first narrative ever took place is not the point: its function here is probably to resonate with Qur᾿ānic verses prohibiting the killing of children in times of poverty (17.31), and of unwanted girl children (16.58–59; 81.8–9) and, in doing so, condemning these alleged Christian practices as barbaric.3 It is interesting, however, that he cannot shake off the fundamental idea that something must be thrown into the Nile in order to secure the inundation. In his time, the Christian Festival of the Martyr (Ar. ῾Īd al-shahīd) continued to be celebrated at the northern Cairo suburb of Shubrā on the eighth day of the Coptic month of Bashans, or 4 May in the Julian calendar. In a rowdy festivity, the relic of a saint was cast into the river to that end. The festival was abolished in AD 1302 for reasons of public order, and again in AD 1354 after a short-lived revival (Popper 1951, 69). What this narrative also inadvertently highlights is the syncretic nature of Nilotic belief, with ‘Umar’s note replacing the putative unfortunate girl. Marcel claims in the Description de l’Égypte that the sacrifice of a girl had originally been an offering to Serapis (cited in Popper 1951, 68). In still earlier times, offerings had been cast into the Nile in honour of the inundation god Hapi (Mercer 1949, 186; Spence 1990, 170; Quirke 1992, 50; Wilkinson 2003, 106–8). This Islamic preservation of pre-Islamic practices through adaptation is also evident in the custom of perfuming the nilometer column. In the Fatimid period, the caliph himself mixed saffron and musk in a cup, and, during a lavish public celebration, gave it to the guardian of the nilometer, who proceeded to jump into the water of the nilometer well, swim to the

3

For a semiotic approach to Ibn ῾Abd al-Ḥakam’s narrative, see Louca 1981.

IS THE TRUE RELIGION’

125

central column, and anoint it with the mixture using his right hand, while others recited Qur᾿ānic verses (al-Qalqashandī; Anon. 1913–1922, 3:516; Ibn Taghrī Birdī; Anon. 1929–1972, 2: 481; al-Maqrīzī; Sayyid 2002, 1: 476). While this perfuming practice has Islamic parallels, it also reflects consecration practices in Coptic churches, and even the ancient Egyptian practice of anointing statues (Popper 1951, 73). The triumph of Islam The final Islamic wresting of control over the ceremony and religious ideology of the nilometer—and indeed of the material cosmology of the monument itself—did not come about for another century and a half after the establishment of the Roda nilometer. In the year 247 AH (AD 861), the Abbasid caliph al-Mutawakkil ῾Alā Allāh (r. AD 847–61) ordered a fundamental reconstruction of the Roda nilometer, overseen by Muḥammad al-Ḥāsib al-Qurṣānī (Ibn Khallikān, in Toussoun 1925, 310 ff; al-Kindī; Guest 1912, 507; al-Maqrīzī; Sayyid 2002, 1: 170). The circumstances and historiography of this reconstruction are detailed extensively by Popper (1951, 19–25). Two things are of particular interest to our case. First, it was with this reconstruction that Christian officiates were, by order of the caliph, finally stripped of their role as custodians of the nilometer: in their place was appointed a Muslim man called Abū al-Raddād, whose office became hereditary (Popper 1951, 19–20). Second, the nilometer was reconstructed as an unambiguously Islamic polemical monument that exuded a specifically Muslim theology and cosmology. The caliph ordered that his name be inscribed on the building, along with suitable verses of the Qur᾿ān. According to al-Ḥāsib (Popper 1951, 49–50), people approaching the nilometer were met with two inscriptions on either side of the outer gate. On one side were two verses from the Qur᾿ān, which challenge the visitor against disbelief in the beneficent creator God: His companion said to him, in the course of the argument with him: ‘Dost thou deny Him Who created thee out of dust, then out of a sperm drop, then fashioned Thee into a man?’ (Sūra 18 [The Cave]: 37 tr. ῾Alī 1989, 719).

126

J. P. COOPER

Yet when We bestow Our favours on man, he turns away and becomes remote on his side (instead of coming to us), and when evil seizes him he gives himself up to despair! (Sūra 17 [The Night Journey]: 83 tr. ῾Alī 1989, 697).

The reference in the first of these verses to creation from a drop can be read as a reference to the popular belief, already mentioned, that the flood was instigated by a drop falling from heaven, itself a reformulation of the tears of Isis narrative. The second seems to remind the reader of the blessing that the flood constitutes. On the other side of the outer gate was the first of three mentions of the caliph: an accompanying statement read that, in the year of the construction of ‘the blessed Mutawakkil nilometer’, the Nile reached a bountiful seventeen cubits and eighteen digits—thus asserting to the visitor the divine affirmation that the structure and its secular patron had received on its inauguration. Once through the gate, a sixty-six-word inscription above the nilometer door incorporated the names of al-Mutawakkil and al-Ḥāsib, and recorded the year, 247 AH, reminding the visitor of the secular patronage underpinning what they were about to encounter within. Inside the nilometer itself, the crossbeam securing the capital of the central column carried the first inscription to be encountered. This was the familiar Throne Verse (sūra 2 [The Cow]: 255), which lauds the omnipotence and omniscience of the Almighty: Allah!—There is no god but He—the Living, the Selfsubsisting, Eternal. No slumber can seize him, nor sleep. His are all the things in the heavens and on the earth. Who is there can intercede in His presence, except as He permitteth? He knoweth what (appeared to His creatures as) before or after or behind them. Nor shall they encompass aught of His knowledge, except as he willeth. His throne doth extend over the heavens and the earth, and he feeleth no fatigue in guarding and preserving them. For he is the Most High, the Supreme (in glory) (tr. ῾Alī 1989, 105–6).

Immediately below the crossbeam, at the level of cubit eighteen, a platband running continuously around the walls of the square well carried an inscription that began with sūra 14 (Ibrāhīm), verse 37, which reminds the visitor of the barren nature of the Makkan landscape, and the duty of others—implicitly, in this situation, fertile Egypt—to supply food to its inhabitants, who were the descendants of Ibrāhīm, so that they can sustain the prayers there:

O our Lord, I have made some of my offspring to dwell in a valley without cultivation, by Thy Sacred House; in order, O our Lord, that they may establish regular prayer. So fill the hearts of some among men with love towards them, and feed them with fruits, so that they may give thanks (tr. ῾Alī 1989, 614).

Deployment of this verse reintegrates the rituals of the Nile into the task of maintaining the cosmic order: not directly, as before, but now in sustaining the divinely ordained rituals of Makkah. The inscription continues with the name of the caliph, ‘builder of this Hāshimī nilometer’, followed by that of al-Ḥāsib, and the date: Rajab, 247 AH (10 September–9 October AD 861). At sixteen cubits, a second, uninscribed, platband marked the plenitude level. Above it, at seventeen cubits, four Qur᾿ānic sūras were inscribed, one on each wall, which al-Ḥāsib says ‘became lines on the surface of the water when it reached seventeen cubits, for this is the average rise’. The inscriptions of al-Ḥāsib survive, although they do not reproduce the entire verse in each case: 1) That on the east wall, facing the entrance, emphasizes that it is God who causes the flood to happen: [In the name of God the Compassionate, the Merciful:] And We send down from the sky rain [water] charged with blessing, and We produce therewith gardens and grain for harvests (Sūra 50 [Qāf]: 9; tr. ῾Alī 1989, 1347).

2) The inscription on the north wall again asserts the agency of the Islamic God in sending the ‘swelling’ water that irrigates the harvest: But when We pour down rain [water] on it, it is stirred, it swells, and it puts forth every kind of beautiful growth in pairs (Sūra 22 [The Pilgrimage]: 5; tr. ῾Alī 1989, 822).

3) On the west wall was verse 63 from the same sūra: Seest thou not that Allah sends down the rain [water] from the sky, and forthwith the earth becomes clothed with green? For Allah is He Who understands the finest mysteries, and is well-acquainted (with them) (tr. ῾Alī 1989, 838).

4) Finally, the inscription on the south wall emphasizes God’s beneficence in sending the rain, while referencing the anxiety and uncertainty that was part of the annual Nile flood cycle: He is the one that sends down rain (even) after (men) have given up all hope, and scatters His Mercy (far and wide). And He is the Protector, Worthy of all Praise (Sūra 42 [Consultation]: 28; tr. ῾Alī 1989, 1254).

‘WE DESIRE TO KNOW WHICH These four inscriptions survive today. Meanwhile, the inscription of the platband at cubit eighteen has been changed, if the account of al-Ḥāsib is to be believed, with the part containing the names of al-Mutawakkil and al-Ḥāsib and the date replaced—presumably by a successor regime—by the following Qur᾿ānic verses: It is He who sends down rain [water] from the sky. From it ye drink, and out of it the vegetation (grows) on which ye feed your cattle (16 [The Bee]: 10; tr. ῾Alī 1989, 639). And We have distributed the (water) amongst them, in order that they may celebrate (our) praises, but most men are averse (to aught) but (rank) ingratitude (25 [The Deliverance]: 50; tr. ῾Alī 1989, 901).

Conclusion The density of inscription on al-Mutawakkil’s ‘new’ nilometer at Roda represents a triumph of Islamic cosmology and theology over the sacred landscape of the Egyptian Nile. Coming barely a century after the ascribed date of the stand-off between the Christian clergy and Muslim authorities, it is surely indicative of the extent to which the centralizing interventions of the late Umayyad and early Abbasid imperial governments had transformed the monumental and ceremonial religious culture of Egypt from explicitly Christian to explicitly Muslim—at least in its formal public manifestations. The Arab-Islamic conquest itself had had an almost immediate impact on the geopolitical landscape of the Nile with the re-excavation in the early 640s of the Roman Amnis Traianus—renamed the Canal of the Commander of the Faithful—from al-Fusṭāṭ to the Red Sea at al-Qulzum (Suez) (Cooper 2008, 198; 2014, 230–31; Trombley 2008, 99). But this was a politically and economically motivated act that did not challenge the prevailing indigenous Nile cosmology. What is apparent from the structure of the nilometer of al-Mutawakkil is that the Abbasid state, by the mid-ninth century AD, felt itself confident enough to assert in monumental and administrative form the authority of Islam as the temporal and spiritual lens through which Egyptians should experience the cycle of the Nile. While this was clearly a top-down, imperial initiative in its genesis, the conversion of the nilometer to reflect a Muslim world view can only have been achieved with the tacit consent, or at least resignation, of a significant part of the broader population, given that the nilometer persisted as the ritual focus of popular anxieties over the flood, and of the relationship between the ruler, the

IS THE TRUE RELIGION’

127

religious establishment, and the ruled. This transformation was to be taken even further in the Fatimid period, when the caliph al-Mustanṣir bi-llāh incorporated the nilometer into a new adjacent mosque complex—echoing the ancient relationship between nilometer and temple. With Islam supreme with respect to ritual ownership of the Nile, the Coptic clergy had to make the best of things. Although ejected from the rituals surrounding the principal nilometer of Egypt, they continued to pursue a ritual relationship with the Nile by alternative means. The Epiphany tanks already mentioned are one manifestation of this. Elsewhere, practices continued to pursue the measuring function of a formal nilometer, albeit with a predictive dimension. To this day, at the village of Ishnīn al-Naṣāra, near al-Bahnasā (ancient Oxyrhynchus) on the Baḥr Yūsuf waterway in Upper Egypt, the church of St George has a well that was in former times the centre of Nile-measuring rituals. This may be the same well that The history of the churches and monasteries of Egypt says was at that time in the Bīsūs monastery near Ishnīn, and which was used to predict the flood: Christ visited it, and a church was built [lit. became] there in that monastery. And in the middle of it is a well of running water, over which prayers are said during the rise of the Nile every year, and then the water in the well rises. And in it there are marks of cubits of the Nile, contrived, and when the water of the well rises and stands still at one of the marks, it is known thereby up to what height the rise of the Nile will reach, in terms of cubits (Evetts and Butler 1895, 96 [transl. 220]).

It may also be the well to which the late Mamluk historian Ibn Iyās refers, citing Ibn ῾Abd al-Ḥakam: In some of the areas of al-Bahnasā … there is a hamlet called Manīl Abū Sha῾rah. In it there is a well called the Well of ῾Īsa [Jesus]. In this well are graded steps. On the twenty-fifth day of Bashans (20 May in the Julian Calendar), at night, the water enters that well, and however many steps are covered by the water, that will be the [extent of the increase of the] Nile that year. Hence the extent of the flood is known. This mystery continues at this well to this day, and it has never lapsed or failed, for it is thus. It is said that ῾Īsa … washed from that well on that night, and this mystery has come to persist to this day’ (Ibn Iyās; Muhammad et al. 1931–1992, 1: 1.22).

What John’s narrative of the events of AD 753 demonstrates is how the ideological struggle between Muslims and Christians for religious vindication was played

128

J. P. COOPER

out through the natural phenomenon of the Nile flood, just as similar contestations had been pursued in previous centuries between Christians and pagans. The sense of anticipation and jeopardy surrounding it inevitably made it a high-stakes drama in which all had a part. Controlling the interpretation of the flood, whether through narrative or through control over its administration and central monument, the nilometer, was therefore a risky affair. The transition process we see in the Medieval period from Christian to Muslim control of the nilometer and its rituals is of course specific to its own time and religious contexts. But, given the deep syncretism of the Nile rituals, we may also draw on the experience of the Christian–Muslim transition in imagining the transitions of earlier eras, as well as appreciating the profundity of the pre-modern relationship between Egyptians and the river that sustains them.

Bibliography All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed 8 February 2019). Epigraphic abbreviations follow are cited according to F. Bérand et al. 2000. Guide de l’épigraphiste: Bibliographie choisie des épigraphies antiques et médiévales. Paris. ῾Alī, A. Y. 1989. The Holy Qur’ān: Text, translation and commentary. Brentwood, MD. Amélineau, É. 1888. Contes et romans de l’Égypte chrétienne. 2 vols. Paris. ———. 1914. Les derniers jours et la mort du khalife Merouân II, d’après l’Histoire des patriarches d’Alexandrie. Journal Asiatique 11 (4): 421–49. Amidon, P. R. (ed.). 1997. The church history of Rufinus of Aquileia: Books 10 and 11. New York. Anonymous (ed.). 1913–22. Al-Qalqashandī: Ṣubḥ al-a῾shā fi sinā῾at al-inshā᾿. 14 vols. Cairo. ———. (ed.) 1929–1972. Ibn Taghrī Birdī: Al-Nujūm al-zāhira fī mulūk miṣr wa-l-qāhira. 16 vols. Cairo. Barbier de Meynard, C. and A. Pavet de Courteilles (eds). 1863. Al-Mas῾ūdī: Murūj al-dhahab. 9 vols. Paris.

Acknowledgements I am indebted to Prof. Johannnes den Heijer of the Université catholique de Louvain (UCL) for his generosity of time and expertise in commenting on this paper, in particular on the textual history of the HPA in general and the place of the central flood narrative within it. I am grateful to him also for providing English translations based on Seybold’s 1912 Arabic edition that render a more authentic picture of the passages under consideration. I wish also to thank him and Dr Perrine Pilette (also UCL) for checking the relevant passages in the primitive recension of the HPA, and for ensuring referencing conventions with respect to Evetts and Butler’s 1895 edition. Their research is conducted under the auspices of the International Copto-Arabic Historiography Project based at UCL. I am also grateful to Dr Alison Gascoigne (University of Southampton), Professor William Gallois (University of Exeter) and Dr Alex Mallett (University of Exeter) for reading and commenting on drafts of this paper. I alone am responsible for its final form.

Basilios, Archbishop. 1991. Epiphany, Liturgy of the. In Coptic encyclopedia, A. S. Atiya (ed.), 967–68. New York. Berggren, J. L. and A. Jones. 2000. Ptolemy’s Geography: An annotated translation of the theoretical chapters. Princeton, NJ. Borchardt, L. 1906. Nilmesser und Nilstandsmarken. Abhandlungen der Königlich Preußischen Akademie der Wissenschaften 1: 1–55. Boutros, N. H. D. 2013. Le calife et le patriarche. Entre autorité politique et autorité de la mystique. In Les mystiques juives, chrétiennes et musulmanes dans l’Égypte médiévale (VIIe–XVIe s.). Interculturalités et contextes historiques, G. Cecere, M. Loubet and S. Pagani (eds), 379–92. Cairo. Butler, A. J. 1884. The ancient Coptic churches of Egypt. 2 vols. Oxford. Cameron, A. and S. G. Hall. 1999. Eusebius: Life of Constantine. Oxford. Černý, J. 1952. Ancient Egyptian religion. London. Cooper, J. P. 2008. Egypt’s Nile–Red Sea canals: Chronology, location, seasonality and function. In Connected hinterlands: Proceedings of Red Sea Project IV, held at the University of Southampton September 2008, L. K. Blue, J. Cooper, R. Thomas and J. Whitewright (eds), 195–209. Oxford.

‘WE DESIRE TO KNOW WHICH ———. 2014. The Medieval Nile: Route, navigation and landscape in Islamic Egypt. Cairo. Daressy, G. 1900. Le Nilomètre de Kom El Gizeh. Annales du Service des antiquités de l’Égypte 1: 91–96. ———. 1915. L’eau dans l’Égypte antique. Mémoires présentés à l’Institut Égyptien 8: 201–14. de Goeje, M. A. (ed.). 1906. Al-Muqaddasī: Kitāb aḥsan al-taqāsīm fī ma῾rifat al-aqālīm. Leiden. den Heijer, J. 1984. Sawirus Ibn Al-Muqaffa῾, Mawhub Ibn Mansur Ibn Mufarrig et la genèse de l’Histoire des patriarches d’Alexandrie. Bibliotheca Orientalis 41: 336–47. ———. 1989. Mawhub Ibn Mansur Ibn Mufarrig et l’historigraphie copto-arabe. Étude sur la composition de l’Histoire des patriarches d’Alexandrie. Louvain. ———. 1991a. History of the patriarchs of Alexandria. In Coptic encyclopedia, A. S. Atiya (ed.), 1239b–42b. New York. ———. 1991b. Mawhub Ibn Mansur Ibn Mufarrij Al-Iskandarani. In Coptic encyclopedia, A. S. Atiya (ed.), 1573a–74b. New York. ———. 1993. The composition of the History of the churches and monasteries of Egypt: Some preliminary remarks. In Acts of the Fifth International Congress of Coptic Studies, Washington, 12–15 August 1992, Vol. 2, Papers from the Sections 1–2, T. Orlandi and D. W. Johnson (eds), 209–19. Rome. ———. 1994. Apologetic elements in Coptic-Arabic historiography: The Life of Afrahām ibn Zur῾ah, 62nd patriarch of Alexandria. In Christian Arabic apologetics during the Abbasid period (750–1258), S. Kh. Samir and J. Nielsen (eds), 192–202. Leiden; New York; Cologne. ———. 2015. The martyrdom of Bifām Ibn Baqūra al-Ṣawwāf by Mawhūb ibn Manṣūr ibn Mufarrij and its Fatimid background. Medieval Encounters 21: 452–84. den Heijer, J.,Y. Lev and M. N. Swanson. 2015. The Fatimid empire and its population. Medieval Encounters 21: 323–44. den Heijer, J. and P. Pilette. 2013. Transmission et diffusion de l’historiographie copto-arabe: Nouvelles remarques sur les recensions primitive et vulgate de l’Histoire des patriarches d’Alexandrie. In Cultures in contact. Transfer of knowledge in the Mediterranean context, S. Torallas Tovar and J. P. Monferrer-Sala (eds), 93–140. Selected Papers (Series Syro-Arabica 1). Cordoba; Beyrouth. Donner, F. M. 2005. The Islamic conquests. In A companion to the history of the Middle East, Y. M. Choueri (ed.), 28–51. Oxford. Dunand, F. 2004. Ptolemaic and Roman Egypt. In Gods and men in Egypt: 3000 BCE to 395 CE, F. Dunand and C. Zivie-Coche (eds), 197–341. Ithaca, NY.

IS THE TRUE RELIGION’

129

Du Roy, G., M. Makhoul and P. Pilette. 2018. Le récit du miracle du Muqattam à travers les siècles: Approches interdisciplinaires. Annales Islamologiques 52: 193–218. Evetts, B. T. A. (ed.). 1907a. History of the patriarchs of the Coptic church of Alexandria I: Saint Mark to Theonas (300). Patrologia Orientalis 1: 3–116 [101– 214]. Paris. ———. (ed.). 1907b. History of the patriarchs of the Coptic church of Alexandria II: Peter I to Benjamin I (661). Patrologia Orientalis 1: 119–254 [383–518]. Paris. ———. (ed.). 1910. History of the patriarchs of the Coptic church of Alexandria III: Agathon to Michael (766). Patrologia Orientalis 5: 1–215 [257–471]. Paris. ———. (ed.). 1915. History of the patriarchs of the Coptic church of Alexandria IV: Mennas I to Joseph I (849). Patrologia Orientalis 10: 358–551. Paris. Evetts, B. T. A. and A. J. Butler (eds). 1895. The churches and monasteries of Egypt and some neighbouring countries: Attributed to Abû Sâlih, the Armenian. Oxford. Farag, F. R. 1977. The technique of presentation of a tenthcentury Christian Arab writer: Severus Ibn Al-Muqaffa῾. Arabica 24 (1): 66–87. Favard-Meeks, C. 1996. Mediating between the gods and humankind. In Daily life of the Egyptian gods, D. Meeks and C. Favard-Meeks (eds), 111–98. London. Frankfurter, D. 1998. Religion in Roman Egypt: Assimilation and resistance. Princeton, NJ. Friedman, Z. 2008. Nilometer. In Encyclopaedia of the history of science, technology and medicine in non-Western cultures, H. Selin (ed.), 1751–60. Dordrecht. Gayraud, R. P. 1998. Fostat: Évolution d’une capitale arabe du VIIe au XIIe siècle d’après les fouilles d’Istabl̓Antar. In Colloque international d’archéologie islamique, R. P. Gayraud (ed.), 435–60. Cairo. Godley, A. D. (ed.). 1966. Herodotus: The histories. 4 vols. Cambridge, MA. Graf, G. 1947. Geschichte der christlichen arabischen Literatur. 5 vols. Vatican City. Grossman, P. 1991. Epiphany tanks. In Coptic encyclopedia, A. S. Atiya (ed.), 3: 968. New York. Guest, R. (ed.). 1912. The governors and judges of Egypt or Kitâb El ‘Umarâ (El Wulâh) Wa Kitâb El Quḍâh of El Kindî. London. Halm, H. 2003. Die Kalifen von Kairo. Die Fatimiden in Ägypten 973–1074. Munich. Hardy, E. R. 1952. Christian Egypt: Church and people. Oxford. Hornung, E. 1982. Conceptions of gods in ancient Egypt. Ithaca, NY. Hurst, H. E. 1952. The Nile: A general account of the river and the utilization of its waters. London. Johnson, D. W. 1977. Further remarks on the Arabic History of the patriarchs of Alexandria. Oriens Christianus 61: 103–16.

130

J. P. COOPER

Jones, H. L. (ed.). 1917–1932. Strabo: Geography. 8 vols. Cambridge, MA. Kubiak, W. 1987. Al-Fustat: Its foundation and early urban development. Cairo. Kupelian, M. 2015. Festivals. In Egypt: Faith after the pharaohs, C. Fluck, G. Helmecke and E. R. O’Connell (eds), 182–87. London. Lewis, N. 1983. Life in Egypt under Roman rule. Oxford. Lindsay, J. 1968. Men and gods on the Roman Nile. London. Louca, A. 1981. Le moment inaugurateur en histoire: Analyse d’un texte d’Ibn ‛Abd al-Ḥakam (187–257 H./803– 871) sur la conquête musulmane de l’Égypte. In Proceedings of the Ninth Congress of the Union européenne des arabisants et islamisants: Amsterdam, 1st to 7th September 1978, R. Peters (ed.), 181–92. Leiden. Maguire, H. 1999. The Nile and the rivers of Paradise. In The Madaba Map centenary, 1897–1997, M. Piccirillo and E. Alliata (eds), 179–84. Jerusalem. Makhoul, M. 2015. Le déplacement d’al-Muqaṭṭam dans l’Histoire des patriarches d’Alexandrie: Débat sur la nativité et sur d’autres dogmes chrétiens. Acta Orientalia Belgica 28: 283–94. Martin, M. M. 1985. Une lecture de l’Histoire des patriarches d’Alexandrie. Proche Orient Chrétien 35: 15–36. Meeks, D. 1996. The gods among themselves. In Daily life of the Egyptian gods, D. Meeks and C. Favard-Meeks (eds), 13–107. London. Mercer, S. A. B. 1949. The religion of ancient Egypt. London. Mikhail, M. S. A. 2014. From Byzantine to Islamic Egypt: Religion, identity and politics after the Islamic conquest. London. ———. 2016. The legacy of Demetrius of Alexandria 189– 231 CE: The form and function of hagiography in Late Antique and Islamic Egypt. London; New York. Morimoto, K. 1981. The fiscal administration of Egypt in the early Islamic period. Kyoto. Muhammad, M., M. Sobernheim, A. Schimmel and P. Kahle (eds). 1931–1992. Ibn Iyās: Badā᾿i῾ al-zuhūr fī waqā᾿i῾ Al-duhūr. 9 vols. Wiesbaden. Naṣṣār, H. 1959. Al-Kindī: Wulāt miṣr. Beirut. Neugebauer, O. 1942. The origin of the Egyptian calendar. Journal of Near Eastern Studies 1 (4): 396–403. Oestigaard, T. and G. A. Firew. 2013. The source of the Blue Nile: Water rituals and traditions in the Lake Tana region. Newcastle-upon-Tyne. Papaconstantinou, A. 2010. The multilingual experience in Egypt, from the Ptolemies to the Abbasids. Farnham. Petrie, W. M. F. 1924. Religious life in ancient Egypt. London. Pilette, P. 2013. L’Histoire des patriarches d’Alexandrie. Une nouvelle évaluation de la configuration du texte en recensions. Le Muséon 126 (3–4): 419–50.

Popper, W. 1951. The Cairo nilometer: Studies in Ibn Taghri Birdi’s Chronicles of Egypt 1. Berkeley; Los Angeles, CA. Quirke, S. 1992. Ancient Egyptian religion. London. Rubenson, S. 1996. Translating the tradition: Some remarks on the Arabization of the patristic heritage in Egypt. Medieval Encounters 2: 4–14. Sayyid, A. F. (ed.). 2002. Al-Maqrīzī: Al-Mawā῾iz wa-li῾tibār fī dhikr al-khiṭaṭ wa-l-āthār. 4 vols. London. Seybold, C. F. (ed.). 1904–1910. Severus ben al-Muqaffa῾, Historia patriarcharum Alexandrinorum. Corpus Scriptorum Christianorum Orientalium, Scriptores arabici, Textus, s. III, 9. Beyrouth; Paris; Leipzig. ———. 1912. Alexandrinische Patriarchengeschichte von S. Marcus bis Michael I. (61–767), nach der ältesten 1266 geschrieben Hamburger Handschrift im arabischen Urtext herausgegeben. Hamburg. Shaheen, M. 1985. Hydrology of the Nile basin. Amsterdam. Sheehan, P. 2010. Babylon of Egypt: The archaeology of Old Cairo and the origins of the city. Cairo; New York. Spence, L. 1990. Ancient Egyptian myths and legends. New York. Springuel, I. and O. Ali. 2005. The River Nile basin. In The world’s largest wetlands: Ecology and conservation, L. H. Fraser and P. A. Keddy (eds), 347–92. Cambridge. Swanson, M. N. 2010. The Coptic papacy in Islamic Egypt (641–1517). Cairo. Thomas, D. 2015. Introduction. In Christian–Muslim relations 600–1500, D. Thomas and A. Mallet (eds), 1–20. Leiden. Torrey, C. (ed.). 1922. Ibn ῾Abd al-Ḥakam: Futūḥ miṣr wa-l-maghrib. New Haven, CT. Toussoun, O. 1925. Mémoire sur l’Histoire du Nil. 2 vols. Mémoires présentés à l’Institut d’Égypte 8 and 9. Cairo. Trombley, F. R. 2004. Sawīrus Ibn Al-Muqaffa῾ and the Christians of Umayyad Egypt: War and society in documentary context. In Papyrology and the history of early Islamic Egypt, P. M. Sijpestein and L. Sundelin (eds), 199–226. Leiden. ———. 2008. ‘Amr b. al-‘Ās’s refurbishment of Trajan’s canal: Red Sea contacts in the Aphrodito and Apollōnonas Anō Papyri. In Connected hinterlands: Proceedings of Red Sea Project IV, held at the University of Southampton September 2008. L. K. Blue, J. Cooper, R. Thomas and J. Whitewright (eds), 99–109. Oxford. Walker, P. E. 2009. Caliph of Cairo: Al-Hakim bi-Amr Allah, 996–1021. Cairo. Wild, R. A. 1981. Water in the cultic worships of Isis and Sarapis. Leiden. Wilkinson, R. H. 2003. The complete gods and goddesses of ancient Egypt. London.

‘WE DESIRE TO KNOW WHICH Winlock, H. E. 1940. The origin of the ancient Egyptian calendar. Proceedings of the American Philosophical Society 83 (3): 447–64. Witt, R. E. 1971. Isis in the Greco-Roman world. London.

IS THE TRUE RELIGION’

131

Wüstenfeld, F. 1866–1873. Jacut’s Geographisches Wörterbuch. 6 vols. Leipzig. Zivie-Coche, C. 2004. Pharaonic Egypt. In Gods and men in Egypt: 3000 BCE to 395 CE, 1–191. Ithaca, NY.

SYNCHRONIZATION AND ITS DISCORDS: CALENDRIC REFORM AND IMPERIAL POLITICS AT THE END OF THE NINETEENTH CENTURY On BARAK

Preface It is little wonder that the institution of the museum captures a modern sensibility about time—one that is sometimes at odds with previous understandings and practices of timekeeping. Allowing museumgoers to frequent, during a single visit and within a single space, artefacts from multiple contexts, the institution stimulates cultural translation that reinforces a presupposition of temporal commensurability. As we move within and between exhibitions it is implicitly suggested that different cultures might be distant in space or distinct in manner, but, as the anthropologist Johannes Fabian famously put it, they are nonetheless coeval, or simultaneous in time (Fabian 2002). Such assumptions of coevalness—themselves useful antidotes to stigmatization of cultural ‘backwardness’ or ‘advancement’— neatly map onto other expectations of commensality animating visions of people naturally sharing space, food or aspirations. In the catalogue for the 2015 British Museum exhibition ‘Egypt: Faith after the pharaohs’, this assumption is revealed in a technical prefatory ‘note on conventions’, indicating that ‘Dates are given in BC/AD throughout, and the Hijra date (AH) given when it is written on the object…’ (Fluck, Helmecke and O’Connell 2015, 4). Calendric conversion is indeed one of the main vessels that allow temporal coevalness. Rather than presuppose the possibility of such seamless cultural translation, this essay aims to historicize it, attending to the effort exerted in bringing calendric coevalness about and to the political, and particularly imperial, aspects of instigating and maintaining it. Based on part of a broader project historicizing timekeeping in modern Egypt (Barak 2013), what follows traces modern calendric commensurability in Egypt back to the last third of the nineteenth century, when this Ottoman province was absorbed into the British empire. Synchronization, I show, cannot be reduced to a synonym of painless coexistence and its colonial legacies are enduring even today.

Introduction Once, in [Caliph] ꜥUmar’s time, when the Month of Fast came round, some people ran to the top of a hill, in order to have the luck of seeing the new moon; and one of them said, ‘Look, there is the new moon, O ꜥUmar!’ As ꜥUmar did not see the moon in the sky, he said, ‘This moon has risen from thy imagination. Otherwise, since I am a better observer of the heavens than thou art, how do I not see the pure crescent? Wet thy hand and rub it on thine eyebrow, and then look for the new moon.’ When the man wetted his eyebrow, he could not see the moon. ‘O King,’ said he, ‘there is no moon; it has disappeared.’ ‘Yes,’ said ꜥUmar, ‘the hair of thine eyebrow became a bow and shot at thee an arrow of false opinion.’ Jalāl al-Dīn Rūmī, ‘The man who fancied he saw the new moon’ (al-Rūmī 1995, 27)

Accounts of the relations between timekeeping and authority divulge a politics of fact-setting, whose blackboxed mechanisms this essay aims to probe. How is the world made objective? And how is this process related to the emergence of a sphere of culture designated as ‘subjective’? ‘The man who fancied he saw the new moon’, by the Sufi poet Jalāl al-Dīn al-Rūmī, sets the stage: the Islamic calendar. This lunar calendar is based on direct observation of the night sky. Its proper name, the ‘Hijrī calendar’, denotes the fact that the first Islamic year marks the hijra, the Prophet’s emigration from Mecca to Medina in AD 622. In the text ‘Umar (579–644), the caliph who established this calendar and under whose reign Egypt was occupied and integrated into this Islamic temporal sphere, has the authority to command the observer to wipe his eye, thereby revealing the absence of the moon and leading to the man’s realization that he should be wiser than to claim to spot the moon before the most observant caliph. The opposite could not have been the case: had ‘Umar himself seen a hair rather than the moon, none of his subordinates would have told him to wipe his eyes, and the hair would have become fact, launching the month of Ramaḍān. Fact-setting, timekeeping included, involves

134

O. BARAK

hierarchies of power. When facts are contested, authority is what distinguishes a crescent in the sky from a hair in the eye of the beholder. After a remarkable longevity from the seventh century on, in the nineteenth century ꜥUmar’s lunar calendar was eclipsed in key spheres of life in colonial Egypt by the Gregorian calendar, the result of a series of shifts over a period of several decades beginning in the 1870s. A new division of labour between these temporal systems resulted in the transformation of the Hijrī calendar—once the primary timekeeping scheme in a comprehensive textual universe predicated on the logic of the transmission of the ḥadīth, or accounts of the deeds and sayings of the Prophet Muhammad—into a cultural artefact, a mere religious calendar recording festivals and holidays. From a framework mediating the facts of nature to the sphere of sociability, this calendar was demoted to a subjective matter of faith and ritual. Yet, rather than an already present domain, the independent existence of ‘culture’ and ‘religion’ cannot be presupposed; rather, ‘culture’ was itself in the making. Calendric reform at this very period involved the emergence of such supposedly insulated domains as free-floating spheres wherein human belief, solidarity, manners and customs could be divorced from a host of new technologies that profoundly structured these arenas, apparatuses that wrote themselves under erasure, slyly removing their footprints from the picture. This relegation of a previously operative form of social timekeeping, organization and expression into a purely ‘cultural’ domain took place just as Egypt was affixed to Europe by means of newly introduced steamer, railway and telegraph lines, as well as the Suez Canal (inaugurated in 1869), which together replaced the long sea voyage to India around the Cape of Good Hope. This new infrastructure transformed Egypt into the geographical centre (literally a ‘Middle East’) and simultaneously an economic and political periphery of the British empire. Peripheralizationthrough-centralization also entailed, at multiple levels, the temporal standardization and harmonization required to make all these technologies work in synch.

Large-scale temporal schemes structure and are structured by quotidian temporalities. The astronomical time of months and years that calendars mark informed both the point at which the day began and the length of the hour.1 Time signals measured in seconds and transmitted through submarine intercontinental telegraph cables, and the train schedules they punctuated on shore, were thus connected at the navel to the Gregorian solar calendar, which contrasted them with the ‘Arabic day’ that started at sunset, or the uneven hours belonging to the universe of the lunar Islamic calendar. Promoting the Gregorian calendar over competing temporalities meant enabling and defending the hegemony and efficacy of Western mechanical standard time. Timekeeping is thus also a perspective that reveals Egypt’s gradual removal from the sphere of the Ottoman empire, and its absorption into the British one. In this sense the politics of fact-setting is also the setting of political facts on the ground. This historical transformation will be examined here in an especially revealing arena, the newly established Arabic press, whose history begins in the 1870s after a few earlier stutters.2 In practically all Arabic textual production preceding the last third of the nineteenth century, the Hijrī calendar was the undisputed organizing principle. Be it in historical texts, biographical dictionaries, or historiography, Hijrī dates ordered a text’s internal structure (by offering the framework in which events and people were related to one another), informed its diachronic position in a tradition or a canon (which, especially in the case of religious literature, connected all texts to the moment of hijra), and defined its synchronic relations with contemporaneous works similarly organized. The degree to which everyday life followed this lunar calendar was probably quite limited. Yet, more practical calendars, such as the Coptic solar one (the main temporal scheme punctuating agricultural life in Egypt), made only unassuming incursions into written texts. Like classical Arabic—nobody’s mother tongue, but the only proper medium for approaching written texts—the Hijrī calendar was the lingua franca of Arabic letters until it was

1

2

On the connection between the daily schedule and the monthly calendar, see Zerubavel 1985.

For a history of the Arab press, see Ayalon 1995.

SYNCHRONIZATION AND ITS DISCORDS

135

Fig. 1: Astrolabic quadrant, made from lacquered paper-covered fruitwood by Muhammad al-Sakasi al-Jarkasi, in Damascus, AD 1891–92/1307–8 AH. This side shows a correspondence table between four different calendars: the Islamic, the Turkish Maliyye year, the Christian Coptic and a French Julian year, all rendered into Ottoman Turkish transcription. BM 1997,0210.1 (Courtesy of the Trustees of the British Museum).

dislodged by the Frankish Gregorian calendar. This new solar calendar, unlike the Coptic calendar it formally replaced in 1875, did not shy away from texts (Fig. 1). On the contrary, it arrived with an entirely new, telegraphic textuality, the newspaper, which was connected to a global economy and a global communications network that required meeting global synchronization standards. The replacement of the Coptic time of cotton agriculture with the Gregorian time of cotton finance and news had sweeping implications for the Hijrī calendar.3

This double-pronged focus on textual and calendric reform allows us to probe technology’s role in shaping new ‘chronotopes’—ways that newly introduced temporal conventions restructured communication and discourse, new modes whereby technology textualized time.4 If the Hijrī calendar was the key system for organizing the pre-modern textual universe, how did this universe respond to the combined calendric and textual transformation brought about by telegraphy?

3

4

In one textual genre, astronomy, the Coptic calendar had a high status. This was the genre in which calendars and timekeeping were traditionally studied.

According to Mikhail Bakhtin, a chronotope is a spatiotemporal matrix that defines a narrative (Bakhtin 1981).

136

O. BARAK

Lunar eclipses The Hijrī calendar is a purely lunar calendar without intercalation and is thus independent of the seasons; it is determined by observation of the evening sky and is therefore unpredictable. For this reason, Muslims have always also relied on solar and quasi-solar calendars for agriculture or taxation (Van Dalen et al. 1955– 2005). Al-Jabartī’s chronicles of the French invasion of Egypt (1798–1801) and police and court records throughout the nineteenth century contain multiple dating systems side by side (al-Jabartī 2004, 49). Labour migration from Southern Europe and increasing interference from Western Europe were among the factors making this multiplicity of calendars increasingly common and also increasingly contested during the nineteenth century, eventually recasting difference and multiplicity as cultural opposition and dichotomy. In this context, 1870 marks both the beginning of a process of harmonization of calendric systems and the eventual decline of the Hijrī calendar. The first printed calendar to be widely distributed for private use was published in the year 1870 by Maṭba῾at Wādī al-Nīl (Wādī al-Nīl Printing House), one of the first semi-private Arabic printing houses in Cairo. The appearance of the calendar was tied to developments in printing technology in the last third of the nineteenth century, particularly to new possibilities for fitting more words into a line and more lines on a page, which made printed products significantly cheaper and more affordable for mass consumption (Wādī al-Nīl, 1 Ṣafar 1287). Printing presses facilitated the concentration of information in legible formats that gradually replaced hand copying (Hanna 2004). The first Arabic printing press to be brought into Egypt was carried by Napoleon’s army of occupation in 1798. The orientalists accompanying the troops used the press to print pamphlets about the compatibility of the French Revolution and Islam. These printouts revealed, however, the incommensurability of the timekeeping systems used by occupiers and occupied: for instance, the first pamphlet was printed on either the thirteenth or the

fifteenth day of the revolutionary month of Messidor, the sixth year of the republic, which al-Jabartī thought occurred ‘toward the end of the month of Muḥarram [1213 ah]’—about ten days off the mark (al-Jabartī 2004, 27; Moiret 2001, 42).5 The subsequent adoption of the printing press by Egypt’s rulers, from Mehmet ‘Ali on, offered a prêt-àporter technological connection between the ideals of the French Revolution and the Nahḍah—the literary ‘awakening’ of the second half of the nineteenth century, often understood as a response to European influence. Throughout the century, the governmental Būlāq Press, founded in 1820, printed hundreds of Arabic translations from European languages, including several almanacs harmonizing the Coptic year with the Hijrī one.6 Even as far as the press’s efficiency and professional standards were concerned, European printing houses were used as the yardstick for quality and speed.7 Yet technology is always anchored in a particular setting, from which it derives much of its meaning. In their early decades, Arabic printing technologies were closely tied to the calligraphic culture that they later replaced. Wādī al-Nīl’s calendar is a good example of a product tailored for a particular readership in a market dominated by hand-copied texts, such as Dalā’il al-Khayrāt (Guidelines to the blessings), a popular almanac stipulating prayer times (and a bestseller of the eighteenth and nineteenth centuries) (Hanna 2004, 94–96). Before the middle of the nineteenth century, almanacs including the Hijrī and Coptic calendars seem to have been the most popular paper products after the Qur’ān, circulating in high-quality calligraphy copies among elites and in low-quality commercial copies among commoners (al-Jabartī 2006, 1: 276, 2: 298, 2:279–90; Hanna 2004, 90). Another example of the tension between the intended use of new printing technologies and their eventual local use is provided by newspapers. In 1867 Wādī al-Nīl Printing House started printing the pioneering, semi-private, biweekly newspaper Wādī al-Nīl (The Nile valley) (Fig. 2). The newspaper was subsidized by

5

7

6

Al-Jabartī dates the pamphlet’s printing to the thirteenth day. Joseph-Marie Moiret dates it to the fifteenth. From 1822 to 1842, three such calendars were printed by Būlāq (Heyworth-Dunne 1940).

For example, in a decree launching a wave of reforms in the printing house during 1860, its performance is measured against foreign printing houses (Sāmī 1936, 356–57; Lane 2003, 226–27).

SYNCHRONIZATION AND ITS DISCORDS

Fig. 2: Front page of Wādī al-Nīl, the first semi-private Egyptian newspaper (Friday, 26 Muharram 1286 AH, corresponding to 7 May, AD 1869).

137

138

O. BARAK

the Khedive Ismā῾īl, for whom the absence of a private press was a lacuna in Egypt’s modernization. The paper was modelled after the European newspapers that had proliferated in Alexandria (including Wādī al-Nīl’s namesake, Le Nil), which were the paper’s main sources of inspiration and news. Like contemporary European periodicals, Wādī al-Nīl was serialized, and its readers were encouraged to leather-bind the sections together. Such hybridization helped familiarize the newspaper form to a reading public accustomed to the non-secular and less-ephemeral textual form of the book (a form wedded to Hijrī time). Yet, European influence was again only one part of the story. Just as important was the fact that the biweekly issues of Wādī al-Nīl regularly reproduced portions of such landmark Arabic works as Ibn Baṭūṭah’s fourteenth-century travel narrative, which was serialized in 1870 (later that year, the press sold a printed compilation of the entire book) (Ibn Baṭūṭah 1870–71). Whereas in Victorian Britain, Germany and the United States serialization was a format for science, politics, and new, often innovative literature of varying quality, in Egypt serialization was also used for the ‘classics’. And while that misnomer incorrectly assumes that such books had not been constantly read by modern readers, even though hand-copied versions of texts by Ibn Khaldūn and Ibn Baṭūṭah were never out of vogue, it is possible that printing and serialization themselves transformed these texts into classics.8 In 1870 the newspaper offered a ready-made platform for advertising the press’s newly printed calendars. As the first advertisement stated, ‘Wādī al-Nīl Printing House launches a meticulous and systematic rendition of calendar time as is the practice in the European countries. It is a presentation of the year 1287 AH, including a juxtaposition of the correct lunar Arabic months with the Coptic, Frankish and Roman [rumi] months.’ After indicating the novelty of this temporal device in Egypt, the advertisement further suggests how to use the calendar (glue it on two sides of a piece of cardboard or hang it on a wall) and who might benefit from it (bankers and employees of the Egyptian administration). As the advertisement makes

8

The agenda of reintroducing the classics of Arabic literature in print form was avidly promoted by al-Ṭahṭāwī during this period (see Hourani 1983, 72).

clear, several temporal systems coexisted in Egypt. The Coptic solar year regulated agriculture and taxation, the ‘Frankish’ Gregorian calendar was used in banking and cotton exchange, and the Hijrī calendar was used by the administration and the educated public. Finally, the ‘rumi months’ referred to the Seleucid calendar or possibly the Julian calendar. Both the Julian and Seleucid calendars served Christian communities in Egypt and the Ottoman lands, while the Julian calendar also had an Ottoman administrative purpose. The new printed calendar promised to help navigate this multiplicity. However, calendric harmonization was unravelling even in its festive inception. The first advertisement was published on ‘Friday, 21 Muḥarram 1287, corresponding to April 19, 1875, the fourth year of the newspaper’. Yet, if Wādī al-Nīl started printing in 1866–67, the fourth year should have been 1870–71, the year corresponding to 1287 AH. A computerized date converter reveals that 21 Muḥarram 1287 corresponds to 22 April 1870, rather than to 19 April 1875. These mismatches appeared frequently on the header of the front page of the newspaper. In each case an incorrect Gregorian date was coupled with the correct Hijrī one. For example, the calendar was advertised again on a Hijrī date ‘corresponding to April 45’ (Wādī al-Nīl, 10 Muḥarram 1286). Such mismatches in calendar dates reveal synchronization as a laborious and effortful process. These breakdowns expose the fragility of an ostensibly seamless temporal grid. They stand in sharp contrast to the claim by Johannes Fabian in Time and the other (Fabian 2002) that cultural difference is coeval or simultaneous. Rather than a natural state of coevalness that in turn gets denied, or the plurality of a multicultural world at the end of a liberal horizon, we see a radical alterity made commensurable only with difficulty and partial success (Povinelli 2001, 327–28). Admittedly, 1287 AH was a confusing year for calendar conversion. Consider the Ottoman financial calendar: the Maliyye ‘fiscal year’ was a scheme based on the Julian calendar that attempted to keep the counting of tax years in line with the years of the hijra by

SYNCHRONIZATION AND ITS DISCORDS

omitting one year for every thirty-three. In 1287 the system broke down when the omission scheduled for that year was not implemented, creating a discrepancy between the Maliyye and Hijrī years (Van Dalen 2000). Egyptian almanacs referenced the Ottoman Maliyye year until the First World War. Up to 1875, Egypt conducted its financial affairs according to the Coptic calendar (Zwemer 1913, 270). That year the country’s connection to a network of intercontinental telegraph lines instigated the replacement of this calendar with the Gregorian one. Telegraphic connectivity and instantly available global commodity prices allowed Egypt to quickly take advantage of the US Civil War. With the temporary dwindling of American cotton production and trade, European markets shifted to Nile Basin cotton, creating an Egyptian cotton boom during the 1860s and early 1870s. Yet the same wired, global cotton market shifted back to American cotton after the war, dragging Egypt into escalating indebtedness. The pressures of debt repayment for European creditors and state bankruptcy forced the Egyptian government to adopt the Gregorian calendar, severing the time of cash from that of cash cropping and agriculture, which continued to follow the Coptic calendar: Whereas the ministries’ engagements with Europeans are mostly conducted according to the Frankish months while budgets and calculations follow the Coptic months, and even though in both systems the annual number of days is the same, to prevent date disagreement we decree that the government will conduct its financial affairs according to the Frankish months (Sāmī 1936, 3: 1251).

In 1876, Al-Ahrām, a private Egyptian newspaper founded by two Syrian Christian brothers across the street from Alexandria’s Cotton Exchange, adopted a dating procedure that employed a Gregorian date as the standard. In the newspaper, which was initially devoted to telegraphic news about things such as commodity prices, the Gregorian date appeared on the right side of the page, with the corresponding Hijrī date on the left. (Because Arabic is read from right to left, placing the Gregorian date on the right gave it primacy.)

139

In the mid-1870s, similar shifts in standards took place in other texts. Consider the autobiography of the champion of timetables, ῾Alī Mubārak, the railway manager responsible for the introduction of train schedules into Egypt. In his narration of his childhood and early government service, Mubārak deploys the Hijrī calendar. Yet when he first mentions Egypt’s debt, in 1876, he suddenly adopts the Gregorian calendar, which he then uses for the remainder of the text (Mubārak 1989, 57). In such shifts, debt provides the particular context for the introduction of the equation of time and money into Egypt. The Gregorian calendar (and the monetized quotidian temporalities associated with it) indexed, and was tainted by, the beginning of an epoch that began with imperial debt collection and management, ushering in ever-more-invasive forms of control and interference. Because it was calendrically synchronized with the global economy, Egypt was already behind—on its payments, among other things. Once again, commensurability revealed itself to be a protocol of differentiation. What I call ‘the time of money’ has a particular history: according to Jacques Le Goff, the rise of commercial capitalism in Medieval Europe involved a transformation in the telling of the hour from the unequal hours of the monastic day to the precision of the clock, a shift from ‘church time’ to ‘merchant time’ (Le Goff 1980, 29–42).9 The Hijrī calendar, by contrast, was connected to a different system of quotidian timekeeping. Because the Hijrī month begins with a moon sighting in the evening sky, the ‘Arabic day’ starts at sunset, as opposed to the ‘Frankish day’, which was believed to start at high noon (‘Al-Sā῾āt al-῾Arabīyah wa’l-Ajnabīyah’, Al-Muqtaṭaf 52 [1918] 128; ‘Al-Sā῾āh al-῾Arabīyah wa’l-Ifranjīyah,’ Al-Muqtaṭaf 32 [1907] 132). Thus, for Egyptians the twelve-hour day was divided into ‘evening’ and ‘morning’ rather than A.M. and P.M., as was the case with train schedules, which were also introduced in 1870 and printed in Wādī al-Nīl (on 16 September 1870). Because the sun sets and rises at different times depending on the season, the length of every hour during the ‘Arabic

9

For a critique of this thesis, see Dohrn-van Rossum 1996, 138–71.

140

O. BARAK

day’ varied seasonally, with the result that watches and clocks had to be reset daily (Baedeker 1902, lxvii). By contrast, the ‘Frankish day’ occurred without variation and was divided into twelve even hours (Al-Hilāl, 1 November 1902). Such differences in timekeeping systems were repeatedly discussed in the new scientific journals that were published starting in the mid-1870s, such as Al-Muqtaṭaf (1876) and Al-Hilāl (1892). Often, the readers raising the issue of conflicts in different temporal systems in question-and-answer columns were employees of the Egyptian administration or the Egyptian State Railways (Al-Hilāl, 1 November 1902; Al-Hilāl, 15 November 1901). The logics of train schedules and debt management required a stable timeto-money conversion rate and seemed to favour the Frankish day and Gregorian year. As the Gregorian calendar was gradually yet firmly established in Egypt, dating errors in Hijrī dates in the press slowly became more common than for Gregorian dates. Disagreements over the determination of the lunar month acquired a new visibility. The journal Al-Siḥāfah, for example, issued this apology on 6 January 1905: ‘Whereas the previous edition carried the date Friday, the first day of the month of Dhu al-Qa῾dah, it was in fact 30 Shawwāl, even though some astronomers say the former date is correct.’ The journal requested that readers stop alerting its editors about such mishaps—a request suggesting that more than a few of these complaints had been filed. In 1916, young ‘Abd al-Razzāq al-Sanhūrī, the future Egyptian legal reformer, wondered just before leaving to study in France why he should remember the Islamic date of his birthday. In a diary entry from 14 August, he wrote about the day before yesterday, his twenty-second birthday: ‘I don’t know why I do not know my birthday according to the Arabic calendar. Why does it matter to me if I knew I was born in Rajab or Shawwāl or Dhu al-Hijjah as long as I know I was born on August 12, 1895 AD [Mīlādī]. . . Why should I want my birthday to be Arabic?’ However, to indicate that these were not merely rhetorical questions, he

10 11

The word Mīlādī denotes the birth of Christ. Other examples are Ḥasan al-Bannā’s father, a watch repairer (al-Bannā, 1990); Muṣṭafā Kāmil’s father, an engineer who built

concluded the note with a resolve not to submit to the erasure of his Arabic birthday (which had never existed), or at least to try: ‘I want to strengthen my will power; will I succeed?’ (al-Sanhūrī and al-Shāwī 2005, 54).10 Al-Sanhūrī belonged to a generation of effendis (educated middle-class professionals) born in the 1880s and 1890s (after the mid-1870s calendar shift), whose fathers were the first to document the birth dates of their children according to the Gregorian calendar or with both the Gregorian and Hijrī calendars.11 Al-Sanhūrī was not questioning the importance of having ‘a birthday’ per se or of knowing the exact moment of his birth. The celebration of the birthday, a personal nativity scene, became popular in Egypt during the first decades of the twentieth century (Amīn 2000). Premodern Islamic scholars sometimes also recorded their birth year—and less frequently also the month and day—but they did so in Hijrī time and for reasons having to do with the need to situate a ḥadīth transmitter in time and place. Because the teachings of the Prophet were transmitted orally from person to person, the key protocol used by later analysts trying to ascertain the soundness of a tradition was to calculate whether every two interlocutors in the chain of transmitters were able to share the same time and space. For members of the effendiyya, knowing one’s exact age distinguished oneself from the lower classes and provided an apt response to British assumptions about Egyptian attitudes to time. ‘Few uneducated Egyptians’, wrote Lord Cromer, the British consul-general and the de facto architect of Egyptian colonialism, ‘know their own age. The usual reply of an Egyptian, if asked the age of some old man, is that he is a hundred years old’ (Baring 1908, 152). In a system wherein middle-class Egyptians internalized and extended such admonitions, birth dates became class markers wedded to colonial renditions of the trope of Oriental time—mindlessness, a longstanding view of Egyptians as indolent, slothful and incapable of rational thinking. In Western Europe, the practice of recording births started in parish churches, which registered candidates

railway stations (Kāmil 1908); and Tawfīq al-Ḥakīm’s father (al-Ḥakīm 1992, 140). See also Jacob 2011.

SYNCHRONIZATION AND ITS DISCORDS

for baptism, thereby signifying ‘the appearance of Christian souls in new corporeal forms’ (Anderson 1998, 69). In the nineteenth century, compulsory registration of births became the practice by which an infant was included in citizenship in many places in Western Europe (Anderson 1998, 69). The secular registration of births in modern nation-states had distinct Christian origins. The secularized Gregorian calendar became post-Christian in a context whose significance outshines the mere fact that this calendar bears the name of Pope Gregory XIII: its eventual adoption even by Protestants forged an inter-confessional unity predicated on separating social harmony from religion. The calendar united European Christendom while simultaneously secularizing it. In Egypt this calendar had a similar secularizing effect, splitting ‘the social’, which it now organized, from ‘the religious’, which was relegated to the Hijrī calendar. The calendric shifts in Egyptian newspapers and in the writings of figures such as Mubārak and al-Sanhūrī offer scattered signposts of a standard shift whose telos is familiar: the hegemony of Arabic dates as points of reference was undermined and eventually overridden. As Abdelfattah Kilito sardonically writes: Arabic literature is subject to a double chronology. At first, and for a long time, it was tied to the Islamic calendar, then one day, without warning, it moved to the Christian calendar! One day, after seven centuries of recumbency, it leaped up suddenly and gracefully over six centuries, and found itself in the middle of the nineteenth century (Kilito 2008, 8–9).

Like Kilito, I am concerned here less with trying to date this outdating—an incomplete and messy process that happened differently in different spheres—than with tracing some of its mechanisms, implications and contexts. According to Talal Asad, the emergence of secularism in Egypt involved relegating a new object —‘religion’—to the private sphere (Asad 2003, 205– 57). Asad’s analysis of family law reform during the last third of the nineteenth century may be complemented by stressing the colonial origins of the notion that Islam, like European Christianity, had two dimensions: it was both a benign ‘religion’ and also a ‘social system’ in serious need of reform (Lane-Poole 1883, 101; Baring 1908, 134). This may be a key explanation for (and one of the outcomes of) a new division of labour between the Gregorian and Hijrī systems, wherein the latter’s role increasingly shrank to regulating religious festivals and holidays.

141

Yet even in the limited sphere of ‘religion’, the Hijrī calendar did not remain intact. Consider the practices of Ramaḍān moon sighting. In 1903, Islamic reformer Rashīd Riḍā (1865–1935), known as a key turn-of-thecentury synthesizer of Islam and modern technoscience, issued two fatwās (i.e., responsa) that indicate how these protocols had changed. The first came in response to a question about varying moon sightings before Ramaḍān and the resulting differences in the start of fasting. The inquirer asked whether, to avoid such discrepancies, actual sightings could permissibly be replaced with printed calendars. Riḍā replied that temporal incongruity among communities located in relative proximity could be explained only by false sightings. But printed lunar calendars could not solve the problem because they disagreed on the beginning of the lunar months. Riḍā’s solution was to adopt the time dictated by the authorities in the capital (al-Munajjid and Khūrī 1970, 1: 45). What needed no mention in this early twentiethcentury fatwā, though it vitally conditioned it, was that since the early 1870s, in tandem with the new train schedules, Cairo time had been disseminated telegraphically to the Egyptian provinces. This allowed Riḍā to assume in 1903 that a moon sighting in the capital could instantly initiate the month of Ramaḍān even in the remotest corner of Egypt. This was by and large a safe assumption. But already in 1873, a belated telegram from Cairo about the sighting of the Ramaḍān crescent had caused the Muslims of Port Said to miss the first day of the fast (Najm 1987, 94). Beyond the suboptimal performance of the telegraph, such mishaps reveal the extent to which people relied on this device as a new timekeeping technology. In the second fatwā, Riḍā made clear that the start of Ramaḍān stipulated in Egyptian newspapers applied only inside Egypt and should not be followed by readers in other countries, where direct sighting of the moon should remain the yardstick. Riḍā explained that it was important for all Egyptian Muslims to begin and conclude the fast together—because collectivity and concord (al-ijtimā῾ wa’l-ittifāq) in performing religious rituals are essentials of Islamic dogma—but that other countries must adopt their own procedures (Al-Manār 6 [1903], 862; Riḍā; al-Munajjid and Khūrī 1970, 1: 67). What needed no mention was the fact that Egyptian newspapers were circulated on board trains and steamers quickly enough to raise the question (posed to Riḍā) of whether their calendric information should be followed abroad.

142

O. BARAK

In both fatwās, Riḍā answered a political concern involving disagreements about moon sightings by offering a political solution, one that accepted the centralizing logic of his day. Rather than resolving disagreement locally, Riḍā succumbed to the authority of the central government of the nation-state, thus ensuring temporal harmony.12 If, in Benedict Anderson’s Imagined communities, temporal simultaneity is what provides the conditions of possibility for the nationstate, for Riḍā the nation-state guarantees religious simultaneity (Anderson 1983). Simultaneity inside the community hinges on the community’s temporal difference from other communities. The national homogeneity of time is always, in comparison, constantly supported by temporal heterogeneity. Riḍā did not renounce the need to physically sight the moon. Though he implicitly relied on the telegraphic transmission of a centrally determined Ramaḍān time to the provinces, the moon still had to be properly spotted in the capital. Though the project of harmonizing Islam and technoscience usually served to make Islam compatible with techno-logics, converting new technologies to Islam and understanding them in religious terms was just as important. For the telegraph, this task was carried out in the first two decades of the twentieth century by Shaykh Muḥammad Bakhīt al-Muṭī῾ī (d. 1354 AH/AD 1935), the qāḍī (judge) of Alexandria and later the grand mufti of Egypt. In his 1911 Kitāb Irshād ahl al-Millah ilā Ithbāt alAhillah (The book on guiding the religious community to the verification of the crescents), al-Muṭī῾ī made an analogy between telegraphic transmission of moonsighting news and the transmission of the ḥadīth accounts, both denoted by the same word, akhbār. Placing the telegraph in the framework of ḥadīth transmission was crucial to allowing the technology to be used for the dissemination of Cairo time: according to Islamic law, for a sighting of the Ramaḍān crescent to count, it has to be reported by an upright (῾adl) Muslim. But what about the mediation of telegraph operators who might be unjust or non-Muslim? Should

the number of telegraphers involved in transmitting a sighting report matter? Should the procedures of court testimony, requiring two witnesses, be applied to telegraphy? Such questions were addressed to al-Muṭī῾ī and to Riḍā before him (Al-Manār 7 [1904], 575–76) (Fig. 3). Al-Muṭī῾ī’s solution was to regard telegraphers as passive ‘mediators’ (singular: wasīṭah) rather than as ‘transmitters’ of telegraphic news (al-Muṭī῾ī 2000, 144–69). Bracketing operators made telegraphing a moon sighting comparable not to testifying in court but to narrating a ḥadīth, requiring only one source. Further, if several telegrams were received, even through the same telegraph line, they should be regarded as akhbār mutawātirah—a category of ḥadīth analysis denoting independent reports that corroborate one another (al-Muṭī῾ī 2000, 144–69). Unlike Riḍā, who did not question the need for an initial physical moon sighting in Cairo, al-Muṭī῾ī followed the opinion of the Shāfi῾ī jurist Taqī al-Dīn al-Subkī (AD 1284– 1355), according to which testimonies of crescent sighting should be rejected if they contradicted astronomical calculations.13 The credibility of evidence derived from observable natural phenomena was beginning to erode. In 1913 Samuel Marinus Zwemer, an American missionary in Egypt, recorded a suggestion by a certain ‘al-Zarqāwī’, printed in the nationalist newspaper Al-Sha῾b, to introduce a ‘solar Hijrī year’.14 Using the Gregorian calendar, al-Zarqāwī determined that the hijra took place on 22 September AD 622. He suggested adopting this date as the beginning of the Muslim calendar for everything but religious festivals, which would be determined by moon sighting (Zwemer 1913, 262–74). By 1357 AH (AD 1939) the importance of the moon was definitely waning. That year, the Supreme Sharī῾ah Court in Egypt determined that the month of Dhu alHijjah began on Saturday, 20 January. ῾Īd al-Aḍḥā (the Festival of Sacrifice) was hence celebrated in Egypt ten days later, on Monday, 30 January. But

12

14

13

Ahmad Dallal makes a similar argument about how the nationstate framed Riḍā’s legal thought (Dallal 2000, 357). Al-Muṭī῾ī appended al-Subkī’s manuscript on the verification of Hijrī months in printed form to his own guide (al-Subkī 1911).

Probably Aḥmad Mūsā al-Zarqāwī, author of Al-Adillah al-Islāmiyyah ꜥAlā Taḥarruk al-Kurah al-Arḍīyah, Cairo, 1913.

SYNCHRONIZATION AND ITS DISCORDS

143

Fig. 3: Remains of the railway embankment with telegraph lines near Kafer Zayat shown in ‘The Overflow of the Nile’, Illustrated London News (London, England), 21 November 1863 (Courtesy of Mary Evans Picture Library).

Egyptian readers of Al-Muqaṭṭam knew that the Saudi Arabian government had decided that the first of the month was not Saturday but Sunday, 21 January, and the ῾Īd was thus celebrated in the Arabian Peninsula on Tuesday, 31 January. And readers of Al-Balāgh discovered that the Muslims of Bombay celebrated the festival on Wednesday as a result of the establishment of the beginning of Dhu al-Hijjah on Monday, 22 January (Shākir 1986, 3–4). According to jurist Aḥmad Shākir, a member of Riḍā and al-Muṭī῾ī’s milieu, such discrepancies were not the exception but the rule: in some Muslim countries crescent sightings resulted in some people sighting it while others were unable to do so. As a consequence, the religious festivals differ from one Muslim country to another: some countries fast while others do not, some celebrate the Festival of Sacrifice, while on that very day others observe a fast (Moosa 1998, 69). Given that the moon sets progressively later than the sun as one goes west, more

westerly Muslims were likely to observe a new moon earlier than their eastern coreligionists, as this instance indicates. But in the age of telegraphy and steam navigation, Muslims in Cairo, Mecca and Bombay experienced the tensions of a new connectivity. The telegraph was not only disseminating the homogeneous time of the capital; through the newspaper, it also spread the word about temporal heterogeneity, thereby bolstering national togetherness at the expense of a larger religious concord. What began as a seemingly pure technological disjunction now acquired a social dimension, one that would soon override and occlude its technical infrastructure. Demonstrating that a new standard was emerging, Shākir’s solution to these discrepancies was to abandon the principle of sighting in favour of a single calendar based on scientific computation (Moosa 1998, 69). This was the explicitly logical conclusion of the telegraphic dissemination of Ramaḍān time and the

144

O. BARAK

successful attempts to give temporal homogeneity official Islamic sanction. To make his case that in its current form the Hijrī calendar was unruly, Shākir resorted to the standard of Gregorian dates. Shākir’s view remains a minority opinion on the commencement of Ramaḍān. Yet, if the resilience of physical moon sighting is taken as an indication of the autonomy of the Islamic calendar, this resilience should also be seen as reinforcing its new and limited scope as a religious calendar only. Dissenting views like Shākir’s reveal that if Europe shifted in the Middle Ages from church to merchant time, in modern Egypt even the religious establishment faced significant pressures to adopt monetized time. Other domains were even less resilient. Telegraphic space, time and text The telegraph was a key culprit in the rearrangement of calendric timekeeping in Egypt. Shifting from calendars to the newspapers that advertised and followed them, we can now examine the implications of the telegraphic reshuffling of temporal systems. How did telegraphy affect the textualization of time? Wādī al-Nīl, the first Arabic newspaper regularly printed in Egypt, was launched in June 1867 during an official visit of Egypt’s ruler, Khedive Ismā῾īl, to France and England (Al-Waqā’i῾ al-Miṣrīyah, 23 Rabī῾ al-Awwal [25 July 1867]; Sāmī 1936, 2: 713). The highlight of the royal trip was the signing of two treaties to sink submarine telegraph cables between Alexandria and the Italian shore and to connect the Malta– Alexandria–Cairo telegraph to a new London–Bombay network (Sāmī 1936, 2: 713). This second attempt at intercontinental telegraphy (after the first underwater cable succumbed to sea termites) was partly financed by news agencies operating in key nodes of this grid, including since 1865 the Reuters office in Alexandria (Storey 1951, 95). Khedive Ismā῾īl actively participated in this process: several months after signing the aforementioned treaties and establishing Wādī al-Nīl, he started subsidizing Reuters.15 The genesis of the private press embedded Egypt in these new communication networks. In 1870, Wādī al-Nīl subscribed to Reuters’ telegram service.

15

Allocating to it 20,000 francs a year (Sāmī 1936, 2: 782).

In the closing months of 1870, telegraphic Reuters news started appearing in Wādī al-Nīl’s foreign news section, bearing Gregorian dates. Domestic Egyptian news items kept their Hijrī dates. The newspaper thus revealed a temporal schism whereby foreign and local news occurred in different temporal (and spatial) domains. The Gregorian dates of foreign news were often accompanied by the corresponding Arabic date in parentheses. The telegraph thus promoted a standard shift whereby Arabic dates were for the first time bracketed, relegated to a parallel realm that required agreement. This ‘outdating’ happened just as a correspondence was established between these two incommensurable time systems. Before the paper subscribed to Reuters, foreign news—translated from the European-language newspapers proliferating in Alexandria—hardly ever occupied more than half a page in Wādī al-Nīl’s three to four pages. But the subscription to the agency’s service quickly transformed the Egyptian newspaper into one mostly devoted to foreign news. Such shifts demonstrate how telegraphy reshaped the conditions of knowledge acquisition and dissemination even before the British occupation. In pre-telegraphic Egypt, proximity roughly translated to familiarity: one knew more about one’s immediate surroundings than about faraway places. With the advent of telegraphy, an excess of foreign news and a ‘thick description’ of the alien quickly overclouded local knowledge. Wādī al-Nīl thus became one of the technologies that formed the worldview of the colonial subject, characterized by an out-of-focus world picture that was sharp around the edges and fuzzy in the centre. This had to do not only with the fact that in the modern world accelerated time was divorced from space, but also with specifically how this delinking was mediated in a colonial setting. The imbalance of local and foreign news produced an imbalance of dates: a larger portion of the news was happening in Gregorian time, which required translation into Hijrī, and not the other way around. This protocol, wherein Gregorian dates were the source or yardstick and Hijrī dates were derivative, quickly became the rule. Wādī al-Nīl’s editors attempted to deal with the excess of telegraphic information by creating foreign news summaries. They approached the matter with unease:

SYNCHRONIZATION AND ITS DISCORDS

In the previous editions of Wādī al-Nīl we have so far made an effort to translate the telegraphic news accumulating until July 8 (9 Jumādā al-Ūlā) and we have transmitted them in their original texts, quoting and presenting them one by one, despite their excess, so that the reader could have the choice and select the news he deems sound from which he can get a true understanding of current affairs. However, the volume of the telegraphic news amassed on July 9, 10, and 11 (10, 11, and 12 Jumādā al-Ūlā) [forces us] to render them in a summary (Sāmī 1936, 1: 240).

This editorial comment captures some of the concerns regarding telegraphy’s ability to collapse a multiplicity of voices into a single flattened narrative, a common trope not only among postcolonial theorists and historians of technology but also for the historical actors themselves. The thesis of ‘flattening’ had much to rely on. But, being predicated on technological determinism—which was itself a historical force (and not merely a faulty analytical framework with which to understand the history of technology)—it blinded many observers, both then and now, from recognizing the multifarious forms and inflections of technological modernity. Evidence of such multiplicity was thus understood with the new framework of ‘cultural difference’, which can be seen as technical determinism’s monozygotic twin. The necessity of devising new tactics for handling information excess generated other changes in important procedures of textual production. One striking development was that foreign news became shorter with the telegraph, because the newspaper omitted the news items’ chain of transmitters. Pre-telegraphic foreign news included an internal history detailing the circumstances of its own production, a preface modelled on the isnād (the chain of transmitters of a ḥadīth). Thus, a translated news item about violence in Mecca from 9 May 1870 opened as follows: ‘Translated from the journal L’Égypte: the following text appeared in a journal titled Alimbrsial Dosmir distributed on April 27 (26 Muḥarram): several newspapers discussed oral reports about what happened in Mecca’ (Wādī al-Nīl, 18 Jumādā al-Ūlā, 1287). This introductory paragraph situates Wādī al-Nīl at the end of an elaborate sequence of sources, following L’Égypte, Alimbrsial Dosmir, anonymous newspapers, and unspecified oral sources. Beyond the fractured and intricate process of news transmission, this paragraph exposes the many

145

temporal delays involved in news circulation. With the telegraph, such introductions disappeared, making room for actual news stories that now stood as independent pieces of information. Traces of ḥadīth-like textual conventions and terminology in early newspapers help us appreciate how their textual extinction interfaced with calendric reform. The Hijrī calendar was connected to a particular paradigm of textuality and knowledge transmission revolving around the ḥadīth. The science of ḥadīth and the imperative to ascertain whether a reported prophetic tradition was sound or spurious were a main driving force in the development of Islamic geography, biography and historiography. These auxiliary disciplines provided information about ḥadīth transmitters, their reliability and the probability that they could occupy the same time and space to pass information from one to another. Reviewed by ḥadīth critics in retrospect, this textual universe was diachronically indexed by the Hijrī calendar in a perfectly reliable and legible manner. Newspapers, by contrast, were media of synchronic information transfer. While the same word, akhbār, denoted journalistic as well as prophetic pieces of information, synchronic and diachronic times were for the first time competing to set the tone for textual information exchange. By effacing chains of transmission and compressing news into summarized narratives, the telegraph severed the connection between message and messenger, transmitter and text. Readers could no longer actively choose sound reports. Such a critical reading—involving constant evaluation of the genealogies of texts that lay bare the devices of their making— was replaced by a passive intake of ‘news’ without circumstances of production, mechanically produced and reproduced, immaculately conceived like the event that launched the calendar that organized them. Moreover, these new media infrastructures linked Egypt to the newly commensurable world in a subsidiary fashion. Consider Al-Ahrām: during 1876, the newspaper’s distribution, carried out by the mail, followed a geographical logic whereby proximity to the head office in Alexandria meant a cheaper subscription. Al-Ahrām was cheapest in the port city. Subscribing from Cairo and Syria (including Lebanon and Palestine) was more expensive than an Alexandria subscription, yet cheaper than a subscription from Europe or India.16 Moreover, inside Egypt, the newspaper covered

16

Subscription rates appeared on the header of every newspaper.

146

O. BARAK

and was distributed almost only in places connected to the railway. Territorially speaking, the Egypt of Al-Ahrām was that of the railway map. But if the railway made Egypt into a unified territory, the telegraph line that ran parallel to the railroad introduced heterogeneity into this supposedly standardized space. The direct telegraphic connection linking Cairo, Alexandria and Europe via Malta provided urban readers with fresh daily foreign news. By contrast, reports that were sent by mail inside Egypt sometimes took several days to find their way into the newspaper, and news from neighbouring countries without a railway connection with Egypt took even longer. For example, on 7 October 1876, Al-Ahrām printed a letter sent on 29 September by its Beirut agent: ‘We do not have anything new to inform you of: all matters are peacefully following their usual course, civic serenity prevails, and everybody is happy. … Rumour has it that His Holiness the Roman Catholic Patriarch is expected to arrive at Beirut from Damascus in the beginning of Tishrīn al-Awwal [the Levantine month corresponding to October] and will continue by sea onboard the Austrian [steamer] toward you, arriving at Alexandria on Monday, October 9.’ This typical item is revealing in several respects. First, it is a report of a non-event, an account of an undisturbed routine. Second, it reveals two temporal systems, the lunisolar month of Tishrīn al-Awwal, at which time the patriarch is expected to arrive at Beirut, and the Gregorian October, when he is to arrive at Alexandria. Finally, the item reveals several degrees of specificity: the patriarch is expected at Beirut during the vague ‘beginning’ of Tishrīn al-Awwal, but exactly on Monday, 9 October, at Alexandria. Clearly, the schedules of Austrian steamers were more exact than those of Roman Catholic patriarchs. Though the correspondence of the agents retained its personal nature (the patriarch was sailing ‘toward you’), telegrams adopted the monetized, compact and impersonal language of the medium: ‘Security in place. Attention is paid to the crops’ reads a terse Ministry of Interior report from al-Daqhalīyah (Tashrī῾āt wa-Manshūrāt, 19 May 1889, 414); ‘Security in place and health is fine’ reads another from Banī Suwayf (Tashrī῾āt wa-Manshūrāt, 19 May 1889, 414). Information from the Egyptian countryside came to newspapers

either by mail or by telegraph. Mailed reports often stressed non-events and were written in a personal and unhurried style. Telegrams, especially governmental ones, exemplified the new logic of importance: they were terse, specific and fresh. This configuration of news reportage positioned Egyptian urbanized newspaper readers in an uneven and uneasy relationship vis-àvis the seemingly action-packed and ‘close’ European centres and their slow, stagnant, uneventful and ‘remote’ immediate surroundings. Countertempos According to a familiar narrative of modernity, the telegraph introduced new forms of monetized textuality and temporality into the places it connected, decommissioning older ones such as the Hijrī calendar and the quotidian temporalities associated with it, the Arabic day and the uneven ‘temporal hour’.17 Telegraphy contributed to the formation of a new modern standard Arabic and had a prophylactic effect on linguistic ornamentalism and embellishment. And yet, while depleting the practical import of various traditional protocols of expression and synchronization, the telegraph charged these protocols with new energies and logics. As far as the Hijrī calendar was concerned, the telegraphic metamorphosis from a scheme indexing the facts of nature (like a clear moon in the night sky) into a matter of faith (a subjective eyelash) transformed the calendar into a free-floating, powerful cultural symbol, one whose very impracticality made it a suitable vessel for new ideological substance. For middle-class urban newspaper readers, modernized enough to know they were not modernized enough, telegraphy and its temporality fuelled inferiority complexes that the Hijrī calendar and similar ‘cultural’ forms of timekeeping in turn helped alleviate. By the first decade of the twentieth century, such cultural forms began coagulating into a so-called Egyptian time that was contrasted with Western alienating time, a temporality understood to be disenchanted and empty—vacant from metaphysics and devoid of the divine. Only against such ‘others’ could abstract mechanical time emerge as such, and in this sense ‘Egyptian time’ and similar colonial theories of relativity were conditions of possibility for the status

17

See, for example, Kern 1983.

SYNCHRONIZATION AND ITS DISCORDS

147

enjoyed by Western mechanical time as the golden standard. By the end of the decade, the 1908 Young Turk Revolution prompted in Istanbul the replacement of alla turka with alla franka time, marking the Ottoman empire’s embrace of Western time only a few years before the empire’s dissolution (Wishnitzer 2015). Yet, in Egypt, still formally an Ottoman province, the year was celebrated with a neoclassical poem titled ‘Taḥīyat al-῾Ām al-Hijrī’ (Long live the hijrī year). (The word taḥīyah comes from the root Ḥ-Y-Y, to revive, resurrect) (Ibrāhīm 1980, 2: 37–42). The 1909 poem by Ḥāfiẓ Ibrāhīm reclaims for hijrī time various developments during what modern historians recognize as the tumultuous year of 1908: constitutionalism in Turkey, civil unrest and struggle for political rights in Iran, anticolonialism in Afghanistan and scientific progress in India. Situating Egypt in this continuum, the poet applauds the new spirit animating the nation and proclaims that the days of slumber are gone. The year 1908 also saw the beginning of a labour militancy and then political agitation culminating a decade later in an anticolonial revolution directed against Egypt’s entire technical infrastructure—telegraphs, railways, tramways, telephones—and the mechanical time it held in place.

The indigenous ‘countertemporality’ of Egyptian time was a modern creation, but it retroactively sank roots in the ancient past; it was associated with slowness and imprecision but also with patience, authenticity, tradition and counterhegemonic modes of togetherness, thus offering a powerful critique of the time of empire and its technologies of rule. The calendric manifestations of this Egyptian time were not insulated from the techno-logics of the devices that decommissioned them: through its contact with the telegraph, the Hijrī calendar absorbed various technical suppositions about, and features of, mechanical time. Yet the interface of lunar- and techno-logics gave rise to a host of new ways for time to be experienced and new hierarchies among these ways. This hierarchization was an unruly process, and the very gestures that demoted certain temporalities in practice invigorated their symbolic import. Thus, even as the newly emergent ‘cultural forms’ became limited in their ability to play a structuring role in the social and political order, they could now offer new means for critiquing and resisting that order.

Bibliography

Barak, O. 2013. On time: Technology and temporality in Modern Egypt. Berkeley; Los Angeles, CA. Baring, E., Earl of Cromer. 1908. Modern Egypt. New York. Dallal, A. 2000. Appropriating the past: Twentieth-century reconstruction of pre-modern Islamic thought. Islamic Law and Society 7: 325–58. Dohrn-van Rossum, G. 1996. History of the hour: Clocks and modern temporal orders. Translated by T. Dunlap. Chicago, IL. Fabian, J. 2002. Time and the other: How anthropology makes its object. New York. al-Ḥakīm, T. 1992. The prison of life: An autobiographical essay. Edited and translated by P. Cachia. Cairo. Hanna, N. 2004. In praise of books: A cultural history of Cairo’s middle class, sixteenth to the eighteenth century. Cairo. Heyworth-Dunne, J. 1940. Printing and translations under Muḥammad ꜥAlī of Egypt: The foundation of modern Arabic. Journal of the Royal Asiatic Society of Great Britain and Ireland 3: 325–49. Hourani, A. 1983. Arabic thought in the Liberal Age, 1798– 1939. Cambridge.

Amīn, G. 2000. Whatever happened to the Egyptians? Changes in Egyptian society from 1950 to the present. Cairo. Anderson, B. 1983. Imagined communities: Reflections on the origin and spread of nationalism. London. ———. 1998. The specter of comparisons: Nationalism, Southeast Asia and the world. London. Asad, T. 2003. Formations of the secular: Christianity, Islam, modernity. Stanford, CA. Ayalon, A. 1995. The press in the Arab Middle East: A history. Oxford. Baedeker, K. 1902. Egypt: Handbook for travelers. Leipzig. Bakhtin, M. 1981. The dialogic imagination: Four essays, M. Holquist (ed.). Translated by C. Emerson and M. Holquist. Austin, TX. al-Bannā, J. 1990. Khiṭābāt Ḥasan al-Bannā al-Shābb ilā Abīhi: Ma῾ Tarjamah Mus’habah wa-Muwaththaqah li-Ḥayāt wa-῾Amal al-Wālid al-Shaykh Aḥmad al-Bannā. Cairo.

148

O. BARAK

Ibrāhīm, Ḥ. 1980. Dīwān Ḥāfiẓ Ibrāhīm, ꜥA. Amīn, ꜥA. alZayn (ed.). Cairo. Ibn Baṭūṭah. 1870–1871. Kitāb Riḥlat Ibn Baṭūṭah al-Musammāh Tuḥfat al-Nuẓẓār fī Gharā᾿ib al-Amṣār wa-ꜥAjā᾿ib al-Asfār. Cairo. al-Jabartī, ꜥA. 2004. Napoleon in Egypt: Al-Jabartī’s chronicle of the first seven months of the French occupation, 1798. Translated by S. Moreh. Princeton, NJ. ———. 2006. Al-Jabartī’s history of Egypt, J. Hathaway (ed.). Princeton, NJ. Jacob, W. C. 2011. Working out Egypt: Effendi masculinity and subject formation in colonial modernity, 1870– 1940. Durham, NC. Kāmil, A. F. 1908. Muṣṭafá Kāmil Bāshā fī Thalāthah wa-Arba῾īn Rabī῾: Sīratuhu wa-A῾māluhu min Khuṭab wa-Aḥādīth wa-Rasā’il Siyāsīyah wa-῾Umrānīyah. Cairo. Kern, S. 1983. The culture of time and space, 1880–1918. Cambridge. Kilito, A. 2008. Thou shalt not speak my language. Translated by W. S. Hassan. New York. Lane, E. W. 2003. An account of the manners and customs of the modern Egyptians. Cairo. Lane-Poole, S. 1883. Studies in a mosque. London. Le Goff, J. 1980. Time, work and culture in the Middle Ages. Translated by A. Goldhammer. Chicago, IL. Moiret, J. M. 2001. Memoirs of Napoleon’s Egyptian expedition, 1798–1801. Edited and translated by R. Brindle. London. Moosa, E. 1998. Shaykh Ahmad Shakir and the adoption of a scientifically-based lunar calendar. Islamic Law and Society 5: 57–89. Mubārak, A. 1989. Ḥayātī: sīrat al-marḥūm ꜥAlī Mubārak Bāshā. Cairo.

al-Munajjid, S. and Y. Khūrī (eds). 1970. Muḥammad Rashīd Ridā: Fatāwā al-Imām Muḥammad Rashīd Ridā. Beirut. al-Muṭī῾ī, M. 2000. Kitāb Irshād Ahl al-Millah ilā Ithbāt alAhillah. Beirut. Najm, N. 1987. Būr Sa῾īd: Tārīkhuha wa-Taṭawwuruha Mundhu Nash’atihā 1859 Ḥatā ꜥĀm 1882. Cairo. Povinelli, E. 2001. Radical worlds: The anthropology of incommensurability and inconceivability. Annual Review of Anthropology 30: 319–34. al-Rūmī. 1995. Tales of mystic meaning: Selections from the Mathnawī of Jalāl-ud-Dīn Rūmī. Translated by R. A. Nicholson. Oxford. Sāmī, A. 1936. Taqwīm al-Nīl. Cairo. al-Sanhūrī, N. and al-Shāwī, T. (eds). 2005. Al-Sanhūrī min Khilāl Awrāqihi al-Shakhṣīyah. Cairo. Shākir, A. M. 1986. Awā’il al-Shuhūr al-῾Arabīyah: Hal Yajūz Shar ꜥAn ꜥIthbātuha bi’l-Ḥisāb al-Falakī? Giza. Storey, G. 1951. Reuters: The story of a century of newsgathering. New York. al-Subkī, T. 1911. Kitāb al-῾Ilm al-Manshūr fī Ithbāt al-Shuhūr. Cairo. Van Dalen, B. 2000. Ta’rīkh. In The encyclopedia of Islam. 2nd ed. 10: 257–71. Leiden. Wishnitzer, A. 2015. Reading clocks, alla Turca: Ottoman temporal culture and its transformation during the long nineteenth century. Chicago, IL. Zerubavel, E. 1985. Hidden rhythms: Schedules and calendars in social life. Berkeley; Los Angeles, CA. Zwemer, S. M. 1913. The clock, the calendar, the Koran. Moslem World 3: 262–74.

III TOPOGRAPHY AND COSMOLOGY

WHAT’S FAITH GOT TO DO WITH IT? A DIACHRONIC PERSPECTIVE ON EMPIRE, LAND AND RELIGION Katherine BLOUIN

The capacity to represent, portray, characterize and depict is not easily available to just any member of just any society; moreover, the ‘what’ and ‘how’ in the representation of ‘things’, while allowing for considerable individual freedom, are circumscribed and socially regulated. E. Said, Culture and imperialism, New York, 1993

The inspiration for this paper comes from Edward Said’s concept of ‘imaginative geographies’ and, more broadly, from his reflection on the intimate connection between imperialism and geography: Underlying social space are territories, lands, geographical domains, the actual geographical underpinnings of the imperial, and also the cultural contest. To think about places, to colonize them, to populate or depopulate them: all of this occurs on, about, or because of land. The actual geographical possession of land is what empire in the final analysis is all about. […] Imperialism and the culture associated with it affirm both the primacy of geography and an ideology about the control of territory. The geographical sense makes projections—imaginative, cartographic, military, economic, historical, or in a general sense cultural. It also makes possible the construction of various kinds of knowledge, all of them in one way or another dependent upon the perceived character and destiny of a particular geography (Said 1993, 93).1

While Said’s Culture and imperialism focuses on the British, French and American novel from the eighteenth century onwards, his reflection on the manifold interactions between land, empire and culture is a theoretically suitable starting point for this chapter. For indeed, as anthropology has shown, what we commonly call ‘religions’ are also cultural systems (Geertz 1973), whose constitutive ideas and disciplinary practices stem from specific, historical conditions (Asad 1983, who critiques Geertz 1973). Bearing that

1

See also Said 1978, Part One; Gregory 1995; Said 2000; and, regarding the closely connected concept of ‘imagined communities’, Andersen 2006 [1983].

in mind, what do ancient representations of the Nile Delta tell us regarding the interplay between imperialism, land and religious dynamics from the early Roman to the Late Antique period? And how did the transition from a mostly polytheistic to a mostly Christian society impact upon these dynamics? Although less richly (and spectacularly) documented than the Nile valley, Fayum, Eastern Desert or the Western Desert and its oases, the Nile Delta has been at the heart of Mediterranean and Near Eastern connectivity networks since the Predynastic period (see O’Connell, this volume, Fig. 1). In addition to factual data, primary sources also convey ideas of Lower Egypt that are infused with both imperial and religious motifs: a land of environmental wonders and riches;a cosmopolitan and commercial hub; a breeding ground for spiritual regeneration and political resistance; the port of entry into/exit from Egypt. Such topoi recur in depictions of the Delta from ancient to modern, be they Egyptian or not, from the Inaros-Petubastis cycle to Claudius Ptolemy, the Bible to al-Maqrīzī, Homer to the Genizah archive, the Palestrina mosaic to Lawrence Durrell (cf. Vasunia 2001 regarding ancient Greek sources). For the purpose of this paper, I wish to identify and analyse the role of the Nile Delta and the ‘imaginative geographies’ associated with it in four descriptions of the region dated from the Principate to the Byzantine periods: a partially preserved, early second-century AD Greek invocation to Isis (P.Oxy. 11.1380); passages from the writings of Cassian and Egeria (fourth century

152

K. BLOUIN

AD); and the Madaba Mosaic Map (late sixth to early seventh century AD, henceforth MMM). Although of diverse origin, date and type, these artefacts all share four fundamental characteristics (or, one could argue, biases): they were produced within the Roman empire; they stem from educated individuals who belonged to the socio-economic elite of their time; they are filled with religious content; and they offer a description of the Delta. To date, the four sources I shall focus on have been chiefly studied individually. They are all commonly seen as religious objects that document the expansion of Isiac or Christian cults within (and beyond) Egypt and the Roman empire, and articulate the place of the Nile Delta within the broader religious landscapes of the time. It is my intention here to look at this corpus as a sample of not only religious, but also social and imperial,2 constructs, whose very production stems from an interaction, real or imagined, with the deltaic land. For these four sources also include interpretations of the Nile Delta, its land and its people, an act which, as emphasized by R. Mairs and M. Muratov in their 2015 study of interpreters in modern Egypt and Near East, ‘is never one of transparent mediation’ (Mairs and Muratov 2015, 10). By identifying the attributes, real and imagined, ascribed to Lower Egypt in these depictions, and by noticing elements of continuity and rupture between these three chronologically spread sets of evidence, I hope to highlight the power mechanisms at play behind these representations of deltaic space, and to reflect on the impacts of the Christianization of Egypt on these mechanisms. I shall first introduce each source or set of sources, before discussing the three main imaginative geographies they convey.

2

3

I follow here D. Mattingly’s (2011, 6) definition of empire (‘the geopolitical manifestation of relationships of control imposed by a state on the sovereignty of others’) and imperialism (‘both the process and attitudes by which an empire is established and maintained’) and I, like him, agree with the idea according to which imperialism is a dynamic phenomenon. For an extensive bibliography, see the Mertens-Pack 3 database record, no. 2477, http://cipl93.philo.ulg.ac.be/Cedopal/MP3/ displayMultipleNotices.aspx?selId=97938184 (last accessed 11 Nov. 2019). Most articles interested in this text focus on specific epithets associated with the goddess, the most commented upon by modern scholars being agapê, which is otherwise associated with Christian texts (Roberts 1953; Cazzaniga 1965; West 1967 and 1969; Witt 1968; Griffiths 1978 and 1985; Veligianni-Terzi 1986; Berenguer 1999). See also

Source: An invocation to Isis (P.Oxy. 11.1380) This invocation (ἐπίκλησις) to Isis has attracted some scholarly attention since its publication by B. P. Grenfell and A. S. Hunt in P.Oxy. 11 (1915) but, surprisingly, it has not yet been the object of any in-depth historical analysis.3 The roll was found in Oxyrhynchus in 1903 during the excavation of one of the ancient town’s rubbish mounds located outside the ancient city.4 It is now stored in the Egyptian Museum, Cairo. The invocation is written on the recto of the long roll (21.8 × 112.5cm), whose first columns are missing. The verso contains a praise of Imouthes-Asclepios, the hellenized version of the Egyptian god Imhotep. According to Grenfell and Hunt, palaeography indicates that the recto is ‘not later than Trajan or Hadrian’, and that the verso was written ‘under the Antonines’, that is roughly at the same time.5 As for the composition of the invocation, the editors believe that it ‘can hardly be placed later than the first [century AD]—a date supported by the evidence of some of the place names, which suggest the period between Strabo and Ptolemy, contemporary with Pliny’ (P.Oxy. 11.1380, intro., p. 191). The author of the invocation is unknown. Grenfell and Hunt and, after them, several scholars have noticed how parallels can be made with Egyptian, Greek and Latin religious and magical texts, the most famous of which is the invocation to Isis in the last book of Apuleius’ Metamorphosis (10.5).6 According to Grenfell and Hunt: It is an intentionally archaic kind of composition, as is clear on comparison with [P.Oxy. 11] 1381, which, though also a composition in praise of a Graeco-Egyptian deity and professing to be concerned primarily with the

4

5

6

Bricault 2006 regarding Isis’ maritime, naval and water-related epithets. More generally, see Totti 1985, no. 20. Preface to the edition P.Oxy. 11 = Grenfell and Hunt 1915. On Oxyrhynchus, see inter alia Bagnall and Rathbone 2004, 158–61 and Parsons 2007, 14–30. According to Grenfell and Hunt (1915, 190), the roll was reused. Yet the fact that P.Oxy. 11.1380 and 1381 are written in opposite directions means that both texts could be read one after the other. Their related nature and content, and their contemporary palaeographical dating, reinforce this hypothesis. We might, therefore, not be in the presence of a reused roll. Whether the two texts were copied both at once or one after the other (by different hands, if one relies on Grenfell and Hunt’s assessment) is, however, impossible to tell. P.Oxy. 11.1380, intro., p. 190. See also Magnani 2001.

WHAT’S FAITH GOT TO DO WITH IT?

translation of a hieroglyphic roll, is much more Greek than Egyptian in character and style […] The author of 1380 was no doubt a priest of Isis, possibly at Oxyrhynchus, where Isis had a separate temple […], but more probably at Memphis, which not only is dignified by an unusual name in l. 2 [‘House of Hephaestus’], and singled out in l. 249, but affords a connecting link with the text on the verso.7

The hypothesis that this text stems from a religious context seems quite solid. Since the roll was found among the rubbish mounds of Oxyrhynchus, the archaeological context (which is not detailed in the edition) is of no help. In any case, one can assume that the invocation’s author and audience were Greek-speaking followers (priests, initiates, worshippers?) of the goddess. Twelve columns of the invocation are preserved. Overall, they offer what J. Assmann considers both a translation and an interpretation of foreign divine names seen as comparable to that of Isis, a tradition that can be traced back to third-millennium BC Mesopotamian texts.8 They consist of two sections. The first one is a list of titles and other goddesses’ names used to refer to Isis in several dozens of localities in (1–76) and outside (76–119) of Egypt (‘the names of nations’; Assmann 1997, 49–50); the second one, which begins with an enumeration of general titles that are not associated with a specific location (119–42), is a prose hymn to the goddess (142–298). Given that the first preserved toponym of the section (Ἡφαίστου οἴκου9) is to be located in Memphis, at the apex of the Delta, and that only what appears to be the preamble and the very beginning of P.Oxy. 11.1381 remain on the verso,

7

8

9

10

P.Oxy. 11.1380, intro., p. 195. On the scholarly consensus regarding the role played by Memphis in the development of Isis’ cosmopolitan aretalogy, see Sfameni Gasparro 2007, 59–62. One should also note the Isis self-revelation text in Greek known through multiple inscribed versions that are said to have been copied from an original in Memphis (see on the matter Totti 1985, no. 1; I thank the anonymous referee for this reference). Assmann 1997, 45–46; see also, for similar dynamics regarding Hermes Trismegistos, Fowden 1986. Another toponym is partly preserved on l.1: [ ]πόλει. The editors have reconstructed [τὴν ἐν Ἀφροδίτης ?] πόλει, but they admit in the note to this line that other toponyms would also be possible. Accordingly, I prefer to leave this toponym aside. ‘The list of Egyptian places that occupies 1380, 1–76 covers only the Delta, but the towns of Upper Egypt on the same scale would not have taken up more than three or four preceding

153

Grenfell and Hunt deduce that several columns of text are missing on both sides. In the case of P.Oxy. 11.1380, the missing columns would correspond to the beginning of the invocation, which, given the geographical logic of the remaining columns, included a segment of section 1 dedicated to Upper Egypt.10 From this it can be deduced that the section on Lower Egypt was originally in the middle of a wider enumeration that covered the whole province, as well as areas located beyond its borders. The listing in section 1 seems partly to obey a hydrographical logic, whereby the author followed the course of the Nile from south to north (if we assume that the lost portion of the invocation included a list of Upper Egyptian locations), before moving to sites and areas located along the eastern Mediterranean coast, and beyond. Otherwise, though, the enumeration does not seem to follow a systematic order: Deltaic areas covered sometimes ‘jump’ from west to east, and the same goes for Mediterranean ones; the order does not seem to proceed from a fluvial logic within the Delta itself, nor does it follow the traditional numbering of the nomes; areas or cities located beyond the Roman empire appear mixed in with toponyms referring to places located within the empire, which indicates that the latter was not the main articulating principle of the list. Finally, since this is a prose text, motivations stemming from metrics must be excluded. The depiction of the Nile Delta offered by this invocation consists essentially of toponyms. In total, sixty-eight or sixty-nine place names located in Lower Egypt are preserved.11 They refer to towns, metropoleis and nomes (used as a complement to town names,

11

columns, and what preceded these is unknown’, P.Oxy. 11.1380, intro., p. 191. This number includes Kasion (74–75) and Ekregma (75–76), which appear in a cluster just after Pelousion (74) all located in the northwestern Sinai. I also wonder whether the Arabia (77) that follows refers to the southeastern deltaic city of the same name mentioned by Egeria instead of to the province or peninsula of Arabia (hence the ‘68 or 69’). The town of Rhinokoloura, which is also located in the northern Sinai and is pictured as part of Egypt in the MMM, appears later (93) in a cluster with Dora, Kaisareia and Askalon, all located in Judaea. According to Verreth (2006, 56–59), this indicates that the invocation’s author(s) saw the northwestern Sinai as part of Egypt, a divide that agrees with Pliny the Elder (Natural history 5.14.68). See also the section below on the MMM (though of a later period).

154

K. BLOUIN

apparently as a way to distinguish a particular town from a homonymous one located in another nome). Two sanctuaries—Hephaisteion (2) and Ision (54)—are also used as toponyms, and one entry concerns the Boukoloi (41–42).12 No hydronyms and no specific references to features of the landscape are found. The general order of the list, which follows the Nile from upstream to downstream, and then covers different areas along the eastern Mediterranean Sea, is evocative of the area’s hydrography. The same goes for some of the epithets associated with Isis (Bricault 2006). As H. Verreth underlines, the association of Isis with a place does not necessarily mean that there was a temple to the goddess there (Verreth 2006, 58). What such associations tell us—and what matters in the context of this chapter—is how Isis was represented by those who copied, read and listened to the invocation. In P.Oxy. 11.1380, the goddess appears as the ruler of a vast oikoumene, which extends from the city of Rome to the Ganges, and from the Hellespont to the Red Sea (Sfameni Gasparro 2007, 54–55). The multiple epithets and assimilations of which she is the object not only emphasize her multifaceted and highly syncretic nature, but they also offer a ‘relativization of all cultural and national differences as mere surface phenomena to be set off against the background of a common universal religion’ (Assmann 1997, 50). The text, therefore, acts as a dictionary thanks to which her universal appeal becomes—through translation and interpretation— understandable and relatable to all. The territory covered by the invocation can be viewed in two further ways. First, it is Egyptocentric. Not only does Egypt appear first, but the concentration of toponyms is not matched by any other extra-Egyptian area. In fact, the more remote an area is from the province, the less numerous the toponyms are and the larger the territories they refer to. To put it metaphorically, one could compare Isis’ cultic impact in the invocation to a seismic map: the closer to the epicentre, the higher the impact. Moreover, the invocation’s geography is closely tied to water and land transportation

networks: the Nile, the Mediterranean and the Red Sea on the one hand; and on the other, the coastal road running eastward from Libya to Syria-Palestine and beyond.

12

13

On the historical and literary Boukoloi (lit. ‘cattle herders’), see notably Blouin 2014, ch. 9.

Source: Christian travellers, Cassian and Egeria The writings of Cassian and Egeria offer two consistent yet complementary perspectives on Christian travellers who sojourned in Egypt in the late fourth century AD. Cassian: A novice among Lower Egypt’s monks What we know about Cassian’s life and the fifteen years or so he spent in Egypt essentially comes from his monastic writings, The Institutes of the Cœnobia and the Confessions of the Desert Fathers.13 He was born c. AD 365 in Scythia Minor (today Dobruja, a region including the Danube Delta that is shared between Bulgaria and Romania). Thanks to his wealthy family, he benefited from an education in both Greek and Latin. Called by the monastic lifestyle, he travelled as a teenager to Palestine with his friend Germanus. After spending a little while in a monastery in Bethlehem, the two young men left for Egypt around AD 385–86 and—apart from a hiatus during which they went back to their cenobium in Bethleem—remained there until 399–400. They first reached the island of Thenessus (Kom Tennis or Tinnis) by sea, and then travelled by foot, with the aim of reaching Upper Egypt and its hermits. As often happens in the course of travels, however, their itinerary changed and instead they ended up spending several years in the northern Nile Delta (which they crossed from east to west) and Scetis. The instigator of this change of plan was Archebius, the archbishop of Panephysis (el-Manzala), a town located on the shores of the Manzala Lake, whom Cassian and Germanus had met upon their arrival in Thenessus. The man encouraged the two novices to take the opportunity to visit some anchorites

For a detailed biographical summary with references to Cassian’s writings, see Pichery 1955, 7–23 and Stewart 1998, 7–19.

WHAT’S FAITH GOT TO DO WITH IT?

living in the deserted salt marshes located in the vicinity. After spending some time there and meeting with three men named Chaeremon, Nesteros and Joseph, Cassian and Germanus sailed to Diolkos, a monastic centre located in the central Delta, by a mouth of the Nile. There, they visited local coenobitic communities and, also, ascetic monks living on an otherwise deserted littoral island, before heading to the desert of Scetis (today, the Wadi Natrun), which became their monastic base and where they stayed for seven years. They also probably visited Kellia and Nitria during this period. Cassian and Germanus left Scetis for good in the context of the first Origenist controversy (late fourth century AD). Their path led them to Constantinople, where John Chrysostom, then bishop of the imperial capital, welcomed them together with other monastic ‘refugees’. It was he who ordained Germanus priest and Cassian deacon. In 405, after John’s exile, the two friends were sent to Rome to meet its archbishop, Innocent I. Cassian, who was probably ordained priest during that time, stayed in the city—and in the archbishop’s entourage—for a decade or so, while Germanus, who disappears from the sources, seems to have died there. It was, therefore, alone that Cassian moved to Marseilles shortly after AD 415. In that city, he founded two monasteries and, following a request from Castor, the bishop of Apt, wrote his two monastic works. These later achievements of his led him to be considered the ‘importer’ of Egyptian monasticism to the western Mediterranean. Scattered in a few passages of both of Cassian’s works are some rather pragmatic descriptions of his journey through the Delta (places visited and, explicitly or implicitly, means of travel) as well as a few short, but highly interesting, descriptions of the landscapes he encountered. His relationship with Lower Egypt is one motivated by ascetic pursuit; hence his focus on monastic rather than secular sites and his preference for anchoreticism over coenobitism. Accordingly, he mostly spent his time in agriculturally marginal or sterile environments such as the salty marshes of the Manzala Lake area (whose transformation from an inhabited to a deserted region he writes about), an isolated

155

island off Diolkos and the desert edges of the western Nile Delta. While the contribution of the Egyptian desert to early monasticism has been amply recognized and studied by scholars, the comparable role of ‘wet deserts’—be they salty marshes or inhospitable islands—which Cassian’s testimony illuminates, deserves to be explored further (Blouin 2019). Cassian’s descriptions of the Nile Delta must, of course, be understood within the wider context of his work. They were also written fifteen years after his departure from Egypt—that is, at a time when his memories of the place were no longer fresh (whether he relied on earlier notes, which he would have had to carry with him throughout his Mediterranean journey, is unknown). As C. Stewart underlines, ‘details about landscapes and people are primarily illustrative, included to lend both atmosphere and authority to his own synthesis of monastic theology’ (Stewart 1998, 7). Yet, as we shall see, these scattered memories—coming from someone who had himself lived in and moved throughout Lower Egypt for fifteen years—are also imbued with pieces of information thanks to which we can grasp how the region was integrated within the imperial and Christian geopolitics of the time. Egeria: A female traveller to the Holy Land The female writer Egeria was rediscovered in the late nineteenth century AD, when part of an eleventhcentury copy of her Travels was found in Italy by G. M. Gamurrini (Wilkinson 1971, 7). Because only the middle part of it remains (approximately a third of the original text), our knowledge of the exact identity and trajectory of this woman is fragmentary. Like Cassian, she was born in the fourth century outside Egypt; in her case, either in Aquitaine (Gaul) or in Galicia (Spain). We know that she was Christian and that she visited the east between AD 381 and 384, that is, at the same time as Cassian. Her ability to do so indicates that she came from a wealthy family, while her encounters, stays and travels with religious men (bishops, monks) and the military indicate that she belonged to the elite. Whether or not she was a nun is still subject to debate.14

14

On ‘who’ Egeria was, see notably Sivan 1988a and 1988b and Wilkinson 1971, introduction.

156

K. BLOUIN

Unlike Cassian and Germanus, Egeria travelled to both Lower and Upper Egypt. Her deltaic journey was in two parts. First, she sailed from Palestine to Alexandria and proceeded from there to the monastic sites of Nitria, then to the Thebaid. From there, she travelled back north along the Nile and, after passing through the ‘Land of Goshen’, an area located in the southeastern Delta, followed the easternmost (Pelusiac) branch of the Nile, before proceeding through the northern Sinai, and from there headed to Galilee. Later, following an undetermined period of time spent in Jerusalem, she left for a second trip to Egypt, by land this time. She visited Mount Sinai first, before coming back to the Nile valley via the Wadi Tumilat, and, just like the last leg of her first journey, reached Syria-Palestine via the northern Sinai. Egeria’s testimony is all the more valuable in that it is unique for its period. Not only is she one of the very rare ancient women whose writings have come down to us (and one, in this case, who specifically writes for a female, Christian audience), but she is also the only fourth-century pilgrim to leave us a first-hand account. What survives of it includes her journey from the vicinity of Mount Sinai to Constantinople, as well as her description of the services performed in Jerusalem on the occasion of the Christian new year. Like Cassian, her motivations are chiefly religious:

narrative whose aim is to promote varied forms of worship and the imperial-like might of one specific god, be it the multifaceted Isis, or the God of the Bible (see Assmann 1997, 44). Source: Madaba Mosaic Map

This remark by J. Wilkinson highlights one of the important characteristics of all the sources examined in this study: they all offer a selective, religiously exclusive portrayal of the Nile Delta. Indeed, the places mentioned, territories crossed and features depicted are all justified by their contribution to an overarching

In the nineteenth century AD, Madaba, an artificial mound that had been the site of a flourishing city from the Roman to the Byzantine and Umayyad periods,15 was practically deserted. Things changed in the early 1880s, when Greek-Orthodox Christians from al-Karak (a city located to the east of the Dead Sea) settled there. In 1884, while removing the debris covering the foundations of the city walls, they discovered a large mosaic map that was once part of the floor of a large church (often called ‘cathedral’ by scholars). After a rough repair, the map was incorporated into their new, smaller church to Saint James (now dedicated to Saint George; Meimaris 1999). Following the publication of the first sketches and description of the map in 1897 by K. Koikylides, librarian of the Patriarchate in Jerusalem, the map started garnering more scholarly attention and attracting pilgrims and tourists to Madaba. It was thoroughly restored by a German team in 1965 and remains to this day one of the main tourist attractions of Jordan (Donner 1992, 11–12; Meimaris 1999). Apart from the Tables of Peutinger (a copy based on a fourth-century AD original), the MMM is the earliest known map of Palestine and, according to H. Donner, ‘the most exact example of ancient cartography before the beginning of modern cartography in the nineteenth century’ (Donner 1992, 80).16 The precise date of its creation is unknown, but its stylistic and topographical features indicate that it was made between the late sixth and early seventh century AD (Avi-Yonah 1954; Piccirillo 1989; Donner 1992). The identity of the artist(s) who conceived and made the mosaic is lost to us, but the excellent quality of the work indicates that the makers were highly skilled. As for the sources of the map itself, a distinction must be made between the section on the Nile Delta (Fig. 1) and the rest: while the map

15

16

Egeria, as a Roman citizen within a Roman and still largely pagan world, displays an almost complete indifference to anything non-Christian. It is of course impossible to say with confidence what Egeria did not describe, since we now have only a fraction of what she wrote. But the part we have is enough to show that for her the non-Christians are usually important only when they affect the Church (Wilkinson 1971, 4).

The site was occupied as early as the fourth millennium BC; see Piccirillo 1989 for a summary.

On a discussion regarding both maps, see Weber 1999.

WHAT’S FAITH GOT TO DO WITH IT?

157

Fig. 1: The Nile Delta according to the Madaba Mosaic Map (Courtesy of the Photo Archives of the Studium Biblicum Franciscanum, Jerusalem).

158

K. BLOUIN

in general seems to rely on a combination of Christian sources (the Septuagint, Eusebius’ Lexicon of biblical place names) and secular ones (Late Antique road maps and itineraries), as well as on the cartographer’s own knowledge of (part of the) region, the section on the Delta is based on another set of sources (Donner 1992, 21–27). Indeed, not only are biblical sites one would expect to find on such a map conspicuously absent (for instance Pithom, Pi-Ramses or the Exodus route), but of the fourteen toponyms preserved, only two are included in Eusebius’ Lexicon. Furthermore, whoever conceived the map resorted to either Herodotus’ description of the deltaic hydrography, or else to a later source based on Herodotus. Evidence of this is found in the six extant branches of the Nile and their names.17 From this, it is evident that the cartographer was learned in both Classical and biblical literature. The surviving portion of the MMM is 10 × 5m (i.e., 50m2), but its original size is thought to have been 15.6 × 6m (i.e., 93m2; Donner 1992, 16). Based on what survives of it and on its position within the church, the mosaic seemingly allowed worshippers to situate themselves both literally and spiritually within the Christian landscape of the wider biblical lands:

The preserved fragments cover an area roughly encompassing the River Jordan running into the Dead Sea (north to centre), the mountainous area east of the Dead Sea up to al-Karak (east), the Palestinian coastal area (west), and northern Sinai and the Nile Delta, including a little remaining part of the Mediterranean Sea (south).

Two smaller fragments, representing respectively a part of Galilee and an area called Zebulun, also remain to the north of the main fragment. According to H. Donner, the original map may have had the following border: % North: the southern part of the Phoenician coast % South: the Sinai and Egypt, perhaps all the way upstream to Thebes % East: the desert edge located just beyond al-Karak, with maybe Bostra, Amman and Petra (the latter based on the restitution of a toponym of which only 2.5 letters remain; Donner 1992, 41–42) % West: the Mediterranean Sea (Donner 1992, 16). The depiction of the Nile Delta is located on the southwestern edge of the main fragment, not too far from the putative southern edge of the original map. The Delta is bordered by the Sinai to the east, the Palestinian coast to the north and the Mediterranean to the west. The southern lacuna breaks just after the apex of the Delta, where part of a crocodile and a fish are pictured; more fish are found in the Canopic and Pelusiac branches, while a boat floats at the starting point of the Sebennytic branch. North of the Saitic branch, one sees a rectangular shape the colours of which are similar to those used to picture waterways. In his discussion of Thenessos (Tinnis), which is located on the rectangle’s edge, Donner wonders: ‘Is the rhombus above the inscription a hint of the lagoon region in which the city was situated? Or an indication of the “field of Zoan” in Ps. 78:12?’ (Donner 1992, 84–85). I would favour the first hypothesis, and suggest that this body of water represents the Manzala Lake. In addition to four hydronyms (Pelus[iac], Sai[tic], Sebenny[tic], Bolbiti[c]), fourteen toponyms are also preserved in totality or in part (with or without an article), and all of them are accompanied by representations of cities in varying sizes: Pelusium, Nikiou, Athribis, Sethroïtes, Tanis, Thmuis, Thennesos, Sais, Xois, Paulinos (Kafr Bulim?),18 Hermoupolis (Parva), Chortaso, Kainoupol (= Kaine) and Chạ[ireos]. Apart from

17

18

The surviving remains of the mosaic map are most impressive and instructive. The Holy Land with Jerusalem in its center—the ‘navel of the earth’—is seen from the west. It is represented in such a way that the viewer could fancy himself standing on a high position in the west or hovering above the Mediterranean Sea. The map is oriented to the east, not to the north, corresponding to the orientation of nearly all Christian churches. […] A screen separated the mosaic map [on the east] from the so-called schola cantorum (Donner 1992, 16).

The Bucolic branch, which is not visible on more recent pictures of the map, can be seen on older photographs. As for the Mendesian branch, which is the seventh and only branch missing (it is presumably lost in the lacuna), it would have flowed from the Saitic branch; Donner 1992, 68.

See TM 11403 (http://www.trismegistos.org/place/11403, 23 June 2017).

WHAT’S FAITH GOT TO DO WITH IT?

Sethroïtes, which (mistakenly?) refers to a nome, all are names of towns. Several of them (Xois, Sais, Nikiou and Pelusium) are located inaccurately with regards to the deltaic hydrography, a phenomenon that contrasts with the general accuracy of the map (Donner 1992, 80). As for what part of Egypt is missing, according to Donner: ‘To the south it is self-evident that Mount Sinai, Alexandria and the pilgrimage centre of St Menas, Memphis and Heliopolis were represented, perhaps even Thebes’ (Donner 1992, 16). There are three main problems with this hypothesis: 1. Αlthough some locations such as Mount Sinai and Alexandria could well have been pictured, the large number of deltaic towns that are not mapped and the absence of apparent, straightforward religious logic behind the cartographer’s choices make any such reconstruction highly speculative. 2. St Menas, or Abu Mina, was a pilgrimage centre, not a monastery, and it is unclear why it would have been included in the map, all the more since its development postdates both biblical and Classical works. 3. The fragment of the map stops shortly after the apex of the Delta, and the space between the edge of the fragment and the southern border of the mosaic makes it difficult for the Nile to be pictured all the way to the Eastern Desert and Upper Egypt, unless, as proposed by Donner, the Nile runs from east to west. This orientation would match the biblical tradition, according to which the Nile was one of the rivers of Paradise and therefore flowed from the east (Donner 1992, 67). Still, since no biblical event is specifically set in Upper Egypt, one might suspect that there was no need for the map to incorporate that region, unlike the Wadi Tumilat, the biblical Land of Goshen, and the Red Sea. For now, let us conclude that the surviving portion of the MMM that has come to us includes the Nile Delta as one of the liminal areas of a wider map of the biblical eastern Mediterranean, whose centre is Jerusalem, and that the cartography of the Delta includes both iconographic and written elements that refer to the physical (waterways and, in the margin, the Sinai), animal (fishes, crocodiles) and human (boat, depictions of towns, hydronyms, toponyms) landscape of the region. While the map has been a topic of interest for art historians and scholars of early cartography and ancient Jordan, it has, like P.Oxy. 11.1380, attracted very little attention among historians of the eastern Mediterranean and of Egypt. Yet beyond the fact that it is an

159

invaluable testimony of early Christian iconography, architecture and spatial representations, this mosaic is also the product—and therefore the testimony— of a wider socio-economic and imperial context that shaped its very creation and reception. The same applies to P.Oxy. 11.1380 and to Cassian’s and Egeria’s writings. Imagined geographies In spite of their chronological and structural differences, the four sources discussed above all include religiously motivated and, therefore, subjectively informed portrayals of the Nile Delta that are embedded within a wider imperial context. That being the case, their authors also resort to some overlapping sets of representations that both proceed from and reproduce an imagined geography of Lower Egypt. In this section, I will focus on three representations that, in my view, are central to our understanding of the relationship between environment, faith and imperialism in the region: the Nile Delta as an anthropicized, as a liminal and as a connected land. An anthropicized land One striking feature of all four sources on which this chapter focuses is that their description of the Nile Delta is a functional one. What stands out are places, people and the divine, be they expressed through toponymy in the invocation of Isis, mentions of/excursus on encounters with religious and imperial officials ([arch]bishops and monks, soldiers) in the cases of Cassian and Egeria, or depictions of towns. The land itself does not matter as much as who lived on it, be they Isis worshippers, Christian pilgrims, monks, longdeparted biblical figures or Christian communities. As such, the natural environment acts as a background: implied in the case of the invocation to Isis; as a means to an end (i.e., isolated and deserted solitudes, suitable for monasticism) for Cassian; as fetishized settings of biblical episodes for Egeria; as ‘Holy Land’ navigator in the MMM. The result is a sacred geography that is significant to the authors and readers/viewers of our sources because it is the stage where the human and one divine power interact. Indeed, while Isis’ assimilation with many goddesses may be interpreted as witnessing syncretic practices, these other divinities are relevant inasmuch as they are seen as kaleidoscopic components of one, universal Isiac essence. As for

160

K. BLOUIN

Cassian, Egeria and the creator of the MMM, they make no mention of non-Christian communities (be they polytheistic or Jewish) still present in the Nile Delta in Late Antiquity. Just as it is subject to the imperium of an earthly ruler, so the Nile Delta is a physical manifestation of the all-encompassing, divinely ruled mundus (Brown 1992). The focus of these representations of the Delta is also essentially urban. The deltaic toponyms preserved in P.Oxy. 11.1380 and the MMM only designate towns, metropoleis and nomes and, although Cassian and Egeria’s respective journeys were motivated by a will to reach marginally located monastic settlements situated at the edges of, or beyond, the Delta, both their itineraries are punctuated by urban stopovers in towns that, just like the places named in the invocation to Isis and the MMM, appear to be transport and religious hubs. The vast deltaic countryside (already famous in antiquity for its size and fertility), its agricultural landscape and its rural inhabitants are not alluded to, except in an excursus of Egeria regarding the Land of Goshen (see below). This is not surprising given the nature of our evidence and, more generally speaking, the universal role of cities as centres of religious, economic and political power, which contrasts with the more limited (yet no less significant) role of individual villages within the Roman provincial power structure. On this matter, of our four sources, only Egeria mentions Alexandria. In the case of the MMM, Alexandria could have been featured in the lacuna; if so, it would have been located at the very edge of the map, a position that, in itself, contrasts with the central position of Jerusalem. As for P.Oxy. 11.1380, different motifs could have been involved: the toponym Nêsos (τῇ Νήσῳ, 68–69) is believed to be a reference to Alexandria. In that case, we would be in the presence of a metonymic association between the Greek capital of the province of Egypt, the famous island of Pharos located at the entrance of its maritime harbour and, more importantly perhaps, its famous lighthouse. We can also wonder whether the use of such a toponym, like that of Hephaisteion (2) to refer to Memphis, also proceeds from the (real or pretended) archaic nature of the composition. Although Cassian is believed to have

been to Alexandria, nothing in his writings confirms it. As for Egeria, in a passage related to her second journey through the Land of Goshen, she briefly mentions having seen it during her first trip to Egypt, when she visited Alexandria and the Thebaid (Travels 9.1.6).19 No other detail of her stay there is preserved. The passage also implies that she did not go to Alexandria or to Upper Egypt on her second trip, when she reached Egypt by land via Pelusium. More generally, that city seems not to have been a must-visit destination for Late Antique pilgrims and monks entering Egypt from Palestine. Indeed, the eastern Delta was a closer, and therefore more convenient, port of entry by land and by sea (as was the case for Cassian and Germanus). The peripheral role of Alexandria in our corpus contrasts with the otherwise important political and economic status of the city. One must understand this as a result of the religious context from which each depiction of the Delta emanates: Oxyrhynchite priests or worshippers of Isis; a novice monk interested in anchoreticism; a Christian, female pilgrim to the ‘Holy Land’; the floor-map of a church. Yet, partly thanks to Alexandria’s unassuming role, these representations of the Delta illuminate the diachronic conspicuousness and socio-economic drive of Lower Egypt’s many urban centres, a fact that has traditionally been underemphasized in favour of the more stereotypical image of the Delta as a vast ‘granary’ giving way on its northern fringe to wild, isolated marshes (Blouin 2017). The lesser presence of monumental and papyrological remains in the region and the scholarly focus put on Alexandria have certainly contributed to the perpetuation of these topoi, some of which go back to antiquity. Yet recent scholarship has shown how biased and, therefore, erroneous they are. Most recent archaeological and historical work on the region has indicated that, by ancient standards, Lower Egypt was a relatively densely populated, occupied and urbanized territory, whose people and resources were fully integrated into the province’s economy. This appears to have been the case even in so-called marginal areas, and on those grounds too, the present corpus brings forward further pieces of evidence for this phenomenon (Blouin 2014).

19

See also Peter the Deacon Y1-3 (in Wilkinson 1971, 203–5).

WHAT’S FAITH GOT TO DO WITH IT?

A liminal land The concept of liminality has been used in anthropology following A. van Gennep’s theory and, mostly, its use by V. Turner in his 1967 study on rites de passage (van Gennep 1909; Turner 1967). Understood at first mostly within ritual contexts, it came to be associated with a variety of transitory states, processes and spaces. For the purpose of this paper, I am interested in spatial liminality. Generally speaking, the Nile Delta is a liminal territory in two ways: 1. From a geopolitical point of view, it has become a cliché to say that Egypt was a crossroads. Yet this cliché is anchored in fact, and especially so when it comes to the Nile Delta. The region was an inescapable part of the journey for anyone/anything travelling between the Mediterranean and the Syria-Palestine corridor on the one hand, and the Red Sea and Indian Ocean on the other. In the Roman period, Egypt was one of the frontier provinces of the empire. This led the Nile Delta and the adjacent Wadi Tumilat to play a crucial role in the so-called international Red Sea trade, be it, in the pharaonic period, between Egypt and Punt, or, later, between the Roman empire, the Indian Ocean region and China (Aubert 2004; Tomber 2008; Sidebotham 2011). The region was also the transit point for all the local commodities (notably, but not exclusively, the wheat) paid as tribute or sold on imperial and international markets. 2. From an environmental point of view, the abundance of wetlands and lakes in the northern Delta, as well as the extended deserts on its eastern and western edges, are agriculturally marginal ecosystems that have since the pharaonic period been exploited in a variety of ways and associated (rightly or wrongly) with such topoi as wilderness, barbarism, religious and political resistance, and regeneration (Blouin 2014, chs 7 and 9). Both these aspects of the Delta’s liminality emerge from our corpus. First, the very position of the region is liminal in all four sources. In the invocation to Isis, the deltaic section is positioned between an almost completely lost section that most probably focused on the southern portion of Egypt, and another one dedicated to places outside Egypt. Both Cassian and Egeria entered Egypt from Palestine via the Delta, with the aim (fulfilled or not) of travelling to Upper Egypt. Whether they went there or not does not matter so much for present purposes. What matters is that the Delta was not their destination, but rather a stage on their respective journeys towards Upper Egyptian sites.

161

The depiction of Lower Egypt in the MMM is liminal in several ways: it is correctly located between SyriaPalestine, Upper Egypt and the Mediterranean Sea; it is one of the littoral zones separating the sacred (Holy Land) and profane (Mediterranean) sections of the mosaic; it stands close to the southern edge of the mosaic itself. On a more tangible level, agriculturally marginal environments, be they damp or dry, are present to varying degrees in all four sources. In P.Oxy. 11.1380, some settlements are to be located in the northern, marshy edge of the eastern Delta, where they acted as fishing and commercial harbours. This is the case for Thonis (28), the area inhabited by the ‘Boukoloi’ (42), and Phernouphis (57), all of which I have discussed elsewhere (Blouin 2014). The semi-arid littoral areas that border Lower Egypt on its northeastern and northwestern sides are also represented with a few settlements located west of Alexandria (Taposiris [67], Plinthine [73], Katabathmos Mikros [43], Apis [44], Leuke Akte [45]) and east of Pelusium, along the northern edge of the Sinai Peninsula (Kasion [74–75], Ekregma [75–76]). Although set in areas that were at the time structurally marginal for ‘mainstream’ agricultural activities such as grain farming or fruit- and vegetable-growing, these towns were deemed important enough to be included in the invocation to Isis. They were, therefore, not really marginal, but rather integrated within the wider deltaic landscape, be it religious or socio-economic. My work on the Mendesian Nome, as well as recent archaeological work at Taposiris-Plinthine and in the northern Sinai, also shows a greater integration than previously imagined (Blouin 2014 and 2017; see also Verreth 2006; Boussac, Dhennin and Redon 2015). The same general conclusions can be drawn from the limited evidence related to liminal environments pictured in the MMM. The preserved portion of the map contains only one allusion to Lower Egypt’s marshy environments. As proposed above, the rectangular shape located north of the Saitic branch, next to Thenessos (Tinnis), most probably represents the Manzala Lake. However, since the whole northern shore of the Delta is lost, we must consider it a possibility that other littoral lakes of the area were pictured, together with more sites. The only coastal area preserved corresponds to a portion of the northern Sinai, which was considered to be part of Egypt at that time, as specified on the map (Donner 1992, 77, no. 124). The names of five settlements are preserved partly or completely before the mention of the border: (Ὅροι Αἰγύπτου

162

K. BLOUIN

κ(αὶ) Παλαιστίνης): τὸ Ἀ[φνάιον], τὸ Πεντάσχοινον (el-Uqsor), τὸ Κάσιν (el-Qeds), Ὀστρακίνη (el-Felusiyat), ῾Ρινοκόρυρα (Rinokolura = el-Arish).20 Of these, Aphnaion, Ostrakine and Rinokolura were bishopric seats at some point (Verreth 2006, 350, 526–27). What remains after that indicates that more coastal settlements were on the map. Overall, one gets the impression of a well-occupied littoral, despite the roughness of the environment (a dangerous coast with sandbanks and shallow water, and generally arid, salty soil). This seems accurate, since, as archaeological work has shown, many military garrisons were located in the area throughout antiquity. Moreover, as has been observed before, the highly frequented coastal road leading from Egypt to Palestine ran through the area, which suggests that these places also acted as road-side stops, with all that such a function required in terms of infrastructures and commercial services. The most detailed evidence regarding the religious dimension of Lower Egypt’s liminal environments comes from Cassian and Egeria. In their works, the Delta’s margins are described as thriving monastic centres (both coenobitic and anchoretic).21 Far from being peripheral to the itinerary of pilgrims and monks visiting Egypt, these regions rather ended up being Cassian’s and Germanus’ destinations, and an important part of Egeria’s tour. Cassian’s Egyptian journey led him to four main places, all located on the edge of the Delta: Thenessus (Tinnis), Panephysis, Diolkos and Scetis (Martin 2015, 23–24). The first two are located in the area of the Manzala Lake, the third one at a mouth of the Nile in the central Delta, and the last one in the Wadi Natrun, southeast of Alexandria. In a passage that has been interpreted as describing some of the local effects of the earthquake and tsunami that hit the eastern Mediterranean in AD 365 (Henry 2012; Blouin 2019), Cassian describes the area close to Panephysis where an abbot named Paul lives:

20

Verreth 2006, 255–588, provides a thorough review and analysis of the sources related to all toponyms attested in the area.

For the former dwelled in the desert which is right near the city of Panephysis and this desert we know had been created not long ago by a flood of very salty water. This covers the whole surface of that region everytime that the north wind blows, surging out of the swamps and so overflowing the nearby land that it makes the old villages there, which for a long time had been utterly deserted for that very reason, look like islands (Cassian, Conf. 7.26, trans. Ramsey 1997).

Eremus, solitudo, salsissima inundatio, stagnum, desertus, insula: the Latin vocabulary used by Cassian shows how the conditions behind the constitution of sandy, salty and swampy areas were understood by his contemporaries as closely interconnected, something which, from an environmental point of view, is accurate. Indeed this, like the fact that all such types of ecosystems are ill-suited for agriculture, makes them, in different ways, deserts. It is precisely this deserted and desert dimension of the recently abandoned vicinity of Panephysis that attracted Abbot Paul and other monks. To them, the area’s agricultural marginality was, literally, central. The same can be said of Diolkos, a coastal settlement to which Cassian and Germanus headed after their stay in Panephysis: [W]hen we desired with still greater ardor to seek out also the more distant parts of Egypt in which a larger and more perfect number of holy men were dwelling, we came to a town called Diolcos, located at one of the seven mouths of the River Nile. In this we were urged on not so much by the requirements of our journey as by a yearning for the holy men who were living there (Cassian, Conf. 18.1.1, trans. Ramsey 1997).

As in the case of the depiction of the Nile Delta in the MMM, Cassian’s mention of ‘one of the seven mouths of the River Nile’ (uni ex septem Nili fluminis ostiis) must be seen as a direct reference to his ‘Classical’ paideia (Brown 1992). So must, in the previously quoted passage, his comparison of the villages in the area of Panephysis to islands, an image that recalls a

21

For a full survey of fifth- to nineteenth-century AD sources on the monastic sites of the Nile Delta, including the area of Alexandria and the Mediterranean shore but excluding the Wadi Natrun, see Martin 2015, ch. 1.

WHAT’S FAITH GOT TO DO WITH IT?

famous description of the effects of the Nile flood on the deltaic landscape found in Herodotus’ book II (97). This is hardly surprising, since apart from Ptolemy, all Greek and Latin descriptions of the Delta follow Herodotus and describe the region as being made of seven branches (Blouin 2014, 26–35). Furthermore, the toponym Diolkos appears in Ptolemy’s second-century AD Geography, where it is described as a ‘false mouth’ (Δίολκος ψευδόστομον) located on the northeastern edge of the Delta (Ptolemy, Geogr. 4.5.5; see Ball 1942 and Blouin 2014, 29–32 and map 3). Cassian’s ‘town’ was no doubt the eponymous settlement built on that location.22 Finally, the desert of Scetis, where the two friends were to settle for several years, is located in a salty desert—as the area’s Arabic name, Wadi Natrun, indicates. The region is a 25km-long depression made of sandy areas and dotted with several salt lakes and saltpans. Of the fifty or so monasteries known to have existed in the area, only four remain active today.23 There, as in Nitria and Kellia, the absence of any possibility for agriculture was seen as an ideal setting for anchoritic introspection: Therefore, the person who keeps constant watch over the purity of the inner man must seek out places that do not draw his mind to the distraction of cultivating them because of their abundant fertility and that do not make him leave the fixed and set location of his cell and force him to do work outside, thus scattering his thoughts at large, as it were, and, by all sorts of things, utterly diverting the aim of his mind and that most delicate focus on his goal. These cannot be provided for or seen by anyone, however careful and watchful, who has not kept his body and mind constantly enclosed within the confines of four walls. Thus, attentive and unmoving, like a clever fisherman looking out for his food with apostolic skill, he may catch the swarms of thoughts swimming about in the calmest depths of his heart and, like someone gazing intently into the depths from a jutting promontory, may with wise discretion judge which fish he should draw to himself with his saving hook and which ones he should let go and reject because they are wicked and harmful (Cassian, Conf. 24.131, trans. Ramsey 1997).

22

See TM 4061, https://www.trismegistos.org/geo/detail. php?quick=4061 (2 July 2019).

163

One can appreciate Cassian’s use of highly evocative and contrasting farming and fishing metaphors to describe the ideal work of the mind monks can hope to achieve while living in the arid solitudes of the desert of Scetis. The region’s sterility—no doubt enhanced by the saltiness of the soil—was such that absolutely no farming whatsoever seems to have been possible, something which, to Cassian, makes this desert preferable to the eastern deserts of Calamus and Porphyrion (located in Upper Egypt, west of Hurghada). Although much more isolated geographically, these monastic sites were fertile. Monks would, therefore, spend some time farming instead of remaining constantly by themselves in their cells. As a result, they would prove unable to deal with true solitude and silence when visiting the monasteries of Scetis (Conf. 24.4). The peculiar liminality of Scetis, located closer to the secular world yet more isolated from it environmentally, hence appears to Cassian as one of the causes of its higher spiritual status. For him, as for a great number of early Christian monks living in Egypt, this type of environment, like the marshy, brackish edges of the northern Delta, was a space where the human could best approach the divine. Finally, Egeria’s description of the Land of Goshen deserves a few words. According to the Bible, the Land of Goshen was allocated to the sons of Jacob (who had come to Egypt following a famine in Hebron) by Joseph. The area has been located by scholars around Phakoussai, in the Arabic Nome, that is close to the edge of the southeastern Delta: Our whole journey on from there [City of Arabia = Phakoussai] led through the land of Goshen, past vineyards (of grapes as well as balsam), orchards, well-kept fields, and many gardens. All the way the road followed the bank of the Nile past the extremely fertile farms which had once been the estates of the children of Israel, and really I do not think I have ever seen a landscape better kept than Goshen (Egeria, Travels 9, trans. Wilkinson 1971).

The sharp contrast between the desert road that leads to Phakoussai and the farmed countryside that starts at

23

The monasteries of the Arab Desert and Wadi Natrun: http:// whc.unesco.org/en/tentativelists/1827/ (23 June 2017).

164

K. BLOUIN

the border of the Delta must indeed have produced a lasting impression on anyone coming from the desert. Yet Egeria’s enthusiasm and idyllic description no doubt partly results from a mixture of retrospective idealization and biblical projection. Through a process that very much resembles the one common among nineteenth- and early twentieth-century travellers to Egypt (Mairs and Muratov 2015), she interprets the agricultural landscape she crosses according to what it means to her (the contemporary reproduction of the Land of Goshen), rather than to what it means to the locals (a succession of plots that have been made fertile through a combination of propitious environmental conditions and hard work). The very fact that she names the area ‘Land of Goshen’ is actually a token of her lack of interest in what the area was at that time. A connected land The four sources discussed all share another important feature: the Nile Delta they describe is embedded in geographically wider contexts. As seen above, the invocation to Isis covers an area stretching from Italy to India, and from the Hellespont to the Red Sea; both Cassian and Egeria travelled from Europe to Palestine, and from there to Egypt; the MMM depicts the biblical land as a whole, from the Syro-Palestinian coast to Egypt. None of these zones corresponds exactly to the Roman empire. This imperfect superposition of religious and imperial geographies highlights the precedence of the religious one, a fact that is not surprising given the context of each source. Yet this is not to say that imperialism did not play a role in the practices and motivations behind their creation: on the contrary. The close relationship between imperialism, commercial networks (both internal and external), and the movement of peoples and ideas (including religious ones) is a universal phenomenon. Although the Mediterranean and Red Sea/Indian Ocean trades are by no means an Augustan novelty, the unification of the shores of the Mediterranean Sea under one power created conditions—notably the pacification of the Mediterranean Sea and the development of a huge road

24

25

On roads in Roman Egypt, see Adams 2007; in the pre-modern world, see Alcock, Bodel and Talbert 2012. For instance, recent, groundbreaking work on the sites of Plinthine and Taposiris Magna has shown how the wider Alexandrian

network—that facilitated connectivity over enormous expanses of territory. It is in this context that the Oxyrhynchus invocation to Isis, an all-powerful goddess whose earthly realm was at the time conceived as extending over three continents, was written; it is in this context that wealthy, educated Christians such as Cassian and Egeria could safely travel, by sea and land, from areas as remote as the western Black Sea coast, Spain or France to Palestine, and from there to the Sinai, Upper Egypt and back;24 it is in this context that Cassian could export monasticism from Egypt to Marseilles, via Constantinople and Rome; it is, finally, also in this context that the pilgrims who visited Madaba could locate themselves on the church’s mosaic map, their feet atop the dark tesserae that represented the Mediterranean Sea. The extra-Egyptian locations listed in P.Oxy. 11.1380 can be understood as nodes within the connectivity networks that linked Egypt to the eastern empire (the absence of any locations west of Rome would be worth further exploration) and the Eurasian regions. The papyrus as a whole contains fifty-nine or sixty such references (the exact location of the toponym Arabia– Arabia Petraia [Jordan] or Yemen/Saudi Arabia? being uncertain), compared to seventy-three or seventy-four located within Egypt (of which, as noted above, sixtyeight or sixty-nine designate deltaic settlements). We might, however, expect the lost part of the invocation to have contained many Upper Egyptian ones. Most of what remains is made of toponyms (cities, provinces, regions), but a couple of hydronyms, such as the Ganges (226) and the Eleutheros rivers (225–26), and an ethnonym, the Amazones (102), are preserved. The vast majority of the geographical references are located in the Greek world at large (continental and insular Greece, modern Turkey) as well as in the Near East (Table 1). This should not come as a surprise given the intense degree of integration and connectivity between Egypt and the rest of the eastern Mediterranean throughout antiquity (if not, in certain respects, as early as the Predynastic period).25 The poor representation of the western part of the empire, of which only Cyrenaica (Kyrene) and Italy (Italy, Rome) are mentioned, is

area was already a centre of international exchanges long before the Macedonian conquest; see notably on the matter Boussac, Dhennin and Redon 2015; Dhennin 2015.

WHAT’S FAITH GOT TO DO WITH IT?

NORTH AFRICA AND CENTRAL MEDITERRANEAN Kyrene Rome Italy

3 toponyms TOTAL: 59

OR

60

GREECE (INCLUDING CYPRUS) AND TURKEY Lycia Myra Lycia Knidos Creta Chalkedon Cyclades Pathmos Paphos Chios Salamis Cyprus Asia Sinope Delphoi Thracia Delos Amazones (people) Thessalia Samothraike Pergamon Pontus Samos Hellespont Myndos Bithynia Tenedos Caria Troas Didyma Olympos 30 toponyms 1 ethnonym

NEAR-EAST (INCLUDING THE SINAI) Arabia? Chalkis Pieria Petra Hypsele? Rinokoloura Dora Kaisareia Askalon Raphia Tripolis Gaza Bambyke Phoenicia Syria Berytos Sidon Ptolemais Akko Tripolis Eleutheros

18 or 19 toponyms 1 hydronym

165 OUTSIDE

THE

ROMAN EMPIRE

Arabia? Nesos? India Persis Sousa Sousa on the Red Sea Ganges

5 or 6 toponyms 1 hydronym

GEOGRAPHICAL REFERENCES

Table 1: Extra-Egyptian geographical references in P.Oxy. 11.1380.

somewhat surprising, inasmuch as we know that the cult of Isis gained popularity there as well. Why is the popularity of Isis in these areas not better represented in the invocation? Does the composition of the text go back, perhaps, to before the Roman annexation of Egypt? I would be inclined to think so. Indeed, it is striking to see how, if one excludes Italy, Rome, Petra and, should it refer to the Roman province, Arabia, the territories covered by the invocation—the Greek world, the Near East, Persia, India, Libya—are all to be

located within Alexander the Great’s empire. I therefore suggest that the composition of the invocation took place in the Hellenistic period, if not before.26 Indeed, the separation between Egyptian locations and extraEgyptian ones could mean that the Hellenistic, Greek text relied on an earlier Egyptian original. As for the two scattered references to Rome (83), Italy (109), Petra (91) and Arabia (77) they could easily have been added after the annexation of Egypt by Octavian and the conquest of Arabia under Trajan. If this hypothesis

26

On this, compare the first hymn of Isidorus at Medinet Madi (early first century BC), esp. lines 14–23 (Moyer 2016, esp. 218–19). I thank the anonymous referee for this reference.

166

K. BLOUIN

proves valid, then the version of the invocation preserved in P.Oxy. 11.1380 could be seen as resulting from the superimposition of three maps of the cult of Isis, each of which was associated with a particular imperialism: an Egyptian, a Graeco-Macedonian and a Roman one. The ‘archaic’ nature of the text observed by Grenfell and Hunt would then be not so much a stylistic affectation as a token of the text’s age and ever-evolving nature. Just like Isis’ ‘empire’, the religious geography of the ancient Holy Land was, too, shaped by an imperial millefeuille. For the purpose of the present analysis, two main dynamics are crucial: 1. Most of the events narrated in the Bible took place within imperial states (Egyptian, Assyrian, Persian, Hellenistic, Roman) and involved interactions with (including resistance to) imperial authorities or policies. If one were to superimpose maps of these empires, the area with the most overlap would correspond to a great extent to the ancient ‘Holy Land’, including Egypt, where three fundamental episodes (the story of Joseph, the Exodus and the Flight to Egypt) take place (Römer 2014). Egypt’s holiness is one of the chief reasons behind Cassian’s and Egeria’s journeys there, and the reason why Egypt features in the MMM. 2. The composition of the Jewish and Christian Bibles took place within these same imperial contexts (essentially from the Persian to Roman periods). The same can be said of the three Christian sources on which this paper focuses, all of which were conceived thanks to the support of institutions (monasteries, churches, army), infrastructures (forts, roads, inns) and officials ([arch]bishops, monks, possibly nuns, priests, soldiers). In many ways, therefore, one can argue that the Judaeo-Christian notion of the Holy Land is in part an outcome of the region’s imperial histories. Where does the Nile Delta stand in this broader picture? To use a knitting metaphor, it is one of the intricate stitches that made up the religious fabric of the

27

R. Tomber (2008) has shown how these complex, far-reaching commercial networks were not so much ‘controlled’ as of particular ‘interest and profit’ to the Roman state, which levied a 25% state importation tax. Beyond some degree of diplomatic and infrastructural support (building, maintenance and securing of roads, canals, road stations, watchtowers and harbours) that was performed through a poorly documented combination of

eastern Mediterranean during antiquity. As part of a succession of empires, the ancient Nile Delta both connected and was connected to different socio-cultural and religious worlds: Egypt, the Syro-Palestinian corridor, the Near East, the Greek world, the Roman empire, the Red Sea, the Indian Ocean. In Egypt, as in the Delta in general, there was never any such thing as real isolation, even in agriculturally marginal areas where monks sought solitude, as the flow of pilgrims who, like Cassian, Germanus and Egeria, travelled there shows. The opportunistic and multifaceted connectedness of the deltaic margins is well described by Cassian in a passage dedicated to the town of Thenessus, where he and Germanus arrived by boat from Palestine: Therefore, having completed our voyage [from Syria], we came to an Egyptian town named Thennesus. Its inhabitants are so surrounded by the sea and by salt swamps that, because there is no land, they have devoted themselves to commerce alone and get their wealth and substance from sea trade. Indeed, when they want to build houses there is no land, unless it is brought from far away in boats (Cassian, Conf. 11.1; trans. Ramsey 1997).

Thenessus, more commonly known as Tinnis, was a city-island located in the Manzala Lake. Throughout Late Antiquity and the Arab period, it was a thriving commercial hub, and became one of the main textile centres of the eastern Mediterranean and Arab world (Gascoigne 2003; 2005; 2007; Blouin 2012). Cassian’s passage shows how connected the Nile Delta —including its northern, lacustrine edges—was to the rest of the Mediterranean. It also illuminates the role played by deltaic harbours in both Mediterranean and international traffic to and from Egypt. This is notably the case in the context of the Indo-Roman trade, whose systematic exploitation is attested from Augustus to the early to mid-seventh century AD.27 As for Egeria, she highlights how the northeastern desert (including the Wadi Tumilat) and the northern

military work, taxation and compulsory public services (including the delegation of responsibilities to communities or consortia of merchants), ‘evidence from documents, sites and finds all point to the Roman component of trade being undertaken by private individuals’ or entities (including, in Late Antiquity, the Christian Church); Tomber 2008, 152–53 cit. See also Adams 2007 and Sidebotham 2011 (esp. 150).

WHAT’S FAITH GOT TO DO WITH IT?

Sinai were well served with imperially secured roads and staging-posts that facilitated pilgrims’ land journeys to and from the Delta. After having walked from the Sinai through the Wadi Tumilat, Egeria’s group reached Arabia, a town located at the southeastern edge of the Delta. She then writes: At this point we dismissed the soldiers who had provided us with an escort on behalf of the Roman authorities when we went through the danger areas. We no longer needed military protection, since we were on the state highway from the Thebaid to Pelusium, which passed through the city of Arabia [=Phakoussai]. […] From the City of Arabia we travelled for two days through the land of Goshen, and then we arrived at Tathnis [Daphnai], the city which was once Pharaoh’s capital, in which holy Moses was born. […] From Tathnis we went by the way I already knew, and arrived at Pelusium. And from there we went on through all the staging-posts of our outward journey till we reached the frontier of Palestine. And from there, in the name of Christ our God, I returned after several staging-posts in Palestine to Aelia, which is Jerusalem (Egeria, Travels, 9, trans. Wilkinson 1971).28

Egeria’s testimony provides us with a first-hand account of what it was like for a member of the Christian elite to travel from Jerusalem to the southern Sinai and back in the late fourth century AD. More generally, the landmarks she uses to render her journey to her readers—towns, biblical sites, staging-posts—are indicative of her relationship to the territories she crossed. To her, the land is both Christian and Roman. Her pilgrimage in the past Holy Land is made logistically possible thanks to the present Roman imperial control over this same land and the communities who lived on it. Accordingly, her travel is one through both space and time. The same logic is visible on the MMM, where several of the sites depicted in the eastern Delta and northern Sinai were staging-posts of the sort alluded to—and no doubt visited by—Egeria. Conclusion: What’s faith got to do with it? This paper asked the following question: what do ancient representations of the Nile Delta tell us

28 29

See more generally Travels 7–9. For a criticism of the traditional representation of the ‘narratives of the end of paganism’, see Brown 1997, ch. 1. See also Chuvin 1990 and Cameron 2011.

167

regarding the interplay between imperialism, land and religious dynamics from the early Roman to the Late Antique period? The comparable, imagined geographies underlying the invocation to Isis preserved in P.Oxy. 11.1380, the writings of Cassian and Egeria, and the MMM indicate that Lower Egypt was understood and experienced as a sacred region, where the human and the divine were tightly intertwined. This religious ‘fertility’ manifested itself in tillable as well as in sterile areas. It was also rooted in and dependent upon a large number of urban centres that acted as religious and commercial hubs. Although the invocation to Isis antedates the rise of Christianity, its structure and content do not differ drastically from Cassian and Egeria’s works, nor from the MMM. In all four sources, the landscape, be it fertile or sterile, urban or rural, made of soil or water, was worth being described, mentioned or pictured inasmuch as it was a locus of human-to-God interaction. The gradual transition from a polytheistic to a monotheistic society and state did not change this fundamental logic. This is all the more true for our corpus, given that substantial parts of the province and the empire’s population were still polytheistic during the period it covers.29 Further, the various socio-cultural, ethnic and linguistic vantage points from which each source originates do not seem to have affected the way the Nile was represented. For that matter, the fact that the cartographer of the MMM relied in large part on the work of Herodotus for his rendering of the branches of the Nile Delta testifies to the enduring nature of Greek representations of Egypt, even within the context of the gradual ‘Christianization’ of the empire’s population. Such continuities no doubt proceeded from the authoritative nature of Herodotus (and other earlier Greek and Latin writers) among educated Christians and non-Christians alike, and, also, from pragmatic reasons. No matter how the religious map of the empire or the microecologies of the Nile Delta evolved between the fifth century BC and the fifth century AD, Lower Egypt retained its general topographical, hydrological, agrarian and liminal features. Furthermore, just as the rise of Christianity in Egypt (and beyond) did not bring about

168

K. BLOUIN

any significant change in the agricultural economy of the province, neither did it lead to any sharp disruption in the conception of the relationship between the land and the sacred. There is, however, one notable exception: anchoritic monasticism. As shown by P. Brown in his seminal book The body and society (1988), the pursuit of the divine in remote, sterile and deserted settings is a phenomenon that developed in Egypt as a response to the challenges posed by the ascetic quest for sexual renunciation:

That is to say, the innovation lay not so much in the act of retreating into the desert as in the institutionalization and scale of the model developed by Anthony. This is true for dry deserts such as Nitria or, in Upper Egypt, the Theban mountains; it is also true of ‘damp deserts’, whose representations as loci of spiritual regeneration go back to the pharaonic period (Yoyotte and Chuvin 1983, 53–58). How did the imperial variable intervene in these complicated, diachronic processes? First, I would like to suggest that it acted as a facilitator of religious connectivity. The unification of vast expanses of land and water under one imperial power—or, in the Hellenistic period, several—brought about important movements of peoples, goods and ideas, which not only fostered the ‘diasporic’ diffusion of both the Isiac cults and Christianity (to focus only on the specific religious representations discussed in this paper), but also led to different sets of associations between such divinities and imperial rulers.30 Imperial powers also had the means to integrate (through pacified sea routes or road networks), maintain and police the vast territories and

water bodies under their authority. Both Cassian and Egeria’s pilgrimages benefited from the pacified Mediterranean of their time. Egeria’s text also indicates that fourth-century Christian pilgrims used the imperial road network and that Roman soldiers were in charge of escorting them whenever they were heading off the beaten track. Although Cassian does not mention such infrastructures and escorts in his account, we can safely assume that he, too, resorted to them on more than one occasion. This supposes a certain degree of logistical sophistication, as well as the recognition by the state of the legitimacy of these religious movements within, and well beyond, the Nile Delta. It also implies that the pilgrims who benefited from such official protection and guidance belonged to a privileged, well-connected imperial elite. Second, to come back to the quote from Edward Said highlighted at the start of this paper, I would like to suggest that each of the four religious sources analysed here are, per se, demonstrations of power. The ‘capability to represent, portray, characterize, and depict’ to which they testify supposes a combination of knowledge (literary, religious, of Greek and/or Latin), financial means, mobility and time that was the prerogative of only a few inhabitants of the Roman empire. For, despite their respective singularities, the Oxyrhynchus invocation to Isis, Cassian’s and Egeria’s works and the MMM were all produced by members of the educated, religious elite who were under the protection of powerful patrons31 (the Isis clergy, Christian lay people, officials or institutions). Further, the sacred geography they promote was ‘in sync’ with the imperial religious policies of the time, be they support or sympathy for polytheistic (including Isiac) cults or for (a particular type of) Christianity. Such dynamics imply that the selective representations they convey are partly standardized and streamlined, and that this process was, consciously or not, motivated by a sense of accountability and a quest for status preservation or promotion. Accordingly, they should not be taken at face value when it comes to documenting the whole spectrum of individual religious practices, identities

30

31

The myth of the desert was one of the most abiding creations of late antiquity. It was, above all, a myth of liberating precision. It delimited the towering presence of ‘the world’, from which the Christian must be set free, by emphasizing a clear ecological frontier. It identified the process of disengagement from the world with a move from one ecological zone to another, from the settled land of Egypt to the desert. It was a brutally clear boundary, already heavy with immemorial associations (Brown 1988, 216).

See notably Bricault, Versluys and Meyboom 2007 as well as Bricault and Versluys 2014 regarding the Isiac cults.

On the role of lay people in the early history of Christianity, see notably Bowes 2008.

WHAT’S FAITH GOT TO DO WITH IT?

169

and geographies (what one could call lived religion; McGuire 200832) associated with the Roman Nile Delta, whether among worshippers of Isis, Christians, or the population of the time as a whole.33 Seen in this light, the link between imperialism, power and religious representations operates not so much on the act of representing, but rather, to refer to Said once again, on the what and how of the representations. Can religious movements survive without political patronage? Yes. Can religious texts, buildings and works of art be produced without the (un)official support—logistical, legal, religious, political—of the imperial state their authors live in? Of course. But whenever that is the case, their number, quality, outreach and ability to last through time tend to be

considerably reduced, and therefore harder for us to trace.34 This is in great part why papyri related to ‘Christians’ are almost unknown before Constantine (Choat 2012). And this is also why our latest Isis aretalogy dates from the fourth century AD (Martzavou 2012, 268). Yet when it comes to religiosity and cultural dynamics, there is no such thing as a deep, overnight paradigm shift (Brown 1997, ch 1). No matter what official, upper-class discourses (in other words, the voices of those who have the means to speak the loudest, both in their own days and in historical records) may boast about or conveniently ignore, human history is ultimately about overlaps, diversity and convoluted webs of slowly shifting continuities.

Bibliography

Bagnall, R. S. and D. W. Rathbone. 2004. Egypt from Alexander to the early Christians: An archaeological and historical guide. Los Angeles, CA. Ball, J. 1942. Egypt in the Classical geographers. Bulaq [Cairo]. Beard, M., J. North and S. Price. 1998. Religions of Rome I: A history. Cambridge. Berenguer, J. A. 1999. Isis como ἀγάπη θεῶν en P.Oxy. 1380. In Τῆς φιλίης τάδε δῶρα: Miscelánea léxica en memoria de Conchita Serrano, 43–54. Madrid. Blouin, K. 2012. Minimum firmitatis, plurimum lucri: Le cas du ‘lin mendésien’. In Actes du 26e Congrès international de papyrologie. Genève, 16–21 août 2010, P. Schubert (ed.), 83–89. Geneva. ———. 2014. Triangular landscapes: Environment, society and the Nile Delta under Roman rule. Oxford. ———. 2017. Beyond the Nile: Orientalism, environmental history and ancient Egypt’s Mareotide (northwestern Nile Delta). History Compass 15.10 (https://doi. org/10.1111/hic3.12397, last accessed 11 Nov. 2019).

All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http:// papyri.info/docs/checklist (last accessed 8 July 2019). Adams, C. 2007. Land transport in Roman Egypt. Oxford. Alcock, S. E., J. Bodel and J. A. Talbert (eds). 2012. Highways, byways and road systems in the pre-modern world. Oxford. Andersen, B. 2006 [1983]. Imagined communities: Reflections on the origin and spread of nationalism. London. Asad, T. 1983. Anthropological conceptions of religion: Reflections on Geertz. Man, NS 18 (2): 237–59. Assmann, J. 1997. Moses the Egyptian. Cambridge, MA. Aubert, J.-J. 2004. Aux origines du canal de Suez? Le canal du Nil à la Mer Rouge revisité. In Espaces intégrés et ressources naturelles dans l’Empire romain, M. Clavel-Lévêque and E. Hermon (eds), 219–52. Besançon. Avi-Yonah, M. 1954. The Madaba Mosaic Map, with introduction and commentary. Jerusalem.

32

33

For an exploration of this topic on the basis of Late Antique data, see Rebillard and Rüpke 2015 (esp. 3–12). On the importance of, yet challenges associated with, the distinction between public and private religious identity in Late Antiquity from a legal perspective, see Noethlichs 2015. For a study of ‘popular’ Christianity in Roman times on the basis of archaeological evidence, see MacMullen 2009; for a more sociologically integrated approach to Christian identity in Late Antiquity, see Burrus and Lyman 2005.

34

Similarly, Mattingly (2011, 29) highlights the paucity of Roman-period sources from ‘subject people’. On Roman imperial approaches to religious control and integration, see Beard, North and Price 1998, 339–48: ‘Roman authorities moved to suppress (or “emend”) religious forms that seem to be a focus of opposition to Roman rule – whenever and wherever they found them’ (339).

170

K. BLOUIN

———. 2019. God is in the marshes: Late Antique asceticism and the northeastern Nile Delta. Journal of the Canadian Society for Coptic Studies 11: 45–64. Boussac, M.-F., S. Dhennin and B. Redon. 2015. Recent work at Taposiris and Plinthine. In Alexandria: Current archaeological activities and future perspectives, M. Haggag (ed.), 189–217. Alexandria. Bowes, K. 2008. Private worship, public values and religious change in Late Antiquity. Cambridge. Bricault, L. 2006. Isis, Dame des flots. Liège. Bricault, L. and M. J. Versluys. 2014. Power, politics and the cults of Isis. Leiden; Boston, MA. Bricault, L., M. J. Versluys and P. G. P. Meyboom (eds). 2007. Nile into Tiber: Egypt in the Roman world. Leiden; Boston, MA. Brown, P. 1988. The body and society: Men, women and sexual renunciation in early Christianity. New York. ———. 1992. Power and persuasion in Late Antiquity: Towards a Christian empire. Madison, WI. ———. 1997. Authority and the sacred. Cambridge. Burrus, V. and R. Lyman. 2005. Shifting the focus of history. In Later Christian antiquity, V. Burrus (ed.), 1–24. Minneapolis, MN. Cameron, A. 2011. The last pagans of Rome. Oxford. Cazzaniga, I. 1965. Intorno ad alcuni epiteti di Iside nella litania di P. Oxy. 1380. Parola del passato 20: 233–36. Choat, M. 2012. Christianity. In The Oxford handbook of Roman Egypt, C. Riggs (ed.), 474–89. Oxford. Chuvin, P. 1990. Chronique des derniers païens. Paris. Dhennin, S. 2015. Plinthine et la Maréotide pharaonique. Bulletin de l’Institut français d’archéologie orientale 115: 15–35. Donner, H. 1992. The Mosaic Map of Madaba: An introductory guide. Kampen. Fowden, G. 1986. The Egyptian Hermes. Cambridge. Gascoigne, A. L. 2003. The Medieval city of Tinnis. Egyptian Archaeology 22: 25–27. ———. 2005. An archaeological survey of Tell Tinnis, Manzala, Egypt. Antiquity 79: 303. ———. 2007. The water supply of Tinnīs: Public amenities and private investments. In Cities in the pre-modern Islamic world: The urban impact of religion, state and society, A. K. Bennison and A. L. Gascoigne (eds), 161–76. London. Geertz, C. 1973. Religion as a cultural system. In The interpretation of cultures: Selected essays by Clifford Geertz, 87–125. New York. van Gennep, A. 1909. Les rites de passage. Paris. Gregory, D. 1995. Imaginative geographies. Progress in Human Geography 19 (4): 447–85. Grenfell, B. P. and A. S. Hunt. 1915. The Oxyrhynchus papyri XI. London. Griffiths, J. G. 1978. Isis and the ‘Love of the Gods’. Journal of Theological Studies 29 (1): 147–51.

———. 1985. Isis and agape. Classical Philology 80: 139– 41. Henry, M. 2012. Cassien et les autres: Les sources littéraires et les événements géologiques du 21 juillet 365. Klio 94: 175–96. MacMullen, R. 2009. The second church: Popular Christianity A.D. 200–400. Atlanta, GA. Magnani, A. 2001. Iside, Apuleio ed il P. Oxy. XI 1380. Analecta Papyrologica 13: 107–13. Mairs, R. and M. Muratov. 2015. Archaeologists, tourists, interpreters: Exploring Egypt and the Near East in the late 19th–early 20th centuries. London. Martin, M. 2015. Monastères et sites monastiques d’Égypte. Cairo. Martzavou, P. 2012. Isis aretalogies, initiations and emotions. In Unveiling emotions: Sources and methods for the study of emotions in the ancient world, A. Chaniotis (ed.), 267–91. Stuttgart. Mattingly, D. 2011. Imperialism, power and identity: Experiencing the Roman empire. Princeton, NJ. McGuire, M. B. 2008. Lived religion: Faith and practice in everyday life. Oxford. Meimaris, Y. 1999. The discovery of the Madaba Mosaic Map: Mythology and reality. In The Madaba Map centenary, 1897–1997, M. Piccirillo and E. Alliata (eds), 25–36. Jerusalem. Moyer, I. 2016. Isidorus at the gates of the Temple. In Graeco-Egyptian interactions: Literature, translation and culture 500 BC–AD 300, I. Rutherford (ed.), 209–44. Noethlichs, K. L. 2015. The legal framework of religious identity in the Roman empire. In Group identity and religious individuality in Late Antiquity, É. Rebillard and J. Rüpke (eds), 13–27. Washington, DC. Parsons, P. 2007. City of the sharp-nosed fish. London; Phoenix, AZ. Piccirillo, M. 1989. Madaba: Le chiese di mosaic. Milan. Pichery, E. (ed.). 1955. Jean Cassien: Conférences. Paris. Ramsey, B. 1997. John Cassian: The Conferences. New York; Mahwah, NJ. Rebillard, É. and J. Rüpke (eds). 2015. Group identity and religious individuality in Late Antiquity. Washington, DC. Roberts, C. H. 1953. Ἀγάπη in the invocation to Isis (P.Oxy. XI, 1380). Journal of Egyptian Archaeology 39: 114. Römer, T. 2014. Bible et Égypte. In Le livre des Égyptes, F. Quentin (ed.), 293–306. Paris. Said, E. 1978. Orientalism. New York. ———. 1993. Culture and imperialism. New York. ———. 2000. Invention, memory, place. Critical Inquiry 26 (2): 175–92. Sfameni Gasparro, G. 2007. The Hellenistic face of Isis: Cosmic and saviour goddess. In Nile into Tiber: Egypt in the Roman world, L. Bricault, M. J. Versluys and P. G. P. Meyboom (eds), 40–72. Leiden.

WHAT’S FAITH GOT TO DO WITH IT?

Sidebotham, S. E. 2011. Berenike and the ancient maritime spice route. Berkeley, CA. Sivan, H. 1988a. Holy Land pilgrimage and Western audiences: Some reflections on Egeria and her circle. Classical Quarterly 38 (2): 528–35. ———. 1988b. Who was Egeria? Piety and pilgrimage in the age of Gratian. Harvard Theological Review 81 (1): 59–72. Stewart, C. 1998. Cassian the monk. Oxford. Tomber, R. 2008. Indo-Roman trade: From pots to pepper. London. Totti, M. 1985. Ausgewählte Texte der Isis- und SarapisReligion. Heidelsheim; Zurich; New York. Turner, V. W. 1967. The forest of symbols: Aspects of Ndembu ritual. Ithaca, NY. Vasunia, P. 2001. Hellenism and empire: Reading Edward Said. Parallax 9 (4): 88–97. Veligianni-Terzi, C. 1986. Bemerkungen zu den griechischen Isisaretalogien. Rheinisches Museum für Philologie 129: 63–76.

171

Verreth, H. 2006. The northern Sinai from the 7th century BC till the 7th century AD: A guide to the sources. Leuven, Trismegistos Online Publications, http://www. trismegistos.org/sinai/. Weber, E. 1999. The Tabula Peutingeriana and the Madaba Map. In The Madaba Map centenary, 1897–1997, M. Piccirillo and E. Alliata (eds), 41–46. Jerusalem. West, S. 1967. An alleged pagan use of agape in P.Oxy. 1380. Journal of Theological Studies 18: 142–43. ———. 1969. A further note on ‘agape’ in P.Oxy. 1380. Journal of Theological Studies 20: 228–30. Wilkinson, J. 1971. Egeria’s travels. London. Witt, R. E. 1968. The use of agape in P.Oxy. 1380: A reply. Journal of Theological Studies 19: 209–11. Yoyotte, J. and P. Chuvin. 1983. Strabon: Le voyage en Égypte. Paris.

BEYOND EMPIRE: KOSMAS INDIKOPLEUSTES, THE CHURCH OF THE EAST AND THE INDIAN TRADE IN THE SIXTH CENTURY AD Maja KOMINKO

Since Augustus’ conquest, Egypt and Alexandria constituted a major gateway for Rome’s eastern trade. Written documents, such as the mid-first-century AD Periplus Maris Erythraei and the mid-second-century Muziris papirus (P.Vindob. G 40822), give an image of booming commerce, with Roman merchants undertaking the long trip to India (Casson 1990; Rathbone 2000). Evidence from excavations in the Red Sea ports indicates that the trade continued until the fifth century, but that by the middle of the sixth most Late Roman ports were much reduced or entirely abandoned (Sidebotham and Wendrich 2001–2002, 23–50; Tomber 2008, 154–70; for an overview see Tomber 2018). It is often assumed that Aksumites took an increasingly significant role as mediators of Rome’s trade with India (Sidebotham 1991, 34; Mayerson 1996, 122; Morony 2004, 186–87), but with diminishing success as the Sasanians dominated the Indian Ocean (Whitehouse and Williamson 1973, 43–44). Discussion of Roman maritime trade with the east in this period often relies on the Christian Topography, a sixth-century anonymous treatise written in Alexandria and attributed from the eleventh century onward to Kosmas Indikopleustes, that is, Kosmas who sailed to India.1 Evidence extracted from this work is usually analysed without consideration for its literary form or cultural and religious context, and as a result it is frequently misinterpreted. Kosmas’ interest in the eastern trade, the information he provides on the spread of Christianity from Persia, and his declared reliance on the teaching of a katholikos of Persia led scholars to assume that his mercantile and religious interests were interwoven, and that his commercial pursuits were facilitated through networks of ‘Persian Nestorian Christian’ traders operating in the east (Whitehouse and Williamson 1973, 44; MacCoull 2007, 67–82;

1

For an edition of the text, see Wolska-Conus 1968–1973. In quoting the Christian Topography I follow Wolska’s numbering of paragraphs.

Tomber 2008, 23). Such a hypothesis ignores the scope of Kosmas’ work, and vastly exaggerates his engagement with contemporary Christological debates. It also relies on an anachronistic identification of the sixthcentury East Syrian church (the Church of the East) as Nestorian and dogmatically opposed to Chalcedonian orthodoxy. Finally, it suggests that mercantile networks relied on ecclesiastical allegiances transcending political boundaries. This chapter attempts to clarify the intellectual and religious context of Kosmas’ work and analyse his insights into mercantile contacts with the east in light of written and archaeological evidence. Such close reading suggests that not only the notion of the ‘Nestorian networks’ but also the assumption of Sasanian dominance in the Indian Ocean may require revision. Moreover, considering India’s western trade solely in terms of the competing interests of Rome and Sasanian Persia, with Aksumites presented merely as ‘middlemen’, is too reductive. Finally, while the volume of the eastern trade may have dwindled in the sixth century, it appears that the types of imported commodities changed, and their diversity may have, in fact, increased. The Christian topography and Book XI on the plants and animals of India The work known as the Christian Topography is not homogeneous. It consists of the Christian Topography proper, plus a treatise on cosmography made up of ten books, to which two more books by the same author, Book XI on the plants and animals of India and Book XII containing excerpts from ancient authors, were added at a later date. The Christian Topography was written to disprove the idea that the earth and heaven are spherical. The

174

M. KOMINKO

author of the treatise opposed this theory because, following Aristotle, the spherical and perpetually rotating universe was believed to be eternal, a notion unacceptable from the point of view of Christian dogma (I.5–8; II.40). Instead, he proposed a universe based on a flat, rectangular earth and divided by the firmament into two superimposed spaces. The lower space houses the present, earthly condition and serves as a school in which humankind gradually progresses from Adam to the Second Coming of Christ. The upper space, destined for the future condition in heaven, will open after the Second Coming (Kominko 2013, 42–48) (Fig. 1). The universe is based on the flat earth, which consists of the rectangular oikoumene surrounded by the Ocean, which is in turn encompassed by another land, with Paradise in its eastern part. The Ocean is not navigable, but the four gulfs—the Roman Gulf (the Mediterranean), the Red Sea, the Persian Gulf and the Caspian Sea—are open to seafaring (Kominko 2013, 52–65) (Fig. 2). Kosmas declares that all his theories and his entire learning come from his master Patrikios (II.2), identified with Mar Aba, a teacher at the school of Nisibis and a katholikos of the Church of the East. He also claims to lack a complete pagan education (τῆς ἔξωθεν έγκυκλίου παιδείας) and familiarity with rhetoric (II.1), but in reality he is conscious of rhetorical tools and strategies, repeatedly strives to give his argument a classical guise, and draws ample inspiration from classical geographies and histories (Kominko 2013, 14–15). Within the Christian Topography all information, no matter how tangentially related to his theories, serves to corroborate his cosmography. From the text we learn that the Christian Topography was written in Alexandria and that it had two editions. The original one, comprising the first five books, can be dated to shortly after AD 543.2 The second edition included an additional five books, written in response to the criticism encountered by the original publication. Book VI must have been written after AD 547.3 It does not seem possible to date Books VII–X more precisely and some of them could have circulated as independent publications before they were incorporated into the

2

Kominko 2013, 12. The evidence for dating is as follows: the elevation of Mar Aba to the episcopal see of Chaldea (AD 540) in II.2; an earthquake which destroyed Corinth (AD 543) in I.22; Kosmas’ journey to Aksum ‘some twenty five years before, at the beginning of the reign of Justin, emperor of the Romans, when Ellatzbaas [’Ella-’Aṣbeḥā or Kaleb], then king

second edition of the treatise, but it is likely that they postdate the Christian Topography: each of them contains references either to the treatise, or to the system of the universe described in the original publication (Wolska-Conus 1973, 304, n. 67.2; Kominko 2013, 12). Book XI and Book XII of the modern edition were appended to the Christian Topography at a later stage, but before the ninth century, when the treatise is quoted by Photius (Photius, Bibliotheca 36). They appear only in two out of three complete codices of the Christian Topography; absent from the earliest, ninth-century manuscript (Vaticanus Gr.699), they are included in two eleventh-century codices (Sinaiticus gr.1186 and Laurentianus Plut.IX.28). Book XI combines descriptions of plants and animals of Ethiopia and India with a short discussion of Sri Lanka, India and a description of trade in the region. Book XII, not covered by this chapter, contains a discussion of excerpts from ancient writers, who, according to Kosmas, confirm the antiquity and veracity of the Bible (XII.1). Book XI is likely an abbreviation of a lost original. It lacks the introductory paragraphs that open all the other books. Many chapters are very short and the text shifts abruptly between subjects. It is often thought that Book XI is an abridgement of Kosmas’ lost treatise on geography, mentioned in the prologue to the Christian Topography in the two eleventh-century codices (McCrindle 1897, VIII; Wolska-Conus 1968, 256, n. 1.4; 1973, 314 note; Schwarz 1975, 475). According to this prologue, the geographical treatise described the whole earth, that is, the earth beyond the Ocean, and the oikoumene with all the countries. The text further specifies that the geography provided an account of the region south of Alexandria, along the Nile to the Ocean, the peoples of Egypt, Ethiopia and beyond, the Arabian Gulf and the lands and peoples around it, with the intention of further corroborating theories explained in the Christian Topography. Such a description makes the identification of Book XI as a summary of the geographical treatise questionable: not only are India and Taprobane (Sri Lanka), on which much of Book XI

3

of the Aksumites, was about to begin the war against the Ḥimyarites’ (II.56), an event dated to AD 518 (Beaucamp, Briquel-Chatonnet and Robin 1999; Hatke 2011, 66–71, 75–84). As indicated by a reference to two eclipses, see VI.3; Grumel 1958, 202; Oppolzer 1962, 168.

BEYOND EMPIRE

Fig. 1: Christian Topography. The Universe, Laur.Plut.IX.28, fol. 95v (Courtesy of Biblioteca Medicea Laurenziana).

175

176

M. KOMINKO

Fig. 2: Christian Topography. The Earth, Laur.Plut.IX.28, fol. 92v (Courtesy of Biblioteca Medicea Laurenziana).

BEYOND EMPIRE

focuses, not mentioned, but also the text implies that the geographical work expounds on the cosmographic theories of the Christian Topography. Book XI does not contain any references to the biblically inspired geography and cosmography described by Kosmas. While in the Christian Topography information on trade and geography is always presented with reference to Kosmas’ biblical system, in Book XI such framing is absent. The few biblical references, relegated to the end of the book, are limited to the identification of India with the biblical Evilat (Genesis 2.10–12), the country of Ham (Genesis 10.6–7), and Indus with the biblical Phison (Genesis 2.11; XI.16), a river identified in the Christian Topography with both the Indus and Ganges, which are thought to be the same (II.81). There is little doubt, however, that the Christian Topography and Book XI were written by the same author. There is an overlap in information, for example, in the description of Sri Lanka and the Christian community on the island (II.45, III.65 and XI.13), and in the outline of trade with China (II.45–46 and XI.15– 16). There is also a similar tendency to include colourful and only tangentially relevant stories with a classical pedigree. No references in the text allow the dating of Book XI. It has been suggested that the king of Sri Lanka mentioned in Book XI, in a story which pre-dates the writing of the book by several decades, is Moggallana I (r. c. 495–515), but nothing in the text allows us to make such an identification (XI.17–19; Schwartz 1975, 484). Likewise, the identification of Gollas (Γολλᾶς), the ruler of the White Huns, with Mihirakhula (r. c. 510–540) based on the alleged similarity of the name, is not entirely convincing (XI.20; Schwartz 1975, 471). Because Book XI contains no references to the cosmographical system of the author, it seems likely that it was written before the Christian Topography, probably before Kosmas met his teacher Mar Aba and developed his cosmography inspired by the latter’s teaching. Such a tentative chronology would mean that Book XI was written before Kosmas came into contact with the teaching of the Church of the East, a dating that undermines his purported connection with Persian Christian ‘Nestorian’ networks, in which Kosmas allegedly participated in his mercantile dealings with the east. The ‘Nestorian’ problem In the late fifth and early sixth century the miaphysites often levelled accusations of the Nestorian heresy

177

against their Chalcedonian opponents, and where these occur they should be treated with caution (Gray 1979, 35; Dijkstra and Greatrex 2009, 228). Such an accusation has been often levelled against Kosmas by modern scholars (MacCoull 2007, 69–72), but never by any Byzantine authors. Photius criticizes Kosmas’ language and outlandish theories, but does not question his theology (Bibliotheca, cod. 36). The Christian Topography is frequently quoted in manuscripts from the eleventh century onward in introductions to Gospels and Psalms, which suggests that its orthodoxy was not in question (Wolska-Conus 1968, 94–116). We should also note that, while Kosmas’ work does not focus on Christology, where the natures of Christ come into a symbolic discussion of cosmography, he does not see them as separate (II.99, 102; V.62, 254; VII.65; X.48). The only author who undermines Kosmas’ orthodoxy is his contemporary opponent, John Philoponus, a miaphysite writer educated under Ammonius in the school of Alexandria. Philoponus was the first Christian author to successfully undermine the notion of the eternity of the universe, demonstrating on the grounds of ancient science that such a notion cannot be correct (Sorabji 1987, 5–18). Kosmas opposes Philoponus’ acceptance of the spherical universe and throughout the Christian Topography describes him as dangerously detached from the Bible. Philoponus took this accusation seriously enough to respond in his treatise De opificio mundi, written between 547 and 560 (Scholten 1996, 65; Elweskiöld 2005). While ridiculing Kosmas’ theories, he also undermines his opponent by stressing his dependence on Theodore of Mopsuestia, newly condemned as a heretic (Wolska 1962, 161–67). Despite the polemical context, most modern scholars have taken Philoponus’ statement at face value, and have gone even further, identifying Kosmas not only as the follower of Theodore, but also as a Nestorian. Theodore of Mopsuestia (d. 428), confirmed as orthodox at the Council of Chalcedon, was seen as a Nestorian by anti-Chalcedonian miaphysite writers, but his works were read and accepted by the Chalcedonians (Pseudo-Zachariah of Mytilene, Historia ecclesiastica 39, 27). Theodore was condemned by Justinian’s edict of 543 and the anathemas at the third council of Constantinople in AD 553, but both of these were motivated by the desire to coax miaphysites into accepting Chalcedon rather than by recognition of Theodore’s affinity to Nestorius (Gray 1997; Price 2009, 319–20). In the Christian Topography, the discernible influence of Theodore of Mopsuestia is combined with

178

M. KOMINKO

notions that derive from East Syrian exegesis and were probably mediated through East Syrian sources. This is in keeping with Kosmas’ statement that his theories depend on the teaching of his East Syrian teacher (Kominko 2013, 19–23). Kosmas’ dependence on the teaching of Mar Aba and his link with the East Syrian Church are often taken as further confirmations of his Nestorian affiliation. Nevertheless, identification of the sixth-century East Syrian Church as ‘Nestorian’—that is, adhering to the diophysite Christology attributed to Nestorius—is anachronistic. It is only under Babai, in the 610s, that the Church of the East adopted an extreme diophysite Christology, with its assertion of two qnōmē (hypostases) in the Son (Brock 1985, 125–42; Chediath 2002, 59–65; 2007, 205–14). In the first half of the sixth century, the East Syrian Church was not Nestorian and its orthodoxy clearly did not appear suspect. East Syrian clergymen not only participated in theological debates at the court in Constantinople, but also acquired recognition among influential spheres in the capital. Paul the Persian, a winner of the three-day debate against a Manichaean opponent organized by Justinian in 527, may be the same person as the teacher mentioned by Junillus Africanus in the introduction to his Instituta regularia, written in the early 540s in Constantinople (Maas 2003, 16; Bruns 2010, 263–68). Junillus Africanus, the chief legal minister of the Roman empire, goes so far as to present his work as notes from Paul’s lectures. Although this is an exaggeration, Junillus, like Kosmas, had no qualms about declaring reliance on the teaching of East Syrian masters, which indicates that the orthodoxy of the school of Nisibis was not in question in the early 540s (Maas 2003; Becker 2006b, 29–47). Kosmas writes that Patrikios/Mar Aba came to Alexandria from the ‘land of the Chaldeans’ accompanied by his student, Thomas of Edessa (II.2; V.1; VIII.25). Mar Aba’s journey west is known from Syriac sources, most importantly the Vita written shortly after his death (Becker 2006a, 35–38, 113–14; Izdebski 2014, 185–204; Jullien 2015, 3–43). The text confirms that Mar Aba visited Alexandria with Thomas of Edessa and that he engaged in teaching on Holy Scripture. It adds that Mar Aba was able to turn many from the foreign (i.e., pagan) wisdom, a goal that Kosmas identifies as an aspiration of his own work (Life of Mar Aba 219; Christian Topography I.1; II.1; III.87; IV.15; VI.13; VII.82; IX.10). This may suggest that Mar Aba’s lectures were concerned with the interface

between ancient science and Christian dogma, rather than with internal Christian debates. We know that Mar Aba was interested in Neoplatonic philosophy: with his return to Nisibis, the works of Nisibean authors begin to show influence from Neoplatonic commentaries, which indicates that he brought some Neoplatonic literature with him (Becker 2006a, 128–29). Later sources claim that in Alexandria Mar Aba lectured on the works of Theodore of Mopsuestia in Syriac, while Thomas of Edessa translated his lectures into Greek, but this is probably a revision based on the increased importance of this theologian for the East Syrian Church (Chronicle of Seert 7 2). There is no indication that Kosmas knew Syriac, and it seems more likely that he and Mar Aba interacted in Greek. Kosmas does not say how he met Mar Aba and neither the Christian Topography nor the Vita give any indication of networks of contacts that could have assisted Mar Aba in Alexandria. Some network of contacts may have existed, however, considering that Kosmas was able to keep himself informed about his teacher’s movements after the latter left Alexandria—he knows, for example, that Mar Aba visited Constantinople and then returned to Persia, where he became the katholikos (II.2; Vööbus 1965, 163, 170). Mar Aba was not the only significant East Syrian figure to travel to Egypt in the first half of the sixth century: Abraham of Kashkar, the great reformer of East Syriac monasticism, travelled to Egypt to learn there the rules of ‘true’ monasticism, but sources reporting this journey give few details (Nau 1918– 1919; Tamcke 2007; Jullien 2008). These two journeys were sometimes seen as exceptional undertakings, at a time when contacts between these two worlds appear to have been quite limited (Izdebski 2014, 203–4). Nevertheless, the view that they were part of a continuous traffic between Mesopotamia and the Roman world throughout the sixth century seems more likely (Walker 2006, 164–204). The Fourth Canon of Narsai for the school of Nisibis, established in AD 496, at the beginning of the school’s existence, forbids the members of the school to go to the Roman empire ‘for the cause of instruction, nor because of a pretence of prayer, also not to buy or to sell’ without the consent of the heads of the community, suggesting that such practices were common enough to require regulation (Statutes of the School of Nisibis 31–32, 93). A growing theological rift in the second half of the sixth century is reflected in a record of a theological dispute, in which a delegation of East Syrian clergy

BEYOND EMPIRE

participated, held in Constantinople shortly after the Byzantine–Persian peace treaty of 561 (Chronicle of Seert 187–88; Uthemann 1999, 77–79). This is not to say that contact ceased: the successor of Mar Aba, katholikos Joseph, received medical training in the Roman empire (Chronicle of Seert 176). An anonymous seventh-century chronicle mentions Peter of Qatar, who studied philosophy in Alexandria ‘during his youth’, which from the context would be in the second half of the sixth century (Khuzistan Chronicle 25–26). Contact and intellectual exchange is also illustrated by the impact of John Philoponus’ argumentation, from his treatise De opificio mundi, against the eternity of the stars in the early seventh-century East Syrian dialogue of Mar Qardagh (Walker 2006, 164–204). The Church of the East and eastern trade in the Christian Topography Much scholarly belief in the existence of a Nestorian religious network in the Indian Ocean has relied on Kosmas’ interest in the spread of Christianity from Persia. Kosmas mentions the Christian Church—the clergy and the faithful—on the island of Taprobane (Sri Lanka), and writes that bishops are ordained in Persia for the Indian regions of Kalliana and Male and for the island of Dioskorides (Socotra), where the inhabitants were descendants of Ptolemaic colonizers and spoke Greek (see also Periplus Maris Erythraei 30). Kosmas enumerates these dioceses within a longer list of nations and regions where one can find ‘innumerable churches, bishops, Christian population, many martyrs and hesychast monks’ (III.65). This list parallels his description of the division of the earth between the descendants of Noah, and largely serves to make the point that the Christian Church unites the dispersed nations of the world (III.64–66 and II.26–27; WolskaConus 1968, 502–7). The emphasis on the spread of the Church of the East may serve to reinforce the authority of Kosmas’ master, who represented the Church, which was particularly successful in its apostolic mission. Perhaps because of his loyalty to Mar Aba, Kosmas presents the Persian empire as second only to Rome, and connects the Persian royal dynasty with the three kings of the Epiphany (II.76–77). There is, nevertheless, no doubt that Rome, where Christianity spread first, is God’s chosen empire, the final of the four worldly empires, the one destined to fulfil God’s plan to unify dispersed humankind (Daniel 2.44; II.75–77).

179

Kosmas argues that the unifying work of the Christian Church is supported by the universal use of the Roman currency, which in itself is a clear sign of God’s favour to Rome (II.77). He also gives an important role to trade, and writes that God created the four navigable gulfs of the Ocean with the specific purpose of facilitating commerce and transport of commodities across the oikoumene, in order to establish contact between distant communities (III.25). Kosmas does not connect trade with any specific branch of Christianity nor with Christianity as a whole, but rather gives religion, trade and the empire parallel and complementary functions in God’s plan to unify humanity. Despite the fact that commerce is given such cosmic significance, within the Christian Topography information on trade is rare, and where it does appear it serves to corroborate biblical and cosmographical points. A good example is his report on the silk trade. Kosmas places the silk country, Tzinista (ἡ Τζινίστα), beyond the Persian Gulf, ‘in the most interior India’ (ἐν τῇ ἐσωτέρα πάντων Ἰνδίᾳ), beyond Taprobane, at the eastern confines of the oikoumene. He describes the land route, by which silk is traded from one people to another, as well as the sea route, where he notes that the silk country is more distant from Taprobane than Taprobane from the top of the Persian Gulf. The land route, which leads to Persia, is considerably shorter and consequently there is always an abundance of silk in Persia (II.45–47). This seems to imply that Kosmas saw the Central Asian caravan route as primary for trading silk (de la Vaissière 2012, 142–70, esp. 148– 49; but see also Whitfield 2007; Graf 2018). It is difficult to draw secure conclusions from a small number of finds, but the numbers of Late Roman coins found along the land route to the east into China appear to expand in number around the reign of Justin II (AD 565–74), suggesting that the route remained relevant for Rome (Morrisson 2002, 963; Wang 2004, 34). Kosmas’ discussion of the silk trade serves to emphasize the great distance and the arduous journey that the merchants need to undertake. This in turn serves to corroborate a point of Kosmas’ cosmography, namely the notion that the earthly Paradise exists but is inaccessible. Kosmas writes that if merchants travel to the ends of the earth for ‘miserable trade’, surely they would travel to Paradise if it could be done (II.45). Similarly, when discussing the trade of Aksum, Kosmas enumerates the commodities to stress that the objects traded and produced in the region are in keeping with biblical descriptions. Kosmas writes that

180

M. KOMINKO

inhabitants of Barbaria, that is, northern Somalia, purchase spices (ἡδυσμάτα)—frankincense (λίβανος), cassia (κασία) and calamus (κάλαμος) (Song of Songs 4.14)—from the inhabitants of the inland regions (μεσόγεια), and send them by sea to Adulis, which resells these products to Ḥimyar, inner India (ἐν τῇ ἐσωτέρᾳ Ἰνδίᾳ) and Persia (II.49). Although the passage has been interpreted as referring to the inhabitants of Barbaria buying spices in India and selling them to Arabia (Power 2010, 75–76), it is clear that Kosmas refers to the trade in spices bought in the frankincense-growing region of Arabia and sold, among other places, to the Indian subcontinent and the lands further east, consistently described throughout the Christian Topography as ‘inner India’ (for example, the silk-growing country of Tzinista is described as ‘innermost India’ II.45). All three commodities—frankincense, cassia and calamus—were known to originate from this region of Arabia, although cassia is sometimes attributed to Ethiopia or India (Schneider 2004, 195–201).4 There is no doubt that Kosmas attributes these three products to the lands around the southern end of the Red Sea, which to him is the southern end of the oikoumene. He points out that the Queen of Sheba, called by the Gospels ‘the Queen of the South’, brought to Solomon spices from Barbaria when she came from the ends of the earth (Matthew 12.42; 3 Kings 10.1). In his view, it corroborates the notion that Ethiopia and adjoining lands constitute the ends of the oikoumene (II.50). He further states that he can confirm this notion from his own observations and from what he heard from the merchants who operate in these areas (II.64). In an even more elaborate argument supporting the notion that the region of Ethiopia constitutes the southern limit of the inhabited earth, Kosmas transcribes inscriptions covering the throne and the stela at the entrance to Adulis, the port of Aksum (II.58–64) (Bowersock 2013; Kominko 2013, 24–35) (Fig. 3). Kosmas stresses that he has transcribed this text himself, at the request of the governor of Adulis, who

4

On the problem of terminology of cassia and cinnamon and for calamus, that is, sweet flag or sweet cane, see Miller 1969, 42–47, 92–94; Schneider 2004, 199–200. While this may be a description of the actual trade, the combination of the three does have a biblical ring to it, as the three spices, although with

provided these transcriptions to the king of Aksum Ellatzbaas, when the latter was preparing an expedition against the Ḥimyarites (II.56). While this is not clear from the text, the king might have intended to use this inscription as a justification for his territorial interests in Arabia (Hatke 2011, 86–87, 109–10, 155–56). In the context of the Christian Topography, Kosmas argues that the text, which lists the lands conquered by an anonymous Aksumite king, corroborates the notion that there is no land south of Ethiopia. For him, this shows that ancient sources agree with his cosmographical theories, an idea that he stresses throughout his treatise (II.65, see also III.57; VIII.3). Kosmas’ discussion of the Adulean inscription fits with a long tradition of including inscriptions in Greek historical writing (Whittaker 1991). His presentation of the throne and the stela, and especially his identification of figures represented on them, resembles Herodotus’ discussion of the stela of Sesostris (Herodotus, History 2.102–6). It seems a part of Kosmas’ effort to give his work a respectable classical pedigree. Indeed, other classical inspirations are discernible in Kosmas’ account of Ethiopia, for example in his description of silent barter involved in the commerce of gold with the tribes in Sasou (II.51–53; Wainwright 1942). Although ostensibly Kosmas gives a fair amount of detail on real mercantile expeditions organized biannually by the Aksumites, who trade livestock, salt and iron for gold, the description of silent barter between the Aksumite merchants and the African tribes has numerous literary predecessors starting with Herodotus (Giardina 1993, esp. 248).5 Overall trade and more detailed geographical description take up relatively little space in the ten books of the Christian Topography. It is interesting to note, however, that when it comes to the regions at the southern end of the Red Sea and beyond, the geography is both more detailed and a little divergent from Kosmas’ biblical system. For example, throughout the treatise Kosmas writes that the Red Sea and the Persian Gulf open directly into the Ocean, and this is how they

5

cinnamon rather than cassia, are mentioned together in the Song of Songs 4.14. On mute trading, see Herodotus, History 4.196 (Carthaginians with Libyans); Solinus, Polyhistor 50.2ff (Seres); Philostratus, Life of Apollonius 6.2 (between Ethiopians and Egyptians).

BEYOND EMPIRE

Fig. 3: Christian Topography. The Monuments of Adulis, Laur.Plut.IX.28, fol. 28r (Courtesy of Biblioteca Medicea Laurenziana).

181

182

M. KOMINKO

are represented on the map in all three codices (II.29– 30). Yet, when discussing Ethiopia and the eastern trade, he writes that the gulfs are connected by an intermediary body of water, which he calls Ziggion (Ζίγγιον) and which is open for navigation but dangerous because of its proximity to the Ocean (II.50). He also mentions the Indian Sea, which does not fit into his description of the earth (Ἰνδικὴ θάλαττα II.29; Ἰνδικὸν πέλαγος II.45–46; II.81; III.65, terms which correspond to Book XI.13). This may suggest that Book XI not only pre-dates the development of Kosmas’ cosmography, but also provided some material for inclusion in the Christian Topography. Book XI. India In keeping with the title Description of animals and trees of India and of the island Taprobane (Καταγραφή περὶ ζῷων ἰνδικῶν καὶ περὶ δένδρων ἰνδικῶν καὶ περὶ τῆς Ταπροβάνης νήσου), Book XI is divided into two parts: the description of plants and animals (XI.1–12), followed by the description of Taprobane and trade in the region, with a small admixture of information and stories about India (XI.13–24). The chapters with descriptions of animals have a very similar structure: in every one of them an image is followed by a short text.6 Kosmas identifies animals, a rhinoceros (ῥινόκερως, XI.1) and a giraffe (καμηλοπάρδηλις, XI.4) as Ethiopian. He writes that two animals, a wild bull or yak (ἁγριόβους XI.5) and a musk deer (μόσχος XI.6) and two plants, pepper and coconuts (πιπερέα, κάρυα ἰνδικά, XI.10) are Indian. One animal, a buffalo (ταυρέλαφος, XI.3), is said to live in both Ethiopia and India though Kosmas specifies that in India the buffalos are domesticated, while those in Ethiopia are wild. The habitats of six animals, hippopotamus (ἱπποπόταμος, IX.9), hogdeer (χοιρέλαφος, XI.8), seal, dolphin and turtle (φώκη, δελφῖνος, χελών, XI.12) as well as a unicorn (μονόκερως, XI.7) are not specified in the text. Kosmas’ first-hand experience seems limited to Ethiopian animals, but he does not claim to have seen all the animals he describes. He clarifies, for example,

6

The same structure is followed by Book IV of the Christian topography, and serves as a set of illustrations for Kosmas’ cosmographical exposition (Kominko 2013, 83–85).

that he only saw a bronze statue of a unicorn (μονόκερως), and of a hippopotamus, he only saw teeth, sold in Egypt and Ethiopia (XI.7 and XI.9). Two chapters, on a hog-deer (χοιρέλαφος), and on a dolphin, turtle and seal (φώκη, δελφῖνος, χελώνη), describe the culinary experience of eating these animals (XI.8; XI.12). While the second of these gives some idea of how the meat tasted, the chapter on hog-deer is so uninformative as to be puzzling: the drawing is accompanied by a line of text which states that Kosmas saw and ate a hog-deer (τὸν δὲ χοιρέλαφον καὶ εἶδον καὶ ἔφαγον) (XI.8). Several identifications have been suggested (Faller 2011, 212–13), but there is simply not enough information in the text to confirm or reject them. An attempt at identification on the basis of the image seems unwise, considering that the hippopotamus is depicted as a river-dwelling horse in the extant manuscripts (Sinaiticus gr.1186, fol. 202v; Laurentianus Plut.IX.28, fol. 268v). The information Kosmas provides appears to be a mixture of animal lore and local information, especially local names. For example, in the chapter on the rhinoceros, Kosmas describes its animosity towards elephants, which is a common topos in Greek literature (XI.1; Gowers 1950), but also provides the Ethiopian name of the animal, arue harsi (ἀρουὴ ἅρισι), which corresponds to the Ge’ez term for this animal. The description of the wild bull (ἀγριόβους), which is very likely the yak, is similar to the one given by Aelian (De natura animalium 16.11). Both Aelian and Kosmas describe a large herbivorous animal hunted for its beautiful tail, and both make the tail the focus of the story. Nothing in Aelian’s text allows us to identify the animal, but the idea of the animal’s sensitivity about its tail, especially Kosmas’ story that it prefers to die rather than damage its tail, corresponds closely to the animal lore on yaks in classical Indian literature of Late Antiquity (Rau 1986, 196 with references). The second of the Indian animals, μόσχος, that is, a musk deer, is native to northern Afghanistan, Pakistan, Tibet, Nepal, Bhutan and northern parts of India and was hunted for its musk, μόσχος, a substance that enters Greek materia medica in the sixth century (see

BEYOND EMPIRE

below). Kosmas provides a local name of the animal, kosturi, which corresponds to the Sanskrit term (XI.6). Kosmas describes two Indian plants. The pepper tree is delicate and weak and grows like a vine, climbing on other trees and producing red fruit. The accompanying image appears to represent harvesting pepper, a process which is not discussed in the text (Sin.gr.1186, fol. 203r; Laur.Plut.IX.28, fol. 270r; McCabe 2009, 282). In the same chapter he describes Indian nuts (κάρυα ἰνδικά), that is, coconuts, which are large trees with big foliage and fruit, full of sweet water. He also gives an additional name of this plant, ἀργέλλιον, for which a Sanskrit origin (from narikela, a coconut tree) has been proposed (Winsteadt 1909, 170). This name seems akin to the Indian term anārgīl recorded in Pahlavi in Khusro and the Page, together with the corresponding Persian term gōz ī hindūg, that is, Indian nut (Monchi-Zadeh 1982, 74; Daryaee 2010, 402). Kosmas also mentions a wine made by the locals out of coconut water, ῥογχοσοῦρα, but the origin of this term is unclear. Inclusion of examples of fauna and flora in a work of geography is not unusual (Strabo, Geography 17.1.15; 17.1.51; 17.1.39; 17.2.2; 17.2.4; Darley 2014, 117), but because the text of Book XI is so limited, it is difficult to speculate on the original goal of this presentation. It is fairly clear, however, that the text is not a record of a personal trip. As noted above, Kosmas does not claim to have seen any of the Indian animals. Moreover, the geographical regions of the Indian plants and animals he describes are widely different. Yak and musk deer come from the Himalayas, while pepper grows only on the southwestern coast of India. A description of such geographically dispersed fauna and flora is unlikely to be based on firsthand observation (Darley 2014, 117). The fact that for several plants and animals Kosmas uses terms which may derive from local Ethiopian and Indian vocabulary has led some scholars to suspect that he transcribed phonetically names he heard during his travels (Dalby 2007; Faller 2011). This certainly is a possibility, but hardly constitutes proof of his travels to India, as he may well have heard these terms in Adulis or even

7

Σελεδίβα in Book II.45. For the suggestion that the name Sielediva derives from Sihala-dipa (Simhala-dvipa), the ancient Indian name for the island see Weerakkody 1997, 23–28.

183

Alexandria. It may be worth noting, however, that loan words are not unusual in the sixth-century koine (Wittman 1913). The stories about animals must have been in wide circulation, as the similarity with Aelian indicates. Yet, the variant preserved by Kosmas is closer to the Indian lore and on the whole the text—or what we have of it—appears to document genuine interest in Ethiopia and India and possibly access to information which was new to the Greek-speaking environment. From describing animals and plants Kosmas moves to the discussion of Sri Lanka. He writes that the island, called by the Indians Sielediva (Σιελεδίβα) and by the Greeks Taprobane, is located in the Ocean, in the Indian Sea, beyond the pepper country.7 Surrounded by small islands with drinking water and coconuts (ἀργέλλια), Taprobane is 300 gaudia, that is, 900 miles long, and 300 gaudia wide, and divided between two kings who are enemies. One rules the part of the island which produces hyacinth (ὁ λίθος ὁ ὑάκινθος), the other the part where a big trading centre and the port are located.8 Further, Kosmas says that the island has a community of Christians from Persia, but the king and the locals are pagans and there are many temples. He gives a brief description of one such temple with a hyacinth (blue gemstone) as large as a great pinecone, fiery and flashing from a distance. Hyacinth, identified by Periplus Maris Erythraei 56 as an Indian stone, and also attributed to the island by Ptolemy (Geography 7.4.1), is usually believed to be blue. Kosmas’ description seems akin to that by a seventhcentury Buddhist monk, Hsuang-Tsang, who relates that a ruby elevated on a spire surmounting one of the temples at Anuradhapura, the island’s capital, was so bright that it illuminated the whole heaven (Beal 1911; Weerakkody 1997, 138). This similarity seems to suggest that the lore on Taprobane had a wide circulation. The division of the island between two kings is sometimes thought to reflect the late fifth-century political struggles, possibly Dhatusena’s opposition to the South Indian invaders, or Moggallana’s wars with his brother Kassapa I, but the text is too vague to allow

8

For hypotheses on the identification of gaudia see WolskaConus 1973, 342, n. 13.4.

184

M. KOMINKO

such a hypothesis to be either accepted or disproved (Weerakkody 1997, 138). The large port Kosmas mentions could well be the northern harbour of Mantai, connected to Anuradhapura (Whitehouse and Williamson 1973, 44; Weerakkody 1997, 138), but such an identification is only tentative, largely made on the assumption that Mantai was the largest port of the island. It may be worth pointing out that Kosmas does not provide the name of the port, although he names several ports in India. Kosmas writes that Taprobane is located in the centre of all the Indias and serves as an intermediary receiving many ships from, and sending many ships to, India, Persia, Ethiopia and China (XI.15–16). From China, and the regions between China and Taprobane, the island receives silk, aloes, cloves (καρυόφυλλον), clove wood (ξυλοκαρυόφυλλον) and sandalwood (τζανδάνα). These commodities are re-exported to Persia, to the land of the Ḥimyarites and to Adulis, as well as to the pepper-growing region of Male, to Kalliana and Sindu. Kosmas lists five trading centres along the western coast of India: Sindu, on the border with Persia, exports musk (ὁ μόσχος), costum (τὸ κοστάριν), and spikenard (τὸ ναρδόσταχυν). Then there are Orrotha and Kalliana, the latter producing copper, sesame wood (σησάμινα ξύλα), and fabrics. Finally, there is Sibor and Male with its five markets which export pepper: Parti, Mangarouth, Salopatana, Nalopatana and Poudapatana. Kosmas also lists two ports to the east of Taprobane, ‘in the interior’: Marallo, which exports spiral shells (κοχλίοι), and Kabe, which exports almandine (ἀλαβανδηνόν). Of these ports and towns only two, Male and Kalliana, are mentioned in the Christian Topography, where Kosmas describes Male as a pepper-growing place and says that there are churches in both Male and Kalliana (III.65). Kalliana is sometimes identified with Elephanta near Mumbai (Power 2010, 59 and 316), and Kabe with Kaveripattinam (present-day Poompuhar) at the mouth of the Kaveri River (Roth 1980, 318–19; Borrell 2008, 13). Further to the east Kosmas places the lands producing cloves and Tzinista, which produces silk. Beyond Tzinista, there are no other countries, a

statement that also appears in the Christian Topogaphy (XI.16 and II.46). After describing the centres and commodities of the east, Kosmas writes that Ethiopia sends large elephant tusks to Rome, India, Persia and Ḥimyar (XI.23) and trades in emeralds (σμάραγδον λίθον), which Ethiopians buy from the Blemmyes, and sell to Indians, who in turn sell them to the White Huns (XI.21). The only Persian export that Kosmas lists is horses, imported by Taprobane (X.22). Kosmas lists mercantile contacts between Taprobane and Ethiopia, Ḥimyar, Persia, India, the Huns and Tzinista, but does not mention any direct exchange with Rome. The only instance where a direct contact occurs is the famous anecdote of Sopatros (XI.17–19). Sopatros was a Roman merchant who visited Taprobane, and who died thirty-five years before Kosmas put down his story in writing in Book XI. Sopatros’ visit to the island coincided with the arrival of a Persian ship, with an ambassador (ὁ πρεσβύτης Περσῶν).9 Both crews were led to the king who wished to know which of the rulers, the Roman or the Persian, was more powerful. Sopatros suggested that the king should find the answer by examining the coins of the two dominions. On examining a golden Roman solidus and a silver Persian drachm, the king recognized the superiority of Rome. The anecdote echoes the notion of the primacy of Rome and its currency from the Christian Topography, discussed above. It has been long noted that the story of Sopatros bears a similarity to Pliny’s account of Annius Plocamus, a freedman of Emperor Claudius, carried away by a gale to Taprobane. There he was received by a king who, upon realizing that the coins found on the captive were all of equal weight, despite being struck in the reigns of several emperors, was impressed by Roman order and justice (Pliny, Natural History 6.24.84–85). The similarities of these narratives notwithstanding, the story of Sopatros has often been seen as corroborating the spread of Roman and Persian currency, and as evidence of direct contact between Taprobane and Byzantium and the strong position of Roman and Aksumite merchants (Bopearachchi 1998, 158; Power 2010, 127–28). The

9

For the discussion of the translation see Wolska-Conus 1973, 349, n. 17.2.

BEYOND EMPIRE

evidence of coins hardly supports such a narrative: the overall number of Late Roman (fourth to seventh century) gold coins found in India is estimated at around 200 (Darley 2014, 211) and only three of these can be securely attributed to Sri Lanka (Walburg 2008, 122, 139, 145; Darley 2017). Coins are the largest category of Roman finds in India. Early coins present a mixture of gold and silver, while Late Roman coins are predominantly bronze (Tomber 2008, 30–37; Nappo 2018). The majority has been found in southern India, with very few finds in the north. The chronological analysis is complicated by the fact that the date at which a coin was struck does not necessarily correlate with its movement to India. Moreover, drawing conclusions about temporal changes in trade using the distribution of coins in India would require accounting for their movement within India, which was likely considerable, but which is still poorly understood (Tomber 2008, 37; Darley 2014, 211–13). The problem is compounded by the fact that most of the Indian coin finds do not come from stratified contexts. Late Roman coin finds in India, predominantly bronze, come from Gujarat, areas along the Ganges, Uttar Pradesh and the Malabar coast, with concentrations in Tamil Nadu, at Madurai, Karur and Alagankulam (Tomber 2008, 165 and fig. 21). The South Indian finds include a hoard of early fourth-century Aksumite and fifth-century Byzantine coins from Mangalore, which must have been deposited in the early sixth century, considering that the latest coin is an Indian imitation of Anastasius I (Darley 2014, 259). Imitation Aksumite coins have been found in Karur and Madurai (Hahn 2000, 287–88). No Sasanian coins have been found in South India, although unconfirmed finds were reported in Gujarat (Darley 2014, 313). The distribution of Late Roman pottery finds in India does not follow the distribution of Late Roman coins. With the exception of a few rare finds on the Malabar and Coromandel coasts, Late Roman amphorae cluster in the north, from Gujarat and Masharashtra, where amphorae finds in Elephanta (which may be Kosmas’ Kalliana) suggest the peak of trading activity was in the sixth century (Tripathi 2004, 122; Tomber 2008, 128, 165 and table 3). Aksumite pottery has been found only in Kamrrej in Gujarat together with the Late Roman amphorae (Tomber 2005). Many ceramic finds in India previously thought to be Roman are now identified as Mesopotamian torpedo jars, spanning the Parthian, Sasanian and early Islamic periods (Tomber

185

2007b). Torpedo jars are found primarily in North India and Sri Lanka with some South Indian examples from Pattanam and Alagankulam (Tomber 2008, fig. 21). Because Late Roman amphorae and torpedo jars in India are sometimes found in the same region, but rarely on the same site, it has been suggested that two separate strands of trade may have existed, one from the Red Sea, the other from the Gulf (Tomber 2008, 167). In Sri Lanka coins are rarely found in scattered single finds, but rather in hoards, often in the vicinity of sites of religious significance, and mostly in the south of the island (Walburg 2008, 312–14). Analysis of these finds is complicated by wildly different estimates of their numbers (Bopearachchi 1998, 70; Walburg 2008; Tomber 2008, 145; Darley 2017). For example, the local copper alloy imitations of Roman fifth-century coins previously estimated at 200,000 but currently placed at 30,000–40,000 (Bopearachchi 1992, 113; Walburg 2008, 57–84; Darley 2017). The view that the large numbers of fifth-century imitation coins reflect directly the scale of the trade with Rome has been contested. An analysis of the chronology of the Roman coins and their imitations on the island led to a suggestion that original Roman coins may have reached the island in only a few cargoes, within a short period of time (AD 425–50) and that imitations were made shortly after (Walburg 2008, 343). This is a hypothesis only, but one in keeping with the absence of any major Roman finds on Sri Lanka (Darley 2014, 257). In Mantai, in addition to five copper alloy coins, of which two are probably local imitations, the only Roman objects found are two beads (Francis 2013, 352–53; Walburg and Graham 2013, appendix 3, 1–3). The Sinhalese royal capital of Anuradhapura yielded five Late Roman bronze coins. Early and Late Roman amphorae were discovered at Tissamaharama, together with 470 Early and Late Roman coins from its greater hinterland (Schenk 2001, 74). Tissamaharama has also yielded the only Aksumite coin found on the island (Walburg 2008, 54). The story of Sopatros is often quoted as evidence for the use of Sasanian coins in Sri Lanka (De Romanis 1997, 187–88; Bopearachchi 2006, 195; Choksy 2013, 380–81). Yet, only ten Sasanian coins, dating between the fourth and sixth centuries, have been found on the island and a sole Sasanian imitation coin does not suggest that large numbers of prototypes reached the island (Darley 2014, 313–17; 2017). On the other hand, ceramics from the region of the Persian Gulf, the

186

M. KOMINKO

torpedo jars and glazed ware, have been discovered at Mantai and Anuradhapura and Tissamaharama (Tomber 2008, 126 and fig. 21, 146–47; Graham 2013, 206–8). As noted above, their dating is broad, and they may be Parthian and Islamic, as well as Sasanian. Mantai has also yielded a black clay bulla, dated by excavators to the sixth or seventh centuries, with a seal impression of the two-humped Bactrian camel, a Persian inscription and a cross (Carswell 1991, fig. 11.2). The known archaeological evidence does not support the view of Sri Lanka as a large international hub of commerce. We should note, however, that the archaeological record on the island is sometimes sparse even where we know the contacts existed: accounts of pilgrims and inscriptions discovered elsewhere in the eastern Indian Ocean, as well as increased interest in the laws practised on Sri Lanka, testify to the visits of Buddhist travellers, which so far do not seem to be reflected in archaeological finds on the island (Legge 1886; Sen 2014, 39–62). Sasanian trade and the Persian Church It is often assumed that by the sixth century the Sasanians dominated trade in the Indian Ocean (Whitehouse and Williamson 1973, 29; Daryaee 2003, 8–14, 16; Tampoe 1989, 2, 119). Yet, evidence from India is scarce and that from the Persian Gulf is more limited still. Although in large part this is due to the dearth of excavation, even the excavated sites do not yield evidence of significant trade. Siraf, the major port in the ninth century, was already settled in the second century, but until the ninth century foreign finds are relatively few. A coin of Constans II suggests contact with the west, while eastern exchange is attested by Chinese pottery (Whitehouse and Wilkinson 2009, 9, 98–99, 103, 113). In Rishahr, located in southern Iran, red polished ware from northern India provided evidence for long-distance commercial ties with the east (Whitehouse and Williamson 1973, 39). Ceramic analysis from Ras al-Khaimah (UAE) indicates a rise in Indian trade in the fourth and fifth centuries with a subsequent lull until the ninth (Kennet 2004, 69–70).

10

The fourth-century Palladius is often quoted as providing the first reference to Sasanian ships on the Indian Ocean (Whitehouse and Williamson 1973, 43, repeated by numerous later

The majority of written sources describing Sasanian maritime involvement are late.10 The tenth-century Chronicle of Seert speaks of Sasanian maritime eastern trade in pearls in the fifth century (see below). Also in the tenth century al-Tabari reports that Bahram V (r. 421–38) married an Indian princess and received as a dowry the port of Daibul in the Indus Delta, together with the adjacent parts of Sind and Makran, but this is otherwise unconfirmed (History of prophets and kings 868; Whitehouse and Williamson 1973, 43). Al-Tabari’s information should probably be taken with caution, considering that he also reports Khusro’s invasion of Sri Lanka (History of prophets and kings 965). Though this account is fictitious, it has been suggested that it captures the echo of Persian influence on the island, but again, there is little evidence to support it (Whitehouse and Williamson 1973, 45; Weerakkody 1997, 136). The best indication of Persian maritime presence in the east is given by Chinese sources referring to Possǔ, which is very likely Persia (Wolters 1967, 129–58; Whitehouse and Williamson 1973, 46–48). For example, in AD 717 or 719 the Buddhist monk Vajrabodhi sailed from Sri Lanka to Canton with a fleet of thirty-five Possǔ ships (Colless 1969–70; Whitehouse and Williamson 1973, 46–47). The references to the maritime presence of Possǔ appear only from the late seventh or early eighth centuries onward, and analysis of Chinese sources in the context of commodities they mention strongly suggests that neither Persian nor Indian merchants were prominent in the maritime trade with China before then (Wolters 1967, 150). In effect, the only earlier evidence of Sasanian maritime activities is provided by Kosmas and Procopius. In the passage that came to define the notion of the Indian maritime trade in the sixth century, Procopius reports Justinian’s attempt to break Rome’s dependence on Persia for the supply of silk. The emperor convinced the Ethiopians to purchase silk from India and sell it to the Romans. The Ethiopians were unable to deliver because the Persian merchants who lived in the country adjoining India were able to meet the Indian ships in the harbours where they arrived and buy

authors), but in reality such a reference appears only in the problematic Latin version of the text, attributed in manuscripts to St Ambrose (Weerakkody 1997, 23–24).

BEYOND EMPIRE

the whole cargoes, as they were accustomed to do (Procopius, De bello Persico 1.20.9–12). Procopius’ account implies Sasanian merchants’ established position in the silk trade in the region, but taking it to mean that they dominated the entire western trade of India may be going too far. As already noted, Kosmas does mention the silk trade and, while estimating the great distance between Taprobane and the Persian Gulf, he suggests that this route was far longer and less convenient than the overland one through which the majority of silk was traded (II.45–47). Kosmas also writes that Taprobane, where silk, aloes, cloves, clove wood, sandalwood and other products arrived from the east, shipped these commodities to Male, Kalliana, Sindu, to Persia, to the country of the Ḥimyarites and to Adulis (XI.15). His account conflicts with that of Procopius in that it suggests that, along with other eastern commodities, silk was distributed by Indian merchants, rather than bought directly and in its entirety by the Persians. Kosmas gives no indication that Sasanians held a dominant position in the trade: the only Persian export he lists are horses, bought by the king of Taprobane and free from custom charges (XI.22). This statement has entered the scholarship, without much scrutiny (Kröger 1979, 447; Ghosh 2000–2001; Daryaee 2003, 9). Nevertheless, not only is such long-distance maritime transport of animals from Persia otherwise unattested, but also it seems unlikely, especially considering that Sri Lanka had a long tradition of importing horses from southern India, documented by numerous inscriptions and written sources (Weerakkody 1997, 142–43). Overall, Kosmas implies that the western trade was just one part of a complex web of mercantile relationships spanning the subcontinent and lands further east. Procopius’ short statement reduces it to a picture in which the westerners—the Sasanians and the Aksumites—were the only customers for the eastern commodities, which seems unlikely. Kosmas writes that while the natives of Taprobane and their king are pagans (ἀλλόφυλοι), there are Persian Christians settled on the island, with a presbyter appointed from Persia, a deacon and liturgy (Περσῶν χριστιανῶν καὶ πρεσβύτερον ἀπὸ Περσίδος χειροτονούμενον καὶ διάκονον καὶ πᾶσαν τὴν ἐκκλησιαστικὴν λειτουργίαν). The contrast between the native pagans and the Persian Christians suggests that the church was established for the needs of the Persian community settled on the island (Weerakkody 1997, 135). Kosmas is the only known author to mention the Persian Christian community in Sri Lanka in

187

Late Antiquity, but the presence of Persian Christians in India is confirmed by several later Syriac sources. The Chronicle of Seert reports that David, the bishop of Basra, left his see to preach in India in the early fourth century (Chronicle of Seert 292, 25). The Acts of the Synod of AD 410 mention the metropolitan of Fars in Rew Ardashir, who had under his jurisdiction bishoprics along the Persian Gulf and beyond (Synodicon Orientale 77). By the seventh century this jurisdiction extended to India, as we know from the conflict between the metropolitan of Fars Simeon of Rew Ardashir and the patriarch Īšō‘yabh III (649–660) who raised the provinces of Bēt Qaṭrāyē and India to metropolitan status (Mingana 1926, 464; Synodicon Orientale 215–26; trans. 480–490; Kozah 2015, 43–88). Ma‘nā, metropolitan bishop of Fars in the late fifth century, produced Persian ‘hymns, homilies, and responses to be sung in church’ and distributed them to the maritime lands (Bēt Qaṭrāyē) and India (Chronicle of Seert 117.9). In the sixth century Job of Rew Ardashir prepared translations of the monastic rules for use in Fars and among Christians in India (Chronicle of Seert 173; Wood 2013, 168). This suggests the existence of a sizeable Christian community in India, one that used Pahlavi rather than Syriac as their liturgical language. The near-total loss of this literature, attested only in a fragmentary Pahlavi Psalter from the Turfan Oasis, has made it difficult to recapture its history (Buck 1996, 81–84). The use of Pahlavi and the presence of Christians in India are documented by the Middle Persian inscription repeated on six crosses in southern India, one in Mylapore near Madras (Chennai), two crosses in Kottayam, one in each Muttuchira, Kadamattom and Alangad. Both the reading of the inscriptions and the dating are controversial. The latter ranges between the seventh and the ninth centuries, although sometimes the crosses are dated as early as the fifth or as late as the tenth century (Gropp 1991, 85; Cereti, Olivieri and Vazhuthanapally 2002, 296–98 with earlier literature; Varghese 2015). No Christian inscription has been found in Sri Lanka and the archaeological evidence of Christian settlement is notoriously difficult to date. The stone cross discovered in 1912 in Anuradhapura is dated between the fifth and the seventeenth century (Devendra 1957, 85–89; Weerakkody 1997, 135). Another cross unearthed in Anuradhapura, one in Sigiriya, and a fifth-century baptismal font found near Mannar are mentioned in the literature but prove elusive and difficult to confirm (Tang 2014, 381–92).

188

M. KOMINKO

Moreover, a stone pillar with a stone cross, perhaps going back to the early Middle Ages, was found in Mantai (Baum and Winkler 2003, 153 and fig. 2.6), but again, the dating and specific context are unclear. The only other Christian discovery on the island is a black clay bulla found at Mantai, mentioned above, dated by excavators to the sixth or seventh century, with a seal impression of a Bactrian camel and a cross, bringing us to the problem of Christian involvement in trade (Carswell 1991, 26). The existence of Christian, and specifically ‘Nestorian’, mercantile networks has been postulated by scholars who saw the churches attested archaeologically in the ports of the Persian Gulf and islands of the Indian Ocean as focal points of contact for Christian merchants, allowing them to link up with co-religionists (Tomber 2007a, 219–28; Seland 2012, 72–86). Although Kosmas is often quoted to support the existence of such networks, in reality nowhere in his work does he make a connection between trade and religion. When Kosmas mentions Christians from Persia in Taprobane and India, he does it without any reference to trade. Conversely, when he mentions merchants from Persia, he makes no reference to their religion and at no point does he connect the Christians of Taprobane with trade. There is evidence of Persian Christians participating in trade, but there is nothing to suggest that their business relied on specifically Christian networks. In fact, the existing texts hint that they did not. The tenth-century Chronicle of Seert reports that in the fifth century the East Syrian katholikos Ahai was co-opted by Yazdigird I (r. 399–421) to investigate the problem of piracy affecting trade in pearls with India and China (Chronicle of Seert 1.69;Whitehouse and Williamson 1973, 43). Trade is also frequently mentioned in hagiographic sources. The father of Job of Rew Ardashir is said to have been a ‘trader of pearls and fine stones’ in Rew Ardashir, who dispatched his son to the Roman empire to sell pearls (Īšō‘dnaḥ of Basra, Book of Chastity 44; Beaucamp and Robin 1983, 182–84). Some of the manuscripts of the fragmentary life of Abraham of Kashkar, noted above in the context of his journey to Egypt, refer to his mercantile career and his trip to India (Nau 1915–1917, 30; the passage is absent in others: Nau 1918–1919). According to the ninthcentury Book of Chastity, the seventh-century Bar Sahde made several journeys to India, in the company of merchants, before entering a monastery after his ship was boarded by pirates (Īšō‘dnaḥ of Basra, Book of Chastity 77; Whitehouse and Williamson 1973, 47;

Mingana 1926, 455–56). Maritime trade, especially trade in pearls, is imbued with symbolic value that should make us cautious about interpreting these stories as straightforward recordings of historic reality (Walker 2018). Importantly, even these anecdotes give no indication that specifically Christian trade networks existed. Studies of Buddhist networks in the Indian Ocean have emphasized the relationship between traders and religious travellers in this period, but this is not to say that trade relied on religious networks (Ray 1994; Glover 1996; Morrison 1997). Trade and religion certainly spread along maritime networks, but it is difficult to prove an association between them beyond the convenience of shared travel (Colless 1969–1970, 17–38; Darley 2014, 305–6). One marked difference between Christianity and Buddhism in Sri Lanka is that, unlike Buddhists or Hindus, Christians had no reasons to visit the island for pilgrimage. Many Christians on the island were likely there for commercial reasons, but this is not to say that all Christians were necessarily merchants or that their mercantile operations relied on exclusively Christian networks. The Pahlavi inscriptions from Chennai (formerly Madras) dated to the ninth century, which document extensive commercial contacts between Fars, the Gulf and India, illustrate how closely intertwined Christian, Zoroastrian and Jewish communities were as trading partners (Gropp 1991; Panaino 2004; Payne 2011, 105–6). Legal evidence indicates that Sasanian trade was conducted by partnerships designated by the Middle Persian legal term hambāyīh, many of which were likely based on religious association, but even then they dealt with other religious groups for long-distance trade (Pigulevskaya 1956, 60–81; Lim 2018). On the other hand, it has been suggested that because the Zoroastrian merchants were looked down on by their co-religionists and treated differently from the three traditional classes of priests, warriors and farmers, there were fewer Zoroastrians involved in trade, which may be one reason why the Sogdian Christians and Manichaeans were the main traders in Central Asia (Daryaee 2010, 403–4). Neither in Book XI nor in the Christian topography does Kosmas refer to networks which rely on a specific brand of Christianity. Kosmas travels to Adulis with a fellow Alexandrian merchant, Menas, who later retires to the monastery at Raithou in Egypt, and with other merchants from Alexandria and Aila (II.54). Sopatros makes his purported journey to Sri Lanka in the

BEYOND EMPIRE

company of men from Adulis, and it is from these people that Kosmas learns the story of his encounter with the king of the island (XI.17, XI.19). Whether the story is real or not, it resembles that of the fourth-century Scholasticus of Thebes, who found a passage to India on an Indian (which may be Ethiopian) ship he found in Aksum (Palladius, De Gentibus Indiae e Brahmanibus 1.4 and 1.10; Derrett 1961; Schneider 2004, 357– 59). If these stories are anything to go by, the choice of travelling partners seems to have been based on convenience only. Aksum As already noted above, it is usually assumed that from the fourth century Aksum increasingly served as an intermediary in Rome’s eastern trade and by the sixth century mediated the majority of commerce (Kobishchanov 1979, 172; Sidebotham 1991, 34; Bopearachchi 1992, 117; Mayerson 1996, 122). Already in Book II Kosmas references Aksumite trade with Barbaria, Sasou and Ḥimyar. In his discussion of trade in India, he writes that Aksum sends elephant tusks to Rome, India, Persia and Ḥimyar (XI.23). Although no finds of Ethiopian ivory are confirmed from the Persian Gulf region or India, the export of Aksumite ivory to Rome is amply attested (Phillipson 2009, 357–60). Kosmas specifies that Ethiopia sends the tusks by ship (ἐκ τῆς γὰρ Αἰθιοπίας καὶ εἰς Ἰνδίαν πλωΐ- ζονται ὀδόντες καὶ ἐν Περσίδι καὶ ἐν τῷ Ὁμηρίτῃ καὶ ἐν τῇ Ῥωμανίᾳ) and the presence of Aksumite sailors in the Arabian Sea is attested by sixth-century inscriptions from the Hoq cave in Socotra (Strauch 2013, 93). Kosmas writes that the Ethiopians purchase emeralds from the Blemmyes, whose settlements in the region of the mines and involvement in extraction of emeralds is recorded in the fifth century by Olympiodorus of Thebes (Olympiodorus, Frag. 35.2 in Eide et al. 1998; Dijkstra 2012). Emeralds were excavated in quarries of Mons Smaragdus, the mountains around wadis Sikait, Zabara, Nugrus and Umm Kabu, about 90km northwest of Berenike (Sinkankas 1989, 542– 48). Emerald is by far the most abundant gemstone found at Berenike, but because it is not mentioned by the Periplus Maris Erythraei and because Pliny describes the Egyptian variety of emerald as rarely flawless, it is usually assumed that they were not exported to India, which had its own pale greenish-blue aquamarine beryls (Sinkankas 1989, 445–555, esp.

189

507–8; De Romanis 1997, 95–98). In contrast, Kosmas writes that Ethiopians bring emeralds to India and sell them to Indians, who in turn sell them to Huns. He thus provides rare written western evidence of objects not only imported to India from the west, but also traded further within the subcontinent (XI.21). In addition to listing contacts between Aksum, Rome, Persia, India, Ḥimyar, Barbaria and the African interior, Kosmas indicates commercial contacts between Ḥimyar, Barbaria, Persia and India. Neither Rome nor Persia is given any particular prominence. In other words, what he describes is not a system dominated by a competition between two superpowers, but a system of interconnected trade. The archaeological evidence seems to corroborate a complex system of trade connection rather than trade dominated by Rome or Persia. Large quantities of Aila amphorae were discovered in sixth-century contexts at Adulis (Phillipson 1998, 63–70; 2014, 195–200; Peacock 2007, 96–97; Tomber 2008, 90–93) and recovered from the sixth- or seventh-century shipwreck off Black Assarca in the Dahlak archipelago north of Adulis (Pedersen 2008, 84–87). In Aksum itself, the Middle (c. 350–500/50) and Late Aksumite (c. 500/50– 80) assemblage is dominated by Aila amphorae with a growing proportion of blue-green Sasanian sherds, while the numbers of Mediterranean amphorae diminish (Manzo 2005, 59–63; Hatke 2013, 25–27). Sasanian glazed sherds have been also discovered at Maṭarā, and a green glass bowl of possibly Mesopotamian origin has been found at Emba Derhō, attesting to the Sasanian trade with Aksum (Manzo 2005, 59, 60). In addition to Roman and Sasanian amphorae, Aksum and Adulis preserve Indian and Chinese finds in fifthand early sixth-century contexts (Kobishchanov 1979, 188; Mundell-Mango 1996, 153; Morony 2004, 186– 87). Although there are no ceramic and botanical Indian finds, beads likely to originate from India occur in great numbers in Aksum and Maṭarā (Tomber 2008, 93). Among the sites along the coast of Somalia only Hafun Main was active in the Late Roman period. It yielded no Late Roman finds, but sizeable amounts of torpedo jars and glazed ware from the Gulf, along with pottery that is likely Aksumite. The presence of Persian ceramics has been interpreted as reflecting the rise of Sasanian power in the Arabian Sea (Tomber 2008, 96–100). Nevertheless, the situation is different on the southern coast of Arabia. Fifth- to seventhcentury levels in Qāni’ in Yemen yielded large

190

M. KOMINKO

numbers of Aila amphorae, along with Aksumite pottery, Mesopotamian torpedo jars and glazed ware. A full range of Aila and Aksumite kitchenware found on the site has been interpreted as evidence of settlement rather than seasonal trade (Sedov 2007, 84–89). Although Aksumite coinage is often believed to have been developed in the context of Indian trade and designed with the aim of interacting with Roman currency (Metlich 2006), a detailed examination has challenged this thesis (Darley 2014, 189–206). Moreover, not only are finds of Late Roman coins on Aksumite South Arabian sites extremely rare,11 but also finds of gold Aksumite coins in Roman ports along the Red Sea are few, which undermines the notion that the Aksumite currency served primarily as the interface, facilitating the role of Aksumites as the middlemen in Rome’s trade with the east (Darley 2014, 204). In general, Rome’s Indian trade is often seen as a factor which determined much of Aksumite history in Late Antiquity. This led to a bias in the interpretation of the sixth-century Aksumite invasions of Ḥimyar, often understood in the light of later diplomatic attempts by Justinian to draw Aksum and Ḥimyar into supporting Rome against the Sasanians (Procopius, De bello Persico 1.20.40; Rubin 2008, 195). The conquest of Yemen by the Aksumite king Ellatzbas during the reign of Justin gave rise to extensive discussion in the secondary literature, attempting to resolve discrepancies in the historical record, including the discrepancies of date (Power 2010, 105–18). Recent scholarship has demonstrated that the sources describe two different invasions: one in 518, reported by Kosmas in the Syriac Book of Himyarites, and reflected in epigraphic evidence; and a later invasion, following the massacre of the Christians of Najrān by the Jewish king Yūsuf, reported by Procopius, John Malalas and Martyrium Arethae among others (Beaucamp, Briquel-Chatonnet and Robin 1999; Hatke 2011, 18–28, 66–84, 117–18). The silence of Roman sources strongly suggests that the campaign of AD 518 was of little geopolitical interest to Rome. The second invasion of Ḥimyar in 525 receives more attention, but while it is clear that Justin supported Aksumite efforts to protect and promote Christianity in Arabia, there is no indication that this

11

An exception is a hoard of 326 Late Roman and 868 Aksumite gold coins, dating between the fourth and sixth centuries, found at al-Madhāriba in southwestern Yemen, some 70km west of

was viewed as a step towards undermining Sasanian influence in the region or creating an anti-Sasanian Romano-Aksumite axis (Hatke 2011, 182–85). The image that emerges from the sources is one of a war between Aksum and Ḥimyar, not a Roman attempt to establish a sphere of influence in South Arabia with the help of an Ethiopian ally. While it is likely that much eastern trade reached Rome through Aksum, we cannot view Aksumite activities in the Red Sea region solely in the context of Roman maritime trade. The analysis of Roman written sources referring to Aksumite participation in the Indian trade is complicated because Aksumites and Ḥimyarites are often designated as ‘Indians’ and the term ‘India’ is employed to refer to all lands bordering the southern Red Sea and western Indian Ocean, making it impossible to distinguish between direct contact with India and Indians, and contact mediated through Aksum and Ḥimyar (Mayerson 1993, 169–74; Schneider 2004, 221–364). An insight into geographical participation in the Red Sea trade is provided by a list of sixty ships docked likely at Gabaza, the harbour of Adulis, probably in the course of trading, and drafted into operation during the war with the Ḥimyarites in AD 525. These included twenty ships from Klysma, fifteen from Aila, seven from Berenike, two from Iotabe as well as seven from the Farasan Islands and nine from India (Martyrium Arethae 747; Kobishchanov 1979, 78–82). It is unclear, however, if the nine ships designated ‘from India’ should be interpreted as Indian rather than as Aksumite. Where the text refers to the preparation of Eletzbaas’ own ships, they are described as ‘Indian ships’ (ἰνδικὰ δέκα πλoῖα) (Martyrium Arethae 29.9), although some recensions of the Martyrium read ἴδια πλoῖα, ‘his own ships’ instead (Beaucamp 2007, 262 n. 177). The terms are equally unclear when it comes to designations of the Roman travellers to India: a sixthcentury graffito in Abu Shaar documents Andreas Indikopleustes, and Olympiodorus uses the term ἰνδικοπλεῦσται to denote those who travel to India, but whether he means Ethiopia or the Indian subcontinent is again open to debate (Bagnall and Sheridan 1994, 109–20; Olympiodoros, In Aristotelis Meteora commentaria 81.26; 163.3).

Aden. The unique scale of this hoard and the chronological span of the coins make it difficult to interpret (Munro-Hay 1989).

BEYOND EMPIRE

Rome’s Indian trade in the fifth and sixth centuries

191

Trade continued to reach Egypt via the southern maritime route for much of the fifth and sixth centuries (summary of evidence in Mundell-Mango 1996). The fifth-century date is borne out by an inscription from Bostra dated to the reign of Anastasius. It records an edict that the dux Palaestinae be financed by the commerciarus, or ‘controller of foreign trade’ at Klysma, indicating that the tax revenue generated by the maritime trade with the east remained considerable (Sartre 1982, 112, no. 9046; Mayerson 1996, 123). The presence of an alabarch at Antinoopolis on the Nile in 568 suggests that the overland route from there to the Red Sea might still have been active (Johnson and West 1949, 229). In about 570, a pilgrim from Piacenza recorded that ships from India, that is, probably from Ethiopia, brought aromatics to Aila and sailed to Klysma (Wilkinson 2002, 147–48; Tomber 2008, 80–81). Towards the end of the sixth century, a man from Aila offered a monk of Enaton monastery near Alexandria three silver coins, so that the monk would pray for his ship, which he had sent with a cargo to Ethiopia (Regnault 1970, 191; Morony 2004, 188). The archaeological finds confirm a substantial trade. Analysis of coin and pottery evidence from Klysma suggests a period of intense activity in the fourth and fifth centuries (Tomber 2008, 66; Power 2010, 47–48). The same period is documented at Berenike and Sheneshef by Indian ceramics and black pepper, as well as Indo-Pacific beads, teak and sapphire (Francis 2007, 254–55; Tomber 2008, 76). The single Indian coin from the Red Sea coast is a mid-to-late fourth-century Kashatrapa coin of Rudrasena II found at Berenike’s church (Sidebotham 2007, 201). Nevertheless, there is very little firm evidence of any direct contact with India: teak used as building material in Berenike and Shenschef was likely recycled from Indian ships, but unfortunately the dating of the strata in which it was found is not entirely clear (Sidebotham and Wendrich 2001–2002). The epigraphic evidence of Indian merchants in Egypt dates before the third century, and unambiguous references

to Indians travelling to Egypt are rare (Salomon 1991; Vita Isidori 51D.). Eastern imports were subject to duty upon entry to Alexandria. The Alexandrian Tariff, a list of these articles, attributed to the third-century jurist Aelius Marcianus, was included in Justinian’s Digest, promulgated in AD 533, which is often taken to suggest its continued relevance in the sixth century (Digest 39.4.16; Parker 2008, 149–50). Of the fifty items listed, Kosmas mentions only four: silk, pepper, spikenard (ναρδόσταχυν) and aloe (ἀλοη), which arrive at Taprobane from the regions between the island and China.12 In addition to these, the κοστάριν which Kosmas lists as an Indian plant may be the same as costum and costamomum (putchuk) mentioned in the Tariff (Miller 1969, 84–85; Wolska-Conus 1973, 346, n. 15.5). The Tariff does not include almandine (ἀλαβανδηνόν), attributed by Kosmas to Kabe, on the eastern shore of India. The stone seems to be named after a carbuncle described by Pliny as excavated in Caria and treated in the town of Alabanda (Pliny, Natural History 37.25). Almandines from India and Sri Lanka probably arrived in Europe both overland and by the maritime route via the Red Sea and Alexandria to the Mediterranean (Roth 1980). Four almandine beads, likely from India, have been found at Berenike (Francis 2000, 221–22; Borrell 2008, 14). In Carthage the remains of a gem-cutter’s workshop dated to the fifth or early sixth century contained more than a hundred almandines (Roth 1980, 324–28). Mineralogical studies demonstrated that the almandines used in Merovingian cloisonné work of the fifth and sixth centuries are likely from India and Sri Lanka and only during the seventh century is there a shift to garnets from deposits in Bohemia (Quast and Schüssler 2000; Lennartz 2001, 268–70). Among the commodities traded from the lands further east beyond Taprobane, not included in the Tariff, Kosmas lists τζανδάνα, that is probably sandalwood, which he attributes to the lands east of Taprobane. The term may be a Greek transcription of the Pali name candana (Dalby 2007, 51–57) or Sanskrit chandana (Casson 1982, 182); it seems otherwise unattested, and

On spikenard, usually νάρδος, see Miller 1969, 88–89 and for aloes, 63–65. Periplus Maris Erythraei, 39, 49, 56 and 63

attributes aloes to North India and nard to both South and North India. See also Miller 1969, 63–65.

12

192

M. KOMINKO

later Byzantine medical texts refer to sandalwood as σάνταλον (Petit 2013, 73). The σαγαλίνο that the Periplus includes among exports from Barygaza to the Persian Gulf has been sometimes taken to be sandalwood (Periplus 36; Miller 1969, 61, 86–87), but in reality it is more likely to be teak (Casson 1982, 181–83). Kosmas may be, therefore, the first Greek author to mention sandalwood, which nevertheless could not have been unknown considering that traces were found in Berenike (Vermeeren 2000, 340; Cappers 2006, 64, 67–68) The σησάμινα ξύλα Kosmas attributes to Kalliana, on the western coast of India, is sometimes identified as the wood of sesame, but it is more likely sheesham or Dalbergia sisso (Wolska-Conus 1973, 346, n. 15.3) and probably the same as σασαμίνων attributed to Barygaza in Periplus Maris Erythraei 36 (Casson 1982, 181). There is an interesting overlap between several commodities omitted by the Tariff but mentioned by Kosmas and the sixth-century materia medica, especially the works of Aetius of Amida and, to a lesser extent, Alexander of Tralles. The Tariff does not include coconut (‘Indian nuts’ = κάρυα ἰνδικά), which could not have been a rare commodity in Egypt since thirty fragments of coconut (cocos nucifera) have been found in five different trenches in Berenike (Wendrich et al. 2003, esp. 71). They are mentioned by Palladius in the report of Scholasticus of Thebes (Palladius, De gentibus Indiae e Brahmanibus 1.6) called Indian nuts. In the sixth century they appear several times in the work of Aetius (Aetius, Iatricorum 1.132.4; 4.24.90; 16.166.6–7; 16.176.1). The clove (καρυόφυλλον), now cultivated throughout the tropical world, is indigenous only to Moluccas in eastern Indonesia (Miller 1969, 48). Cloves were known in India from the second century BC (Miller 1969, 50), but they are not mentioned in the Periplus, nor included in the Alexandrian Tariff, unless we assume that they are covered by the term aroma indicum (Miller 1969, 51). Pliny, who is the first known classical author to make reference to cloves, uses the Greek term caryophyllon (Natural History 21.30). Cloves appear in texts from the second century onward, but rarely (Cyranides 1.18.32; Aelius Promotus, Dynameron 9.8.1; 44.3.7; 81.1.1; Pseudo-Galen, De remediis parabiblus libri III, 14.507, 512, 529). In the fourth to fifth century Philostorgius mentions cloves as growing in the regions east of Taprobane and writes

that the inhabitants of these regions believe it formerly grew in Paradise (Philostorgius, Ecclesiastical History 3.10.10). It is only in the sixth century that cloves become extensively used in materia medica, frequently recommended by Aetius (Iatricorum liber 1.132.5; 1.133.5; 1.234.20–21; 3.11.8; etc.) and Alexander of Tralles (De febribus 1.431 and Therapeutica 1.613.4; 2.259.24; 2.291.9; 2.525.8; 2.531.3; 2.545.19). In the seventh century Paul of Aegina, who practised in Alexandria, considers them rare in the city (McCabe 2009, 282–83). It is also interesting to note that Kosmas and Aetius seem to be the only Greek authors whose works preserve a reference to clove wood (ξυλοκαρυόφυλλον) (Iatricorum liber 1.132.18; 16.172.4; 16.173.1). Musk (μόσχος) is first mentioned in a list of perfumes by St Jerome (Adversus Jovinianum II. 8 PL 22, col. 311; McCabe 2009, 282), but is absent from medical sources until it appears in the works of Aetius (Iatricorum liber 1.131.36, 40 and 50; 1.132.5 and 18; 1.133.5; 1.234.20 and 21; 2.111.8) and Alexander of Tralles (De febribus 431; Therapeutics 1.2). The similarity between the eastern aromata listed by Kosmas and included—either for the first time, or with far greater frequency than before—in the sixth-century medical treatises suggests a change in commodities imported from the east. It would be unwise to draw conclusions on the basis of the short survey above, but overall, it may suggest a greater diversity of eastern regions with which Rome was connected through a web of mercantile networks. The change in the patterns of long-distance maritime trade with the east merits a more in-depth investigation, one that accounts for a variety of local, regional and long-distance partnerships, rather than presenting India’s western trade purely in terms of competition between Rome and Aksum with Sasanian Persia. We should not overstate the importance of Kosmas’ account, nor assume that his information is always correct, but what he describes is undoubtedly a more complex constellation of contacts and trade networks than the majority of modern scholarship allows. The mercantile contacts between India and the west were only a part of this system, and viewing it from a western perspective can give us only a limited understanding. Political changes on the subcontinent and further east are likely to have had a rippling effect on the trade with the west and need to be reintegrated into this narrative.

BEYOND EMPIRE

Bibliography All papyrological abbreviations are cited throughout the chapter in accordance with the conventions established in J. F. Oates, R. S. Bagnall, S. J. Clackson, A. A. O’Brien, J. D. Sosin, T. G. Wilfong and K. A. Worp, Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets. http://papyri.info/docs/checklist (last accessed 13 Nov. 2018) and P. M. Sijpesteijn, J. F. Oates, A. Kaplony, E. M. Youssef-Grob and D. Potthast, Checklist of Arabic documents. http://www.naherosten.lmu.de/isapchecklist (last accessed 13 Nov. 2018). Aelian. 1958. De natura animalium, A. F. Scholfield (ed. and trans.). 3 vols. Cambridge, MA. Aetius. 1935–1950. Aëtii Amideni libri medicinales, A. Olivieri (ed.). 2 vols. Corpus medicorum Graecorum 8. Leipzig. Alexander of Tralles. 1878. Alexander von Tralles 1, T. Puschmann (ed.). Vienna. Bagnall, R. S. and J. A. Sheridan. 1994. Greek and Latin documents from ‘Abu Shaar, 1992–1993. Bulletin of the American Society of Papyrologists 31: 109–20. Baum, W. and D. Winkler. 2003. The Church of the East: A concise history. London. Beal, S. (trans.). 1911. The Life of Hiuen Tsiang, translated from the Chinese of Shaman (monk) Hwui Li. London. Beaucamp, J. (trans.). 2007. Le martyre de Saint Aréthas et de ses compagnons. Bibliotheca Hagiographica Graeca 166. Paris. Beaucamp, J., F. Briquel-Chatonnet and C. J. Robin. 1999. La persécution des chrétiens de Nagrān et la chronologie Himyarite. Aram 11: 15–38. Beaucamp, J. and C. Robin. 1983. L’Évêché nestorien de Mâšmâhig dans l’archipel d’al-Bahrayn (Ve–IXe siècle). In Dilmun, new studies in the archaeology and early history of Bahrain, D. T. Potts (ed.), 171–96. Berlin. Becker, A. H. 2006a. Fear of God and the beginning of wisdom. Philadelphia, PA. ———. 2006b. The dynamic reception of Theodore of Mopsuestia in the sixth century: Greek, Syriac and Latin. In Greek literature in Late Antiquity: Dynamism, didacticism, classicism, S. F. Johnson (ed.), 29–47. Aldershot. Bopearachchi, O. 1992. Le commerce maritime entre Rome et Sri Lanka d’après les données numismatiques. Revue des études anciennes 94: 107–21. ———. 1998. Archaeological evidence on changing patterns of international trade relations of ancient Sri Lanka. In Origins, evolution and circulation of foreign coins in the Indian Ocean, O. Bopearachchi and D. P. M. Weerakkody (eds), 133–78. New Delhi. ———. 2006. Circulation of Roman and Byzantine gold coins in Sri Lanka: Fact or fiction? In Dal Denarius al dinar: L’oriente e la moneta Romana. Atti dell’incontro

193

di studio, Roma 16–18 settembre 2004, F. De Romanis and S. Sorda (eds), 181–200. Rome. Borrell, B. 2008. The early Byzantine lamp from Pong Tuk. Journal of the Siam Society 91: 1–26. Bowersock, G. 2013. The throne of Adulis: Red Sea wars on the eve of Islam. Oxford. Brock, S. 1985. The Christology of the Church of the East in the synods of the fifth to early seventh century: Preliminary considerations and materials. In AksumThyateira: Festschrift for Archbishop Methodios, G. Dragas (ed.), 125–42. London. Bruns, P. 2010. Wer war Paul der Perser? Studia Patristica 45: 263–68. Buck, C. 1996. The universality of the Church of the East: How Persian was Persian Christianity? Journal of the Assyrian Academic Society 10 (1): 54–95. Cappers, R. T. 2006. Roman footprints at Berenike: Archaeobotanical evidence of subsistence and trade in the Eastern Desert of Egypt. Los Angeles, CA. Carswell, J. 1991. The port of Mantai, Sri Lanka. In Rome and India: The ancient sea trade, V. Begley and R. D. De Puma (eds), 197–215. Madison, WI. Casson, L. 1982. Periplus maris Erythraei 36: Teak, not sandalwood. The Classical Quarterly, n.s. 32 (1): 181–83. ———. 1989. Periplus maris Erythraei. Princeton, NJ. ———. 1990. New light on maritime loans: P. Vindob. G 40822. Zeitschrift für Papyrologie und Epigraphik 84: 195–206. Cereti, C. G., L. M. Olivieri and J. Vazhuthanapally. 2002. The problem of the Saint Thomas crosses and related questions: Epigraphical survey and preliminary research. East and West 52 (1– 4): 285–310. Chediath, G. 2002. The three crucial terms in Syriac theology – Kyana, Qnoma, and Parsopa. The Harp 15: 59–65. ———. 2007. The Christology of Mār Bābai the Great. In Inkulturation des Christentums im Sasanidenreich, M. Arafa, J. Tubach and G. S. Vashalomidze (eds), 205–14. Wiesbaden. Choksy, J. K. 2013. Sailors, soldiers, priests and merchants: Reappraising Iran’s early connections to Ceylon. Iranica Antiqua 48: 363–91. Chronicle of Seert. 1907–1919. Histoire nestorienne inédite: Chronique de Séert, A. Scher, J. P. Périer, P. Dib and R. Griveau (eds and trans.). Paris. Colless, B. C. 1969–1970. The traders of the pearl. Abr Nahrain 9: 17–38. Dalby, A. 2007. Some Byzantine aromatics. In Eat, drink and be merry (Luke 12:19): Food and wine in Byzantium, L. Brubaker and K. Linardou (eds), 51–57. Aldershot. Darley, R. 2014. Indo-Byzantine exchange 4th to 7th centuries: A global history. PhD diss. University of Birmingham. ———. 2017. ‘Implicit cosmopolitanism’ and the commercial role of ancient Lanka. In Sri Lanka at the cross-

194

M. KOMINKO

roads of history, Z. Biedermann and A. Strathern (eds), 44–65. London. Daryaee, T. 2003. The Persian Gulf in Late Antiquity: The Sasanian era (200–700 CE). Journal of World History 14 (1): 1–16. ———. 2010. Bazaars, merchants and trade in Late Antique Iran. Comparative Studies of South Asia, Africa and the Middle East 28 (1): 401–9. Derrett, M. D. 1961. The history of ‘Palladius of the races of India and the Brahmans’. Classica et Mediaevalia 21: 64–135. De Romanis, F. 1997. Romanukharaṭṭha and Taprobane: Relations between Rome and Sri Lanka in the first century A.D. In Crossings: Early Mediterranean contacts with India, F. De Romanis and A. Tchernia (eds), 161– 237. New Delhi. Devendra, D. T. 1957. The date of the Anuradhapura Cross. Journal of the Royal Asiatic Society, n.s. 5 (1): 85–96. Dijkstra, J. H. F. 2012. Blemmyes, Noubades and the Eastern Desert in Late Antiquity: Reassessing the written sources. In The history of the peoples of the Eastern Desert, H. Barnard and K. Duistermaat (eds), 238–47. Los Angeles, CA. Dijkstra J. and G. Greatrex. 2009. Patriarchs and politics in Constantinople in the reign of Anastasius (with a reedition of O.Mon.Epiph. 59). Millennium 6: 223–64. Eide, T., T. Hägg, R. Holton Pierce and L. Török (eds). 1998. Fontes Historiae Nubiorum. Textual sources for the history of the Middle Nile region between the eighth century BC and the sixth century AD. Vol. 3: From the first to the sixth century. Bergen. Elweskiöld, B. 2005. John Philoponus against Cosmas Indicopleustes: A Christian controversy on the structure of the world in sixth-century Alexandria. PhD dissertation, Lund University. Faller, S. 2011. The world according to Cosmas Indicopleustes: Concepts and illustrations of an Alexandrian merchant and monk. Transcultural Studies 1: 193–232. Francis, P. 2000. Human ornaments. In Berenike 1998: Report of the 1998 excavations at Berenike and the survey of the Egyptian desert, including excavations in Wadi Kalalat, S. E. Sidebotham and W. Z. Wendrich (eds), 211–25. Leiden. ———. 2007. Personal adornments. In Berenike 1999/2000, S. E. Sidebotham and W. Z. Wendrich (eds), 200–10. Los Angeles, CA. ———. 2013. The beads. In Mantai: City by the sea, J. Carswell, S. Deraniyagala and A. Graham (eds), 349– 73. Aichwald. Ghosh, S. 2000–2001. The route of horse trade early in India (up to c. 500 AD). Proceedings of the Indian History Congress 61: 126–31. Giardina, A. 1993. The merchant. In The Romans, A. Giardina (ed.), 245–71. Chicago, IL.

Glover, I. 1996. The archaeological evidence for early trade between India and Southeast Asia. In The Indian Ocean in antiquity, J. Reade (ed.), 373–78. London. Gowers, W. 1950. The classical rhinoceros. Antiquity 24: 61–71. Graf, D. F. 2018. The Silk Road between India and China. In Trade, commerce and the state in the Roman world, A. Wilson and A. Bowman (eds), 496–506. Oxford; New York, NY. Graham, A. 2013. Earthenware pottery. In Mantai: City by the sea, J. Carswell, S. Deraniyagala and A. Graham (eds), 191–212. Aichwald. Gray, P. T. R. 1979. The defense of Chalcedon in the east (451–553). Leiden. ———. 1997. Covering the nakedness of Noah: Reconstruction and denial in the age of Justinian. Byzantinische Forschungen 24: 193–205. Gropp, G. 1991. Christian maritime trade of Sasanian age in the Persian Gulf. In Golf-Archäologie: Mesopotamien, Iran, Kuweit, Bahrain, Vereinigte Arabische Emirate und Oman, K Schippmann, A. Herling and J. F. Salles (eds), 83–133. Espelkamp. Grumel, V. 1958. La Chronologie. Paris. Khuzistan Chronicle. 1893/1903. Un nuovo testo siriaco sulla storia degli ultimi Sassanidi, I. Guidi (ed.). In Actes du Huitième Congrès International des Orientalistes, tenu en 1899 à Stockholm et à Christiana I. Section sémitique (B), 3–36. Leiden 1893; repr., with Latin translation in Chronica Minora I, I. Guidi (ed.). Corpus Scriptorum Cristianorum Orientalium 1–2: 15–32. Paris 1903. Hahn, W. 2000. Aksumite numismatics: A critical survey of recent research. Revue numismatique: 281–311. Hatke, G. 2011. Africans in Arabia Felix: Aksumite relations with Himyar in the sixth century CE. PhD diss., Princeton University. ———. 2013. Aksum and Nubia: Warfare, commerce and political fictions in ancient Northeast Africa. New York. Herodotus. 1920. Herodotus: History, A. D. Godley (ed. and trans.). Cambridge, MA. Īšō‘dnaḥ of Basra. 1896. Le Livre de la Chasteté, composé par Jésusdenah, évêque de Baçrah, J. B. Chabot (ed.). Mélanges d’archéologie et d’histoire 16 (3–4): 1–80, 225–91. Izdebski, A. 2014. Cultural contacts between the superpowers of Late Antiquity: The Syriac school of Nisibis and the transmission of Greek educational experience to the Persian empire. In Cultures in motion: Studies in the Medieval and Early Modern periods, A. Izdebski and D. Jasiński (eds), 185–204. Krakow. Johnson, A. C. and L. C. West. 1949. Byzantine Egypt: Economic studies. Princeton, NJ. Jullien, F. 2008. Le monachisme en Perse: La réforme d’Abraham le Grand, père des moines de l’Orient.

BEYOND EMPIRE

Corpus Scriptorum Christianorum Orientalium 622, Subsidia 121. Leuven. ———. (ed.). 2015. Histoire de Mār Abba, catholicos de l’Orient. Martyres de Mār Grigor, général en chef du roi Khusro Ier et de Mār Yazd-panāh, juge et gouverneur. Corpus Scriptorum Christianorum Orientalium 658–659, Scriptores Syri. 254–55. Leuven. Justinian. 2009. Digest of Justinian, A. Watson (trans.). Philadelphia, PA. Kennet, D. 2004. Sasanian and Islamic pottery from Ras alKhaimah: Classification, chronology and analysis of trade in the western Indian Ocean. Oxford. Kobishchanov, Y. 1979. Axum. Philadelphia, PA. Kominko, M. 2013. The world of Kosmas: Illustrated Byzantine codices of the Christian topography. Cambridge. Kozah, M. 2015. Isho‘yahb III of Adiabene’s letters to the Qataris. In An anthology of Syriac writers from Qatar in the seventh century, M. Kozah, A. Abu-Husayn, S. S. al-Murikhi and H. al-Thani (eds), 43–88. Gorgias Eastern Christian Studies 39. Piscataway, NJ. Kröger, J. 1979. Sasanian Iran and India: Questions of interaction. In South Asian archeology, H. Härtel (ed.), 441– 48. Berlin. Legge, J. 1886. A record of Buddhistic kingdoms, being an account of the Chinese monk Fa-Hien of his travels in India and Ceylon (A.D. 399–414) in search of the Buddhist Books of Discipline. Oxford. Lennartz, A. 2001. Die Rolle Ägyptens im mediterranen Fernhandel vom Ende des 6. Jahrhunderts bis zu seiner arabischen Eroberung. In Archäologisches Zellwerk: Beiträge zur Kulturgeschichte in Europa und Asien. Festschrift für Helmut Roth zum 60. Geburtstag, E. Pohl et al. (eds), 267–80. Rahden. Lim, R. 2018. Trade and exchanges along the silk and steppe routes in late antique Eurasia. In Empires and exchanges in Eurasian Late Antiquity: Rome, China, Iran, and the Steppe, N. Di Cosimo and M. Maas (eds), 70–83. New York; Cambridge. Maas, M. 2003. Exegesis and empire in the early Byzantine Mediterranean: Junillus Africanus and the Instituta regularia divinae legis. Tübingen. MacCoull, L. 2007. Philosophy in its social context. In Egypt in the Byzantine world 300–700, R. S. Bagnall (ed.), 67–82. Cambridge. McCabe, A. 2009. Imported materia medica, 4th–12th centuries, and Byzantine pharmacology. In Byzantine trade, 4th–12th centuries: The archaeology of local, regional and international exchange, M. Mundell-Mango (ed.), 273–96. Farnham. McCrindle, J. W. 1897. The Christian topography of Cosmas, an Egyptian monk. London. Manzo, A. 2005. Aksumite trade and the Red Sea exchange network: A view from Bieta Giyorgis (Aksum). In People of the Red Sea: Proceedings of Red Sea Project II

195

held in the British Museum October 2004, C. M. Starkey (ed.), 51–66. Oxford. Mayerson, P. 1993. A confusion of Indias: Asian India and African India in the Byzantine sources. Journal of the American Oriental Society 113 (2): 169–74. ———. 1996. The port of Clysma (Suez) in transition from Roman to Arab rule. Journal of Near Eastern Studies 55 (2): 119–26. Metlich, M. A. 2006. Aksumite gold coins and their relation to the Roman-Indian trade. In Dal denarius al dinar: L’oriente e la moneta Romana. Atti dell’incontro di studio, Roma 16–18 settembre 2004, F. De Romanis and S. Sorda (eds), 99–103. Rome. Miller, J. 1969. The spice trade of the Roman empire, 29 BC to AD 641. Oxford. Mingana, A. 1926. The early spread of Christianity in India. Bulletin of the John Rylands Library 10: 435–514. Monchi-Zadeh, D. 1982. Xusrov I Kavatan ut Retak, Pahlavi text: Transcription, transliteration and translation. In Monumentum Georg Morgenstierne 2: 47–91. Acta Iranica 22. Leiden. Morony, M. G. 2004. Economic boundaries? Late Antiquity and early Islam. Journal of the Economic and Social History of the Orient 47 (2): 166–94. Morrison, K. D. 1997. Commerce and culture in South Asia: Perspectives from archaeology and history. Annual Review of Anthropology 26: 87–108. Morrisson, C. 2002. Byzantine money: Its production and circulation. In The economic history of Byzantium: From the seventh to the fifteenth century, A. E. Laiou (ed.), 909–66. Washington, DC. Mundell Mango, M. 1996. Byzantine maritime trade with the East (4th–7th centuries). Aram 8: 139–63. Munro-Hay, S. C. H. 1989. The al-Madhariba hoard of gold Aksumite and Late Roman coins. Numismatic Chronicle 149: 83–100. Nappo, D. 2018. Money and flows of coinage in Red Sea trade. In Trade, commerce and the state in the Roman world, A. Wilson and A. Bowman (eds), 557–78. Oxford. Nau, F. 1915–1917. Résumé de monographies syriaques: Barṣauma, Abraham de la Haute Montagne, Siméon de Kefar ‘Abdin, Yaret l’Alexandrin, Jacques le reclus, Romanus, Talia, Asia, Pantaléon, Candida, Sergis et Abraham de Cašcar. Revue de l’Orient chrétien 10 (20): 3–32. ———. 1918–1919. Histories d’Abraham de Kaskar et de Babaï de Nisibie. Revue de l’Orient chrétien 1 (21): 161–72. Olympiodorus. 1900. In Aristotelis Meteora commentaria, G. Stüve (trans.). Berlin. Oppolzer, T. 1962. Canon of eclipses. New York. Palladius. 1987. De gentibus Indiae et Bragmanibus, W. Berghoff (ed.). Meisenheim am Glan.

196

M. KOMINKO

Panaino, A. 2004. Commerce and conflicts of religions in Sasanian Iran between social identity and political ideology. In Commerce and monetary systems in the ancient world: Means of transmission and cultural interaction, R. Rollinger and C. Ulf (eds), 385–401. Stuttgart. Parker, G. 2008. The making of Roman India. Cambridge. Payne, R. 2011. Monks, dinars and date palms: Hagiographical production and the expansion of monastic institutions in the early Islamic Persian Gulf. Arabian Archaeology and Epigraphy 22: 97–111. Peacock, D. P. S. 2007. Pottery from the survey. In The ancient Red Sea port of Adulis, Eritrea: Results of the Eritro-British expedition 2004–5, D. P. S. Peacock and L. Blue (eds), 79–108. Oxford. Pedersen, R. K. 2008. The Byzantine–Aksumite period shipwreck at Black Assarca Island, Eritrea. Azania 92: 77–94. Petit, C. 2013. The fate of a Greek medical handbook in the Medieval West: The Introduction, or the Physician ascribed to Galen. In Medical books in the Byzantine world, B. Zipser (ed.), 57–77. Bologna. Phillipson, D. W. 1998. Ancient Ethiopia. Aksum: Its antecedents and successors. London. ———. 2009. Ancient churches of Ethiopia: Fourth–fourteenth centuries. New Haven, CT; London. ———. 2014. Foundations of African civilization: Aksum and the Northern Horn, 1000 B.C.–1300 A.D. Woodbridge. Pigulevskaya, N. 1956. Economic relations in Iran during the fourth through sixth centuries AD. Journal of the K. R. Cama Oriental Institute 38: 60–81. Power, T. 2010. The Red Sea during the ‘long’ Late Antiquity (AD 500–1000). DPhil. dissertation, University of Oxford. Price, R. 2009. The Acts of the Council of Constantinople of 553 with related texts on the Three Chapters Controversy. Liverpool. Procopius. 1914. History of the Wars I, H. B. Dewing (ed. and trans.). Cambridge, MA. Pseudo-Zachariah of Mytilene. 2011. Historia Ecclesiastica, R. Phenix, C. Horn and G. Greatrex (trans.). Liverpool. Quast, D. and U. Schüssler. 2000. Mineralogische Untersuchungen zur Herkunft der Granate merowingerzeitlicher Cloisonnéarbeiten. Germania 78: 75–96. Rathbone, D. 2000. The ‘Muziris’ papyrus (SB XVIII 13167): Financing Roman trade with India. Bulletin de la Société Archéologique d’Alexandrie 46: 39–50. Rau, W. 1986. Poetic conventions in Indian Kāvya literature. Adyar Library Bulletin 50: 191–97. Ray, H. P. 1994. The winds of change: Buddhism and the maritime links of early South Asia. Oxford. Regnault, L. 1970. Les sentences des pères du désert: Nouveau recueil. Sablé-sur-Sarthe.

Roth, H. 1980. Almandinhandel und -verarbeitung im Bereich des Mittelmeeres. Beiträge zur Allgemeinen und Vergleichenden Archäologie 2: 309–35. Rubin, Z. 2008. Islamic traditions on the Sāsānian conquest of the Ḥimyarite realm. Der Islam 84 (2): 185–99. Salomon, R. 1991. Epigraphic remains of Indian traders in Egypt. Journal of the American Oriental Society 111 (4): 731–36. Sartre, M. (ed.) 1982. Inscriptions grecques et latines de la Syrie 13.1. Paris. Schenk, H. 2001. The development of pottery at Tissamaharama. In Ancient Ruhana: Sri Lankan–German archaeological project in the southern province, H.-J. Weisshaar, H. Roth and W. Wijeyapala (eds), 59–195. Mainz. Schneider, P. 2004. L’Éthiopie et l’Inde. Rome. Scholten, C. 1996. Antike Naturphilosophie und Christliche Kosmologie in der Schrift De Opificio Mundi des Johannes Philoponos. Berlin. Statutes of the school of Nisibis. 1961. The statutes of the school of Nisibis, A. Vööbus (ed.). Papers of the Estonian Theological Society in Exile 12. Stockholm. Schwarz, F. F. 1975. Kosmas und Selediba. Ziva antika (Antique vivante) 20 (1–2): 469–90. Sedov, A. V. 2007. The port of Qana’ and the incense trade. In Food for the gods: New light on the ancient incense trade, D. Peacock and D. F. Williams (eds), 71–111. Oxford. Seland, E. H. 2012. Trade and Christianity in the Indian Ocean during Late Antiquity. Journal of Late Antiquity 5 (1): 72–86. Sen, T. 2014. Buddhism and the maritime crossings. In China and beyond in the Medieval period: Cultural crossings and inter-regional connections, D. Wong and G. Heldt (eds), 39–62. Amherst, MA; Delhi. Sidebotham, S. E. 1991. Ports of the Red Sea and the ArabiaIndia trade. In Rome and India: The ancient sea trade, V. Begley and R. D. De Puma (eds), 12–38. Madison, WI. ———. 2007. Coins. In Berenike 1999/2000, S. E. Sidebotham and W. Z. Wendrich (eds), 200–10. Los Angeles, CA. Sidebotham, S. E. and W. Z. Wendrich. 2001–2002. Berenike: Archaeological fieldwork at a Ptolemaic– Roman port on the Red Sea coast of Egypt 1999–2000. Sahara 13: 23–50. Sinkankas, J. 1989. Emerald and other beryls. Prescott, AZ. Sorabji, R. 1987. John Philoponus. In Philoponus and the rejection of Aristotelian science, R. Sorabji (ed.), 5–18. London. Strabo. 1917. Geography, H. L. Jones (ed. and trans.). Cambridge, MA. Strauch, I. 2013. Foreign sailors on Socotra: The inscriptions and drawings from the Cave Hoq. Bremen. Synodicon orientale. 1902. Synodicon orientale, J. B. Chabot (ed.). Paris.

BEYOND EMPIRE

al-Tabari. 1999. The history of al-Tabari 5: The Sasanids, the Byzantines, the Lakhmids and Yemen, C. E. Bosworth (trans.). New York. Tamcke, M. 2007. Abraham of Kashar’s pilgrimage. Aram 19: 477–82. Tampoe, M. 1989. Maritime trade between China and the West: An archaeological study of the ceramics from Siraf (Persian Gulf) 8th–15th century AD. Oxford. Tang, L. 2014. Traces of Syro-Persian Christians in ancient Ceylon. In Geschichte, Theologie und Kultur des syrischen Christentums: Beiträge zum 7. Deutschen Syrologie-Symposium in Göttingen, Dezember 2011, M. Tamcke and S. Grebenstein (eds), 381–92. Wiesbaden. Tomber, R. 2005. Aksumite and other imported pottery from Kamrej, Gujarat. Journal of Indian Ocean Archaeology 2: 99–102. ———. 2007a. Bishops and traders: The role of Christianity in the Indian Ocean during the Roman period. In Natural resources and cultural connections of the Red Sea, J. Starkey, P. Starkey and T. Wilkinson (eds), 219–28. Oxford. ———. 2007b. Rome and Mesopotamia – importers into India in the first millennium AD. Antiquity 81: 972–88. ———. 2008. Indo-Roman trade: From pots to pepper. London. ———. 2018. Egypt and eastern commerce during the second century AD and later. In Trade, commerce and the state in the Roman world, A. Wilson and A. Bowman (eds), 531–56. Oxford. Tripathi, A. 2004. Onshore and offshore exploration of Elephanta Island: Evidence for Indo-Roman trade. Journal of Indian Ocean Archaeology 1: 116–23. Uthemann, K. H. 1999. Kaiser Justinian als Kirchenpolitiker un Theologe. Augustinianum 39: 5–83. de la Vaissière, E. 2012. Central Asia and the Silk Road. In The Oxford handbook of Late Antiquity, S. F. Johnson (ed.), 142–70. Oxford. Varghese, B. 2015. The so-called Persian crosses in South India: State of research. In Christsein in der islamischen Welt: Festschrift für Martin Tamcke zum 60. Geburtstag, S. H. Griffith and S. Grebenstein (eds), 389–99. Wiesbaden. Vermeeren, C. E. 2000. Wood and charcoal. In Berenike 1998: Report of the 1998 excavations at Berenike and the survey of the Egyptian desert, including excavations in Wadi Kalalat, S. E. Sidebotham and W. Z. Wendrich (eds), 311–42. Leiden. Vööbus, A. 1965. History of the school of Nisibis. Corpus Scriptorum Christianorum Orientalium 266. Leuven. Wainwright, G. A. 1942. Cosmas and the gold trade of Fazoqli. Man 42: 52–58. Walburg, R. 2008. Coins and tokens from ancient Ceylon. Wiesbaden.

197

Walburg, R. and A. Graham. 2013. The coins. In Mantai: City by the sea, J. Carswell, S. U. Deraniyagala and A. J. Graham (eds). Aichwald. Philostorgius. 1855. The ecclesiastical history of Sozomen … Also the ecclesiastical history of Philostorgius as epitomized by Photius, E. Waldorf (trans.). London. Walker, J. 2006. The legend of Mar Qardagh: Narrative and Christian heroism in Late Antique Iraq. Berkeley, CA; London. ———. 2018. Luminous markers: Pearls and royal authority in Late Antique Iran and Eurasia. In Empires and exchanges in Eurasian Late Antiquity: Rome, China, Iran, and the Steppe, N. Di Cosimo and M. Maas (eds), 253–67. Cambridge. Wang, H. 2004. Money on the Silk Road: The evidence from Eastern Central Asia to c. A.D. 800, with a catalogue of the coins collected by Sir Aurel Stein. London. Weerakkody, D. P. M. 1997. Taprobanê: Ancient Sri Lanka as known to Greeks and Romans. Turnhout. Wendrich, W. Z., R. S. Tomber, S. E. Sidebotham, J. A. Harrell, R. T. J. Cappers, and R. S. Bagnall. 2003. Berenike crossroads: Integration of information. Journal of the Economic and Social History of the Orient 46 (1): 46–87. Whitehouse, D. 1973. Chinese stoneware from Siraf: The earliest finds. South Asian Archaeology 1: 241–55. Whitehouse, D. and T. J. Wilkinson. 2009. Siraf: History, topography and environment. British Institute of Persian Studies Monograph. Oxford. Whitehouse, D. and A. Williamson. 1973. Sassanian maritime trade. Iran 11: 29–49. Whitfield, S. 2007. Was there a Silk Road? Asian Medicine 3: 201–13. Whittaker, H. 1991. Pausanias and his use of inscriptions. Symbolae Osloenses 66: 171–86. Wilkinson, J. 2002. Jerusalem pilgrims before the Crusades. Warminster. Wilson N. G. 1994. Photius: Bibliotheca. A selection. London. Winstead, E. O. 1909. The Christian topography of Cosmas Indicopleustes, edited with geographical notes. Cambridge. Wittman, J. 1913. Sprachliche Untersuchungen zu Cosmas Indicopleustes. Leipzig. Wolska, W. 1962. La Topographie chrétienne de Cosmas Indicopleustes: Théologie et science au VIe siècle. Paris. ———. 1968–1973. Cosmas Indicopleustes, Topographie chrétienne. Sources chrétiennes 141, 159, 197. Paris. Wolters, O. W. 1967. Early Indonesian commerce: A study of the origins of Śrīvijaya. Ithaca, NY. Wood, P. 2013. The chronicle of Seert: Christian historical imagination in Late Antique Iraq. Oxford.

WHERE DID THE CHRISTIANS GO? PEASANTS AND TRIBESMEN OF THE FAYUM, AD 1060–1240 Yossef RAPOPORT

When and how did Egypt become Muslim? In recent decades, historians have disagreed sharply on the time at which Egypt turned into a Muslim-majority country. Some have argued that Muslims became the majority as early as the ninth century, only 200 years after the Arab conquest, as a result of mass conversion that followed the suppression of a Coptic rebellion against Abbasid rule (Lapidus 1972). Others have pushed the Islamization process five centuries forward, highlighting mass conversions of Copts in the fourteenth century, a result of sustained popular and state persecution under the Mamluk sultanate (Little 1976; el-Leithy 2004). Yet a third group has denied that mass conversion occurred at all, arguing that continuous Arab nomadic settlement of areas where Coptic population declined was the key reason for Islamization (Brett 2005; O’Sullivan 2006). Over recent decades, studies of the rich documentary evidence, written on papyrus, dating from the first four centuries of Islamic rule have demonstrated that the early Islamization model is not tenable. The publication of significant amount of documents from tenth- and eleventh-century villages in Upper and Middle Egypt shows the persistence of Coptic rural life. In those village communities that are known to us in any detail, Muslims were still a minority as late as 1050 (Mouton 2003; Lev 2012; Livingstone 2014). The documentary trail for subsequent centuries, however, is not so rich, and in any case has not received as much scholarly attention. We know a lot about the mass conversions of the fourteenth century, but our evidence for these events mainly comes from Cairo, and reflects the pressures applied to a wealthy Coptic elite of bureaucrats and tax officials. The history of the Islamization of Egypt’s peasantry—without a doubt, the majority of its Medieval population—is yet to be told. The following chapter will contribute to the debate about the Islamization of the Egyptian countryside by focusing on one province, the Fayum, for which the evidence is particularly rich and extends up to the thirteenth century. A unique fiscal survey of the province, composed in 1245, shows that by that time the villages of the Fayum had acquired an overwhelming

Muslim majority, with Copts forming a small semiurban minority (Rapoport 2018). The survey also shows that most Muslim village communities claimed a tribal Arab identity, and that Islamization and tribalization occurred more or less simultaneously. Yet, I will reject here the possibility of a sudden influx of tribal populations into the Fayum, a migration that somehow supplanted the indigenous Coptic population. Instead, I will argue that Christian communities took upon themselves Islamic identities as well as tribal ideologies in response to changing political and economic circumstances. This explanation does not exclude the possibility of limited migrations and intermarriage, but fundamentally sees the pervasive concurrent Islamization and tribalization of the Fayum as a result of internal transformation. Al-Nābulusī’s Villages of the Fayum We know more about the history of the Fayum than about any other Egyptian province. It is the provenance of hundreds of thousands of papyri from the Ptolemaic, Roman, Byzantine and Islamic periods, rich documentary evidence that invariably comes from abandoned sites. For all their fertility, the villages of the Fayum were always in danger, a result of the exceptional vulnerability of the ambitious irrigation system put in place by the Ptolemaic kings in the third century BC. Documents from villages covered with sand are complemented by literary texts that describe repeated attempts—in all periods—to sustain the fragile water supply. Because the Fayum has provided us with such exceptionally rich evidence, we can uniquely follow its religious transformations, first from a colony of Greek soldier-settlers to a Late Antique Christian centre, and then its gradual conversion into a Muslimmajority province. In fact, the documentary evidence from early Islamic Fayum attests to a rather slow process of Islamization, whereby the majority of the villagers continued to practice Coptic Christianity well into the eleventh century, even if they had already adopted the Arabic language two centuries earlier.

200

Y. RAPOPORT

One of the most spectacular discoveries of Arabic documents from the Fayum in recent decades has been the private archive of the Banū Bifām, a Coptic family from the village of Damūya near al-Lāhūn, dating from 992 to 1029 (Fig. 1; Gaubert and Mouton 2014). This orderly archive dealing mainly with property and taxation was found in a residential building in the Naqlūn Monastery in 1997. It contains about fifty paper and parchment documents found in a jar and in a leather bag, respectively. The survival of this rich Coptic family archive, nearly four centuries after the Muslim conquest of Egypt, gives us an excellent view of the religious make-up of one small rural community during the first decades of the eleventh century. It leaves no doubt that during the 1020s, this small village of no more than several hundred people was almost entirely Christian (Gaubert and Mouton 2014, 176–78). Gaubert and Mouton have found that the documents mention about 100 Coptic names of local residents. The village had no more than a dozen Muslim local residents, who mostly lived around a mosque on the edge of the village, first mentioned in 1025. Some two centuries later, in 1245, Abū ῾Uthmān al-Nābulusī was sent from Cairo to the Fayum on the orders of the Ayyubid sultan of Egypt, al-Malik al-Ṣāliḥ. Al-Nābulusī, who had been a top official in the administration of al-Ṣāliḥ’s father, Sultan al-Kāmil, was now called out of retirement to report on the agricultural conditions in the province. The exceptional productivity of the Fayum had been renowned since antiquity, but its fiscal revenues were now in decline. Al-Nābulusī proceeded to survey the Fayum by going from village to village, relying on local tax and irrigation officials and paying careful attention to minute details of agricultural production. The resulting account has survived in two manuscripts (Fig. 2), and the treatise was first published by B. Moritz in 1898. It has now been re-edited and translated into English by the present author and Ido Shahar as The villages of the Fayyum (al-Nābulusī 2018). Al-Nābulusī’s work is divided into nine introductory chapters dealing with the geography, climate, irrigation systems and population of the province. These are followed by the main body of the treatise, the survey itself, with entries for more than 100 villages and hamlets. For each village, al-Nābulusī indicates the size of the village and the state of its habitation, its geographical location, the tribal or religious identity of its inhabitants, its sources of water, and local mosques, churches and monasteries. He also invariably indicates

Fig. 1: Packet of paper documents from the eleventhcentury archive of Banū Bifām, discovered in a four-handled jar in the monastery of Naqlūn in 1997 (Courtesy of W. Godłewski).

whether the village’s tax revenues are paid to the sultan, to a religious endowment (waqf), or, most commonly, allocated to army officers in return for military service (iqṭā῾). The lists of taxes levied on each village take up most of the treatise. They relate to the fiscal obligations of the villages, not to actual payments, and are based on the tax registers of a previous year, 641 AH/AD 1243. The lists are divided into taxes in kind, levied on grains, and taxes in cash levied on other agricultural products, such as livestock and cash crops. The category of taxes in cash also includes the poll tax levied on non-Muslims, recording with precision the number of non-Muslim men residing in each village. When al-Nābulusī came to the Fayum in 1245, the village of Damūya appears to have been standing in the same place, although it was now called Dimūh al-Lāhūn or Kūm Darī. It was still a very small village. Its minimal contribution to the provincial labour levy was only 0.08%, well below average; as the provincial population was about 70,000, it may have had no more than 100 inhabitants. But the religious mix of the village had changed completely. Dimūh was now an entirely Muslim community, with not a single Christian man subject to the poll tax. The inhabitants, according to al-Nābulusī, all hailed from the Hawwāra branch of the Lawāta Berber tribe. The identification of the village with the Hawwāra would be confirmed by the early Ottoman period, when the village was renamed Hawwārat ῾Adlān (Gaubert and Mouton 2014, 12). The dramatic transformation of Coptic Damūya to Muslim Dimūh was replicated in at least two other

WHERE DID THE CHRISTIANS GO?

201

Fig. 2: Title page of Abū ῾Amr ῾Uthman Ibn al-Nābulusī, Iẓhār Ṣan῾at al-Ḥayy al-Qayyūm fī Tartīb Bilād al-Fayyūm, MS Ayasofia 2960, copied in the middle of the sixteenth century and presented to Jānim min Qaṣrūh, the inspector of Royal Dikes (kāshif al-jusūr al-sulṭāniyya) in the Fayum and al-Bahnasā under the Ottoman administration (Courtesy of the Topkapı Sarayı Müzesi).

villages of southwestern Fayum, Tuṭūn and Buljusūq. Tuṭūn was a centre of Coptic learning up until the 950s, and a significant number of documents show it still had a Coptic majority in the mid-eleventh century. The nearby village of Buljusūq had a substantial Coptic community as late as the 1130s, as attested by a series of fragments carrying acknowledgements of debts made by peasants with Coptic names. But when al-Nābulusī visited the Fayum in 1245, both villages had relocated to new sites and were now predominantly Muslim. Tuṭūn was identified with the Banū Ḥātim clan, without a single resident Copt. Buljusūq, also occupied by the Banū Ḥātim, had nine non-Muslim men registered to pay the poll tax, but only one of them was present in the village, and the rest had travelled

away from the Fayum to other provinces in Upper Egypt. With practically no Christian community left, the church of Buljusūq had been deserted, and the only religious building standing was the village’s Friday mosque. Al-Nābulusī’s records show that by the thirteenth century, the Fayum was an overwhelmingly Muslim province, where non-Muslims had become a small minority. The poll tax records available to al-Nābulusī indicate that 1,142 non-Muslim male adults were registered as subject to the poll tax. Nearly a third of these non-Muslim men lived in the city of Madinat al-Fayum. The remainder, about 600 non-Muslim men, lived in eighteen different villages across the province. Together, they represented a non-Muslim population of

202

Y. RAPOPORT

some 3,500 (applying a multiplier of 3 to account for women and children). This was only 5% or less of the total provincial population, which was about 70,000 (Krol 2015). The small size of the non-Muslim population is corroborated by the rarity of references to Christian villages, and by the relatively few functioning churches. Out of over 100 villages, al-Nābulusī describes only two as predominantly Christian. These were two small suburban villages, Bāja and Minyat al-Usquf, which were the only two settlements in the Fayum where there were churches but no mosques. Besides these two villages, there were fifteen other active churches in the province, four of them in Madinat al-Fayum. In comparison, the province had at least seventy-eight mosques, of which forty-seven, mostly Friday mosques, were spread around the province’s villages. The majority of the non-Muslim men subject to the poll tax were Coptic Christians. We know from other sources that by the beginning of the Ayyubid period the city of Madinat al-Fayum also had Jewish and Melkite communities. Benjamin of Tudela, travelling in Egypt after Saladin’s rise to power, reports that the Fayum had a community of 200 Jews, presumably all in the provincial capital (Krol 2015, 144). Around the same time the Christian author Abū al-Makārim reports a Melkite church in Madinat al-Fayum alongside three Coptic churches (Abū al-Makārim 1969, 206). Al-Nābulusī, however, mentions only Christians, churches and monasteries, but no Jews, and when he identifies the sectarian affiliation of Christians—admittedly only once—he refers to them as Copts. The map (Fig. 3) visualizes the concentration of non-Muslim minorities in and around the urban and semi-urban centres. The largest community of nonMuslims was in Madinat al-Fayum, where they numbered 269 resident men, a third of the entire nonMuslim population of the province, and possibly as much as a quarter of the city’s population. Other large Christian communities were in the market villages of Bamawayh and Sinnūris, each with over 100 nonMuslim men. The two suburban villages of Bāja and Minyat al-Usquf, identified by al-Nābulusī as predominantly Christian, had non-Muslim communities of 100 and fifty men respectively. Other communities of over fifty Christian men lived in three villages that served as nodes of the emerging sugar-cane industry of the time (Dhāt al-Ṣafā, Fānū and Dimashqīn al-Baṣal). These communities were large enough to sustain functioning churches, sometimes two in the same village.

In addition, some dozen villages had non-Muslim minority communities of less than thirty men, occasionally only a handful of individuals. These scattered groups, presumably all Copts, were generally unable to maintain a functioning local church, and lived in overwhelmingly Muslim villages. As we zoom in on individual villages, we find a strong correlation between Christian communities and specific economic activities. The two villages which al-Nābulusī identifies as predominantly Christian, Bāja and Minyat al-Usquf, were also two of the few villages in the province where there were no taxes in kind at all, indicating no cultivation of grains. These two Christian villages paid all of their taxes in cash, mainly on the cultivation of trees and gardens, which accounted for 50% of the taxes in Bāja and 60% in Minyat al-Usquf. If we exclude the poll tax, then the taxes on orchards in both villages amounted to about 90% of taxes on the village economic activity. There is also a close correspondence between Christian presence and textile production. Taxes on the pits of weavers (ḥufar al-qazzāzīn) are mentioned in seven villages, amounting to 27 dinars. The largest sums were paid in the predominantly Christian villages of Bāja and Minyat al-Usquf, 9 and 6 dinars respectively. A tax of 4.8 dinars on the weavers in Ihrīt may be related to the presence of twenty-five non-Muslim men in this medium-sized village. The bulk of the tax on textile production was levied on weavers in Madinat al-Fayum, amounting to 484 dinars. Given the correpondence between taxes on weaving and Christian communities, it is plausible that many of the urban weavers were Christian too. There is no doubt that Christianity was on the decline in thirteenth-century Fayum. Al-Nābulusī records seven deserted churches, three of them in the village of Fānū, where there was also a high rate of non-Muslim absentees (40%). Al-Nābulusī reports that Fānū had recently suffered depopulation and diminishing tax revenues due to the tyranny of its local iqṭā῾-holder, and the village’s vineyards were particularly affected. As we have noted, the only church in the village of Buljusūq was deserted when eight out of nine registered Christian men in Buljusūq were recorded as absentees. Many more deserted churches either escaped al-Nābulusī’s attention or were by now converted to other uses. Al-Nābulusī himself acknowledges the Coptic past: he tells us that the village of Sīla, where only seven Christian men are mentioned in his tax survey, was once a centre of Coptic Christianity and home to forty churches.

WHERE DID THE CHRISTIANS GO?

203

Fig. 3: Map of villages, active churches and monasteries in the Fayum, 641 AH/AD 1243, based on al-Nābulusī’s Villages of the Fayyum (Map: Max Satchell).

What appears most striking is the disappearance of the Coptic peasantry. The Copts were now a small minority, concentrated in Madinat al-Fayum and its nearest orchards, as well as in a handful of large villages which also served as regional markets or centres for textile production. We can say with near certainty that Ayyubid Fayum had no Copts tending to arable land. The cultivation of wheat and barley was the exclusive domain of Muslim peasants, who formed the vast majority of the population, and whose mosques now replaced the churches as the dominant public spaces in the rural landscape. Al-Nābulusī’s register suggests that the disappearance of the Coptic peasantry was almost certainly related to the weight of the poll tax levied on nonMuslim men. The Islamic legal tradition allows exemptions and lower rates, but these were abandoned by the Ayyubid sultans and the Fatimid caliphs before them, at least since the 1130s. Evidence from the Cairo Genizah shows that the authorities put in place a flat rate, granting no exemptions, and combined it with

reinvigorated tax-enforcement (Alschech 2003). During the 1220s, in response to a sharp monetary crisis, the Ayyubid government announced the poll tax to be 2 dinars per head, with no distinctions between rich and poor (Severus ibn al-Muqaffa῾ et al. 1943–, 4: 95). A flat tax-rate of 2 dinars per head is also reported by al-Nābulusī. The Villages of the Fayyum allows us to quantify the heavy burden of the poll tax on the Coptic rural communities of the Fayum. This is very evident in the two Christian villages of Bāja and Minyat al-Usquf, where the poll tax constituted about a third of the total taxes. In Bāja, the entire taxes of the village amounted to 476 dinars. Of these, a full 180 dinars, or 38%, were levied as poll tax on the ninety resident Christians in the village. In the smaller village of Minyat al-Usquf the situation was only slightly better. Of the total taxes of 343.75 dinars, the poll tax on forty-seven resident Christians was 94 dinars, or 27.3% of the entire tax burden of the village. Keeping the faith did not come cheap.

204

Y. RAPOPORT

For cereal-producing peasants, who paid their taxes in kind and had hardly any access to cash, a payment of 2 gold dinars a year was unsustainable. Most villages in the Fayum paid the majority of their taxes in grains. Based on al-Nābulusī’s data, an average Fayumi village had a population of 1,500 inhabitants, liable to pay 1,400 ardabbs (1 ardabb = 90 litres) in mixed grain, and only 185 gold dinars in cash. Now let us assume this average village was Christian, and that all the men in it were liable to a flat poll tax of 2 dinars. Out of a population of 1,500, some 500 would be men subject to the poll tax, which would amount to 1,000 dinars. The local revenues from cash crops could not have covered even a small part of that amount. If the poll tax were to be paid, peasants would have had to turn their grain reserves into cash. But such grain reserves were scarce. Average taxes in grain of 1,400 ardabbs represented about a quarter of the harvest, and the rest was mostly consumed locally—the annual consumption of grains by a village of 1,500 inhabitants must have been about 4,000 ardabbs (Goitein 1967–99, 4: 235; Sabra 2000, 120). Any grain surpluses were small, hardearned and unreliable. In order to pay a poll tax of 2 dinars per adult male, the peasants would have had to gamble on the limited grains stored for a bad year, and then transport the yield in order to find buyers in a cash-scarce economy. Under these conditions, it is unsurprising that the vast majority of the villages in the Fayum, which primarily relied on grains and livestock for subsistence, were entirely Muslim. The flat-rate, inflexible Ayyubid poll tax made the survival of a Coptic peasantry unsustainable, and led to the concentration of the remaining Copts in the provincial capital, in market villages, in crafts and as orchardists. The poll tax moulded the Coptic communities of the Fayum into a thin layer of clergy and well-off urbanites, whose wider base among the grain-producing villagers has disappeared. Tribal peasantry In al-Nābulusī’s survey, the peasantry was not only Muslim but also Arab and tribal. Nearly all villages and hamlets in the Fayum are identified with a named tribal section or clan, and most of the population of the Fayum consisted of Arab tribesmen, who are also called badw, or Bedouin. The entire tribal structure is described by al-Nābulusī in his introductory chapters, and is then repeated in the individual village entries, where the people of the village (called the ahl) are

nearly always identified as belonging to one of the named clans. Because each village was almost always identified with one clan only, the clan identity defined the boundaries of the population of the village. Most commonly, the tribal people are called ῾arab, or Arabs. In the village entries, al-Nābulusī sometimes identifies the local villagers as ‘Arab, of the such-and-such clan’. The tribal system presented by al-Nābulusī is the standard segmentary model familiar to modern anthropologists. At the top level were three tribal confederacies (aṣl or āl), each inhabiting a large number of villages. First in importance were the Banū Kilāb, with as many as fifty villages, followed by the Banū ῾Ajlān and the much smaller Lawāta, a Berber tribe. The Banū Kilāb dominated the centre, south and west; the ῾Ajlān in the east and the north, while the Lawāta dwelt in villages along the Lāhūn gap (Fig. 4). Each tribal confederacy was divided into sections, which al-Nābulusī usually calls fakhdh (pl. afkhādh), or, infrequently, ‘branch’ (far῾). Each tribal section inhabited a varying number of settlements, from one hamlet (the Banū Muṭayr in Sanhūr) to as many as nineteen villages (the Banū Zar῾a). Al-Nābulusī sometimes calls the tribal people ‘Bedouin’, and he does so when he wants to create a distinction between them and the non-tribal communities whom he calls ḥaḍar; this badw–ḥaḍar dichotomy was later enshrined in Ibn Khaldūn’s (d. 1405) philosophy of history. Unlike the inhabitants of Bedouin villages, the ḥaḍar communities were clan-less, and under the protection of clans based in other villages. Thus, the Christian villagers of Bāja were under the protection of the Banū ῾Āmir, and the Christian ḥaḍar inhabitants of Minyat al-Usquf were under the protection of the Banū Rabī῾a of the Kilāb. The Muslim ḥaḍar village of Akhṣāṣ al-Ḥallāq was protected by the Banū Ka῾b, but when the latter were driven away by the governor, their protection was taken over by the Banū Qayṣar. The inhabitants of Bamawayh, a large village with a weekly market, were ḥaḍar, while their protectors were the Banū Samālūs. The Bedouin–ḥaḍar dichotomy had a clear economic dimension. The purely ḥaḍar villages of Bāja, Minyat al-Usquf—both Christian villages—and Akhṣāṣ al-Ḥallāq, a Muslim village, did not pay any taxes in wheat or barley. Bāja had ‘no field crops worthy of note, except for some sugar canes and green vegetables [irrigated] by waterwheels’. Minyat al-Usquf, whose houses were surrounded by orchards and gardens, paid 216 dinars in land tax on perennial crops, but none on

WHERE DID THE CHRISTIANS GO?

205

Fig. 4: Map of tribes and clans in the Fayum, 641 AH/AD 1243, based on al-Nābulusī’s Villages of the Fayyum (Map: Max Satchell).

grains. Akhṣāṣ al-Ḥallāq, one of the ‘brides of the Fayum’, was due to pay 570 dinars on its gardens, as well as two sheep and four goats as alms-tax, but not a single ardabb in wheat. The absence of taxes in kind, indicating no local cultivation of grains, perfectly overlapped with the absence of clan-based social structure. The opposition between the tribal badw and the nontribal ḥaḍar corresponded to the presence, or absence, of grain cultivation. The few villages which relied solely on gardens and orchards, without any grain cultivation, were ḥaḍar, politically submissive and under protection. On the other hand, villages in which grain cultivation was the predominant economic activity were tribal by default, with Arab (or Berber) identity complementing the Muslim dominance in the village. It is important to emphasize that the Bedouin or Arab tribal villagers of the Fayum were definitely not nomads, but rather cereal-growing peasants, who worked the land as tenants. In nearly all tribal villages, the inhabitants are called the muzāri῾ūn, sometimes mu῾āmilūn, the tenants who are liable to pay land tax

in return for a lease of the village’s arable lands. A common phrase is ‘the land tax from the cultivation of the tenants (kharāj zirā῾at al-muzāri῾īn)’, which clearly identifies them by their fiscal duties (cf. Michel 2000, 526–67). These taxes are never individualized, and were almost certainly levied on the village as a whole through the village headmen. Myths of migration Chronologically and substantively, the tribalization of the Fayum was linked to its conversion to Islam, and the disappearance of the Coptic peasantry was a corollary of the spread of Bedouin tribal identity. In their study of the Banū Bifām archive, Gaubert and Mouton have already identified the Bedouin as ‘les veritables vecteurs de l’islam’, the real agents of Islamization in eleventh-century Fayum (Gaubert and Mouton 2014, 270). In the thirteenth century, however, the Bedouin were no longer agents of Islamization, because the process had been nearly completed. By then, the peasantry

206

Y. RAPOPORT

themselves were Bedouin. A close reading of the Villages of the Fayyum, as we have done here, leaves no doubt that the majority of the rural population in Ayyubid Fayum identified with village clans and subscribed to tribal ideologies. How, then, did Coptic Fayum acquire this Muslim and tribal identity? This is not just about religious identity, but also a question of social organization: how does a society become tribal after it was not? For most of its history, the population of the Fayum did not view itself as tribal, and there is nothing in the topography or location of the Fayum to ‘naturally’ produce tribal identity. Yet in al-Nābulusī’s Fayum, the process of tribalization and Islamization that began in the eleventh century had achieved its climax in the thirteenth, when the Arab Muslim tribesmen had become the majority and the settled Copts a semiurbanized minority. The emergence of tribal military groups in Medieval Fayum was part of a wider pattern in Middle Eastern history, from Syria to the Mahgreb; these were the ‘Bedouin Centuries’ (Brett 1995; Franz 2011). During the late tenth century, and especially in the eleventh, Syria and Palestine saw the rise of several independent states ruled by Bedouin elites, such as the Midrāsid emirate of Aleppo ruled by a clan of the Banū Kilāb, or the less successful Āl Jarrāḥ in Ramla. Their rise to power is often explained by the protection they offered to townsmen. Elsewhere in the Islamic world, the Banū Hilāl invasion of North Africa, beginning in 1050, is seen as a watershed in the history of Tunisia and the wider Maghreb, and Ibn Khaldūn used the Hilalian invasion as an exemplary parable about the destructive power of nomads. The Bedouin push of the eleventh century has been conventionally explained by waves of migration from the desert: a sudden influx of tribal populations, beginning with a few individuals in the eleventh century and culminating with mass migration, eventually leading to wholesale replacement of the indigenous population. The rise of the Bedouin dynasties in Fatimid Syria is thus explained by migrations of the Banū Kilāb from northern Arabia. Thierry Bianquis suggested that nomadic Bedouins invaded agricultural areas in order to find pasturages or were co-opted by the state by assuming formal responsibilities for rural protection (Bianquis 1986, 433, 655). Recent research on the climatic history of the Middle East links the rise of Bedouin dominance with a cold spell from 950 to 1100 that put an end to cotton production in Iran and

triggered crop failures throughout the eastern Mediterranean (Bulliet 2009; Ellenblum 2012). According to this model, waves of migration and settlement by Muslim nomadic tribes supplanted and replaced indigenous non-Muslim inhabitants, taking over lands abandoned by settled cultivators. As put forward by Levtzion in an influential essay, the militant expansion of Islam was closely associated with Arab nomads of the desert or the Turkic nomads of the steppe. Conquest by nomads was followed, where circumstances allowed, by migration and settlement (Levtzion 1990). The concurrent Islamization and Turkification of Anatolia in the later Middle Ages has been explained in a similar fashion. Vryonis (1971) used the appearance of many new Turkish place-names in western Anatolia as an indication of widespread tribal settlement. Brett (2005) and O’Sullivan (2006) extended this model to Medieval Egypt, arguing that Islamization was a result of Arab nomadic settlement in areas where Coptic population declined. With regard to the Fayum, however, this theory of mass migration and settlement of nomadic tribes is untenable. First, unlike in Anatolia, the vast majority of the names of villages in al-Nābulusī’s Fayum come from Coptic or Greek. In fact, only thirteen villages carry unambiguously Arabic names, while the vast majority of villages had names that pre-dated the Arabization of the province, and in many cases the names can be easily traced back to settlements mentioned in Roman and Greek papyri. It is remarkable that not even one of the ninety villages listed in Villages of the Fayyum carries a name that reflects the tribal identity of its residents, despite the fact that nearly every village is identified with an Arab clan. No village is named after the Kilāb, the ῾Ajlān, or any of their sections. Arabic names and sometimes a reference to lineage are common only in smaller hamlets. Even then, only a couple of hamlets are identified by the name of the dominant tribal section, while the rest are identified by reference to a family, such as Munsha᾿at Awlād ῾Arafa, al-῾Athāmina (those descended from ῾Uthmān), or to a named individual, such as Ibn Kurdī or Abū ῾Uṣayya. If the original non-tribal population were driven out by the Bedouin, one would have expected village names to change, and village sites to move. Nicholas Michel also noted this difficulty when commenting on the complete lack of tribal village names in the early Ottoman cadastral surveys of Upper Egypt, which sheds serious doubts on narrative accounts of the settlement of nomads (Michel 2002, 229). Place-names

WHERE DID THE CHRISTIANS GO?

demonstrate the strong continuity of habitation; villages with Arab Muslim inhabitants stood on the same sites as the formerly Christian villages. We have mentioned the village of Sīla, occupied in the thirteenth century by the Banū Zar῾a, which used to be a centre of Coptic Christianity. Of its forty churches, only one church and one monastery remained when al-Nābulusī visited the site. Admittedly, al-Nābulusī does describe desertion of villages at the edges of the Fayum, and our Islamic-era documentary evidence largely comes from sites deserted after the Fatimid civil war of 1068–73, such as Tuṭūn and Buljusūq in the southwest. Krol has used this desertion of sites to argue that the Coptic peasantry of the Fayum took flight, and were replaced by Arab nomads (Krol 2015). But there is no evidence of dramatic rupture in habitation in other areas of the Fayum or that it affected the majority of the villages. Villages which were deserted after 1068 were re-established on new sites nearby, keeping the old names alive. That, again, indicates demographic continuity rather than an influx of new ethnic groups. There was also a continuity of holy sites. Al-Nābulusī mentions a rural Sufi lodge (ribāṭ zāwiya) in a hamlet called Munsha᾿at al-Shaykh Abū ῾Abdallah al-Qaḥāfī, in the land of Iṭfīḥ Shallā, just below the Monastery of Naqlūn. According to al-Nābulusī, the entire hamlet was occupied by Sufis, and had a Friday mosque alongside the lodge, itself supported by the profits from local date palms and fig trees. The location of this Sufi institution points to remarkable continuity with the Coptic past, known to us through documentary evidence. In a deed dated 336 AH/AD 947, a Coptic woman called Tūsānah grants real estate to two monasteries known as Naqlūn and Shallā (Abbott 1937). While Naqlūn had survived to al-Nābulusī’s day, the monastery of Shallā disappeared in the three intervening centuries, only to be replaced by the Sufi lodge of Iṭfīḥ Shallā described by al-Nābulusī. Even the character of the lodge, with its endowed palms and figs supporting an enclosed community of mystics, is reminiscent of Coptic monasteries. The villagers of the Fayum were identified by al-Nābulsuī as Arab or Berber, and that identity implied a genealogy; in order to be an Arab, you had to be a descendant of a tribal group that came from the Arabian Peninsula. But not everyone who claimed to be an Arab was one. Genealogies were not, and are not, meant to be historical records. Tribal myths of origin, still found in peasant communities of the Middle East,

207

invariably include a founder and a time of ancestors followed by migration and settlement, which establishes the right to occupy a specific territory (Franz 2011). The tribe’s migration is a constitutive feature of nearly all Bedouin groups; members are certain that their tribe has not always resided in its present site. The tribal histories of Tunisian village communities, for example, would have us believe that the country as a whole was empty before first inhabited by nomadic tribes (Valensi 1985, 57). But even if we were to take Arabian lineage claims at face value, we are confronted by the problem of finding sufficient numbers of nomads willing to settle down. The sheer number of tribal people in the Fayum in the thirteenth century, a clear majority of its 70,000-strong population, precludes the possibility that all migrated from the surrounding deserts. The natural environment simply cannot support such large transhumant population. In fact, documented historical cases show that nomadic migrations are by necessity small-scale affairs. One can only speak of ‘Bedouin infiltration’ over long periods of time (Franz 2011, 39). That has been a criticism of the nomadic settlement model elsewhere. There is no material evidence of such a process of sedentrization in village sites in Palestine, for example (Walker 2013, 144). And there is one further, ultimate question. If we are to believe that the Muslim majority in al-Nābulusī’s Fayum was indeed descended from Arabian nomadic tribes, we must wonder about the fate of the indigenous population. We are not dealing here with one single village in which nomads have driven out the original cultivators, or with nomads settling on the margins of the cultivated areas, but rather with a transformation of an entire province. We need to ask what happened to the villagers who cultivated those fields and worshipped in these churches. Where did all the Christian peasants go? Forging a tribal identity The answer is, of course, nowhere. The most plausible explanation for the simultaneous Islamization and tribalization of the Fayum is that Coptic peasant communities took upon themselves Islamic identities as well as tribal ideologies. This explanation does not exclude the possibility of limited migrations and intermarriage, but fundamentally sees the pervasive tribalization of the Fayum as a result of internal transformation. Such internal transformation is far more

208

Y. RAPOPORT

feasible in demographical terms, is strongly supported by the continuity of Coptic village names, and has parallels in several other Middle Eastern contexts. Tribes are not biologically defined groups that only emerge in nomadic contexts, nor are they a primitive stage towards higher forms of social organization. They are instead social constructs, with symbolic, economic and political functions. In Frederik Barth’s social constructionist approach, awareness of common ancestry is a consequence of shared interests, not their cause. According to Barth, genealogy belongs to a derivative category of ‘cultural stuff’—together with language, food, and dress—developed in order to maintain the boundaries of the group against outsiders (Barth 1969). This approach has been widely applied to modern Middle Eastern tribal groups, where genealogy is malleable and routinely manipulated to fit actual working arrangements. In the context of Medieval history, Peter Webb’s recent work boldly applies Barth’s social constructionist theory for the emergence of Arab identity between the rise of Islam and the end of the tenth century. Webb argues that a notion of a unified Arab genealogy was constructed in Abbasid Iraq in order to maintain the solidarity of the conquering Muslim elites against a growing number of converts from among the conquered populations (Webb 2017). The social constructionist approach allows us to view the genealogical articulation of social structures as a response to the political, economic and social needs of village communities. Segmentary tribal organization and the Arabian genealogy that came with it served the villagers of the Fayum on several levels. The clan was a collective unit of production; the multivillage sections oversaw allocation of water resources and provided armed protection; and the all-embracing Arabian identity integrated the villagers into the Islamic order, brushing away their recent Christian past. For a Coptic community of peasants, coming under increased fiscal and administrative pressure by the late Fatimid period, the adoption of a tribal, Arab identity made sense. At the village level, as al-Nābulusī’s survey strongly suggests, the function of clans was linked to production of grains: clans regulated access to jointly held arable lands, while villages without arable land were clanless. There is a direct link between the pervasive clan structure of cereal-growing villages and the manner in which lease contracts were handed out by the state administration to village communities. The most important tax in kind, the munājaza land tax, was collectively

imposed on the arable land of the village (al-Qalqashandī 1913–18, 3: 458). Under the munājaza lease, common not only in the Fayum but also in other parts of Egypt, leases were granted to the villages’ representatives, so officials did not have to bother with parcelling out land to individual tenants, nor were they required to list the names of the individual cultivators. In this context, clan affiliation would appear to be the most effective way of defining who belonged to the privileged group of peasants entitled to the lease. This also explains how al-Nābulusī knew of the tribal identity of village communities. As lease contracts were registered in the name of a clan, this information was passed on to the visiting bureaucrat from Cairo. Multi-village tribal sections probably functioned as riparian communities, with the solidarity of lineage used to settle disputes over water rights. In the near absence of state intervention, it was the tribal group that most likely took upon itself the management and the upkeep of the canal system. The distribution of the tribal groups closely overlaps with the irrigation system of the Fayum (Rapoport and Shahar 2012). Tribal sections often united a string of villages along an irrigation canal, rather than simply binding together adjacent villages. Eight of ten villages along the Sinnūris Canal were inhabited by the Qayāṣira (Banū Qayṣar), of the ῾Ajlān tribe. In another example, all five villages lying along the al-Sharqiyya Canal, in the water-scarce area of the eastern Fayum, were inhabited by the Banū Zar῾a clan, also of the ῾Ajlān. Multi-village tribal sections also played a key military role in the relationship of protection between tribesmen and clan-less townsmen and orchardists. The tribal section provided protection to a non-tribal or ḥaḍar community which resided in a nearby village. For example, Bāja was under the protection of the Banū ῾Āmir section of the Kilāb, who lived in eight villages in the southwest of the Fayum; the ḥaḍar village of Minyat al-Usquf was protected by the Banū Rabī῾a, who lived in Dumūshiyya, just to the south. The same tribal sections were also the building blocks of the Arab auxiliary army enlisted by the Ayyubid state. At the end of Villages of the Fayyum, al-Nābulusī records a levy of 400 horsemen to be provided by the tribesmen of the villages of the Fayum in the event of a royal military campaign, divided by tribal sections. The list shows that the state co-opted the tribal sections at all levels: not only were cultivation rights given to village clans, but multi-village tribal sections were formally acknowledged as nuclei for military units.

WHERE DID THE CHRISTIANS GO?

On another level, beyond the material aspects of production and allocation of resources, membership in a tribal confederacy was also an ideological statement about autonomy and resistance to state power. Al-Nābulusī uses the terms Arab and Bedouin to indicate population groups outside the control of the state, whose political autonomy and military potential is a threat to central authority. The irascible tribesmen of al-Qubarā᾿ are min ajlāf al-῾arab, ‘crude Arabs’, a derogatory designation well attested in the writings of other state functionaries (Leder 2015, 114–16). Whether they are Kilāb or ῾Ajlān is of no consequence; it is their tribal identity—or should we say tribal ideology— that matters. The tribesmen had a right to bear arms, acknowledged by the official register of their military contributions. And more than the actual military power, it was the ideology of Bedouin-ness, ideals of autonomy and egalitarianism detached from actual pastoralism, which empowered and nourished armed resistance. Set against a Mamluk elite composed of newcomers both to Egypt and to Islam, Arab lineage expressed rights over the land, and a sense of ownership. When an Arab uprising engulfed large parts of the Egyptian countryside, including the Fayum, in 650 AH/ AD 1252–53, the tribal leader of the rebellion is reported to have scorned the rule of the Turkish slaves, claiming that the Arab tribes were ‘the true owners of the land’ (al-Maqrīzī 1934–72, 1: 386). Most importantly, tribal identity concealed the very fact of conversion. As incisively put by Tamer alLeithy, the ultimate act of conversion is its denial (el-Leithy 2004). Tribal identity, formed by constructed genealogies, distances the convert community from its pre-Islamic past. Sometimes the claims for shared ancestry could be reinforced by intermarriage, and in the fifteenth century al-Maqrīzī specifically linked the conversion of the Egyptian countryside to the marriage of Arab tribesmen to Coptic women (al-Maqrīzī 2002, 1: 218; Lev 2012). At other times, lineages were simply invented, a historical lie turning into a social fact. Most fittingly, the fifteenth-century bureaucrat al-Qalqashandī, who authored a famous tribal genealogy, claimed a Bedouin lineage for himself and for all the inhabitants of the Delta village of Qalqashanda (al-Qalqashandī 1980). The claim to Arab descent by the villagers of the Fayum must be understood as an attempt to negotiate the present rather than as a factual narrative of the past. As argued by Michael Brett, even the Banū Hilāl invasion of North Africa was perpetuated by sedentary

209

tribesmen who wished to present themselves as a class of warriors. He concludes that the Banū Hilāl epos emerged as an attempt to ‘modify the realities of the past in order to meet the exigencies of the present’ (Brett 1995, 265). Another example comes from the traditions concerning Berber origins, weaving a ‘Himyari Myth’ focusing on the Yemen as the place of origin of the Berber tribesmen of North Africa (Norris 1972). In Nasrid Granada, the repeated claims of the elite to be direct descendants of the Arab conquerors six centuries earlier could not have been historically accurate. But they conveyed the message that the elite of Nasrid Granada had nothing to do with a pre-Islamic past being claimed by an encroaching Christian enemy (Fierro 2014). Patronage and clientage I have argued here that tribalization and conversion are the only feasible explanations for the transformation of predominantly Christian villages into a tribal and Muslim peasantry. But how this tribalization happened in the Fayum is difficult to trace. Relative to other periods, there are very few documentary or literary sources for the Fayum between 1068 and al-Nābulusī’s visit in 1245. In some ways, this dearth of textual sources may have been a by-product of the process of tribalization, or even a pre-condition for it. Bedouin identity has an oral, anti-textual component, and genealogies can only be forged away from the limelight of history. It is far harder to invent a lineage in a milieu that keeps written records (Sowayan 1996). What we can do is to read back the process of change from its outcomes, from the functions tribes and clans performed, as well as from comparisons between al-Nābulusī’s Fayum and earlier periods in the Fayum’s history. One possible route by which tribalization could have occurred was through relations of patronage and clientage. Since the early Islamic centuries, the privileges accorded to tribal groups meant that new converts wished to attach themselves to existing clans, often creating legal ties of patronage similar to those tying an ex-slave to his or her former master. Early Islamic law even recognized patronage ties created by conversion, in which the convert to Islam accepts the same conditions as that of a manumitted slave, in return for the protection granted by the patron’s tribal group (Bernards and Nawas 2005). The Fatimid dynasty was equally committed to the political and symbolic value of tribal ideologies, a result of its own dependence on

210

Y. RAPOPORT

North African Berber groups. The Fatimids also tried to regulate membership in the ‘noble’ tribes. An edict of the Fatimid caliph, dated 414 AH/AD 1022, explicitly deals with marginal outlaws who sought protection with Arab tribes ‘whenever a governor earnestly chases them down’ (al-Musabbiḥī 1978, 4–6). The edict refers to rural outlaws who sought refuge with tribes allied with the Fatimid government by making claims about their genealogy—about shared ancestors—that would integrate them into the tribe, and thus shield them from official prosecution. We have some evidence for the development of patron–client ties between tribal groups and Coptic communities under the Fatimids, reminiscent of the patron–client ties that were important to the spread of Islam in earlier centuries. An instructive example is Ibn Ḥawqal’s late-tenth-century account of the region of the Wāḥāt, the Western Oases, which was under the control of the Āl ῾Abdūn, a section of the Lawāta Berbers (Ibn Ḥawqal 1938–39, 105–6). Āl ῾Abdūn collected land-taxes and poll tax alongside the central government, and welcomed the Banū Hilāl who came to the region during the harvest. Ibn Ḥawqal reports that the area as a whole had been originally Christian; in Farfarūn (modern Farfara) and in other localities the population was still majority Copt, even if Friday mosques and village mosques were a common sight. Most significantly for our purpose here, Ibn Ḥawqal states that all the Copts that live in the Oases are clients of the ῾Abdūn in the same way a manumitted slave is a client of his ex-master (wa-jamī῾ man bi-hā min al-qibṭ fī walā᾿ āl ῾Abdūn ῾attāqatan). Ibn Ḥawqal equates here the patronage offered by the dominant Berber clan to the local Christian community with the formal ties of patronage established between a master and his former slave. In Ibn Ḥawqal’s account, although the Copts in the Oases were neither slaves nor converts, they were bound to the tribal patrons by similar patronage ties. Presumably, this made the tribesmen residual heirs and guardians upon the marriage of their Coptic clients, but also meant they had formal responsibilities for protecting the Christians, including the payment of blood-money in cases of homicide. One unique document from the Fayum itself—one of the latest to come to us from the Fatimid period— shows another example of patronage ties. The document, which can be dated to 461 AH/AD 1069 or shortly before that, records a sale of one feddan (6,368 sq. m.) of arable land in the village of Ṭanshā, in the Fayum (Grohmann and Khoury 1993, no. 88). The

buyer is identified as Yūsuf ibn Khayyāṭ, the client (mawlā) of Jirād ibn Ghaylān al-Rab῾ī; and the seller is Ziyād ibn Muslim ibn Qaṭrān, the client of Rāfi῾ ibn ῾Abdallāh al-Rab῾ī. Thus both buyer and seller are clients of different members of the al-Rab῾ī tribal group, i.e., the Banū Rabī῾a. The formal patronage ties here probably emerged from the paid protection (khafāra) imposed by members of the Banū Rabī῾a on Fayumi village communities in the previous decades, well attested in the documentary evidence (Rāghib 1995). The ties of patronage in this document are similar to those recorded by Ibn Ḥawqal in his account of the Oases. Here, both clients have Muslim names. But they both lived in the village of Ṭanshā or Tanchoiris, a well-attested settlement in the Graeco-Roman period. It would make no sense to assume that the original Coptic peasantry deserted Ṭanshā / Tanchoiris, and that settling Arabs brought with them their own clients to cultivate the land. A far more plausible explanation, and one in keeping with Ibn Ḥawqal’s account of patronage in the Oases, is that both Yūsuf ibn Khayyāṭ and Ziyād ibn Muslim ibn Qaṭrān were converts, or direct descendants of converts, and that their conversion to Islam was closely linked to their status as clients of tribesmen of the Banū Rabī῾a. Thus, Copts and Arab tribesmen were tied by relations of patronage already by the end of the eleventh century. It is possible that the concurrent mass tribalization and conversion that followed in the next two centuries was simply an extension of this patronage system. The process may well have accelerated following the deep economic and political crisis of 1068–73, under the reign of the Fatimid caliph al-Mustanṣir, when state control slipped away (as suggested in Lev 2015). As we have noted, some villages on the edge of the Fayum were deserted at the time, including the village of Ṭalīt (Keenan 1999). Elsewhere in Egypt, the crisis hit Coptic communities hard. The chaos in the countryside reached the person of the Coptic patriarch, who was caught by rebel troops in his provincial house in the Delta (Swanson 2010, 62). The History of the patriarchs also reports that the Berber Lawāta tribes took control of the rural economy. They sowed as they wished, without paying land tax, and prevented peasant communities from cultivating fields. They also intentionally avoided the maintenance of irrigation works in order to drive up prices (Severus ibn al-Muqaffa῾ et al. 1943–, 2: 203–4). While the report is likely to be exaggerated, it conveys the sense of overpowering tribal presence that came with the collapse of the central government.

WHERE DID THE CHRISTIANS GO?

211

Conclusion The most plausible explanation for the concurrent Islamization and tribalization of the Fayum in the middle Islamic period, from the eleventh century to the thirteenth, is the construction of tribal identities by new converts to Islam (Fig. 5). Most Christian peasants never left the Fayum or their villages, but stayed on under a new identity, that of Arab Muslim tribesmen. By its very nature, this had to be a collective process, in which village communities of Christian Copts realigned themselves with a tribal group, and simultaneously or consequently accepted Islam. At first, it is likely that the new converts-cum-tribesmen had a lower status vis-à-vis more established tribal elites. Eventually, as in all tribal groups arising in illiterate contexts, the memory of the actual formation of the tribe was suppressed, and replaced with nomadic myths of origins. Successful conversions must be followed by integration. Being clients of tribesmen, and eventually becoming members of a tribe, afforded village communities the protection of a wider tribal confederacy, and the right to carry arms. It integrated villages into the networks of water distribution, which were often linked to clan structures. At the level of the individual household, membership in a clan meant access to arable land, which was held communally and collectively taxed. We have seen that grain-producing peasants did not have the cash or the resources to pay the increased poll tax, and therefore could not remain Christian. Membership in a tribe, on the other hand, allowed the convert a share in the village arable lands. The two identities reinforced each other: one could be a member of a tribe only if one were Muslim. Conversion is surely one of the more elusive questions facing historians of the Islamic world. This is partly because we tend to forget that converts try to cover their tracks in order to gain acceptance and to make their act of conversion socially meaningful. Names of ancestors can be retrospectively Arabicized and Islamicized. The arrival of a Sufi shaykh can conveniently coincide with

Bibliography Abbott, N. 1937. The monasteries of the Fayyūm. The American Journal of Semitic Languages and Literatures 53 (2): 73–96, (3): 158–79. Abū al-Makārim. 1969. The churches and monasteries of Egypt and some neighbouring countries, attributed to

Fig. 5: Photograph by K. A. C. Creswell, captioned ‘Madinat al-Fayyum (El Medineh): Mosque of Princess Asal Bay. Entrance’. The mosque was built in 1498 and named after the wife of Sultan Qaytbay. Gelatin silver print, taken 1916 to 1921 (Courtesy of the Victoria and Albert Museum).

the miraculous disappearance of a Christian community, giving a spiritual lineage to recent converts (Luz 2002). The most potent fiction, however, was, and still is, a claim to lineage, and particularly Arab lineage. Myths of tribal migration provided Islamic legitimacy and prestige in a way that no other fictive identity could. They still do: as el-Leithy (2004) puts it, today the illusion persists that Egyptian Muslims are all of Arab stock, rather than converted Copts.

Abū Ṣāliḥ al-Armanī. Translated by B. T. A. Evetts and A. J. Butler. Oxford. Alschech, E. 2003. Islamic law, practice and legal doctrine: Exempting the poor from the jizya under the Ayyubids (1171–1250). Islamic Law and Society 10 (3): 348–75. Barth, F. 1969. Ethnic groups and boundaries: The social organization of culture difference. London.

212

Y. RAPOPORT

Bernards, M. and J. A. Nawas. 2005. Patronate and patronage in early and classical Islam. Leiden. Bianquis, T. 1986. Damas et la Syrie sous la domination fatimide (359–468/969–1076): Essai d’interprétation de chroniques arabes médiévales. Damas. Brett, M. 1995. The way of the nomad. Bulletin of the School of Oriental and African Studies 58: 251–69. ———. 2005. Population and conversion to Islam in Egypt in the Mediaeval period. In Egypt and Syria in the Fatimid, Ayyubid and Mamluk eras IV: Proceedings of the 9th and 10th international colloquium, Katholieke Universiteit Leuven, May 2000 and 2001, U. Vermeulen and J. Van Steenbergen (eds), 1–32. Leuven. Bulliet, R. 2009. Cotton, climate and camels in early Islamic Iran: A moment in world history. New York. Ellenblum, R. 2012. The collapse of the Eastern Mediterranean: Climate change and the decline of the East, 950–1072. Cambridge. Fierro, M. 2014. Ways of communicating with the past: Genealogies in Nasrid Granada. In Genealogy and knowledge in Muslim societies: Understanding the past, S. Bowen Savant and E. de Felipe Rodriguez (eds), 71–88. Edinburgh. Franz, K. 2011. The Bedouin in history or Bedouin history? Nomadic Peoples 15: 11–53. Gaubert, C. and J.-M. Mouton. 2014. Hommes et villages du Fayyoum dans la documentation papyrologique arabe (Xe–XIe siècles). Cairo. Goitein, S. D. 1967–99. A Mediterranean society: The Jewish communities of the Arab world as portrayed in the documents of the Cairo Geniza. 6 vols. Berkeley, CA. Grohmann, A. and R. G. Khoury. 1993. Chrestomathie de papyrologie arabe: Documents relatifs à la vie privée, sociale et administrative dans les premiers siècles islamiques. Leiden. Ibn Ḥawqal. 1938–39. Opus geographorum auctore Ibn Ḥauḳal (Kitāb Ṣūrat al-arḍ), J. H. Kramers (ed.). 2nd ed. 2 vols. Bibliotheca Geographorum Arabicorum 2. Leiden. Keenan, J. G. 1999. Fayyum agriculture at the end of the Ayyubid era: Nabulsi’s survey. In Agriculture in Egypt from pharaonic to modern times, A. K. Bowman and E. Rogan (eds), 287–99. Proceedings of the British Academy 96. Oxford. Krol, A. A. 2015. The ‘disappearing’ Copts of Fayyūm. In And the earth is joyous…: Studies in honour of Galina A. Belova, S. Ivanov and M. Tolmacheva (eds), 142–62. Moscow. Lapidus, I. 1972. The conversion of Egypt to Islam. Israel Oriental Studies 2: 148–262. Leder, S. 2015. Towards a historical semantic of the Bedouin, seventh to fifteenth centuries: A survey. Der Islam: Journal of the History and Culture of the Middle East 92 (1): 85–123.

el-Leithy, T. 2004. Coptic culture and conversion in Medieval Cairo, 1293–1524 AD. PhD dissertation, Princeton University. Lev, Y. 2012. Coptic rebellions and the Islamization of Medieval Egypt (8th–10th century): Medieval and modern perceptions. Jerusalem Studies in Arabic and Islam 39: 303–44. ———. 2015. The Fatimid caliphs, the Copts and the Coptic Church. Medieval Encounters 21: 390–410. Levtzion, N. 1990. Conversion to Islam in Syria and Palestine and the survival of Christian communities. In Conversion and continuity: Indigenous Christian communities in Islamic lands, eighth to eighteenth centuries, M. Gervers and R. J. Bikhazi (eds), 289–311. Toronto. Little, D. P. 1976. Coptic conversion to Islam under the Baḥrī Mamlūks, 692–755/1293–1354. Bulletin of the School of Oriental and African Studies 39: 552–69. Livingstone, D. 2014. Fāṭimid subjects in light of documentary sources: An integrated view of the life in the Egyptian valley beyond political change. Master’s thesis, University of Oxford. Luz, N. 2002. Aspects of Islamization of space and society in Mamluk Jerusalem and its hinterland. Mamlūk Studies Review 6: 133–54. al-Maqrīzī. 1934–1972. Kitāb al-sulūk li-ma῾rifat duwal al-mulūk, Muḥammad Musṭafā Ziyāda and Sa῾īd ῾Abd al-Fattāḥ ῾Āshūr (eds). 4 vols. Cairo. ———. 2002. Al-Mawāꜥiẓ wa-al-iꜥtibār fī dhikr al-khiṭaṭ wa-al-āthār, Ayman Fuʼad Sayyid (ed.). 4 vols in 5. London. Michel, N. 2000. Devoirs fiscaux et droits fonciers: La condition des fellahs égyptiens (13e–16e siècles). Journal of the Economic and Social History of the Orient 43 (4): 521–78. ———. 2002. Villages désertés, terres en friche et reconstruction rurale en Égypte au début de l’époque ottomane. Annales Islamologiques 36: 197–251. Mouton, J.-M. 2003. L’islamisation de l’Égypte au moyen âge. In Chrétiens du monde arabe, B. Heyberger (ed.), 110–26. Paris. al-Musabbiḥī. 1978. Tome quarantième de la Chronique d’Égypte de Musabbihi, A. Fu᾿ād Sayyid, Ḥ. Naṣṣār and T. Bianquis (eds). Cairo. al-Nābulusī. 2018. The villages of the Fayyum: Thirteenthcentury register of rural, Islamic Egypt. Edited and translated by Y. Rapoport and I. Shahar. The Medieval Countryside 18. Turnhout. Norris, H. T. 1972. Saharan myth and saga. Oxford. O’Sullivan, S. 2006. Coptic conversion and the Islamization of Egypt. Mamlūk Studies Review 10 (2): 65–79. al-Qalqashandī. 1913–18. Ṣubḥ al-a῾shá. 14 vols. Cairo. ———. 1980. Nihāyat al-arab fī ma῾rifat ansāb al-῾arab. Beirut.

WHERE DID THE CHRISTIANS GO?

Rāghib, Y. 1995. Archives d’un gardien du Qalamūn. Annales Islamologiques 29: 25–57. Rapoport, Y. 2018. Rural economy and tribal society in Islamic Egypt: A study of al-Nābulusī’s Villages of the Fayyum. The Medieval Countryside 19. Turnhout. Rapoport, Y. and I. Shahar. 2012. Irrigation in Medieval Islamic Fayyum: Local control in a large scale hydraulic system. Journal of the Economic and Social History of the Orient 55: 1–31. Sabra, A. 2000. Poverty and charity in Medieval Islam: Mamluk Egypt, 1250–1517. Cambridge; New York. Severus ibn al-Muqaffa῾ et al. 1943–. History of the patriarchs of the Egyptian church: Known as the History of the Holy Church. Translated and annotated by Yassā ῾Abd al-Masīḥ, O. H. E. Burmester, A. Khater and A. S. Atiya. 4 vols. Cairo.

213

Sowayan, S. A. 1996. ꜥbadw and ḥaḍar: An alternative to the Khaldunian model. In The anthropology of tribal and peasant pastoral societies, U. Fabietti and P. Salzman (eds), 367–75. Pavia. Swanson, M. 2010. The Coptic papacy in Islamic Egypt (641–1517). Cairo; New York. Valensi, L. 1985. Tunisian peasants in the eighteenth and nineteenth centuries. Cambridge; Paris. Vryonis, S. 1971. The decline of Medieval Hellenism in Asia Minor and the process of Islamization from the eleventh through the fifteenth century. Berkeley, CA; London. Walker, B. J. 2013. Islamization of central Jordan in the 7th–9th centuries: Lessons learned from Hisban. Jerusalem Studies in Arabic and Islam 40: 143–76. Webb, P. A. 2017. Imagining the Arabs: Arab identity and the rise of Islam. Edinburgh.

IV CUSTOMS AND APPEARANCE

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT: CONTINUITY AND CHANGE Cäcilia FLUCK

Preserving the body intact for the afterlife was essential in ancient Egyptian belief, independent of social status. The practice continued when Egypt was under the rule of the Ptolemies from 332 to 30 BC, and later under the Romans, who—apart from two temporary interruptions by regimes from Palmyra in the third century and Sasanian Persia at the beginning of the seventh century—governed the country from 30 BC until the arrival of the Arabs from AD 639. In the early fourth century the imperial family adopted Christianity and, between the fourth and the seventh century, the majority of people in Egypt became Christian. Based on case studies this chapter aims to explore if, and how, burial practices changed at a time when Egypt shifted from polytheism to monotheism in the context of a newly Christian empire. Since earlier cemeteries were usually reused, the variety of graves (from mausoleums standing above underground chambers to simple pits) remained largely the same; however, the preparation of the body of the deceased slowly changed. Wrapping and unwrapping: Textile culture and burial in Egypt Textiles—until Roman times primarily made of linen—always played an important role in funerary contexts in Egypt. A distinction must be drawn between textiles formerly used during the lifetime of the deceased that were reused as grave goods, and those that were necessary for mummification. The former consist of clothing and household fabrics that were folded, batched or bundled and then put in baskets, boxes or chests which were placed in the tombs (Vogelsang-Eastwood 1995, 41–44; 1999, 13, 15–16; David 2000). The latter were required to wrap the mummy and were in large part produced exclusively for that

purpose.1 They deserve special attention because they are intimately connected to the human body and, therefore, mark the first action in the preparation of the dead for inhumation and for the afterlife. Apart from this, it was also customary to dress (cult) statues and other objects serving as grave goods in linen (Riggs 2014, esp. 19–32). Elaborate mummy wrapping reached its peak in Roman times. Less careful wrapping of the body coincided with the ‘Christianization’ of Egypt. Instead, dressing the dead in reused clothing worn in daily life became a common burial practice that lasted until the first centuries of Arab Muslim rule, when burying the dead in simple funeral dress and shrouds became the norm. The complexity and diversity of the topic provides more than enough material to fill books. Renate Germer’s and Christina Riggs’ publications are exemplary and essential for the role of textiles in funerary rites, but they are also very sensitive historical studies about the treatment and handling of mummies from their original (artificial) creation for burial purposes in the past up to the colonial and post-colonial discourse, the scientific investigation methods, museological aspects and display of the present day (Germer 1991; Riggs 2014; see also Cat. Münster 2016, 30–41). From an archaeological viewpoint each burial site has its own traditions and peculiarities depending on the local circumstances, and each deserves attention individually. Nevertheless, there are some features seemingly typical for each period, which will be demonstrated below by selected examples of Roman and Late Antique/early Christian burials. They show evidence of both continuity and change in the use of fabrics in a funerary context, and they cause us to wonder whether these changes were due to shifts in religious beliefs or in economic circumstances.

1

Torn household textiles and garments were often reused for the inner layers of the wrapping (see Montserrat 1997, 37 with further references).

218

C. FLUCK

Investigation of ancient Egyptian mummies From the beginning of dynastic times, and even earlier, burial practices and rites are among the most significant characteristics of ancient Egyptian culture, an aspect which still fascinates people today. In the Medieval period, Egyptian mummies were already appealing to European travellers and scholars. Since the Egyptian expedition of Napoleon Bonaparte and the subsequent decipherment of the hieroglyphic script in the 1820s, ancient Egypt has held the attention of the European public. Along with an abundance of antiquities, many mummies were transported to Europe to enrich newly installed museum collections. It was the beginning of scientific investigations, at first mainly focused on medical aspects, which attracted not only scholars but the wider public. The unwrapping of mummies became a social event in the mid-nineteenth century, which people paid to see (Germer 1991, 15–26).2 Many of the mummies were destroyed during such events by what is considered today inappropriate handling, but thanks to the policies of prominent public collections a considerable number of mummies survive intact. An important step in the development of nondestructive research was the application of X-ray examination (radiology). This technology had been practised since the end of the nineteenth century, but it was not until the 1960s and 1970s that larger collections of mummies were investigated using this method. With the beginning of the twenty-first century, three-dimensional imaging and computerized tomography (CT) scanning became an increasingly popular technology. This method allows the virtual removal of layers ‘to observe the bandages, skin, muscles, skeleton and internal organs, as well as any objects placed inside the mummies’ (Cat. London 2014, 18). The results are impressive with regard to the gender, age, disease, diet and personal belongings of the individuals, but also reveal embalmer’s practices, as early twenty-first-century research projects and exhibitions have showed (see for instance Cat. Stuttgart 2007; Germer, Kischkewitz

2

3

See also https://www.atlasobscura.com/articles/victorian-partypeople-unrolled-mummies-for-fun; Dimitra Nikolaidou, February 23, 2016 (last accessed July 2018). For a critical remark see Cox 2015; see also Riggs 2014, 207–13.

and Lüning 2009; Cat. London 2014; Huppertz 2017, 33–38). Quite recently the CT analysis of the Late Period female mummy Ta-cheru allowed a facial reconstruction, and for the first time the CT data were used to create a three-dimensional animated sequence in a holographic rendering (Weber, Engels and Reinhart 2018, 48–57; Sokiranski and Pirsig 2018, 34–47; Gauert, Pirsig and Sokiranski 2018, 134–41). Despite the technological progress, non-invasive investigations have their limitations,3 for they do not allow in-depth material and technical analyses of objects such as jewellery, ceramics or toilet objects that are found between the layers, nor of the fabrics themselves. CT technology is becoming more and more sophisticated, but the results always depend on the quality of the instruments, and therefore the anthropological results are not always as clear as might be desired, as will be demonstrated below. Unwrapping corpses is today rare, but nevertheless still practised, in particular during fieldwork, when technical equipment is lacking. In these cases, the excavator must decide between ethical considerations and essential scientific research requirements. Ancient mummification The first evidence of natural mummification due to desiccation in the desert sand comes from Predynastic Egypt as early as the fifth millennium BC. The earliest textile finds in connection with burials also date to this period (Dickey 2019, esp. 142–43 and Table 2 on p. 138). It appears that the phenomenon of natural mummification most probably gave way to artificial mummification, seemingly evidenced around the middle of the fourth millennium at the latest (Taylor 1997, 9; Friedman et al. 1999, esp. 1–18; Jones et al. 2018).4 Artificial mummification was an extremely timeconsuming procedure taking approximately fifty days in total or even longer. After the body was embalmed, it was wrapped in shrouds and bandaged, in one case using up to 375 m² of linen fabrics to wrap a single mummy (Winlock 1940, 257). Only a minority of

4

On the topic in general see Germer 1991 and the overview by von Recklinghausen 2007 with references; Jones et al. 2018 argue that the prehistoric mummies were already artificially mummified.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

individuals could afford an elaborate burial in which the body was mummified, placed in coffin(s), supplied with grave goods and inhumed in a mausoleum. Throughout all periods, the graves of the elite are much better preserved than those of the average citizen, and our present knowledge is based on these. They demonstrate what a ‘top quality’ burial looked like, but they do not mirror the habits of the masses, of which we know much less.5 In the fifth century BC the historian Herodotus mentioned three graded price classes for burials (Herodotus, Histories 2.86–90). The most expensive includes removal of internal organs and wrapping them in linen, while the cheapest allows only a treatment with natron. Most likely the majority of the dead were simply buried in the desert sand without any mummification, and were only occasionally and under very specific conditions naturally mummified. Only rare examples have been preserved until modern times.

219

With the arrival of Alexander the Great in Egypt in 332 BC, and during the reigns of the Ptolemaic kings who succeeded him, the cultural landscape in Egypt changed gradually. Alexandria became the capital and developed into a cosmopolitan city (for a map, see O’Connell, this volume). Greek practices were adopted by Egyptians and vice versa, and this pattern also slowly affected traditional burial practices. Although some of the new inhabitants, more precisely some of the Greeks in Alexandria, chose cremation, for the majority of people in Egypt—independent of their ethnic affiliation—preserving the body remained essential.6 Corpses continued to be wrapped in shrouds and elaborately bandaged. But the removal of the brain and internal organs, which was practised previously in dynastic times, was not systematically performed, and the treatment of the body could even be described as careless (Quack 2007, 22–24; Germer, Kischkewitz and Lüning 2009, 139–42). Grave goods became rarer, but the tradition of adorning the dead with garlands and

bouquets of flowers and leaves (known at least since the Middle Kingdom) remained (Germer 1991, 120– 27; Heilmeyer 2015, 242–43; Helmbold-Doyé forthcoming). The religious necessity of mummification was no longer a priority; instead, it became socially and economically motivated (Quack 2007, 20–21; Zdiarsky 2012, 29). The anthropoid coffins typical throughout the pharaonic era until the Late Period and the stylized masks of Ptolemaic times were gradually given more individual traits in the transition between the Ptolemaic and Roman periods. A good example is a mummy case of a bearded man of Late Ptolemaic or early Roman times (first century BC–first century AD) from Panopolis/Akhmim, now in the British Museum (Fig. 1). During the same period, stucco or cartonnage masks that covered only the head and upper body were also employed (Fig. 2). They were mass-produced from moulds with a base of linen or papyrus and glue. Individualizing details were painted on or added later (Nauerth 2003, esp. 231–38; Riggs 2005, 115–16; Willburger 2007, 234–35). These masks were joined to separately-made breast pieces of wood or stucco on which hands and clothing were painted. It became increasingly common to present the deceased as he or she had appeared in life. Some of the ‘individualized’ mummies were also laid in coffins, often reused (Riggs 2000, 121–44; Nauerth 2003, 237, see also below). Contemporary coffins were still decorated with elements of ancient Egyptian funeral art and texts. Even more than Greeks, Romans emphasized knowing what the deceased looked like (e.g., imagines of the ancestors in Roman Italy). Thus, at the beginning of the first century AD when Egypt was under Roman rule, and inspired by the Roman custom of self-presentation, a new type of mummy mask developed: portraits painted on wood, or occasionally on a linen fabric, and inset into the mummy wrappings. They are rare evidence for prestigious panel painting in antiquity and therefore a popular subject matter for publications and exhibitions.7 Most probably, their production ended about the end of the third century AD, when the ancient

5

7

Burial practices in the Ptolemaic and Roman periods

6

With the exception of the burials of ‘workers’, recently excavated at Amarna (see Stevens 2015). Further reading: Willburger 2007 with further references; Riggs 2005; Helmbold-Doyé forthcoming.

A selection of literature includes: Parlasca 1969–2003; Picton, Quirke and Roberts 1992; Doxiadis 1995; Bierbrier 1997; Cat. London 1997; Borg 1998; Cat. Paris 1998; Cat. Frankfurt 1999; Cat. New York 2000; and Willburger 2007.

220

C. FLUCK

Fig. 2: Cartonnage mask of a woman, first century, from Meir, near Koussa (inv. ÄMP 34434, © Staatliche Museen zu Berlin, Ägyptisches Museum und Papyrussammlung. Photo Sandra Steiß).

Fig. 1: Cartonnage mummy case of a bearded man, first century BC to first century AD, from Panopolis/ Akhmim (BM EA 29584, © The British Museum).

8

K. Parlasca postulated that mummy portrait production ended in the late fourth century, when the emperor Theodosius I prohibited the practice of ‘pagan’ cults: Cat. Frankfurt 1999, 33–37;

cults were almost given up and many of the ancient Egyptian temples were abandoned or converted, as for example in Luxor, where a military camp was installed in the ancient temple (Bagnall 1993, 261–68). A social and economic shift in the population may have been caused by changes in the political system introduced by the emperor Septimius Severus (r. 193–212), who placed all political and financial risk on the local elites, but also by local disasters that led to the abandonment of entire villages, for instance in the Fayum (Walker 1997, 16; Borg 1998, 88–101, esp. 96–101; Michaelis 2012, 151–53).8

summarizing the controversial discussion about chronology: Riggs 2002, 93–95.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

For many of the mummy portraits the find context is lacking. From his investigations in Hawara in the Fayum—the most abundant find-spot for these portraits that we know of—W. M. F. Petrie stated, first, that only 2% of the mummies were fitted out with portraits, and second, that most of them were simply buried without any other grave goods in the desert sand. A few were found in reused ancient Egyptian pit tombs (Petrie 1889; 1911; Borg 1998, 15–17).

221

Following in Petrie’s footsteps, the German lawyer, art collector and patron Richard von Kaufmann (1849– 1908) went to Hawara in the spring of 1892 and discovered a tomb with eight mummies (Borg 1998, 17–21; Germer, Kischkewitz and Lüning 2009, 153– 60; Helmbold-Doyé 2017). The exact location of the tomb in the cemetery of Hawara is not known, nor do any plans exist. According to Kaufmann’s description the tomb consisted of a simple chamber of about 3.50m in length and 2.80m wide, which was completely lined with bricks. Nothing is recorded about any superstructure, the entrance to the tomb or its original depth. J. Helmbold-Doyé (2017, 17) recently suggested that, in accordance with adjacent tombs unearthed by Petrie, it most probably had a vaulted ceiling, with niches in the walls, and must have been accessible. Von Kaufmann found the bodies in three layers one above the other: on top, three corpses of which we know only that they were poorly wrapped in linen shrouds; in the middle, the mummies of a man and a girl with gilded cartonnage masks protecting the upper parts of their bodies; and, at the bottom, the corpses of a woman and two children fitted out with painted portraits (Fig. 3; Helmbold-Doyé 2017, 20 fig. 16). The corpses of the man in the middle layer and the woman at the bottom were unwrapped still in situ. The head of the woman, including its covering fabric and

the portrait, was handed over to Rudolf Virchow, Professor of Pathological Anatomy in Berlin, who examined it and who tried a virtual reconstruction of the woman’s physiognomy (Helmbold-Doyé 2017, 25–27). The head was lost in the aftermath of World War II and the whereabouts of the three upper bodies and the corpses of the man and the woman remain unknown. The three children’s mummies, the woman’s portrait and the man’s gilded cartonnage mask came into the Ägyptisches Museum in Berlin in the year of the discovery together with other grave goods: wreaths of flowers, a pot, some clay seals and an inscribed stela (Helmbold-Doyé 2017, 52, 64–65; Moje 2017, 53–54).9 The latter was found standing at the head end of the woman and it is most likely that it refers to her. The inscription, written in Greek, mentions the name of ‘Aline, alias Tenos, the daughter of Herod’, which gave the grave its name. Aline died aged thirty-five in year 10 of an unnamed emperor, most probably Trajan, which would correspond to the year AD 107.10 Aline’s portrait, the children’s mummies (see Fig. 3), and the man’s mask have been on display in the Neues Museum on the Museum Island in Berlin since 2009.11 The three children’s corpses were bandaged with linen bands of different qualities arranged in an intricate, overlapping, diamond-shaped pattern, the socalled ‘cassette’ technique so typical of the Roman imperial period. Some of the linen bands were dyed red, but others were untreated. The bandages of the two children’s mummies with portraits placed within the tomb next to Aline are additionally adorned with stucco buttons painted in ochre and covered with pieces of fine gold leaf. If we trust Richard von Kaufmann’s records, Aline’s body must have been bandaged in the same manner. He also reported that all mummies carried a wax seal showing an imprint of a gem with a representation of Heracles and the lion. Aline’s body was enveloped in seven layers of a linen cloth of which the end was pulled over her face. This part was painted

9

11

The tomb of Aline: A Roman family burial in Hawara

10

Egyptian Museum Berlin, inv. ÄM 11403 (pot); ÄM 11408– 11410 (seals); ÄM 11415 (inscription, lost). Some scholars argue for an earlier date: see the discussion by Helmbold-Doyé and Moje in Helmbold-Doyé 2017, 60–61). Radiocarbon analyses of the bandages of the children’s mummies were carried out in 2017. The results will be published by J. Helmbold-Doyé.

Egyptian Museum, Berlin, inv. ÄM 11411 (portrait of Aline); inv. ÄM 11414 (gilded cartonnage mask); inv. ÄM 12125 (mummy of the girl with cartonnage mask and painted shroud); inv. ÄM 11412–11413 (children’s mummies with portraits).

222

C. FLUCK

Fig. 3: Mummies of three children from the tomb of Aline in Hawara, probably early second century AD (inv. ÄM 12125, 11412 and 11413 © Staatliche Museen zu Berlin, Ägyptisches Museum und Papyrussammlung. Photo Sandra Steiß).

with her portrait, and the same is true for the two children’s portraits. This proves that they were made after death and not during the individuals’ lifetime, as the portraits painted on wood are commonly thought to have been.12 The portraits are executed in tempera technique exhibiting excellent manual skill and without using preliminary sketches, most probably by one and the same painter.

12

See the article by Montserrat (1997, 37), who puts forward the idea of portrait production after the death of the individual.

Aline is portrayed as a well-fed woman with brown almond-shaped eyes, strong dark eyebrows, a luscious mouth and rosy cheeks. Her hair is parted in the middle, laid in waves with undulating curls on the forehead and gathered at the back. She wears a white tunic with pink clavi, golden earrings with rose pearls and a golden necklace with a series of pendants consisting of three beads each.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Both children are depicted in their portraits wearing crowns with gilded laurel wreaths, typical attributes for mummy portraits in general, and a further indication that they were produced post-mortem. The elder child is a girl with the same hairdo as her mother Aline. She is wearing a brown tunic with white clavi, a golden chain with longitudinal beads and golden earrings in the form of small discs. The younger child, dressed in a mauve-coloured tunic, was thought to be a girl, too (Borg 1998, 18, 20–21, 52; Fluck and Finneiser 2009, 9). The child has dark hair falling in curly strands down to the shoulders and is adorned with a neckband with a crescent-shaped pendant, the so-called lunula. Usually, mauve-coloured clothing and lunula pendants are attributes for girls. However, CT analyses carried out in 2016 suggest that the little body is that of a boy, three years old at most (Huppertz 2017, 33, 36; Helmbold-Doyé 2017, 50–51), but lingering doubts still remain. According to the results of the CT analyses the elder child was about four years old, and the child’s mummy with a mask, again a girl, died at the age of six or seven. The internal organs were removed from all three mummies, the empty space in the abdominal cavity was filled with linen rags and resinous anointing oil, and all corpses show post-mortem injuries at the spine. The brain was removed only from the girl with the mask (Huppertz 2017, 33–38). It is striking that the mask covering the head and breast represents a woman instead of a girl. Her face and arms are gilded, but the hair, with little curls at the temples, is dark. She is wearing a white tunic with dark clavi and a mauvecoloured veil around the head and covering the shoulders. Her jewellery consists of convex earrings, a necklace with a lunula-pendant, bracelets in the form of double-headed snakes on both forearms, simple bangles at the upper arms and a ring on the little finger of the left hand. The arms are laid in front of the breast, and the right hand holds a garland imitating the petals of a rose. The lower edge of the mask is decorated with alternating rosettes and stripes while a vulture with spread wings, symbolizing the deity Nut, is painted on its top and back (Helmbold-Doyé 2017, figs 58–59). Remnants of another multicoloured foot-cover that once sheltered the girl’s feet and ankles are still preserved beneath the lower bandages. The painted feet and some ornaments are still visible. Originally, a lavishly painted linen shroud covered the girl’s mummy showing representations from the traditional ancient Egyptian mortuary cult in four horizontal registers (Helmbold-Doyé 2017, figs 60–64). For

223

conservation reasons it was removed from the body and is stored separately. From the mummy of the man in the family tomb only the cartonnage mask survives (Helmbold-Doyé 2017, figs 35–40). Von Kaufmann did not mention any details about his mummification. The mask is covered with gilded stucco. It represents a bearded man with short hair in two rows of short strands brushed on the forehead. The eyes were inlaid with white and black stone and the lashes were carefully cut out of bronze sheet. His arms are crossed upon the breast. Like the girl who once lay next to him, and who was probably his daughter, he is holding a garland of rose petals. On the fourth finger of his left hand he is wearing a seal ring. The man is dressed in a tunic and a mantle that covers his head and upper body. On top of the head the gilded folds of the mantle pass into a painted part with a simple red stripe followed by a row of lotus flowers and circles. This part is so unusual and unparalleled that its authenticity may be doubted (HelmboldDoyé 2017, 31–32). The mummies with their portraits and masks in the family tomb of Aline represent an upper-class burial in the cemetery of Hawara, most probably from the early second century AD. With the foundation of Antinoopolis in Middle Egypt by the emperor Hadrian in the year 130, many of the inhabitants moved from the Fayum to the new city, and they brought their cultural heritage with them. Consequently, Antinoopolis became another place of provenance of larger quantities of masks and mummy portraits. Others were found, for example, in Thebes and Panopolis/Akhmim in Upper Egypt, and at Marina el-Alamein, 100km west of Alexandria. A peculiar Roman mummy from Thebes While the mummies from Hawara—how they were wrapped, their portraits and masks—demonstrate what is considered a ‘normal’ burial for wealthy people in the Roman period, a contemporary mummy said to be from Thebes shows some unusual features (Fig. 4). It comes from the collection of the British consul Henry Salt, and arrived in the British Museum in 1835. It is one of the mummies that were investigated and published as part of the British Museum’s CT project linked to the exhibition Ancient Lives in 2014 (Cat. London 2014, 134–49). The first–third century AD mummy is that of an unidentified man buried in the reused coffin of a woman called Mutemmenu from

C. FLUCK

224

c. 1250 BC.13 His body was not given the shape of a bundle, like the majority of mummies from Egypt; rather, the head—with the exception of the skullcap— the torso and extremities, and even the fingers and toes were wrapped separately with plain linen cloths. Only the forearms show the typical lozenge pattern known from other mummies. Diagonally arranged bands cross the breast and the thighs. It appears that the man is wearing a sort of loincloth. Even the accessories—sandals, bracelets, anklets and a belt—are imitations made from textiles. Details of the face are modelled from linen and painted, as are the nipples. A lotus flower (reminiscent of a tattoo) is painted on each knee. The images obtained by CT scanning and the surface of the wrappings show traces of gilding on the toes, fingers and partly on the torso. Hollows were filled with rags. The chest and abdomen have been packed with a homogeneous mass which probably consists of sand, mud and resin. Between these masses, cylindrical bundles can be seen which might contain the organs. The bones inside the nose are broken, an indication that the brain was removed during embalming. Such lifelike outer treatment of the body was rarely practised in the Roman period, but it is not unique. Attempts are known from a much earlier period, in the Old Kingdom, but with virtually no other examples until the Roman period.14 Shrouds with life-size portraits

Fig. 4: Mummy of a man, first to third centuries, said to be from Thebes (BM EA 6704. © The British Museum).

13

14

The mummy was not necessarily buried in the coffin but could have been transferred into it after discovery. Other mummies, probably from same group, are known in other collections, for instance in the Museum van Oudheden in

Another way to personalize the visual appearance of a mummy was to cover the body with a painted shroud made of linen that displayed a nearly life-size portrait of the deceased. A considerable number of such shrouds is preserved from the Roman period (see, for instance, Riggs 2005, 157–60 and pls 2, 6–9, 12). They served as an additional cover for the mummy, but it also seems that they replaced coffins or masks. On these shrouds the deceased, portrayed in the shape of an Osiris or dressed in a typical Roman tunic with longitudinal stripes running from the shoulder downwards and a mantle draped around his or her body, is often depicted between the gods Osiris and Anubis, who guide the dead into the netherworld, and surrounded by traditional scenes from the ancient Egyptian cult of the dead (Fig. 5).

Leiden: see Raven and Taconis 2005, 191–203, nos 29–31 and 222–23, no. 43. For Old Kingdom mummies, see Cat. London 2014.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Fig. 5: Shroud, second century, from Saqqara near Memphis (inv. ÄM 11651, © Staatliche Museen zu Berlin, Ägyptisches Museum und Papyrussammlung. Photo Sandra Steiß).

225

226

C. FLUCK

From the embalmer to the grave As in dynastic Egypt, quite a long period could elapse between the time of embalming and the actual burial. The first-century BC historian Diodorus assumed that the mummified corpses were temporarily kept in the houses of their families (Diodorus, Bibliotheca Historica 1.92.6). However, it is most likely that mummies awaiting burial were put into chapels in the area of the cemeteries for interim storage, where the bereaved could visit them (Montserrat 1997, 38–40; Lembke 2004, 57–58; 2007, 26–27 with further references). To avoid confusion during transport to the chosen burial site, little mummy labels were attached to the body with the name of the deceased, his or her profession, home town and age. Sometimes these labels also contained small messages to the undertaker to ensure that the funeral went smoothly. For example, on one label, attached to the mummy of Maronas from the village of Philadelphia, today in the Kunsthistorisches Museum in Vienna, the message asks the undertaker to ‘hand him over to Anthestates!’ and to ‘bring him to Kerke!’ (Harrauer 2003, 27–28).15 Such mummy labels are attested until the early fourth century, though a few exceptions inscribed with Christian formulae are dated to the fifth–sixth centuries (Torallas Tovar 2013).

In Alexandria, for instance, the graves consist of underground chambers, hypogea, or shafts. At other places they are carved into rock. Mostly, however, people were buried in simple pits, Latin fossae, in the sandy desert ground, sometimes covered with small brick vaults, stone slabs or cairns. Significant Roman grave sites include Kom el-Shogafa at Alexandria, Terenuthis/Kom Abu Billu in the Western Nile Delta, Tuna el-Gebel, one of the cemeteries of Hermopolis, and the cemeteries of Antinoopolis in Middle Egypt and Kysis/Dush in Kharga Oasis. The design of the tombs is usually a mixture of Roman stylistic elements and traditional Egyptian representations of the death and resurrection of Osiris, but in the necropolis of Tuna el-Gebel there are also purely pharaonic-style tombs side by side with purely Roman-style tombs (Lembke 2004, esp. 59–62; Lembke and Prell 2015). Most of the tombstones show again an assimilation of traditional Egyptian with Graeco-Roman elements, such as a group of stelae from Terenuthis, a village in the Western Delta (Fig. 6; Abd el-Al, Grenier and Wagner 1985). The deceased, facing the viewer and dressed in Greek clothing (chiton and himation), is resting on a kline, a sort of couch that was also used for taking meals by the Greeks and Romans. The ancient Egyptian gods related to funeral rites, often also rendered in frontal view, accompany him.

Cemeteries in the Roman period In general, cemeteries from previous eras located outside settlements were reused for inhumation. Their layout and equipment depended firstly on local geological conditions and secondly on social hierarchy. Some burials are found in mausoleums or chapels, as in the Roman necropolis of Hermopolis at Tuna elGebel in Middle Egypt. In several of the chapels a baldachin was installed with imitations of textile hangings painted on the back wall. One particularly interesting example is one of the chapels, unfortunately destroyed today, but which archival photographs show. In front of the baldachin stands a sort of funerary bed, a kline; a blanket with a lozenge pattern is painted on its surface and on this bed lies the mummy of an individual (Lembke 2007, 30–31, fig. 6).

15

Kunsthistorisches Museum, Vienna, inv. 8024.

Jewish burials Also dated to the Roman period is one of the few well-known Jewish cemeteries of ancient Egypt. It is located in Leontopolis/Tell el-Yahudiya, about 20km northeast of Babylon/Cairo. Here the Jewish high priest Onias IV is said to have founded, around 170 BC, the only temple to Yahweh outside Jerusalem. It existed until the Jewish Revolt in AD 71 (Naville 1890; Petrie 1906, 19–27; Capponi 2007). The temple has not been archaeologically identified, but the existence of Jews in this place is attested by grave inscriptions from the first century BC to the first century AD. Some of them are now in the Egyptian Museum in Cairo, others in the Départment des antiquités égyptiennes of the Louvre in Paris (Griffith 1890, esp. 51–53, pls III–IV, XVI

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Fig. 6: Stela from Terenuthis/Kom Abu Billu in the Western Delta, first to second centuries (inv. ÄM 24142, © Staatliche Museen zu Berlin, Ägyptisches Museum und Papyrussammlung. Photo Sandra Steiß).

227

the departed. The lower part is not decorated and was evidently anchored in the ground. According to the excavator F. Ll. Griffith, at Tell el-Yahudiya there were three adjacent cemeteries with hundreds of Jewish graves of different types: square pits; pits about 3m deep, opening into an elongated cavity; rectangular tombs, carefully cut into the rock; tombs with a groove for a locking plate; and subterranean, mostly undecorated chambers with symmetrical niches, so-called loculi, also familiar from Jerusalem, Jericho and Qumran. Elsewhere there will also have been Jewish cemeteries wherever Jewish communities were established.16 It is probably from such locations that various ossuaries and smaller objects come, such as lamps with depictions of the menorah, grapes and palm branches, now at the Graeco-Roman Museum in Alexandria and the Egyptian Museum in Cairo. They suggest that in the Roman period the Jews in Egypt were buried in accordance with their own contemporary tradition (Fig. 7).17 Their dead were wrapped in linen, occasionally equipped with grave goods, and initially buried in a family grave. After some time, the bones were collected and placed in an ossuary, which was then kept in a separate area in the family tomb (Hachlili 2005, 447–516; Popović and Zangenberg 2013, 88–89).18 Burial practices in Late Antique Egypt

[below]; Leibovitch 1942, 41–47; Bernand 1992, 116– 19, pls 42–43; Horbury and Noy 1992, esp. 51–182, nos 29–105; Hachlili 1998, 263–64). They are characterized by a gable design with a suggestion of a pediment on the upper part and below it a recessed rectangular field with an inscription in Greek. This contains the name of the deceased, typical funerary formulae, the date of death and, also occasionally, brief lamentations or exhortations to the bereaved to mourn for

The third century was characterized by political crisis and economic depression, which nearly led to the collapse of the Roman empire. It consequently affected the entire commonwealth, including Egypt, and all areas of life of its people. In this period of instability, Christian belief gradually evolved and brought new ideas about resurrection and the afterlife. The imperial crisis that caused a shift in the structure of society, in combination with the development of a new religion with a spectacular transition from pantheism to monotheism, might have been the reasons for a modification

16

17

Jewish cemeteries of the Ptolemaic and early Roman imperial periods have been found in Alexandria: Goodenough 1953, 62–63; Hachlili 2005, 264. Jews again settled in Oxyrhynchus, for example, in the third century, following the conquest of Egypt by Queen Zenobia of the Palmyrene empire (c. 240– 272/73). They left written documents in Hebrew and apparently had close ties to Palestine; see for instance Turner 2007, 144– 45; Bowman 2007, 177, 180. For Jewish survival in Late Antique Egypt see Bagnall 1993, 275–78, and Sijpestein, this volume.

18

In the Jewish scriptures, life after death is not mentioned. The focus, rather, is a life pleasing to God. But Persian–Greek beliefs about the beyond were adapted by certain Jewish circles. On ossuaries, Jewish grave stones and grave goods from Egypt, see Goodenough 1953, figs 113, 863, 864, 923–26, 931, 935 (erroneously located in Rome, Berlin Obj. 188 in this volume), 936, 937, 959; on Jewish inscriptions from Egypt: Horbury and Noy 1992.

228

C. FLUCK

Fig. 7: Ossuary of Nicandor the Alexandrian, from Jerusalem, first century (BM 1903,0715.1, © The Trustees of the British Museum).

of burial customs for the people of Egypt (Borg 1998, 88–101). Some of the early church fathers—first and foremost Shenoute (d. 466), the famous abbot and founder of monasteries in the Akhmim/Panopolis region (van der Vliet 1993)—vehemently objected to pagan cults. However, as far as burial rites were concerned, everything that did not contradict Christian teaching was retained. It is said of Saint Antony, the father of monks, that he did not wish to be put on a rack and kept at home like the Egyptians after death; instead he wanted to be buried in a pit in a secret place where he would stay until the day of the resurrection of the dead (Vita Antonii 1917, chs 90–91). Thus, this tradition that includes wrapping the corpses in linen cloths before being displayed was still common practice in his time and adopted in the Christian martyr’s cult. But Antony refused to do this, because he wanted to avoid a personal cult around his shrine (Baumeister 1971, 406). Even three hundred years later, the old customs were still well known, as can be seen from a nice little anecdote about Pisentios, sixth/seventh century bishop of Koptos, written down by his pupil John (Budge 1913, xlviii; O’Connell 2007, 260–62; Torallas Tovar 2017). During the Persian invasion at the beginning of the seventh century, Pisentios took refuge in a dynastic

tomb that was reused as a common burial-place in Graeco-Roman times. One day John witnessed a conversation between his master and one of the mummies whose soul could not find rest in the hereafter. Although the dialogue aims to convey Christian beliefs about purgatory and the salvation of the soul, it clearly demonstrates that the Christians were familiar with the material culture of the past. One of the mummies is described as being bandaged with garments made of the pure silk of the kings with the toes and fingers wrapped separately (cf. Fig. 4). Preserving the body and supplying its needs for the afterlife remained essential at a time when most of the inhabitants were Christians. Last wills and testaments written during this period survive and they contain arrangements to ensure the execution of the funeral with all its connected rites. Bishop Abraham from Hermonthis/Armant (Fig. 8), who also lived in the sixth/ seventh century, for instance, gave clear instructions in his will to obtain the bands (or ‘ribbons’) and shrouds necessary for his funeral, and instructed his heir Victor to bury him according to the customs of the country for the funeral feast and the lamentation of the dead (Krause 1983, 87, 89; 2003, 34–35, 38–39; Grossmann 2002, 315–16). We have to take into consideration that in periods of conversion not all people

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Fig. 8: Portrait of Bishop Abraham of Hermonthis, seventh to eighth centuries (MBK inv. 6114, © Staatliche Museen zu Berlin, Skulpturensammlung und Museum für Byzantinische Kunst. Photo Markus Hilbich).

in Egypt adopted the new religion, but that a part of the population always maintained traditional worship. This applies both to the time when Christianity replaced the belief in the old gods and to the time after the Muslims took control of the country, when a part of the population retained the Christian faith. The burial places remained the same for coexisting religions (Gessler-Löhr 2010, 311; Torallas Tovar 2014, 132–40 with further references).

229

corpses. While in Graeco-Roman times the deceased lay with their heads to the west, where the next world was thought to be located, in Christian times the dead were buried with their heads to the east (Gessler-Löhr 2010, 311). But this cannot be taken as a rule. It is almost impossible to distinguish between the grave of a Christian and that of a ‘non’-Christian, unless they contain such significant symbolic features as crosses, depictions from the Bible or inscriptions with invocations to Jesus Christ. One textile cache from a tomb of unknown provenance of the fourth/fifth century, now in the Abegg-Foundation in Riggisberg, impressively proves the close connection between ‘pagan’ and Christian ideas: attached to the fragments of a large hanging representing Dionysiac initiation rites, a silk fabric with scenes from the life of Mary and the infancy of Jesus Christ was found (McNally 2002; Schrenk 2004, 26–34 no. 1 and 186–89 no. 62 with further references; Willers and Niekamp 2015). Another example is a fragment of a tunic from the Museum of Byzantine Art in Berlin, showing a depiction of a pendant cross hanging down from a panel with nude dancers that are usually associated with the god Dionysos (Fig. 9). The heritage of classical antiquity and Christian belief did not contradict one another; on the contrary, they merged with each other on the same garment. Only the cemeteries that were connected to churches or monastic communities can be regarded as purely Christian, for instance in Kom el-Ahmar/Sharuna near ancient Ankyronpolis or in Naqlun at the entrance to the Fayum (Huber 2004; 2006; 2007; 2009; Godłewski 2005; Dzierzbicka and Ożarek 2012). Burials of monks, mainly from the Theban region, can be clearly identified by their habit, consisting of a plain tunic, a hood and the melote, a goat-skin in the form of a long strap ending in three tongues, whose function has not yet been clarified (Winlock and Crum 1926; Castel 1979; Paetz gen. Schieck 2012, 235–39; Tatz 2017; Fluck 2020, 109–12). Modified methods

Identifying Christians and their burials B. Gessler-Löhr has indicated that a possible characteristic for distinguishing between a traditional and a Christian burial could lie in the positioning of the

By the end of the third/beginning of the fourth century AD the elaborate mummification of the Roman period gave way to ever more practical and inexpensive methods.19 The bodies were no longer eviscerated, and 19

Further reading: Krause 1983; Horak 1995; Davies 1999; Grossmann 2002, 315–47; Gessler-Löhr et al. 2007; GesslerLöhr 2012; Dunand 2007; Förster 2008; Torallas Tovar 2014.

230

C. FLUCK

Fig. 9: Tunic fragment with a large pendant cross in combination with motifs from the retinue of the god Dionysos, probably Akhmim, sixth to seventh centuries (MBK inv. 9205, © Staatliche Museen zu Berlin, Skulpturensammlung und Museum für Byzantinische Kunst. Photo Stefan Büchner).

also a change in preparing the corpses for the funeral can be observed. Portraying the dead went out of fashion by the end of the third century. Their bodies were washed, treated with a natron mixture and occasionally dusted or sprinkled with fragrant essences, such as incense, oils or resins, as in previous centuries. The corpses were wrapped in shrouds, and body cavities were stuffed with rags. Discarded clothing was also occasionally used for stuffing. The shrouds were usually made of a plain weave linen in a rectangular shape with self-bands on the long sides while the short sides ended in fringes.20 Frequently, the corpses were stabilized with a board or simple slats under the back, the shape of which often followed the outline of the human body (Fig. 10). Sometimes they were laid on biers made from vertically and horizontally arranged palm ribs (gerid) bound together with cords of palm bast or halfa-grass (Fig. 11; Huber 2009, figs 3 and 19). These

20

Recent excavations in the cemeteries of Qarara, dated from the fifth/sixth and seventh/eighth centuries, and Fag el-Gamous, dated between the third and fifth centuries, revealed special local

Fig. 10: Wooden boards in the shape of a human body from the Late Antique/early Christian cemetery in Antinoopolis, fifth to seventh centuries (© Istituto Papirologico ‘G.Vitelli’ - Università degli Studi di Firenze. Photo Cäcilia Fluck).

were covered with a length of fabric, which was affixed to the corners of the construction. Such biers were found in the monastic settlement of Kom el-Ahmar/ Sharuna, in the nearby settlement of Qarara and in Kysis/Dush in Kharga Oasis, for example. Finally, the bodies were tied up into bundles with bands of cloth, often bound in complicated and elaborate patterns as in previous centuries (Fig. 12). For this procedure special ribbons were used which were produced only for this purpose. They were made from flax or palm fibres on the rigid heddle; some are bi-coloured in brown and white or red and white, and simple flax cords were also used (Fig. 13; Huber 2007, 51, 58

types of shrouds in different techniques that were used side by side with plain linen cloths: Huber 2015; South 2017.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

231

Fig. 11: Drawing of a bier from el-Kom el-Ahmar/Šaruna near ancient Ankyronpolis (© Béatrice Huber).

with appendix by A. Unger and S. Schwerdtfeger 67–69; South 2013, 56–71). A special kind of binding seems, as far as our current understanding suggests, to have arisen at the end of the fourth century, limited to the Middle Egyptian regions. Here, several mummies were found with a construction over their heads, shaped like a gabled roof and made from textiles and palm leaves (Fig. 14). This triangular construction can be regarded as a continuation of the tradition of the ancient Egyptian mummy mask, which also protected the entire head (Huber 2013a, 85). From Late Antiquity also some coffins survive that were fitted out with a triangular construction on top, which came to lie above the head of the deceased, as a beautifully decorated sixth- to eighth-century coffin

21

Mummies of this form were found in, e.g., Lisht and Dahshur, Deir el-Banat, Fag el-Gamous, Deir el-Qarabin, Kom al-Ahmar/ Sharuna, Qarara, Naqlun and Antinoopolis: see summary in Gessler-Löhr et al. 2007, esp. 256–57 and 265, n. 8. For Dahshur see Cortes 2012, 78–80, p. 84; for Fag el-Gamous see

from Qarara now in Heidelberg demonstrates (Fig. 15; Huber 2013a; Huber and Nauerth 2018). It shows a peacock, grapevines and, exceptionally, the large-scale Christian symbol of a cross.21 In 1997 a similar coffin was found in Qarara, wrapped in linen and bandaged in a lozenge pattern (Huber 2013a, 81–86). Between the wooden bottom and the surrounding tissue corn grains were found, which probably originally lay under the corpse. B. Huber indicates that there might be a connection to the so-called corn mummies which are attested until the Roman period (Huber 2013a; cf. Raven 1982). As far as their find-spot is known, all these corn mummies come from Middle Egypt, precisely the region around Qarara. It seems that here an ancient tradition was still present in early Christian times.

South 2012a and b; on Deir el-Banat see Orfinscaya, Belova, Naunton and Tolmacheva 2015 and Orfinskaya and Tolmacheva 2017; on Naqlun see Godłewski and Czaja-Szewcak 2008; Zych 2008.

232

C. FLUCK

Fig. 12: Mummy from Qarara near ancient Ankyronpolis, sixth to eighth centuries (inv. 1022. © Egyptian Collection, Heidelberg University. Photo Robert Ajtai).

Fig. 13: Typical binding of a burial from Fag el-Gamous in the Fayum, 2005 season (© Brigham Young University, Brigham Young University Egypt Excavation Project).

Fig. 14: Construction over the head of a mummy from Deir el-Qarabin/Kom el-Ahmar Sharuna near ancient Ankyronpolis, fourth to sixth centuries (© Béatrice Huber).

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

233

Fig. 15: Peacock coffin from Qarara near ancient Ankyronpolis, sixth to eighth centuries (inv. 500, © Egyptian Collection, Heidelberg University. Photo Robert Ajtai).

Dressing the dead in Late Antiquity At the turn of the fourth century serious changes in the use of textiles in burials can be identified. From now on clothing played an important role in preparations for the funeral. The dead were dressed before being wrapped in reused cloths and simple shrouds that were fixed by ribbons in the traditional manner. The findings from an excavation in the 1930s conducted by the Metropolitan Museum of Art New York in the third/fourth-century cemetery of el-Bagawat, but also from Kysis/Dush and el-Deir in Khargah Oasis, and those from Kellis in Dakhla Oasis for instance, are among the earliest evidence for dressing the dead in their everyday clothing—foremost, in the tunic, the obligatory garment in the whole Roman world (Bowen 2002; Dunand et al. 2005; Kajitani 2006; Dunand 2007, 167–68; Letellier-Willemin 2011; 2015; Livingstone 2012). From the second half of the third century, it became common to decorate tunics and mantles with symmetrically arranged tapestry elements. Between the third to fifth centuries interlace and geometric ornamental

designs carried out with a fine linen thread on a purplecoloured background were popular, as shown, for instance, by the tunics of the soldiers depicted in the wall-paintings of the imperial chamber in the temple of Luxor (Jones and McFadden 2015). It is exactly this type of tunic in which the dead in the cemetery of el-Bagawat were dressed (Fig. 16). In other cemeteries of a later period, the fifth to seventh centuries, the deceased often wore several tunics, coats, shawls and head coverings one on top of the other; the face was sometimes covered by a sort of padding, often made of reused headdresses. Many examples come from the cemeteries of Antinoopolis in Middle Egypt that were excavated between 1896 and 1916 by a French mission under the leadership of Albert Gayet. Complete grave ensembles came to France and other museums in Europe to be exhibited after each campaign (Lintz and Coudert 2013; Cat. Lyon 2013). Often, Gayet manipulated these ensembles by adding well-preserved textiles and further objects from other contexts in order to make a burial look complete. Nevertheless, they give a good impression of the elaborate burials of wealthy people in that period.

234

C. FLUCK

Fig. 16: Fragment of a tunic similar to those that were found in el-Bagawat, Kharga Oasis, AD 380–540 (radiocarbon date, 95.4% probability) (MBK inv. 9936, © Staatliche Museen zu Berlin, Skulpturensammlung und Museum für Byzantinische Kunst. Photo Antje Voigt).

The mummy of Euphemia in Brussels One of these luxury burials from Antinoopolis is that of the so-called embroideress Euphemia, now in the Royal Museums of Art and History in Brussels (Fig. 17).22 It was purchased in 1901—manipulated by the seller to make it more attractive for the buyer— at one of the public auctions that usually followed Gayet’s campaigns. Gayet’s staff found the mummy in the winter season 1899/1900 near the northeast corner of the ancient city wall (Gayet 1900, 2, 5). The mummy of Euphemia herself is untouched and still dressed in the clothes in which she was buried. The contents of her grave were investigated a few years ago by a team of scholars from the Royal Museums of Art and History and the Royal Institute for Cultural Heritage in Brussels (Van Raemdonck, Verhecken-Lammens and De Jonghe 2011, 222–35; Van Strydonck et al. 2011, 237–54). The re-evaluation of Gayet’s notes one hundred years after the discovery of the mummy has helped to reconstruct the original content of the grave and to correct its date/chronology, which falls in the Roman period according to Gayet, while other scholars dated it to the fourth/fifth century.

22

Royal Museums of Art and History, Brussels, Inv. E.1045.

Fig. 17: The mummy of the so-called embroideress Euphemia from the Northern Necropolis of Antinoopolis, AD 440 and 610 (radiocarbon date, 95.4% probability) (Inv. E.1045 © Royal Museums of Art and History, Brussels).

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

The mummy of Euphemia is dressed in four tunics, one on top of the other. All the tunics were made of a linen ground weave, but in different qualities, and they have round cut neck openings instead of a woven slit. From the fourth tunic only the trimming around the neck opening is visible. The outer tunic is woven to shape in three parts and decorated with clavi extending the full length of the tunic and between them a scattered pattern of red and white buds forming a triangular shape pointing to the waist. Leaf-shaped motifs are woven in on the shoulder-parts and near the lower edges of the tunic. Separately woven bands are sewn around the neck opening, the wrist-part of the sleeves and the lower edge of the tunic. All are carried out in coloured wool and linen in tapestry technique. The second and the third tunic are both bordered with separately made trimmings in the same technique. It is remarkable that the sleeves of the two outer tunics are not put on but rest, empty, on top of the arms, a feature that has been observed with other mummies, too. Euphemia’s head is covered with a cap made of a linen fabric and trimmed with a blue tapestry band with a geometric design around the front edge and along the hair parting. Above the cap underneath the head a reed bundle is visible. A mantle, decorated with tapestry bands showing tendrils and ornamental design, bi-coloured in the upper part and multicoloured in the lower part, and a series of rosettes in between is draped at both sides of Euphemia’s corpse. The bright colour of the linen and the tapestries contrasts with the tunics in which Euphemia is dressed. The upper part of the mantle is fixed with a modern thread to a so-called ‘bourrelet’, a sort of wreath made of multicoloured wool, and probably imitating the ancient flower wreaths. Remnants of an old linen fabric and of the original sewing thread on the ‘bourrelet’ indicate that it was once connected to another cloth, which could not be identified among the textiles of the Royal Museum of Art and History. The textiles along Euphemia’s sides had another function and were reused in order to make the burial look complete. The two pillows under Euphemia’s head and feet in the display case also seem not to belong to the original burial. The pillow under the head shows the typical Late Antique symmetrically arranged interlace design on a purple-coloured ground

235

woven into a linen plain weave. This occurs on the cushion at the feet only on the hidden side, while the visible upper side is decorated with circles with interlace motifs in orange, dark red and blue. Gayet mentioned in his description of the 1899/1900 findings in the northeastern area of the Northern Necropolis a large linen cloth with five woollen tapestry medallions, two shawls with purple-coloured bands and scattered flowers and a resist-dyed fabric. All these pieces were identified among the textiles in the storerooms of the Royal Museums of Art and History, but it is uncertain whether they belong to Euphemia’s burial or another grave nearby, possibly that of Colluthos and his wife Tisoia, from which the well-known hanging with two idealized portraits survives.23 Apart from the textiles, other grave goods belonged to the original burial, mainly weaving instruments and toilet items (Van Raemdonck,Verhecken-Lammens and De Jonghe 2011, 232–34). Gayet also records shoes with a Greek inscription, which have not been found yet, and further lead seals with cords and XP-symbols now stored in the Egyptian department of the museum. According to the museum’s accession books a coin of the time of Constantine the Great is registered as a part belonging to the lot acquired in 1901, which Gayet did not notice. The way in which Euphemia’s tunics were made, with cut-out neck openings and separately made bands, and the style of the tapestry decoration speak for a late fifth- to early seventh-century date, which was confirmed by radiocarbon analyses (Van Strydonck et al. 2011, 242–43). The loose textiles, the reed bundle and the objects of daily use may have come from other graves and are from different periods. All are of a neutral design without any evidence of Christian symbolism. The evaluation of radiographic images leaves no doubt that the dead individual is a woman of more than forty years. Radiocarbon analyses of the woman’s hair and skin delivered a date between AD 440 and 610 with 95.4% probability (Van Strydonck et al. 2011, 238, 242–43, 244 fig. 7) which corresponds to the dates estimated for the tunics and which is later than previously thought. From stable isotope and dietary analyses we can conclude that Euphemia’s diet consisted of

23

Royal Museums of Art and History, Brussels, inv. ACO.Tx.2470.

236

C. FLUCK

bread made of wheat and barley, beans and fruit, meat from herbivores and for a lesser part river fish. She had enough water to drink and nourished herself with highquality food (Van Strydonck et al. 2011, 241). Summarizing comments on Late Antique burials Not all the dead were buried in elaborate garments and with other grave goods during the fifth to seventh centuries. The garments and other textiles that survive are likely to come from those levels of society that could afford to have their deceased wrapped in clothing and to supply them with household textiles that were still usable. During recent excavations in the so-called peristyle court in the northeast of the Northern Necropolis in Antinoopolis several tombs were discovered that prove that people from different social classes were buried there together (Pintaudi 2008; Minutoli 2008; Grossmann 2008). While the more prestigious burials were found in chambers adjacent to the court, other simpler burials were discovered under the floor in the sand. A boy wrapped only in rags, for instance, was lying next to a mother with her infant, who was dressed in tiny, carefully executed garments one on top of the other (Figs 18a and b; Fluck 2013). It is the same place where Tgol and her newborn were buried. Tgol is identified by an inscribed tombstone mentioning her name. The newborn was wrapped in a shroud made of discarded woollen fabrics, while the mother was dressed in a linen tunic, several shawls, partly decorated, an upper garment and a mantle with ‘bourrelet’ in a striking yellow colour—a combination which is not unique in this cemetery and speaks for a local style (Fig. 19; Fluck 2014, 115–23). Also in Antinoopolis a considerable number of graves contained individuals who were buried in a special type of clothing unusual for Egypt. The deceased did not wear the common tunic, but tailored dresses and coats with oversized sleeves made from precious fabrics—silk or a mixture of sheep’s wool and cashmere—as well as gaiters and boots, inspired by oriental riding costume (Fig. 20; Calament 2004; Fluck 2004b; Linscheid 2004). For a long time these high-quality

24

Some fragments with a decoration typical of tailored dresses were known from a few other sites, for instance Arsinoe/ Krokodilopolis, see Linscheid 2000, 81, no. 31 with further references.

garments were considered to be limited to Antinoopolis, but in 2006 elements of oriental riding costume were also found in a cemetery next to the monastery of Archangel Gabriel at Naqlun (Godłewski and CzajaSzewczak 2008, 251–58; Godłewski and Linscheid forthcoming).24 Independent of their find-spot, these garments make us wonder about the identity of the former owners. Were they wealthy Egyptians who favoured oriental fashion, or members of a foreign elite who settled in Egypt? Future anthropological studies may help to clarify this question. As a rule, the garments that dressed the dead were not made especially for the funeral, but had previously been worn in everyday life, as suggested by traces of wear and tear, and repairs (Fluck 2004a, 208; GesslerLöhr et al. 2007, 256). But no rule exists without exceptions. In the cemetery of Kom el-Ahmar/Sharuna some naturally mummified bodies were found in tunics which, according to the excavator Béatrice Huber, were made especially for the burial (Huber 2013b, 12–21). Depending on their personal wealth, the deceased might also be provided with accessories such as sandals, shoes and boots, jewellery and combs, as well as everyday objects including clay lamps, small vessels, votive offerings and toys or such utensils from professional life as instruments for writing or weaver’s tools. Even reading matter (codices) was sometimes provided, loosely inserted into the bindings.25 The ancient Egyptian custom of equipping the deceased with wreaths of blossoms and garlands for eternity was maintained, as can be judged by the finds in quite a number of graves from various sites. Late Antique cemeteries Excavations and the rich archaeological material that they have revealed have highlighted the importance of the late Roman and early Christian necropoleis of Arsinoe/Medinet al-Fayum, Deir el-Banat, Fag el-Gamous and Naqlun in Fayum; Saqqara near Memphis; Ankyronpolis/el-Hibeh, and nearby Qarara, Kom el-Ahmar/ Sharuna, and Antinoopolis/Sheikh Ibada in Middle Egypt; Panopolis/Akhmim and Western Thebes in

25

See numerous examples from Antinoopolis in Lintz and Coudert 2013, 175–253.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Fig. 18a and b: Two of the four garments of a child buried in the so-called peristyle court in the Northern Necropolis of Antinoopolis, fifth to seventh centuries (© Istituto Papirologico ‘G.Vitelli’ - Università degli Studi di Firenze. Photos Harald Froschauer).

Upper Egypt; el-Bagawat, Kysis/Dush and El-Deir in the Kharga Oasis; and Kellis in the Dakhla Oasis. The burial sites of this period do not differ fundamentally from those of the preceding period. The habit of occupying tombs that already existed was retained, as well as the rituals of commemoration in the cemeteries; these services were, much later, moved into the churches. For this purpose buildings were constructed or existing meeting-places near the graves were reused, sometimes equipped with fixed tables and benches for the funeral feast, as in Saqqara, Antinoopolis and elBagawat where the bereaved could pay their respects to their ancestors (Grossmann 2002, 325–42).26

26

Site 1823 in Saqqara, which was initially interpreted as a church, is however clearly a mausoleum surrounded by porticoes: Grossmann 2002, 319–20.

237

238

C. FLUCK

Fig. 19: Yellow tunic from Antinoopolis, fifth to seventh centuries (MBK inv. 9928, © Staatliche Museen zu Berlin, Skulpturensammlung und Museum für Byzantinische Kunst. Photo Antje Voigt).

Fig. 20: Riding coat in oriental style from Antinoopolis, AD 440–640 (radiocarbon date, 95.4 % probability) (MBK inv. 9695, © Staatliche Museen zu Berlin, Skulpturensammlung und Museum für Byzantinische Kunst. Photo Antje Voigt).

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Funerary inscriptions were written firstly in Greek and later in the regional Coptic dialect. They contain an invocation to God and Jesus Christ and the name of the dead man or woman. Sometimes the age of the deceased or the date of death is also stated, as on Roman and Jewish grave stelae. Longer texts with litanies, prayers and salutations to the living are also attested. And although church fathers and prominent monks such as Shenoute opposed the practice, lamentations for the dead were also offered—a custom already found in ancient Egyptian times as well as in the Jewish treatment of the dead (Leibovitch 1942, 41–42, 44–46). Conclusion ‘… in the hundreds upon hundreds of meters of cloth that investigators have “unrolled”, … lies one of the most costly and culturally significant aspects of an Egyptian burial: fine linen, just as the ancient scribes had marked it’ (Riggs 2014, 108). This quote, taken from C. Riggs’ Unwrapping ancient Egypt, underlines the important role that textiles played in Egyptian funerary contexts from dynastic times onwards. Shrouds and bandages were essential to wrap the body and to give it the shape of a mummy, which transformed the dead into a representation of Osiris. This practice of wrapping the body continued throughout the centuries of Egypt’s long history and, strangely enough, reached its qualitative peak at the time when Egypt was under Roman rule, just when veneration of the ancient gods gradually declined, thus losing the original religious aspect behind mummification. Conservation treatments to preserve the bodies of the deceased were carried out more negligently. Instead, more emphasis was put on the outer appearance. It seems that perfect ‘mummy forming’ became prestigious in Roman times, going hand in hand with the Roman custom of selfrepresentation and identity. Numerous painted portraits from various sites in the country bear witness to this. These portraits covered the face of the deceased and were fixed by bands, but left outside the wrapping itself in order to be visible to the outside world. From the fourth century AD onwards burying the dead in garments worn in everyday life became a common burial practice, while the outer layers of the wrapping were carried out in the traditional way. It seems that from that time onwards clothing became an expression of personal identity and self-representation, not only for a small elite but for everybody. It became common to decorate tunics and mantles with patterns,

239

motifs and narratives, mostly carried out in wool in the technique of tapestry—a sort of ‘woven painting’. These garments, often accompanied by dress accessories and household soft furnishings, found their way into the graves. Often individuals wore two or more garments one on top of the other, not only in the tombs of the rich but also in those of the broader masses, as examples in the Northern Necropolis of Antinoopolis, for instance, prove—a new ritual to honour the deceased and to care for his or her burial. In most cases it is not possible to determine who the individuals were, to which ethnic group they belonged or which religious belief they followed. ‘Foreign’ elements, such as clothes in oriental style which appeared in tombs in Antinoopolis and in the Fayum, are no indication of the origin of the people who wore them. Were they officials, merchants or emigrants from the Byzantine empire who settled in Egypt for a while, or natives dressed according to oriental ‘fashion’? It is just as difficult to determine the religious affiliations of the individuals when concrete evidence—a certain cultic environment, specific invocations or symbols—is lacking. The alteration in burial practices in the form of dressing the dead corresponds to the rise of Christianity. But it seems that it was neither due to the new religion nor to a changing regime. There must have been other reasons which are still difficult to grasp. The decline of the temples during the third century owing to the withdrawal of state support by the Roman emperors might be one possible factor for a modification in burial practices. The priesthood, formerly partly involved in funerary rituals, no longer existed and this might have affected the funerary industry (Bagnall 1993, 261–68). On the other hand, the elaborate execution of a considerable number of Late Antique mummies (see Fig. 12) at various places in Middle Egypt proves that there still must have been active workshops connected to funeral cult at that time. Throughout its long history Egypt was repeatedly ruled by foreign regimes, but it seems that they did not have serious influence on the traditional burial norms in the country. Giving the deceased the outer shape of a mummy was an Egyptian phenomenon, practised from the fourth millennium BC onwards and maintained until at least the seventh/eighth century AD in monastic settlements. Burial customs were modified, and sometimes certain cultural features of ‘the others’ (again: Roman portraits) were adopted, but they did not radically change either with the rise of a new

240

C. FLUCK

Fig. 21a and b: Two examples of tiraz from Cemetery A, near the Monastery of Archangel Gabriel in Naqlun in the Fayum, tenth to twelfth centuries (© PCMA Archives. Photo Tomasz Szmager).

religion or when new rulers took power. Even the observation that until Roman times it was customary to bury the dead with their heads towards the west, while in Christian times the head was facing the east, is no reliable indication for a changing religion or political system. Only the Jewish community in Egypt seems to have retained its own traditional burial forms, as the archaeological remains of the Jewish cemetery at Tell el-Yahudiya from Graeco-Roman times and some scattered evidence elsewhere suggest. When the Arabs came to power in Egypt from AD 639 there was no immediate change in burial practices. The custom of dressing the dead was still practised until the Fatimid period (AD 969–1171), as examples from a cemetery for lay people next to the monastery of Archangel Gabriel in Naqlun show (Czaja-Szewczak 2011; Fluck and Helmecke 2015, 236–37). However, the costume changed: instead of tunics and mantles the individuals wore cut-to-shape garments, often decorated with ornamental tiraz-bands (Fig. 21), some of which contain the baraka, the Muslim blessing. Were Christians and Muslims buried side by side or did Christians adopt the Muslim blessing for their own salvation? In the same cemetery other individuals were dressed in simple, carelessly sewn shirts of coarse linen

and then wrapped in shrouds; both were made especially for burial purposes (Czaja-Szewczak 2003). This meets the Muslim burial regulations according to the Hadith (records of Muḥammad) that burials should be simple and without coffin and grave goods. In contrast to Late Antique burials, the Muslim dead lay on their sides with their heads towards Mecca. In the time of the Fatimids the traditional Egyptian burial practice came to an end, with one exception: it is still customary in Coptic churches and monasteries to display the mummies of their holy and revered monk fathers wrapped in bundles in show cases for pilgrims. Acknowledgements I would like to thank Elisabeth R. O’Connell, Carolyn Jones and the anonymous peer-reviewers for their suggestions on earlier drafts of this chapter. Further thanks go to Robert Ajtai, Dina Faltings, Włodzimierz Godłewski, Jana Helmbold-Doyé, Béatrice Huber, Maya Müller, Kerry Muhlestein, Rosario Pintaudi, Kristin South, Sandra Steiß and Olivia Zorn for their valuable comments which helped to improve the text and especially for giving permission to publish the images included in this article.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

Bibliography Abd el-Al, A., J. C. Grenier and G. Wagner. 1985. Stèles funéraires de Kom Abu Bellou. Édition recherche sur les civilisations, Mémoire 55. Paris. Bagnall, R. S. 1993. Egypt in Late Antiquity. Princeton, NJ. Baumeister, T. 1971. Spätägyptische Bestattungssitten und christliche Heiligenverehrung in Ägypten. Studia Patristica 12: 406–12. Bernand, É. 1992. Inscriptions grecques d’Égypte et de Nubie au Musée du Louvre. Paris. Bierbrier, M. L. (ed.). 1997. Portraits and masks: Burial customs in Roman Egypt. London. Borg, B. 1998. ‘Der zierlichste Anblick der Welt …’: Ägyptische Porträtmumien. Mainz. Bowen, G. 2002. Textiles, basketry and leather goods from Ismant el-Kharab. In Dakhleh Oasis Project: Preliminary reports on the 1994–1995 to 1998–1999 field seasons, C. A. Hope and G. E. Bowen (eds), 87–104. Dakhleh Oasis Project Monograph 11. Oxford. Bowman, A. K. 2007. Roman Oxyrhynchus: City and people. In Oxyrhynchus: A city and its texts, A. K. Bowman, R. A. Coles, N. Gonis, D. Obbink and P. J. Parsons, 171–81. London. Budge, E. A. W. 1913. Coptic apocrypha in the dialect of Upper Egypt. London. Calament, F. 2004. L’apport historique des découvertes d’Antinoé au costume dit de ‘cavalier sassanide’. In Riding costume in Egypt: Origin and appearance, C. Fluck and G. Vogelsang-Eastwood (eds), 37–72. Studies in Textile and Costume History 3. Leiden. Capponi, L. 2007. Il tempio di Leontopoli in Egitto. Identità politica e religiosa dei Giudei di Onia (c. 150 a.C.–73 d.C.). Pubblicazioni della Facoltà di Lettere e Filosofia dell’Università di Pavia. Dipartimento di Scienze dell’Antichità. Pisa. Castel, G. 1979. Étude d’une momie copte. In Hommage à la mémoire de Serge Sauneron (1927–1976) II: Égypte post-pharaonique, 121–43. Bibliothèque d’étude 82. Cairo. Cat. Frankfurt 1999 = Parlasca, K. and H. Seemann (eds). 1999. Augenblicke. Mumienporträts und ägyptische Grabkunst aus römischer Zeit (exhibition catalogue Schirn Kunsthalle Frankfurt). Munich. Cat. Hildesheim 2018 = Schulz, R. and O. Gauert (eds). 2018. Ta-cheru: Eine Reise ins Innere einer Mumie (exhibition catalogue Roemer- und Pelizaeus-Museum Hildesheim). Hildesheim. Cat. London 1997 = Walker, S. and M. Bierbrier (eds). 1997. Ancient faces: Mummy portraits from Roman Egypt (exhibition catalogue The British Museum London). London.

241

Cat. London 2014 = Taylor, J. H. and D. Antoine (eds). 2014. Ancient lives – new discoveries: Eight mummies, eight stories (exhibition catalogue The British Museum London). London. Cat. London 2015 = Fluck, C., G. Helmecke and E. R. O’Connell (eds). 2015. Egypt: Faith after the pharaohs (exhibition catalogue The British Museum London). London. Cat. Lyon 2013 = Durand, M. and F. Calament (eds). 2013. Antinoe, à la vie, à la mode: Visions d’élégance dans les sol (exhibition catalogue Musée des tissus Lyon). Lyon. Cat. Münster 2016 = Lichtenberger, A., A. Lohwasser and H. H. Nieswand (eds). 2016. Tod und Ewigkeit: Die Münster-Mumie im Fokus der Forschung (exhibition catalogue Archäologisches Museum der Westfälischen Wilhelms-Universität Münster). Münster. Cat. New York 2000 = Walker, S. (ed.). 2000. Ancient faces: Mummy portraits from Roman Egypt (exhibition catalogue Metropolitan Museum of Art). New York. Cat. Paris 1998 = Aubert, M.-F. and R. Cortopassi (eds). 1998. Portraits de l’Égypte romaine (exhibition catalogue Musée du Louvre Paris). Paris. Cat. Stuttgart 2007 = von Recklinghausen, D., et al. 2007. Ägyptische Mumien: Unsterblichkeit im Land der Pharaonen (exhibition catalogue Landesmuseum Württemberg Stuttgart). Mainz. Cortes, E. 2012. Recovering contexts: The Roman mummies excavated by The Metropolitan Museum of Art at Dahshur, Egypt. In Dressing the dead in classical antiquity, M. Carroll and J. P. Wild (eds), 75–88. Stroud. Cox, S. L. 2015. A critical look at mummy CT scanning. The Anatomical Record 298: 1099–110. doi : 10.1002/ ar.23149. Czaja-Szewczak, B. 2003. Burial tunics from Naqlun. Polish Archaeology in the Mediterranean 14: 177–84. ———. 2011. Funerary textiles from the Medieval cemetery of Naqlun. Polish Archaeology in the Mediterranean 20: 413–20. David, A. R. 2000. Mummies and mummification. In Ancient Egyptian materials and technologies, P. T. Nicholson and I. Shaw (eds), 372–89. Cambridge. Davies, J. 1999. Death, burial and rebirth in the religions of antiquity. London; New York. Dickey, A. 2019. Egyptian textiles, Midgley and Bolton Museum: Exploring a valuable Predynastic and Early Dynastic collection. In Egypt as a textile hub: Textile interrelationships in the 1st Millennium AD. Proceedings of the 10th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2017, A. De Moor, C. Fluck and P. Linscheid (eds), 134–47. Tielt. Doxiadis, E. 1995. The mysterious Fayum portraits: Faces from ancient Egypt. New York.

242

C. FLUCK

Dunand, F. 2007. Between tradition and innovation: Egyptian funerary practices in Late Antiquity. In Egypt in the Byzantine world, 300–700, R. S. Bagnall (ed.), 163–84. Cambridge. Dunand, F., J. L. Heim, N. Heinein and R. Lichtenberg (eds). 2005. La nécropole de Douch II: Tombes 73 à 92. Douch 5; Documents de Fouilles de l’Institut francais d’archeologie orientale du Caire 45. Cairo. Dzierzbicka, D. and Ożarek, M. A. 2012. Two burials from Cemetery A in Naqlun: Archaeological and anthropological remarks. Polish Archaeology in the Mediterranean 21: 233–43. Fluck, C. 2004a. Textilien. In Die Ägyptische Sammlung des Gustav-Lübcke-Museums Hamm, M. Von Falck and C. Fluck, 217–61. Bönen. ———. 2004b. Zwei Reitermäntel aus Antinoopolis im Museum für Byzantinische Kunst, Berlin. Fundkontext und Beschreibung. In Riding costume in Egypt: Origin and appearance, C. Fluck and G. Vogelsang-Eastwood (eds), 137–52. Studies in Textile and Costume History 3. Leiden. ———. 2013. Textiles from Antinoupolis: Recent finds from the so-called peristyle complex in the Northern Necropolis. In Itinerari mediterranei fra IV e IX secolo: Città-capitale e Deserto-monastico, B. Astrua (ed.), 85–104. Turin. ———. 2014. Textiles from the so-called ‘tomb of Tgol’ in Antinoupolis. In Egypt in the First Millennium AD: Perspectives from new fieldwork, E. R. O’Connell (ed.), 115–23. Leuven; Paris; Walpole, MA. ———. 2020. Apa Schenutes Kleider. In The textile centre Akhmîm-Panopolis (Egypt) in Late Antiquity: Material evidence for continuity and change in society, religion, industry and trade, R. El-Sayed and C. Fluck (eds), 105–19. Studia Panopolitana 4. Wiesbaden. Fluck, C. and K. Finneiser. 2009. Kindheit am Nil: Spielzeug – Kleidung – Kinderbilder aus Ägypten in den Staatlichen Museen zu Berlin. Berlin. Fluck, C. and G. Helmecke. 2015. Burial practice. In Cat. London 2015, 232–41. Förster, H. 2008. Mumifizierung von Christen in Ägypten: Eine religionsgeschichtliche Anfrage. Journal of Coptic Studies 10: 167–82. Friedman, R. F., A. Maish, A. G. Fahmy, J. C. Darnell and E. D. Johnson. 1999. Preliminary report on field work at Hierakonpolis: 1996–1998. Journal of the American Research Center in Egypt 36: 1–35. Gauert, O., W. Pirsig and R. Sokiranski. 2018. Die Mumie der Ta-Cheru und ihr Hologramm. In Cat. Hildesheim 2018, 134–41. Gayet, A. 1900. Notice relative aux objets recueillis à Antinoë pendant les fouilles exécutés en 1899–1900 et exposés au musée Guimet du 12 décembre 1900 au 12 janvier 1901. Paris.

Germer, R. 1991. Das Geheimnis der Mumien. Zurich; Munich. Germer, R., H. Kischkewitz and M. Lüning. 2009. Berliner Mumiengeschichten: Ergebnisse eines multidisziplinären Forschungsprojektes. Regensburg. Gessler-Löhr, B. 2010. Two child mummies and some grave goods of the Byzantine period from the Egyptian collection at Heidelberg. In Mummies of the world 1, A. Wieczorek and W. Rosendahl (eds), 310–15. Munich; Berlin; London; New York. Gessler-Löhr, B. 2012. Mummies and mummification. In The Oxford handbook of Roman Egypt, C. Riggs (ed.), 664–83. Oxford. Gessler-Löhr, B., E. Grabbe, B.-W. Raab and M. Schultz. 2007. Ausklang: Eine koptische Mumie aus christlicher Zeit. In Cat. Stuttgart 2007, 255–65. Godłewski, W. 2005. The Medieval Coptic cemetery at Naqlun. In Christianity and monasticism in the Fayum Oasis, G. Gabra (ed.), 173–83. Cairo. Godłewski, W. and B. Czaja-Szewczak. 2008. Cemetery C.1 in Naqlun: Tomb C.T.5 and its cartonnages. Polish Archaeology in the Mediterranean 18: 247–60. Godłewski, W. and P. Linscheid. Forthcoming. Textiles from Burial M.2 in Cemetry C1. In Explorers, first collectors and traders of textiles from Egypt of the 1st Millennium AD, A. De Moor, C. Fluck and P. Linscheid (eds). Goodenough, E. R. 1953. Jewish symbols in the GrecoRoman period 2: The Archeological evidence from the diaspora; 3: Illustrations. New York. Griffith, F. Ll. 1890. The antiquities of Tell el Yahûdîyeh. In The mound of the Jew and the city of Onias, E. Naville (ed.), 33–74. The Egypt Exploration Fund Memoir 7. London. Grossmann, P. 2002. Christliche Architektur in Ägypten. Handbuch der Orientalistik 62. Leiden. ———. 2008. Antinoopolis – Der Komplex des ‘Peristylbaus’. In Antinoupolis I: Scavi e materiali, G. Bastianini and R. Pintaudi (eds), 41–44. Florence. Hachlili, R. 1998. Ancient Jewish art and archaeology in the diaspora. Handbook of Oriental Studies 35. Leiden; Boston; Cologne. ———. 2005. Jewish funerary customs, practices and rites in the Second Temple Period. Leiden; Boston. Harrauer, H. 2003. Tod in den Papyri: Griechische Zeit. In Tod am Nil: Tod und Totenkult im antiken Ägypten, H. Froschauer, C. Gastgeber and H. Harrauer (eds), 21–32. Nilus 8. Vienna. Heilmeyer, M. 2015. Flower garlands for the dead. In Cat. London 2015, 242–43. Helmbold-Doyé, J. (ed.). 2017. Aline und ihre Kinder: Mumien aus dem römerzeitlichen Ägypten. Wiesbaden. ———. Forthcoming. Schöner Schein – was bleibt: Überlegungen zu römerzeitlichen Bestattungen in Ägypten.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

In Making differences matter. Die Macht der Schönheit als Index von sozialem Kapital im Alten Ägypten, N. Kloth and J. Kahl (eds). Papers on occasion of a conference held in December 2012 at the Freie University, Berlin. Horak, U. 1995. Koptische Mumien: Der koptische Tote in Grabungsberichten, Funden und literarischen Nachrichten. Biblos 44: 39–71. Horbury, W. and D. Noy. 1992. Jewish inscriptions of Graeco-Roman Egypt. Cambridge. Huber, B. 2004. Die Klosteranlage bei el-Kom el-Ahmar/ Šaruna (Mittelägypten). Bulletin de la Société d’archéologie copte 43: 45–61. ———. 2006. Die Klosteranlage Deir el-Qarabin bei el-Kom el-Ahmar/Šaruna in Mittelägypten. Bulletin de la Société d’archéologie copte 45: 59–68. ———. 2007. The textiles of an early Christian burial from el-Kom el-Ahmar/Šaruna (Middle Egypt). In Methods of dating ancient textiles of the 1st Millennium AD from Egypt and neighbouring countries. Proceedings of the 4th meeting of the study group ‘Textiles from the Nile Valley’, Antwerp 2005, A. De Moor and C. Fluck (eds), 36–69. Tielt. ———. 2009. The funerary beds from the monastic cemetery at el-Ghalida (el-Kom el Aḥmar/Šaruna). In Clothing the house: Furnishing textiles of the 1st Millennium AD from Egypt and neighbouring countries. Proceedings of the 5th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2007, A. De Moor and C. Fluck (eds), 56–72. Tielt. ———. 2013a. Ein koptischer Sarg zwischen Tradition und Innovation. Bulletin de la Société d’archéologie copte 52: 81–98. ———. 2013b. Eine Tunika fürs Jenseits. In Drawing the threads together: Textiles and footwear of the 1st Millennium AD from Egypt. Proceedings of the 7th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2011, A. De Moor, C. Fluck and P. Linscheid (eds), 12–21. Tielt. ———. 2015. Qarara: Une affaire de linceuls. In Textiles, tools and techniques of the 1st Millenium from Egypt and neighbouring countries: Proceedings of the 8th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2013, A. De Moor, C. Fluck and P. Linscheid (eds), 12–25. Tielt. Huber, B. and C. Nauerth. 2018. Coptic coffins from Qarara. The Pfauensarg (the peacock coffin) in context. In Ancient Egyptian coffins: Craft, traditions and functionality, J. H. Taylor and M. Vandenbeusch (eds), 435–69. British Museum Studies in Egypt and Sudan 4. Leuven. Huppertz, A. 2017. Die Kindermumien. In Aline und ihre Kinder: Mumien aus dem römerzeitlichen Ägypten, J. Helmbold-Doyé (ed.), 33–38. Wiesbaden.

243

Jones, M. and S. McFadden (eds). 2015. Art of empire: The Roman frescoes and imperial cult chamber in Luxor Temple. New Haven, CT. Jones, M., T. F. G. Higham, D. Chivall, R. Bianucci, G. L. Kay, M. J., Pallen, R. Oldfield, F. Ugliano and S. A. Buckley. 2018. A prehistoric Egyptian mummy: Evidence for an ‘embalming recipe’ and the evolution of early formative funerary treatments. Journal of Archaeological Science 100: 191–200. doi : 10.1016/j. jas.2018.07.011 Kajitani, N. 2006. Textiles and their context in the third-tofourth-century CE cemetery of al-Bagawat, Khargah Oasis, Egypt from the 1907–1931 excavations by The Metropolitan Museum of Art, New York. In Textiles in situ: Their find spots in Egypt and neighbouring countries in the First Millennium CE, S. Schrenk (ed.), 95–112. Riggisberger Berichte 13. Riggisberg. Krause, M. 1983. Das Weiterleben ägyptischer Vorstellungen und Bräuche im koptischen Totenwesen. In Das römisch-byzantinische Ägypten, G. Grimm, H. Heinen and E. Winter (eds), 85–92. Aegyptiaca Treverensia 2. Mainz. ———. 2003. Das Totenwesen der Kopten. In Tod am Nil: Tod und Totenkult im antiken Ägypten, H. Froschauer, C. Gastgeber and H. Harrauer (eds), 33–44. Nilus 8. Vienna. Leibovitch, J. 1942. Stèles funéraires de Tell el-Yahoudieh. Annales du Service des antiquités de l’Égypte 41: 41–46. Lembke, K. 2004. Tuna el-Gebel, die Nekropole von Hermopolis. In Ägyptens späte Blüte: Die Römer am Nil, K. Lembke, C. Fluck and G. Vittmann (eds), 59–76. Mainz. ———. 2007. Tod und Bestattung im kaiserzeitlichen Ägypten. In Verborgene Zierde: Spätantike und islamische Textilien aus Ägypten in Halle, G. Brands and A. Preiß (eds), 24–33. Halle. Lembke, K. and S. Prell (eds). 2015. Die Petosiris-Nekropole von Tuna el-Gebel I. Tuna el-Gebel 6. Vaterstetten. Letellier-Willemin, F. 2011. Accessories from the Christian cemetery of El Deir in the Egyptian oasis of Kharga, Western Desert. In Dress accessories of the 1st Millennium AD from Egypt: Proceedings of the sixth conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2009, A. De Moor and C. Fluck (eds), 96–109. Tielt. ———. 2015. The long- and narrow-sleeved tunic of the mummy W14 of El Deir. In Textiles, tools and techniques of the 1st Millenium from Egypt and neighbouring countries: Proceedings of the 8th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2013, A. De Moor, C. Fluck and P. Linscheid (eds), 26–37. Tielt. Linscheid, P. 2000. Die Textilien von Theodor Graf, vormals im Besitz des Kunstgewerbemuseums Berlin (Kat.-Nrn.

244

C. FLUCK

1–78). In Textilien aus Ägypten 1: Textilien aus dem Vorbesitz von Theodor Graf, Carl Schmidt und dem Ägyptischen Museum Berlin, C. Fluck, P. Linscheid and S. Merz (eds), 43–124. Staatliche Museen zu Berlin – Preußischer Kulturbesitz: Skulpturensammlung und Museum für Byzantinische Kunst Bestandskataloge 1. Wiesbaden. ———. 2004. Gaiters from Antinoopolis in the Museum für Byzantinische Kunst Berlin. In Riding costume in Egypt: Origin and appearance, C. Fluck and G. Vogelsang-Eastwood (eds), 153–61. Studies in Textile and Costume History 3. Leiden. Lintz, Y. and M. Coudert (eds). 2013. Antinoé: Momies, textiles, céramiques et autres antiques. Paris. Livingstone, R. 2012. Five Roman period tunics from Kellis. In The Oasis Papers 6: Proceedings of the Sixth International Conference of the Dakhleh Oasis Project, R. S. Bagnall, P. Davoli and C. A. Hope (eds), 317–26. Oxford. McNally, S. 2002. Syncretism in Panopolis? The evidence of the ‘Mary silk’ in the Abegg Stiftung. In Perspectives on Panopolis: An Egyptian town from Alexander the Great to the Arab conquest, A. Egberts, B. P. Muhs and J. van der Vliet (eds), 145–64. Papyrologica LugdunoBatava 31. Leiden; Boston. Michaelis, L. 2012. Zur Datierung eines Mumienporträtfragments. Chronique d’Égypte 87: 150–58. Minutoli, D. 2008. Antinoe, necropoli nord 2007: La tomba di Tgol. Prime informazioni. In Antinoupolis I: Scavi e materiali, G. Bastianini and R. Pintaudi (eds), 61–73. Florence. Moje, J. 2017. Der Grabstein der Aline im Vergleich mit anderen Stelen. In Aline und ihre Kinder: Mumien aus dem römerzeitlichen Ägypten, J. Helmbold-Doyé (ed.), 53–54. Wiesbaden. Montserrat, D. 1997. Death and funerals in the Roman Fayum. In Portraits and masks: Burial customs in Roman Egypt, M. L. Bierbrier (ed.), 33–44. London. Nauerth, C. 2003. Pfauensarg und Mumienhülle. Bestattungsformen im spätantiken Ägypten. In Grab und Totenkult im Alten Ägypten, H. Guksch, E. Hofmann and M. Bommas (eds), 226–39. Munich. Naville, E. 1890. The mound of the Jew and the city of Onias. The Egypt Exploration Fund Memoir 7. London. O’Connell, E. R. 2007. Transforming monumental landscapes in Late Antique Egypt: Monastic dwellings in legal documents from Western Thebes. Journal of Early Christian Studies 15: 239–73. Orfinscaya, O., G. A. Belova, M. Naunton and E. G. Tolmacheva. 2015. Textiles from burial 213 in Deir el-Banat. In Textiles, tools and techniques of the 1st Millennium from Egypt and neighbouring countries: Proceedings of the 8th conference of the research group ‘Textiles from

the Nile Valley’, Antwerp 2013, A. De Moor, C. Fluck and P. Linscheid (eds), 38–47. Tielt. Orfinskaya, O. and E. Tolmacheva. 2017. Textile finds from the Grave 213 at Deir el-Banat: From an object to a loom. In Excavating, analysing, reconstructing textiles of the 1st Millennium AD from Egypt and neighbouring countries: Proceedings of the 9th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2015, A. De Moor, C. Fluck and P. Linscheid (eds), 12–31. Tielt. Paetz gen. Schieck, A. 2012. Zu den Ursprüngen des Mönchshabits und der Textilproduktion in ägyptischen Klöstern des 4. bis 7. Jahrhunderts n. Chr. In Kleidung und Identität in religiösen Kontxten der römischen Kaiserzeit, S. Schrenk, K. Vössing and M. Tellenbach (eds), 226–56. Mannheimer Geschichtsblätter Sonderveröffentlichung 4. Regensburg. Parlasca, K. 1969–2003. Ritratti di mummie, Repertorio d’arte dell’Egitto greco-romano Ser. B, vols 1–4. Rome. Petrie, W. M. F. 1889. Hawara, Biahmu and Arsinoe. London. ———. 1906. Hyksos and Israelite cities. London. ———. 1911. Roman portraits and Memphis (IV). British School of Archaeology in Egypt and Egyptian Research Account, Seventeenth Year. London. Picton, J., S. Quirke and P. C. Roberts. 1992. Living images: Egyptian funerary portraits in the Petrie Museum. Walnut Creek, CA. Pintaudi, R. 2008. Gli scavi dell’Istituto Papirologico ‘G. Vitelli’ di Firenze ad Antinoe (2000–2007): Prime notizie. In Antinoupolis I: Scavi e materiali, G. Bastianini and R. Pintaudi (eds), 1–41. Florence. Popović, M. and J. Zangenberg. 2013. Dagelijks leven in Judea en de materiële cultuur van Qumran in regionale context. In De Dode Zee rollen. Nieuw licht op de schatten van Qumran, M. Popović (ed.), 72–93. Amsterdam. Quack, J. F. 2007. Zwischen Auswahl der Qualität und Zwang zu einer Werkstatt. In Cat. Stuttgart 2007, 19–27. Raven, M. J. 1982. Corn-mummies. Oudheidkundige Mededelingen van het Rijksmuseum van Oudheden te Leiden 63: 7–38. Raven, M. J. and W. K. Taconis. 2005. Egyptian mummies: Radiological atlas of the collection in the National Museum of Antiquities at Leiden. Papers on Archaeology of the Leiden Museum of Antiquities/Rijksmuseum van Oudheden. Turnhout. von Recklinghausen, D. 2007. ‘Man hat Dir eine Bestattung mit Salbe und Binden bereitet’: Die Mumifizierung im Alten Ägypten. In Cat. Stuttgart 2007, 53–61. Riggs, C. 2000. Roman period mummy masks from Deir el-Bahri. Journal of Egyptian Archaeology 86: 121–44.

TEXTILES IN BURIAL PRACTICE IN ROMAN AND LATE ANTIQUE EGYPT

———. 2002. Facing the dead: Recent research on the funerary art of Ptolemaic and Roman Egypt. American Journal of Archaeology 106: 86–101. ———. 2005. The beautiful burial in Roman Egypt: Art, identity and funerary religion. Oxford. ———. 2014. Unwrapping ancient Egypt. London; New Delhi; New York; Sydney. Schrenk, S. 2004. Die Textilien des Mittelmeerraumes aus spätantiker bis frühislamischer Zeit. Riggisberg. Sokiranski, R. and W. Pirsig. 2018. Faszination der medizinischen Bildgebung. Vom Röntgenbild zur 4F-Holographie in Anatomie, Medizin und Wissenschaft. In Cat. Hildesheim 2018, 34–47. South, K. H. 2012a. Roman and early Byzantine burials at Fag el-Gamous, Egypt: A reassessment of the case for religious affiliation. MA thesis. Brigham Young University, Provo. https://scholarsarchive.byu.edu/ cgi/viewcontent.cgi?article=4533&context=etd (last accessed February 2020). ———. 2012b. Minor burial textiles and religious affiliation: An archaeological case study from Roman Egypt. In Dressing the dead in classical antiquity, M. Carroll and J. P. Wild (eds), 62–74. Stroud. ———. 2013. Regarding ribbons: The spread and use of narrow purpose-woven bands in Late Roman Egyptian burials. In Drawing the threads together: Textiles and footwear of the 1st Millennium AD from Egypt. Proceedings of the 7th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2011, A. De Moor, C. Fluck and P. Linscheid (eds), 56–71. Tielt. ———. 2017. The use of basket-weave linen in burials of the necropolis of Fag el-Gamous, Egypt. In Excavating, analysing, reconstructing textiles of the 1st Millennium AD from Egypt and neighbouring countries: Proceedings of the 9th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2015, A. De Moor, C. Fluck and P. Linscheid (eds), 88–107. Tielt. Stevens, A., G. R. Dabbs, M. Shepperson and M. King Wetzel. 2015. The cemeteries of Amarna. Journal of Egyptian Archaeology 101: 17–27. Tatz, S. 2017. The textile finds from the monastery of Deir el-Bachit/Paulos monastery (6th–10th century) in Western Thebes. In Excavating, analysing, reconstructing textiles of the 1st Millennium AD from Egypt and neighbouring countries: Proceedings of the 9th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2015, A. De Moor, C. Fluck and P. Linscheid (eds), 108–23. Tielt. Taylor, J. 1997. Before the portraits: Burial practices in pharaonic Egypt. In Cat. London 1997, 9–13. Torallas Tovar, S. 2013. Egyptian burial practices in Late Antiquity: The case of Christian mummy labels.

245

In Cultures in contact: Transfer of knowledge in the Mediterranean context. Selected papers, J. P. Monferrer and S. Torallas Tovar (eds), 15–26. SyroArabica 1. Córdoba. ———. 2014. Egyptian burial practice in a period of transition: On embalming in Christian times. In Mapping knowledge: Cross-pollination in Late Antiquity and the Middle Ages, P. Mantas and C. Burnett (eds), 129–40. Córdoba. ———. 2017. A lexical note on the Life of Pisentios. In Labor omnia uicit improbus: Miscellanea in honorem Ariel Shisha-Halevy, N. Bosson, A. Boud’hors and S. H. Aufrère (eds), 373–80. Orientalia Lovaniensia Analecta 256. Leuven. Turner, E. G. 2007. Roman Oxyrhynchus. In Oxyrhynchus: A city and its texts, A. K. Bowman, R. A. Coles, N. Gonis, D. Obbink and P. J. Parsons (eds), 141–54. London. van der Vliet, J. 1993. Spätantikes Heidentum in Ägypten im Spiegel der koptischen Literatur. In Begegnung von Heidentum und Christentum im spätantiken Ägypten, 99–130. Riggisberger Berichte 1. Riggisberg. Van Raemdonck, M., C. Verhecken-Lammens and D. De Jonghe. 2011. The mummy of the ‘Embroideress’ and the contents of her grave. In Dress accessories of the 1st Millennium AD from Egypt: Proceedings of the 6th conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2009, A. De Moor and C. Fluck (eds), 222–35. Tielt. Van Strydonck, M., I. Vanden Berghe, M. Boudin and K. Quintelier. 2011. Euphemia: A multidisciplinary quest for the origin and authenticity of a mummy’s clothes and accessories. In Dress accessories of the 1st Millennium AD from Egypt: Proceedings of the sixth conference of the research group ‘Textiles from the Nile Valley’, Antwerp 2009, A. De Moor and C. Fluck (eds), 236–57. Tielt. Vita Antonii 1917 = Stegmann, A. and H. Mertel (ed. and transl.). 1917. Des heiligen Athanasius ausgewählte Schriften aus dem Griechischen übersetzt 2: Gegen die Heiden, Über die Menschwerdung, Leben des heiligen Antonius. Bibliothek der Kirchenväter 31. Munich. Vogelsang-Eastwood, G. 1995. Die Kleider des Pharaos: Die Verwendung von Stoffen im Alten Ägypten. Hanover; Amsterdam. ———. 1999. Tutankhamun’s wardrobe: Garments from the tomb of Tutankhamun. Rotterdam. Walker, S. 1997. Mummy portraits and Roman portraiture. In Cat. London 1997, 14–16. London. Weber, H., S. Engels and C. Reinhart. 2018. In Cat. Hildesheim 2018, 48–57. Willburger, N. 2007. Zeugnisse römischer Mumiendekoration aus Ägypten. In Cat. Stuttgart 2007, 229–35.

246

C. FLUCK

Willers, D. and Niekamp, B. 2015. Der Dionysosbehang der Abegg-Stiftung. Riggisberger Berichte 20. Riggisberg. Winlock, H. E. 1940. The mummy of Wah unwrapped. Metropolitan Museum of Art Bulletin 35 (December): 253–59. Winlock, H. E. and W. E. Crum. 1926. The Monastery of Epiphanius at Thebes I. New York. Zdiarsky, A. 2012. Den Toten entrissen. Textilien im Bestattungswesen des spätantiken Ägypten. In Gewebte

Geschichte: Stoffe und Papyri aus dem spätantiken Ägypten, B. Palme and A. Zdiarsky (eds), 27–35. Nilus. Studien zur Kultur Ägyptens und des Vorderen Orients B. Vienna. Zych, I. 2008. Cemetery C in Naqlun: Preliminary report on the excavation in 2006. Polish Archaeology in the Mediterranean 18: 230–46.

FASHIONING ASCETIC LEADERSHIP: THE ENDURING TRADITION OF MANTLES OF AUTHORITY IN PORTRAITS OF EGYPTIAN MONASTIC FATHERS Thelma K. THOMAS

Egypt looms large in conceptions of early Christian monasticism as the ascetic spiritual authority of the Desert Fathers continues to inspire emulation of their examples (e.g., Dijkstra and van Dijk 2006). One influential Christian monk of modern times, Thomas Merton, chose to become a modern Desert Father by living apart from his brothers in his Trappist monastery, the Abbey of Gethsemani in Kentucky (Schiffhorst 2011). In the preface to his publication of the sayings of the Desert Fathers, Merton pondered the personal relationship of early monks to God, writing of their search for salvation outside of society: In those days men had become keenly conscious of the strictly individual character of ‘salvation’. Society —which meant pagan society, limited by the horizons and prospects of life ‘in this world’—was regarded by them as a shipwreck from which each individual man had to swim for his life. … The fact that the Emperor was now Christian and that the ‘world’ was coming to know the cross as a sign of temporal power only strengthened them in their resolv. … [E]ven the cenobite knew that his Rule was only an exterior framework with which he was to help himself build the spiritual structure of his own life with God (Merton 1960, 3 and 6).

Merton’s image of fiercely independent hermits and other monks parallels a popular view of the early monk as total outsider, lacking the education of the traditional governing elite and cultivating an innocence of the social system, current events and machinations that kept

1

This pairing of commemoration and the evocation of presence is also at the heart of the broadly conceived recent study by Katherine Marsengill (2013), Portraits and icons: Between reality and spirituality in Byzantine art; see 158–81 on portraits of monks, and 172–79 and 249–52 on the Late Antique Egyptian portraits. However, where Marsengill and others have focused on the importance of the face in portraiture, our attention should not neglect the representation of the body, as has been so persuasively demonstrated in Patricia Cox Miller’s (2009) The corporeal imagination: Signifying the holy in Late Ancient Christianity. Indeed, monks were deeply attuned to bodily experience and appearance, including the clothed body, and as I argue elsewhere, following Evagrius Ponticus, monks arrayed themselves

the elite in power. Moreover, this view still obscures branches of some Late Antique ascetic practices despite the influential refutations by cultural historians that began to appear over a generation ago (Goehring 1992 [1999]; Rubenson 1995). One of these practices, I argue here, was the traditional and continually evolving system of dress that characterized the monk in relation to both monastic society and men of the world. Yet, Merton’s phrasing offers an apt alternative to what has become a standard simplistic image of the monk as outsider: clothing functioned as one element of the visible ‘exterior framework’ with which the monk shaped his ascetic self within monastic society (e.g., Coquin 1992; Giorda 2007; 2011; 2012; Krawiec 2014). Monastic mantles, in particular, utilized secular conventions of dress to signal authority and status within the monastic sphere and in the secular sphere. Mantles identified monastic leaders much as they identified elites of the Christian Roman empire, and as they continued to do later during the early Islamic Arab caliphates (Thomas 2018). Monastic mantles expressed with fine distinctions new ways of conceiving authority. Late Antique and Medieval portraits of Desert Fathers are at the heart of this study. Painted on the walls of cells, assembly rooms and churches in Late Antique Egyptian monasteries, the portraits commemorated generations of monks and evoked their presence for subsequent generations of monastic viewers (Fig. 1).1 The painted portraits reveal variations of the

effectively in the symbolic attributes of their character and monastic life (Thomas in progress). In consequence, my subject in this chapter is not the portrait images as a whole. Indeed, much more could be said of these portraits about their realism or abstraction or, more importantly, their attributes, contexts and devotional functions. Instead I focus on the operation of clothing to evoke the presence of monastic teachers from the beginning of monasticism, as evident in the mid-fourth-century Life of Anthony by Athanasius (Thomas 2019 and infra), and as it is related to the gifting of monastic clothing (Thomas in progress), paying special attention to the motif of the mantle as it articulated particular conceptions of authority.

248

T. K. THOMAS

Fig. 1: Map of Egypt, highlighting sites mentioned in this paper (Map: Allyson McDavid).

FASHIONING ASCETIC LEADERSHIP

type, shape, size and ornamentation of mantles that express monastic ideals and respond to the evolving social and political milieu for monastic leadership. The painted portraits from monasteries in Egypt present insider views that are at odds with a now outdated rigid distinction between the pioneering anchoretic asceticism of Antony (251–356) and the communal way of life devised by Pachomius (292–348) as the founding father, or apa, of coenobitic monasticism. That distinction has obscured the remarkable diversity of early monasticism that belongs to the long, sinuously sprawling genealogies of monastic fathers reflected in the portrait programmes of early monastic practice, which encompassed a spectrum from anchorites, that is hermits, to coenobites, those monks who lived by the shared rules of their communities, which were more closely associated with village life (e.g., Wipszycka 2009). Already in the fourth century, pilgrims travelled great distances from around the Mediterranean to see the abbas and ammas of Egypt, to learn from them and write about them to spread the models of their righteous lives and their teachings. Some, like Antony and Pachomius, wielded ascetic and pragmatic authority from their own monasteries whereas others, such as Serapion of Thmuis (bishop c. 339–60), also wielded ecclesiastical authority after their appointments as bishops. In Egypt and across the increasingly Christian Roman empire, the monk-bishop merged ecclesiastical and monastic authority (e.g., Sterk 2004; Rapp 2005; Giorda 2009). In recent years, scholarship across several disciplines has brought to light the singularly important Apa Shenoute (348–466), the archimandrite (i.e., the leader) of a federation of coenobitic monasteries, whose legacy in Egyptian Coptic- and Arabiclanguage traditions came to eclipse that of Pachomius (e.g., Krawiec 2002; Goehring 2006; Emmel 2008; Layton 2014). Through his remarkably extensive writings as well as his leadership, Shenoute breached the walls of his monasteries to be a force in the region. Notables of the city of Panopolis (in whose territory the federation was located), high-ranking military officers and leaders of the Church either sought him out or had to reckon with his imposing personality. Shenoute furnishes a prime example of the coenobitic monastic leader at the intersection of spiritual, ecclesiastical and worldly spheres. Portraits of Shenoute and other monastic fathers spoke, in part through the traditional language of dress, to audiences across these different

249

spheres (Bolman, Davis and Pyke 2010; Bolman and Szymańska 2016). I chart the fuller range of that Late Antique tradition through representations of mantles in the sixth- to seventh-century portraits of the founders of two impressively extensive and closely linked monasteries that combined cenobitic and anchoretic practices: that of Apa Apollo, whose long life spanned the fourth century, and whose monastery was near the ancient city of Hermopolis Magna (and nearer the modern-day village of Bawit); and Apa Jeremiah, who, in the later fifth and early sixth century, led a monastery established within the necropolis for the city of Memphis (modern Saqqara). Not simply literal depictions of actual clothing, these images belong to a fully developed pictorial rhetoric of dress. In conjunction with texts about monastic dress and artefacts of monastic dress, the portraits enhance our understanding of the formative role of dress in the construction, expression and performance of monastic authority. Then, taking up a complementary diachronic view, I turn to a group of later mantles and a series of subsequent portraits of Shenoute to chart changes in the pictorial articulation of monastic leadership in clothing expressions that responded to the political and cultural changes from the Late Roman empire to the early Islamic empire and beyond. Portraits of monastic fathers in Late Antique wall painting At the monastery of Apa Jeremiah, a group portrait with Apollo was painted in the sixth or seventh century fairly high up on the northern wall of Room A, the first room of the monastery to be excavated (e.g., Quibell 1908, 63–64 and pl. XLIV; Rassart-Debergh 1981, 40–41 and fig. 15; Rassart-Debergh and Debergh 1981, 187–92) (Fig. 2). This and many similar rooms likely accommodated assemblies for teaching, prayer and liturgical commemorations of monastic fathers of the past, many of whom were depicted in painted portraits on the walls and in the small apses and smaller prayer niches that faced east. In its present fragmentary state in the Coptic Museum (inv. no. 7951), the composition from Room A presents an array of idealized portraits of monastic leaders who were originally identified by painted inscriptions, that is, dipinti (Dilley 2008; 2016). Signs of distinction among the four standing fathers include scale, pose, attributes and fine clothing, especially their mantles, as shown in this illustration. The

250

T. K. THOMAS

Fig. 2: Wall painting of portraits of monastic fathers from the northern wall, Room A, Monastery of Apa Jeremiah, Saqqara (Memphis), sixth to seventh century (Drawing Allyson McDavid; photograph: Quibell 1908, pl. 44).

FASHIONING ASCETIC LEADERSHIP

251

Fig. 3: Detail of portrait of Apa Apollo in the painted group portrait of monastic fathers from the northern wall, Room A, Monastery of Apa Jeremiah, Saqqara (Memphis), now in the Coptic Museum, inv. no. 7951, sixth to seventh century (Photograph: the author).

two at centre are identified as, at left, ‘Our Father Apa Macarius the Spiritbearer’ and, at right, ‘Our Father Apa Apollo the Great’. The poses of the standing fathers alternate between arms raised in prayer and arms held at the chest to present the bejewelled covers of codices. Three of the standing fathers wear long tunics decorated with small motifs repeated along paired vertical bands that are similar to the ornamental bands (clavi) worn in the secular world, except that these are quite thin and articulated by parallel vertical alignments of dots and crosses made up of dotted or triangular cross arms extended by dots. Their mantles, the large rectangular type known as the himation in Greek, pallium in Latin, are also decorated with these crosses (Fig. 3). The one exception is the hermit clothed only in his hair, which is so thick and luxuriously wavy that it echoes the drapery folds of the others’ garments. Each of the clothed fathers wears his mantle slightly differently: notably, Apa Apollo, singled out for his holiness by three large crosses around his haloed head, wears his mantle crossed over his chest. At right, a young father (no identification survives), much smaller

in scale and, therefore, in relative importance, prostrates himself before four of his precursors, bowing down to their sandal-shod feet. His shawl is decorated with fringe along the edge but is in other respects similar to the mantles worn by the elders. Clothing relates them to each other, following longstanding Roman tradition, here identifying them as members of the same monastic family (Kampen 2009). A contemporaneous painted portrait of Apa Jeremiah from the monastery of Apa Apollo was painted high up on the eastern wall of Room 20 (Thomas 2018) (Fig. 4). Although the painting no longer survives, one detailed photograph from the published excavation report shows Jeremiah wearing a relatively skimpy, light-coloured mantle over a meagre, dark tunic that does not quite reach his ankles (Maspero 1931, 8, 31–32 and pls 31–34) (Fig. 5). The tunic is marked with a cross at the centre of the chest. Jeremiah stands on a tall, narrow podium and holds a staff. Painted in the apse on the eastern wall is a two-zoned composition of Christ in majesty above; below, Mary the mother of God, flanked by apostles; and, behind them monastic fathers dressed

252

T. K. THOMAS

Fig. 4: Eastern wall, Room 20, Monastery of Apa Apollo, Bawit (Hermopolis), sixth to seventh century (Photograph: Maspero 1931, pl. 31, A).

identically to the apostles in the long tunics of ancient Roman days and Late Antique formal ceremony. Dipinti still identify some of these figures (Thomas 2012). These early painted portraits of monastic fathers from semi-coenobitic institutions present a diverse array of ‘habits’ that overlap the boundaries of the too-tidy categories of Antonian-anchoritic and Pachomian-coenobitic monasticism (as noted by Innemée 1992, 91). The idealized portrait with Apa Apollo presents calculated variety in uniform dress (and undress) to signal affinity and lineage. The same holds true for the portrait of Apa Jeremiah, which would have been seen as part of a larger programme including, in the apse next to him, the portraits of monks whose garments signal their affinity to apostolic character. The portraits were also grounded in other longstanding secular traditions and cannot be comprehended fully without knowledge of those traditions as they

continued into the Medieval era and later. Adaptations of the early Egyptian traditions in Europe are evident already in the writings (in Latin) of John Cassian (c. 360–435), a former disciple of the Desert Fathers of Egypt. Cassian’s Conferences, which recounted his studies with the Desert Fathers of Scetis in Egypt, and his Institutes on monastic rules and customs, beginning with his presentation of Egyptian monastic dress, in the sixth century continued to inspire European monastic practice and the systematized habit of Saint Benedict, adopted later by the Medieval Cistercian order, and again in the eighteenth century by the Trappists, joined in the mid-twentieth century by Thomas Merton. Merton, in turn, wrote compellingly about his precursors and located the roots of his own ascetic vocation in early Egyptian monasticism. The remarkably longlived Western traditions were paralleled by developments of monastic dress in Egypt, as is clear in the

FASHIONING ASCETIC LEADERSHIP

Fig. 5: Portrait of Apa Jeremiah from the eastern wall of Room 20, Monastery of Apa Apollo, Bawit (Hermopolis), sixth to seventh century (Photograph: Maspero 1931, pl. 34, B).

series of painted portraits to be discussed below. In Medieval Egypt, the forms of the habit remained fluid, enjoying an instability that allowed for greater nuance and for different understandings from perspectives inside and outside of monastic society. Mantles of authority in Late Antique vestimentary systems These Late Antique monastic portraits distinguished saints of the past by their clothing. The portraits should not be taken as mere documentary snapshots of what the Egyptian monastic fathers actually wore but as reflections of who they were. The images are purposeful

2

The term schema preceded monastic usage, as in Dio Chrysostom, Discourses 72.2, ‘On personal appearance’ (Peri tou schematos) (Crosby 1951). The fullest sustained discussion in relation to Late Antique and Byzantine dress may be found in

253

rhetorical expressions, much like their written counterparts in hagiography. In images, in texts and in actual garments, as discussed below, the rhetoric of the monastic mantle of authority cultivated its roots in more ancient and widespread systems and styles of dress. Together, tunic and mantle constituted the ‘suit’ of Antiquity and Late Antiquity. The mantle, however, defined the outer, visible, characteristic form of the wearer. Of the different types of mantles current in Late Antique secular dress for men, three are particularly relevant to the representation of secular authority: the chlamys of high-ranking military and imperial officials; the civic toga; and the himation–pallium of philosophers and intellectuals (Fig. 6). These three types of mantles were so prevalent among the honorific statues erected in Roman and Late Antique public civic spaces all across the empire, and so important to the characterization of their subjects, that honorific statue types—chlamydati, togati and palliati—were named after the mantles (Smith 1999; Gehn 2012). Indeed, the figure constructed by the outer garment is schematic, diagramming character. Schema in Greek, habitus in Latin, referred to outer visible form, and although the word later became synonymous with monastic appearance, in Late Antiquity not only monks wore ‘habits’.2 Chlamys, toga and himation rendered the authoritative character of the wearer visible. To perceive that character one had to recognize the differences between the shapes of the mantles and how they were worn. As it can be difficult to distinguish between tunic and mantle, especially in a line drawing, the illustration (see Fig. 6) picks out the mantles in grey. Indeed, polychromy of the statues would have rendered clothing motifs more legible (Gehn 2012, e.g., col. pl. and p. 22). The chlamys was semicircular, worn fastened at the right shoulder and opening over the right arm to leave it free, as was appropriate for men of action. In contrast, as shown in another Late Antique honorific statue from Aphrodisias, Late Antique civic dignitaries wore for various formal occasions different forms of another semicircular mantle, the toga, which, intricately draped and folded, was extremely difficult to

Kalamara 1995. As concerns monastic modes of dress, see Thomas 2012; Kiilerich 2015 offers a more systematically comprehensive use of the term in the history of Late Antique art.

254

T. K. THOMAS

Fig. 6: Togatus (a), palliates (b) and chlamydatus (c) statue-monuments from Aphrodisias (Asia Minor), late fourth to fifth century (Drawing: Allyson McDavid after Smith 1999, figs 2, 5, 6).

put on and to wear properly; the several Late Antique versions of the toga had developed from the traditional dress of the Roman citizen (male, of course) (Smith 1999). The himation was a simple rectangular mantle of Greek origin that in Roman culture came to be indicative of a life of learning. Within the rhetoric of the venerable dress system of Late Antiquity, these mantles shared in the long tradition of changes of clothing, often with ritually gifted garments that indicated change of status or role: the chlamys as an imperial gift, whether to high-ranking officers and

governors or to foreign rulers; the toga virilis from father to son when the son came of age; and, more rarely, the philosopher robing his advanced students in the outer garment, himation or tribon, representative of his school (Tellenbach, Schulz and Wieczorek 2013; Urbano 2013; 2014; Rollason 2016). The mantle-statues illustrated here, preserved with their inscribed bases in the city of Aphrodisias in Caria, Asia Minor, present striking comparisons to the painted monastic portrait representing Jeremiah standing on a pedestal that resembles a statue base. Certainly,

FASHIONING ASCETIC LEADERSHIP

255

correlations between statues and statuesque appearances had long been cultivated in public behaviour and represented in art, a trend that intensified in the striking tableaux of Late Antiquity (e.g., MacMullen 1964; Stewart 2003). The three types of statues also present an illuminating comparison to the group portrait with Apa Apollo, in which the fathers wear their himationmantles in different, characterizing configurations: the two carrying books wear their mantles wrapped around the waist and over the shoulder and arm to display affinities to the venerable palliatus type; Apollo wears his wrapped across the chest in apparent allusion to the voluminous old-style togatus type (as opposed to a more streamlined newer-style toga) (Smith 1999).3 Among the fathers in this group portrait, in conjunction with other attributes (e.g., book, pose) and the way it is worn, a single type of mantle expresses varieties of spiritual authority. As in this group portrait, the chlamys and the up-to-date Late Antique variant of the toga, the loros, do not characterize monks in other programmes of monastic wall paintings, but dress instead figures of soldier-saints and archangels. The monastic himation in the portraits with Jeremiah and Apollo displayed no allegiance to the governance of state or city, but did adopt longstanding associations to philosophy, teaching and the authority of wisdom gained from living the life of true philosophy (Urbano 2013; 2014). In his De pallio Tertullian (155–240) had already made this case for the pallium as a mantle for Christians, a redefinition of the garment expressed visually in images, where Christ and his disciples, the apostles, are shown wearing the himation–pallium (e.g., Zanker 1996; Brennan 2008; Hunink 2005). Those associations are evident in the portraits of the northern wall of Room A at Jeremiah’s monastery and especially in those of the painted apse of Room 20 of Apollo’s monastery, where monastic fathers are represented as latter-day apostles by their portrayal in oldfashioned first-century garb amid the ceremonial gathering of apostles. A wide range of Late Antique texts, from pilgrims’ accounts to saints’ lives, indicate that some monks actually wore the himation, whether displaying large and expensive expanses of fabric, like the

formal wear represented in palliati statues and the group portrait from Saqqara, or notably small, like Apa Jeremiah’s mantle in his portrait from Apollo’s monastery, adopting a sign of material poverty to express ascetic renunciation of elevated status in the secular world.4 Monks and monastic fathers are portrayed consistently in the wall paintings wearing clothes in bulky, old-fashioned styles or in skimpy garments that contradict other signs of material wealth and high status (such as fabric, colour or ornament). The monastic adaptation of the himation-mantles and an emphasis on the past are paralleled in the portrait paintings by the depiction of scriptural mantles of authority. Critically important is the mantle of scriptural prophets, the melote, as known from the Hebrew Bible and the Christian Old Testament, as well as the canonical books of the Christian New Testament and apocrypha produced throughout Late Antiquity. As considered below, some monastic leaders are characterized as latter-day prophets by their association with the melote, a mantle made of fleece or leather like those worn by prophets (Innemée 1992, 106–7). A paradigmatic example of this fourth type of mantle was Elijah’s melote, which, as he ascended to heaven, he let fall to his disciple, Elisha. The gift of the melote marked the transfer of prophetic authority from Elijah to Elisha, replacing Elisha’s himation (in the Septuagint II Kings 2:13) (e.g., Carroll 1969; Thomas 2019). One example of the continuing relevance and scriptural importance of these two mantles in Late Antique pictorial tradition is the portrayal of paradigmatic proto-monks in the sixth-century apse mosaic of the church of the Holy Monastery of St Catherine on Mount Sinai, where the prophet Elijah, as the ‘prophet of the desert’ (I Kings 19), is shown wearing his shaggy melote and Moses the himation, likely also an allusion to the wisdom of Moses that was to inspire Greek philosophers (e.g., Hanfmann 1951, 220; Urbano 2016, 30; Thomas 2019) (Fig. 7). As the church and apse mosaics were the product of imperial patronage (commissioned by the emperor Justinian between 548 and 565), this rhetoric of ascetic mantles reflects

3

4

Apollo’s crossed mantle also bears a resemblance to the Medieval monastic garment known as the schema, see, e.g., Innemée 1992, 126.

Most useful for the sheer density of attestations to monks wearing the himation may be the Historia monachorum in Aegypto (Russell 1981; see also Thomas 2012; Thomas 2019).

256

T. K. THOMAS

Fig. 7: Apse mosaic rendered to emphasize mantles the Holy Monastery of St Catherine, Mount Sinai, sixth century (Drawing: Allyson McDavid. Photograph: Michigan-PrincetonAlexandria Expedition to Sinai).

imperial engagement with local traditions, and a viewership of pilgrims to the monastery as well as local monks. Moreover, this sartorial expression of nuances of spiritual authority not only informs the portraits at the monasteries of Apollo and Jeremiah and, as shown below, those at Shenoute’s monastery, but it is also embedded in monastic hagiography and the widespread, long-lived monastic practice of the gifting of garments. In one sense, the old-fashioned style of the himation and the anti-fashion of the melote would appear to refigure the statuary that honoured and commemorated illustrious men in the Roman city, cornerstone of the Roman empire, within the emerging conception of the desert as successor to the city. Against this background, these portraits would have furnished visual proof of Athanasius’ assertion in his Life of Antony (14): ‘and the desert was made a city by monks …’ (Gregg 1980, 42–43). Athanasius, archbishop of Alexandria, likely composed the biography soon after Antony’s death in the mid-fourth century, presenting his subject as the earliest monk to take up both the himation and melote. In the biography, a narrative portrait—as in the story of Elijah and Elisha—of the gifting of mantles signals

the recognition of authority and articulates powerful bonds between men of God. According to the biography, Athanasius had given the himation to Antony in recognition of Antony’s spiritual wisdom and authority. In return, on his deathbed Antony gifted the himation back to Athanasius along with his melote, thereby recognizing Athanasius’ own work on behalf of monasticism, the basis of Athanasius’s ascetic authority. More than a sign of ascetic renunciation, in this context the mantle functioned as a social sign and as an outer reflection of inner character. In addition, the re-gifted mantle, worn out from use, carried with it signs of the life of Antony that enabled it to serve as a sign of the transmission of his wisdom, teaching and monastic practice. Thereafter, across the generations of monastic culture, as monastic father invested monastic son, making him a monk, gifted items of clothing signalled the monk’s monastic lineage (Thomas 2019). In contrast to the transfer of authority by the mantle in the Late Antique Egyptian monastic context, the monastic hood took symbolic meaning from the hoods worn by children, to express rebirth into the monastic family and the virtue of humility (e.g., Innemée 1992). Mantles and hoods were not ubiquitous across all forms

FASHIONING ASCETIC LEADERSHIP

of monastic practice or, it seems, all levels of ascetic spiritual advancement. Both, however, figure in Shenoutean (and Pachomian) monasticism. The following discussion of mantle and hood in coenobitic dress draws upon current understanding of the history and import of Shenoutean monasticism and Shenoutean monastic art as presented so powerfully in Elizabeth Bolman’s important edited volume, The Red Monastery Church: Beauty and asceticism in Upper Egypt (2016). By the time Shenoute began his leadership as archimandrite in the later fourth century, the three congregations of the federation were already quite large, with a population numbering in the thousands. The federation and its territory comprised three monasteries, one for women near the village of Atripe in the south, and two for men, the centrally located White Monastery and the Red Monastery to the north, named after their impressively monumental white limestone and red brick churches. Shenoute himself was responsible for the first and larger of these churches at the main establishment, the White Monastery, for the increasing prosperity of the federation and for the engagement of the federation with the world outside.5 Shenoute led his coenobitic federation as archimandrite from the White Monastery, the largest of the congregations and at the head of the federation, even as he lived most of his life as a hermit in his own cell in the nearby desert. Indeed, the inclusion of advanced, usually elder hermits within an otherwise strictly regulated communal practice is another indication of the creative and distinctive variations possible along the spectrum from the anchoritic to the coenobitic.6

5

6

7

Did Shenoute travel with Cyril of Alexandria to the Council of Ephesus and visit Constantinople? Shenoute’s hagiography stresses his relationship to the imperial court: e.g., Lopez (2013, 28) reports that he did travel to Constantinople; Schroeder (2007, 24) suggests he may have done so, and with high-ranking dignitaries in Egypt, such as Caesarius, the imperially appointed governor and benefactor of Shenoute’s new church at the White Monastery. This is evident from passages of Shenoute’s Canons, as in Layton 2014, 281, Rule 441. E.g., Boud’hors (2009) and Maguire (2003, 43–44, citing Rutschowscaya 1990, 131) connected this element with the mantle handed down to Elisha by Elijah through a Late Antique textile hanging in the Abbegg-Stiftung (inv. no. 2439). That hanging has been tentatively assigned a date between the fifth and eighth centuries based on comparisons to the ornamental vocabulary in the paintings at Bawit and mosaic pavements across the Greek East (Schrenk 2000, 293–306 and colour pls 28–31; see also

257

Portraits of the fathers of the Shenoutean federation in the Red Monastery Church The habit of Shenoute’s coenobitic practice is seen in paintings in the church of Shenoute’s successor, Pshai (or Bishai), at the Red Monastery, a smaller, sixth-century architectural ‘copy’ emulating the triconch design of the enormous fifth-century church building at the White Monastery. At the Red Monastery Church, in the central register of the northern lobe of the sanctuary (Fig. 8), is a portrait gallery of similarly dressed fathers of the federation, each in his own niche, part of a third, later sixth-century phase of painting (Fig. 9). Dipinti around the heads of the figures identify them as the earliest leaders of the monastery: ‘Abba Besa, archimandrite’, who was Shenoute’s successor; ‘Abba Shenoute the archimandrite’; ‘Pcol, the Forefather, archimandrite’; and ‘Abba Pshai archimandrite’ (Dilley 2008; 2016). All wear the same kinds of garments: a voluminous belted tunic, a pocketed apron on a long strap, a version of the melote, that was worn over the shoulders and across the chest—expressly indicative of the monastic state7—and a small mantle, tied at the centre of the chest, ornamented with a crossrune at the shoulders.8 Differences among their tunics and mantles lie in colour only: various shades of browns and yellows might reflect the variation of the natural colours of wool and hides or, perhaps, point to a remembered change in the habit over time. This is what we know of the monastic habit in Shenoutean coenobitic practice. Tunic, belt and bag are attested in writing and images; artefacts of such items have been

8

Schrenk 2004). Eunice Maguire (2003, 44) suggested that these items fulfilled the function of aprons when monks were weaving baskets by protecting their tunics from sharp strips of palm leaves. This garment is often identified as an apron, in reference to the Coptic term rahtou because in examples from some sites, as in later seventh-century burials from the semi-anchoritic site of the Monastery of Epiphanius, the pocket is so small as to be vestigial, making it impossible to carry anything in it, see Winlock and Crum 1926, 76–77; Castel 1979. In the Pachomian Rules (Veilleux 1981, 2:151), monks receive sweets in a goatskin bag (Rules 37–38). Maguire (2003, 44) and others have noted that this element is also called the schema (Torallas Tovar 2007, 222; Innemée 1992, as in fig. 10 and p. 182). Is this the hoeite-garment that remains unidentified in the Canons (Layton 2014)? For a more likely alternative, see n. 14. In some written traditions, the tunic is identified as the schema of the monk (Torallas Tovar 2007, 220).

258

T. K. THOMAS

Fig. 8: Overall view of the northern lobe of the triconch sanctuary, Red Monastery Church, in the monastic federation of Apa Shenoute the archimandrite, Atripe (Panopolis), sixth to seventh century (Photograph: Arnaldo Vescovo, © American Research Center in Egypt).

FASHIONING ASCETIC LEADERSHIP

Fig. 9: Detail of Shenoute from portraits of monastic fathers in the register below the apse, Red Monastery Church, sixth to seventh century (Photograph: Elizabeth Bolman, © American Research Center in Egypt).

found at monastic sites in the same region.9 The small mantle is often attested in visual representations and should probably be identified with the maphorion (Greek), also known as the mafort (Coptic), which covered the shoulders and back in several sayings of

9 10

See note 7. Over the past twenty-five years, a number of scholars have mined textual sources for the terminology of the early monastic habit in Egypt (e.g., Morfin-Gourdier 1991; Innemée 1992 and 2005; Coquin 1992; Torallas Tovar 2007; Giorda 2007; 2011; Layton 2014). Still, study of the habit through textual sources is complicated by multiple terms apparently referring to the same element and, often, by a single term referring to multiple items, and by inconsistent usage within a given language and in translations. Several scholars have noted that the same word may be used to refer to a garment that changed its form over time.

259

the Fathers.10 The hood, which was to become ubiquitous in monastic portraits of the Medieval era (a practice enduring today) is not attested in Late Antique portraits of monastic fathers.11 Within the larger programme, Shenoute is represented not only as one of the founding fathers of the monastery, but also as one among the ranks of the blessed. This is expressed, in part, by the compositional logic of association articulated by placement. In the register of larger niches directly below the portraits of leaders of the monastery are portraits of archbishops of Alexandria, again identified by inscription. Moreover, Shenoute has the white hair and beard of a venerable elder, a monastic Old Man, Father, Abba (see Fig. 9). His lined brow indicates both age and thought. His eyes are wide open—the whites show below and almost all around. These are features long associated with portraits of Late Antique visionaries, including teachers of old, and holy men (Smith 1999, e.g., 143–44). His gesture carries multiple possible meanings: blessing, speaking and pointing. He wears a stole and carries a staff that elders of the monastery carried as a symbol of their authority (Layton 2014, 255, Rule 388). The apse painting presents the court of heaven within a setting of fictive, painted architecture that plays with the elements of actual architecture (Fig. 10). The central motif of Mary nursing the infant Christ is surrounded by figures of angels, Old Testament prophets and New Testament apostles. In this composition, the theme of the habit comments upon the relationship of monks to such saints more explicitly than seen so far in the paintings from the monasteries of Apollo and Jeremiah. The apostle Paul, next to Mary at right, wears the himation and is further characterized by the attribute of scrolls. Peter, next to Mary at left, wears a long, clerical stole over a habit like that worn by the fathers below, only his cross-rune is slightly different. He is not only the rock of Christ’s Church, he appears as a column of this church and, in Shenoute’s symbolic,

11

Moreover, the texts served a wide range of purposes, from terse receipts to detailed dress regulations to sermons. For textual references, see Ward 1975 at, e.g., 116, attributed to Abba Cronius; Innemée 1992, 211; Torallas Tovar 2007. Karel Innemée (1992) drew attention to the ubiquitous small mantle (105) and the missing hood (98 and 101) in monastic portraits, but without acknowledging that the portraits represent monastic fathers rather than their spiritual children. I suggest that the Pachomian rule to wear the hood at synaxis referred to little brothers rather than fathers.

260

T. K. THOMAS

Fig. 10: Detail, left side, of the painted northern lobe of the apse with the apostle Peter as a monk-cleric column-figure, Red Monastery Church, sixth to seventh century (Photograph: Elizabeth Bolman, © American Research Center in Egypt).

corporate conception of the federation, as a member of this congregation.12 As in the apse mosaic at Sinai, mantles are key markers of ascetic authority and genealogy. The column-figure identified by inscription as ‘Apa Elijah the Tishbite’ here wears the himation(!) even as he represents the Old Testament prophet who, as told in Besa’s Life of Shenoute, had given his melote

(meaning the mantle or the apron-bag or both?) to Shenoute through the intermediary of an angel when Shenoute was as a young boy first vested in the monastic habit by his uncle, Pcol.13 The spectacular extended programme of the Red Monastery Church, thanks to meticulous restoration, still encompasses the multiple levels and apses of the

12

13

Indeed, Bolman and Szymanska (2016, 169–70) interpret the column figure as doubling the portrait of Pshai, whose Greek name was Peter.

Bolman and Szymanska (2016, 168–69) note that the doubled portrait of Elijah-Shenoute would have been paralleled by that of Peter-Pshai as in note 12 above. See Bell 1983 on the memory of this miraculous gift in Besa’s Life of Shenoute; and Boud’hors 2009 on the melote referring to both items.

FASHIONING ASCETIC LEADERSHIP

triconch sanctuary, including pendant column-portraits that further enlarge the network of associations linking monastic and ecclesiastical leaders to prophets, patriarchs and apostles of the scriptural past. Throughout the programme, the appearance of the habit can be seen to mark identity across the historical time of the monastery, and across scriptural and eschatological time, and the mantle, in particular—himation, melote and the small monastic mantle, mafort in Coptic and maphorion in Greek—to characterize types of spiritual and ascetic authority.14 The meagre size of the small mantle, signalling ascetic renunciation and ascetic authority, is an inversion of the mantles worn by worldly elites of Late Antiquity. In this, the programme as a more elaborate and specific presentation of the theme of monastic identity articulated by dress in group portraits is closer to the apse composition of Justinian’s church in the monastery at Mount Sinai than to those at the monasteries of Apollo at Bawit and Jeremiah at Saqqara (as discussed above). The Shenoutean habit Monastic portraits all along the length and breadth of Egypt figure monastic leaders through dress, offering insider views of monastic appearance and identity. I turn below to the comprehensive critical edition of the rules of Shenoute’s federation as published recently by

14

15

Although the identification of clothing terms with the garments represented in the paintings is not clear, perhaps the hoeite-garment is to be associated with a mantle of authority, based on such mentions as those found in Rules 100 and 101 (Layton 2014, 129): Rule 100: Therefore, not only shall I command the siblings who are of one mind with us to take care of the words that we write in that book and the things written in all the epistles, but also I shall command them to preserve my hoeite-garments that I have torn many times and finally tore in two when I was greatly grieved even unto death. Rule 101: Therefore, I shall command the siblings who are of one mind with us and who are alive at the time of my death to lay down my torn hoeite-garments in the presence of those who are coming after us and of all who listen then or now to all the words written in that book or those written in the epistles that we have written or shall write; so also they will say, What are these torn hoeite-garments and all these written words and all these curses? The Pachomian heritage of Shenoutean monasticism is discussed throughout Goehring 2006. Layton (2002, 28 n. 19) provides citations of Shenoute’s use of the Pachomian term koinonia to describe his congregations. In the single known mention of Pachomius among Shenoute’s Canons, Shenoute

261

Bentley Layton (2014) to consider social resonances of prescribed garments within the monastic community. This and past studies by Layton (2002; 2007) Rebecca Krawiec (2002) and Caroline Schroeder (2007) have shown that the federation’s founder, Pcol, drew upon the model of regimented coenobitic life that was the innovation of Pachomius, his predecessor in Upper Egypt. The Pachomian model included a similarly rigorously uniform habit, which along with how the monk wore his clothing (following rules scrupulously, for example, in what he wore, where he wore it, and how he behaved at prescribed tasks and events) constructed the monk’s visible outer form, his schema.15 This schema was to be maintained both inside and outside of the monastery, identifying the wearer as a monk and not a secular, and in the details of garments and comportment identifying him as a member of this— Shenoute’s—federation and no other, consequently separating the monk from even closely related monastic cultures, such as that of the Pachomian federation.16 Within Late Antique coenobitic monastic practice, uniformity is key to understanding the habit’s nuanced expression of meaning. Whereas semi-coenobitic and anchoretic practice allowed brothers to differ in appearance, in stricter coenobitic practice variety would have marred the appearance of the pure corporate monastic body (Krawiec 2009, 13 n. 4).

16

names Pachomius an authoritative predecessor (e.g., Layton 2002, 28 n. 19; Layton 2014, 20). Just as Layton noted the striking absence of references to Pachomius, the iconography of the paintings in the Red Monastery Church sanctuary makes no explicit reference to Pachomius. The curse-like phrasing of some of the rules in the Shenoutean Canons differs notably from the phrasing of the corpus of Pachomian rules (e.g., Layton 2014, 42–44). The relation of the Shenoutean to the Pachomian habit is not entirely clear. NB: Karel Innemée (1992, 91), writing prior to the revolution in Shenoutean scholarship, attributed coenobitic monasticism in the Thebaid to Pachomius, characterizing the region as the ‘Pachomian Thebaid’. That said, we have to recognize that Pachomian monasteries were spread across the Qena Bend region and found as far north as the Delta near Alexandria (Tsirpanlis 1986). Layton 2014, 81–84, on ‘Rules and the acquisition of monastic authority’, and 84–88, on ‘Maintenance of monastic identity’. On the constant maintenance of monastic identity even at the expense of speaking with blood relations, see, e.g., Canon 38 at p. 105 and Canon 260 at p. 199; on the strict uniformity of habit and identity, see, e.g., Canons 299–301 at p. 217; despite rank, Canon 390 at p. 257.

262

T. K. THOMAS

From inside the world view of Shenoute’s coenobitic federation, when a monk looked at his brother he saw an image of himself. He also saw a part of his corporate body as constituted by him and his brothers and elders. As Shenoute put it: ‘We and you are limbs to our companions and we form a single body in our Lord Jesus Christ’ (Krawiec 2002, 67). The monk’s outer uniformity was dynamic, too, in that it helped to shape his inner character as he strove to follow the rules, to be like his brothers and to live a life without sin.17 Caroline Schroeder (2004) has written evocatively of the conception of the corporate body in Shenoute’s monastic ideology. If a monk were to commit a sin, for example, he sinned against himself and his fellow monks, infecting the corporate body as much as his own physical body. So, if a Shenoutean monk were to sin without regret and compunction, without a change of heart and behaviour, then he needed to be cut out like an infectious, diseased limb and expelled from the monastery for the health and purity of that corporate body (Krawiec 2002, 8, 29, and 155). The Shenoutean habit known from the apse paintings of the church would have worked not just as a mirror between brothers, it would have acquired meaning as it was actually worn in Shenoutean monastic society and in the built environment of the monastery now known from archaeology and texts. That social setting can be visualized as at least three concentric social zones (Layton 2007) (Fig. 11). The inner zone comprised the church building and main activity centre for the monastery, including a refectory, kitchen area, residence halls and infrastructure built within an enclosing wall (Grossmann et al. 2004). The wall and gatehouse comprised a middle zone. The gatehouse accommodated visitors and prospective monks; only vested members of the community and special guests could

17

18

This idea is deeply embedded in Christian ascetic theology. Clement (150–215), for example, had already linked virtuous dress worn during life with the ‘immortal vesture’ worn in the court of heaven in his The Instructor, e.g., book 2, chapter 11, ‘On clothes’. All Christians, Clement made clear, must confront this choice because what is worn now makes an impression, and not only for a person’s social life, but more importantly, because it makes an impression on the person’s soul. The continuation of this line of thought is addressed in Thomas (in progress). Each monk swore to renounce his prior life in the secular world—possessions and relationships—and agreed to submit to the coenobitic way of life handed down by his monastic fathers through obedience to written rules and the teaching of his elders

advance beyond the gatehouse and wall into the interior of the monastery.18 In the outer zone, outside the wall, were the monastery’s agricultural installations, where many monks worked; monks also travelled on monastery business between the three establishments (on orders or with permission, of course). Here and moving farther beyond the wall into village and civic spheres, Shenoute’s monks were to maintain their appearance at all times, taking care to wear their identical habits according to the rules, just as they would in the monastery, and to walk together in silence. Their quiet synchrony and uniform appearance must have made quite an impression on all who saw them outside the walls (e.g., Layton 2014, 275, Rules 287, 430, 453–54). Thus, Shenoute’s monks were seen from multiple social perspectives. The trained gaze of the monk took in minute details of dress that could easily escape the perception of outsiders. Prospective monks were scrutinized in the gatehouse, first by the gatekeeper, to make sure they were suitable candidates, then by the archimandrite. There, the approved prospective monk was taught the basic rules, which he swore to follow, and vested in the ritual that marked the transformation of the man into a monk of the community (Krawiec 2002, 14 and 29; Layton 2014, e.g., 191, Rules 243, 265, 410). Monastic property, including clothing, was held in common, and thanks to the mediation of monastery elders, all the federation’s ‘siblings’ were to wear the same items according to Shenoute’s sermons and rules (e.g., Layton 2014, 295, Rule 472). As part of the new and totalizing identity of the new monk, within the monastery he learned closely regulated behaviour, often involving the habit, such as how to wear a given garment, what to cover (or keep uncovered) and when and how to care for a garment or request a replacement (e.g., Layton 2014, 217,

(Layton 2007, 47–50; Layton 2014, 78–80, on ‘The experience of conversion to monastic life,’ and, e.g., Canon 472 (at 295) on the new monk surrendering his old garments and Canon 469 (at 295) on the submission of clerical monks entering the monastery ‘just like all the siblings’. No traces of the Late Antique gatehouse survive. The diagram (Fig. 11) suggests a location near the church so that visitors would not have had to walk through the monastery to attend services. Some sections of the Late Antique wall were found in this southeastern area of the site; however, most of the eastern Late Antique walls have been overbuilt by the modern road, village structures and agriculture (Grossmann et al. 2004, 371–72).

FASHIONING ASCETIC LEADERSHIP

Fig. 11: Diagram conceptualizing the concentric social zones of the White Monastery (Plan: Allyson McDavid after Grossmann et al. 2004, 372, fig. A and Bolman, Davis and Pyke 2010, fig. 1).

263

264

T. K. THOMAS

Rules 299–301; 223, Rules 312–21; 235, Rule 343; 239, Rules 352, 354; 257, Rules 390, 394; 291, Rule 463).19 The gatehouse was a liminal place where a visible change of elective and ontological status occurred: where the man of the world became the monk, a member of the monastic family and its corporate body. Visible resemblance created by the uniform of the habit and by uniform behaviour demonstrated close affiliation that underscored the terminology of family, and the social and economic structuring of the monastery as a household (Krawiec 2002, 133–160). Monastic family is part of the framework for the portrayal of the fathers in the Red Monastery Church paintings. The church building could be said to constitute a special innermost zone, the locus of the great assembly of the corporate monastic body and a place of ceremony (Schroeder 2004, 494–500). The church building also came to symbolize the corporate monastic body, the community constructed of the individual bodies of monks. By analogy, the habit encompassing the monk’s

19

20

21

Mauss 1935 (1973); Goffmann 1961 (2007). For Goffman, the monastery is a kind of totalizing institution, in which each individual is stripped of his ‘identity kit’ upon arrival; this identity kit is replaced with the clothing and other apparatus of his new identity (Goffman 1961 [2007], 20). See also Layton 2007, 63–64 and 69–73. As we lack portraits of Pachomian monks, we do not know how the small mantle looked, but it was perhaps similar to the mantles in portraits at the monasteries at Bawit and Saqqara. The Pachomian rules required monks to wear long linen mantles in church at the synaxis. I have not yet found a prescription for a linen mantle in Shenoute’s rules. Nor have I found a reference to the analabos, straps that crossed the chest, which Torallas Tovar (2007, 222), identifies as the scapular. Absent from the paintings of historic fathers in the Red Monastery is the shaggy melote-mantle worn by the fictive column-figure of John the Baptist in the southern lobe of the sanctuary paintings in the Red Monastery Church (Bolman 2016, e.g. 147, fig. 10.26). Karel Innemée (1992) noted as well the very puzzling lack of any written mention of the small mantle that is quite common in Upper Egyptian representations. The written prescription and lack of visible compliance would seem to be an instance of the elements of the habit depicted not being aligned with the written habit: ‘The most striking phenomenon is, that in none of the representations does one come across a hood, while two hoods are mentioned in all of the [Pachomian] texts’ (98). He also noted that in portraits of Upper Egyptian fathers the small mantle is a common element of the habit, but this item is not mentioned in any of the Coptic or Greek texts: ‘[W]e have no representations of Pachomian monks with hoods or any other form of headdress. From Bawit (room XII, southern wall, side E) we know a wallpainting representing two monks with a cloth draped around the shoulders, which has five crosses, arranged in the shape of a

body symbolized community—its uniformity and purity—and the individual monk’s perpetual attention to living a virtuous life in the community. In these painted images, the fathers were fused to the symbolic image of the church. Just as these spatial zones indicate social, ritual and sacral zones, aspects of the habit, such as mantle and hood, indicated monastic status within a spiritual, ascetic and institutional hierarchy. Several mysteries remain about the symbolism of mantle and hood; however, it seems increasingly likely that in these early images the small mantle that appears in the painted portraits is likely the maphorion (Greek) or mafort (Coptic).20 Back in 1992, Karel Innemée noted the lack of any early Egyptian portraits showing a monk wearing a hood, another common item of monastic dress and generally understood as a sign of humility.21 The evidence presented here may help to reconstruct an alternative meaning behind the lack of hoods. In these portraits, they are absent because the subjects are monastic fathers, not their monks.22 A

22

cross on it. We could possibly interpret this as the cloth of the tiara which is draped around the shoulders’. Innemée further noted ‘[t]he small shoulder-mantle’, citing the Latin terms amictus and palliolum, but without noting any Coptic or Greek equivalents (101). He continued: ‘In all representations of Pachomian monks the shoulders are covered by a small mantle, sometimes fastened by a small loop on the chest. It is difficult to give any details about its origin or function, as it does not occur in a context that permits us to draw any conclusions. It is probably the same vestment that Cassianus calls palliolum or mafors, which according to him covers the neck and shoulders and is small to avoid the price and luxury of a wide mantle’ (105). Melote referred to a different garment, a larger item of leather or fleece, used as furnishing and as an outer cloak, see Innemée 2005, 103, 106–7. This function would also address the expectation in the Late Antique vestimentary system that a mantle be worn in public and at formal, ceremonial occasions. I do not think, as he argues, that ‘it would be an iconographical problem to present a monk with his hood on’ because the hood would cover the face and in these iconic portraits ‘the face had to be visible’ (98). Greek and Roman art had long presented an array of solutions to portraying the veiled face. Instead, the key may lie in the fact that Graeco-Roman clothing often served multiple functions. A pallium, for example, might be draped like a toga for expressive effect, in the instance of Apa Apollo from the group portrait at Saqqara, and draped like an analabos. One of the main associations of the monk’s hood, via the hoods worn by children, was to childhood. This association is attested across the full spectrum of monastic practice from anchoritic to coenobitic, and it linked the habit conceptually to the monk’s vesting, when he began his new life and learned anew how to do everything and to lead a life predicated on monastic virtue.

FASHIONING ASCETIC LEADERSHIP

trajectory to praying with uncovered head is attested in the writing of Evagrius Ponticus, who lived in the semi-anchoritic monastic community at Kellia. Evagrius, in his Letter 7 to Rufinus, suggested that the uncovered head is a mark of ascetic advancement towards an angelic state: ‘I am not like one who stands in the presence of God—I who am unworthy!— Since I am not yet competent to pray “with uncovered head” [1 Cor. 11.4] by my own authority. …’ (Casiday 2006, 60). If the small mantle (and bare head) signalled advanced status, as I suspect, then the portraits would present Shenoute and the other founding fathers in a selected, ideal aspect as elder siblings.23 Shenoute’s beautiful mantles

265

hood, or pair of shoes made for them beautifully, saying, They are ours; and without equality obtaining for all, from old to young’ (Layton 2014, 117, Canon 72). A crisis described in one letter hinged on Shenoute dramatically rending his mantle in grief and frustration at the behaviour of one of the women leaders in his federation.24 That behaviour was certainly not in line with expectations for the brothers who wore hoods. A notable trait shared by most uniform systems of dress, however, is that leaders fashion their own characterizing garb (Joseph 1986; Craik 1993; 2003; Fussell 2002). Inscriptions of authority in early Medieval monastic mantles

The painted portraits of the Red Monastery Church, in which Shenoute and the other leaders of the monastery wear the habit of the brothers of the monastery, and the tale of the miraculous gift of Elijah’s mantle to Shenoute, do not reflect the full range of mantles associated with Shenoute. Although he wears the same habit in his fifth-century painted tomb portrait (Bolman, Davis and Pyke 2010, fig. 9), Shenoute’s own writings attest to his possession of larger, decorated mantles (e.g., Krawiec 2014, 69–73). These passages help explain a repeated rule in the Canons about not coveting beautiful clothes—inexplicable unless beautiful clothes were seen within the monastery but not accessible to some brothers: ‘Cursed be any male or female among us who fixes their heart upon having a hoeite-garment, ršōn-garment, prēš-garment/cover,

Written evidence of specially made beautiful mantles in Shenoute’s monastery has corollaries in a larger group of inscribed monastic mantles published by Jacques van der Vliet (2006) from a variety of sites mainly dated between the eighth and eleventh centuries when Egypt was under Islamic rule; one example from Saqqara will suffice to introduce the larger group (Fig. 12).25 This one belongs to a subgroup of mantles decorated and inscribed with verses from Psalms in Greek or Coptic commenting on the things that reward just kingship, the righteous performance of ruling authority. No two mantles bear the same scriptural verse, although some refer to the kingship of David. Addressing the example illustrated here, van der Vliet concluded his discussion (2006, 47–48):

23

24

It is unclear whether, under the Pachomian rules, the hood was to be worn during prayer, as several rules seem to contradict each other (Veilleux 1981, 2). They wear hoods when eating (Rule 29), not during prayer (28); and wear the linen mantle for synaxis, not for work or walking around in the monastery (61); they do not wear the mantle when in the fields or village, when going to synaxis, or at the meal (102); hoods bear the sign of the community and the sign of their house (99); the monk wears the hood and goatskin at all times (91); and receives sweets (37) ‘not in his hood but in his goat skin’ (38). The relationship of hood to small mantle among actual garments is more complicated. Several hoods discovered in burials were no longer sewn, although they preserve traces of that stitching. Petra Linscheid’s (1999) meticulous discussion of hoods in secular- and monasticdress artefacts makes clear their close similarity in shape, ornamentation and size to the small mantles in the paintings. Perhaps hoods worn by new and striving monks were unstitched when those who wore them had demonstrated spiritual attainments that elevated them above novice status?

25

The many scriptural precedents for the act of rending garments in grief are given a particularly patriarchal and prophetic note by Shenoute’s quotation of Joel 2.13. Shenoute’s association to Elijah must have called to mind Elisha rending his garments in 2 Kings 11–14. And the association of garment to community must have threatened the offender with expulsion, much as he explicitly threatened to distance himself and the federation from the woman’s monastery. Krawiec 2002, e.g., 32–34, and see Layton 2014, 129, Rules 100–1. In her most recent discussion of this episode Krawiec (2014) demonstrates the instrumentality of dress in the creation of monastic cultural memory. The textiles from Saqqara were dispersed to several European museums, including the Museum für angewandte Kunst in Vienna, which seems to have the largest collection (Riegl 1889). To my knowledge, the mantles from Saqqara have not been studied as a group; however, their shared use of embroidery and brocading for adornment may well be a distinguishing feature of monastic work and use. See also Krawiec 2014, 71–72.

266

T. K. THOMAS

Fig. 12: Monastic mantle with inscription of Psalm 44 preceded by an unidentified text (MAK T 580), said to be from Saqqara (Photograph: Courtesy of the Museum für angewandte Kunst, Vienna).

The blessing, prestige and authority promised or ascribed to their Master in the Psalms are not only borne out by exhibiting these texts inscribed on luxury dress, but also reflect on the servants themselves who bear the texts on their bodies and who are already a [holy king] by their very nature (1 Peter 2:9). This interpretation is, in fact, strongly supported by the deliberate change of pronouns in one mantle where every ‘him’ of the Psalm was turned into a ‘you’, and by the fusion between royal dress and real dress accomplished by the quotations from Psalm 44 [LXX Psalm 45].

Here, adding another, specific layer of meaning to the mantle of authority as a type of garment, both scriptural authority and the identity of the wearer (ideal biblical persona and actual monastic father) are written on the mantle. The mantle becomes a vehicle for scripturalized social interaction and for the inscription as a kind of speech. Other parts of Psalm 44 (LXX 45) although not quoted speak of embroidered garments and nations that shall fall, perhaps alluding to the current political

26

Tiraz texts, Jacques van der Vliet (2006) noted, ‘contain very specific information that situates the textiles along the parameters of politics, patronage, and economical and status value: names of caliphs, names of donating high officials, names of

situation of the Copts and vowing allegiance to God rather than temporal rulers. The mantles inscribed with the verses of Psalms belong, van der Vliet pointed out, to a larger group of mantles, with some inscriptions functioning as auspicious invocations of blessings and prosperity, while others are apotropaic, that is, they ward off danger, and, as here, yet others are intended to enhance social status and prestige involving literacy and erudition (in this case, scriptural literacy). These mantles are, van der Vliet argued, like contemporaneous tiraz, inscribed textiles in Islamic tradition, at the highest levels produced in caliphal factories and bestowed as signs of honour and loyalty. (Some examples are more like tiraz as they include pseudo-Arabic motifs.) The inscribed monastic mantles also participate, if only by aspiration, in the tiraz tradition of gifts of prestigious clothing because of the high rank of the donors, the restricted production of tiraz and the high value of materials and craftsmanship.26 It should be noted that scriptural inscriptions

workshops and dates’. The monastic mantles, in contrast, avoid that kind of temporal specificity. Some inscriptions include the Coptic version of the Arabic bismallah (in the name of God). Some of these mantles point to their ultimate use as shrouds

FASHIONING ASCETIC LEADERSHIP

267

were employed as well on Jewish prayer shawls in Medieval Cairo (van der Vliet 2006, 42).27 Changing mantles of authority in early Medieval portraits of Shenoute This remarkable shift towards contemporaneous styles of decoration can be traced more closely in portraits of Shenoute in manuscript illuminations, as the elements of the belt, the bag, the small mantle and the stole endure along with the tunic, and all come to be represented as highly ornamented.28 The first portrait, on a single leaf said to have been found at the White Monastery and now in the Coptic Museum in Cairo (MS 3852), likely dates to the ninth–tenth century (Fig. 13). This portrait presents an ornate striped tunic, belt, clerical stole and small mantle, strewn with crosses.29 The second portrait, in a manuscript in the Pierpont Morgan Library in New York (MS M604), belongs to a large group of manuscripts found at Hamouli in the Fayum, dated by their colophons to the ninth–tenth century (Leroy 1974, 32) (Fig. 14). The small mantle had only two reddish crosses. (The black spots are transfers of ink from the facing page.) Again, Shenoute wears a striped tunic. In these first two examples, the mantles retain their small size and shape but are transformed by ornament. The third is from a manuscript dated to the tenth–eleventh century in the Bibliothèque Nationale in Paris (MS Or. Copte 130) that came from the library of the White Monastery (Fig. 15). The limited palette makes it difficult to tell whether the lines on the tunic represent folds or stripes (a popular design, as discussed below). The small mantle is ornamented only with a pair of crosses. The

27

through the inclusion of funerary prayers. One displays verses from Psalm 8 (ibid., Appendix 1, no. 1) for ‘the great master, the Deacon Giorgios’; some inscriptions, as in the Cairo Genizah, document commissions by fathers for their sons. Several identify the honorees as clerics by title (ibid., 40), of which some should be identified as monastic (e.g., van der Vliet 2006, Appendix 2, no. 2, Musée du Louvre, inv. no. E 26792; van der Vliet 2006, 41 and n. 6, and 55–57, Katoen Natie, inv. no. 43). The inscription of one mantle declares itself to be a gift for a monastic brother. The current example, MAK T 580, is described as ‘part of a tunic (?)’ and Psalm verses are cited on van der Vliet 2006, 51. Stillman (2000, e.g., 56) notes that various economic classes among the Jewish community gifted inscribed garments to each other (and see below, note 27). Vivian Mann (2000) at pp. 106–8 cites the Responsa of Moses Maimonides (1135–1204), and at pp. 108–9, David ibn

Fig. 13: Portrait of Shenoute in MS. 3852. Eighth to ninth century(?), Coptic Museum, Cairo (Photograph: Leroy 1974, fig. 27, 1).

fourth manuscript, also in the Bibliothèque Nationale and also from the White Monastery, may be added to this group (MS Or. Copte 129, fol. 134r) (Fig. 16). A colophon dates the manuscript to the eleventh century although this portrait seems to have been added later. The figure I identify as Shenoute, despite the lack of inscription, wears apron-bag, belt and small mantle

28

29

Abi Zimra (Spain c. 1480–1573 Safed), Responsa Radbaz: ‘Concerning what they used to do in Egypt: embroidering sentences from the Torah in gold on prayer shawls’. As concerned shared traditions, this seems to hark back to the origins of Christian liturgical costume in the ancient ritual dress of the Jewish priest (Innemée 1992, 3–6). Stillmann (2000, 18) calls the pallium the Roman counterpart to the Jewish ritual tallith; Maguire (2003, 53) makes a similar comparison. Karel Innemée (1992, 162–85) quite rightly contextualizes this increase in ornamentation in a wider world of Medieval Eastern Christianity. Jules Leroy (1974, 89–91) could not believe the images represented monks because of their ornate clothing. Following Kurt Weitzmann, he wrongly identified this single leaf as a bookcover.

268

T. K. THOMAS

Fig. 14: Portrait of Shenoute (full composition: Shenoute conversing with Christ), Homily on Gilead in MS M604, 1 v., ninth to tenth century, The Pierpont Morgan Library and Museum, New York (Photograph: The Pierpont Morgan Library and Museum, New York).

Fig. 15: Portrait of Shenoute in Oeuvres du Schnoudi, MS. Or. Copte, 130 (5), 128 v (single leaf), White Monastery, eleventh century(?), Bibliothèque Nationale, Paris (Photograph: Anne Boud’hors).

identical to the ones in the Paris manuscript, and a tunic decorated with crosses punctuating faint stripes.30 In this latter pair of examples, the ornamentation of the mantle is more restrained than in the early portraits, a tradition that, at least in images, was not discontinued. Shenoute is again dressed in an ornate mantle and striped tunic in his portrait at the main church of the Monastery of St Antony by the Red Sea dated to 1232/33 (Fig. 17). The multiplication of cross-rune motifs on the mantle finds at least one parallel among artefacts of Medieval (and modern) monastic hoods in Egypt. The motifs of Shenoute’s tunic—a rune-like cross, a line of dots, dots flanking a line—are very like those in the Cairo manuscript and in painted portraits from other monasteries, including those of Apollo and

Jeremiah (see Fig. 2). Taken all together, these examples reflect a specifically monastic ornamental tradition that merits further study. In these later portraits of Shenoute, fashionable stripes have been transformed into emblems of the monastic heritage. Or emblems of a cross-section of semi-coenobitic and coenobitic practice have been treated to a fashionable makeover.

30

Jules Leroy (1974, 192) identified the figure as an anonymous saint in a stereotypical monastic habit. He notes that the colophon gives the provenance as the White Monastery and the date

Ornate mantles of authority in Medieval portraits at the Monastery of St Antony At the Monastery of St Antony by the Red Sea, Shenoute and Pachomius are excerpted from the programme of famed monastic fathers and patriarchs of the past painted in 1232/33 on the walls encircling the

as eleventh century, and describes the drawing as added to the completed manuscript.

FASHIONING ASCETIC LEADERSHIP

nave. On the interior of the wide arch that projects into the nave, Shenoute and Pachomius face each other across the space of the nave (see Fig. 17). They wear tunics with similar stripes, similar clerical stoles, the same kind of belt and a small mantle. By their location within the programme and by their habits, Shenoute and Pachomius are quite deliberately memorialized as the fathers of closely related coenobitic monastic traditions; however, only Shenoute wears the apron-bagmelote seen in his earlier portraits. Despite the damage to these paintings, it can be seen that neither Pachomius nor Shenoute is shown wearing a hood like the other fathers portrayed in this programme. Under his smaller mantle, Pachomius also wears a cape that differs from Shenoute’s, albeit with larger stripes, similar to those worn by some of the other figures. All of the other fathers wear the same kind of monastic belt and a Medieval development of the analabos, seen falling in two divergent arcs from its fastening to a cross just below the belt; by this time, the analabos had come to be called the schema and was indicative of the monastic state (Innemée 1992; Boud’hors 2009). There is still, however, variety among the other fathers’ mantles, many of them articulated with stripes and finished borders, and some wear hoods displaying various ornamental compositions. Here, and in other Medieval Egyptian portraits, change across local traditions towards newer

269

Fig. 16: Portrait of Shenoute in MS. Or. Copte 129 (14), 134 r, eleventh century, Bibliothèque Nationale, Paris (Photograph: Leroy 1974, fig. 108, 2).

Fig. 17: Portraits of Shenoute and Pachomius in the painted programme of the Monastery of St Antony by the Red Sea 1232/33 (Photographs: Elizabeth Bolman, © American Research Center in Egypt; Plan: Allyson McDavid).

270

T. K. THOMAS

ornamental fashions and even different types of garments would also seem to point to broad circumstantial change in the way the habit produced meaning. Given that the likely range of dates of these portraits with ornate mantles begins in the late ninth century, the images suggest a long-term trend of increasing openness to representing Shenoute in more splendid garments. Openness to the ornate may be seen in other monasteries—including that of Apa Apollo—from at least the sixth century on. Moreover, monastic literature from the fifth century presents a number of comments about increasing luxury in monastic life.31 The later portraits, in particular, seem to participate in a change in Egyptian secular dress known from documentary texts from the Cairo Genizah (a room for proper storage of any papers and books bearing the name of Yahweh no longer in use)—texts assigned to the Fatimid caliphate (909–1171). Yedida Stillman (2000) analysed the burgeoning of specialist terms in Genizah documents to describe patterns on secular garments and furnishings, such as terms for striped cloth with borders and striped cloth without borders; stripes, Stillman noted, comprised a popular pattern in the Cairo Genizah documents for Muslims, Jews and Christians (59). These later portrayals of Shenoute’s habit, then, seem to have taken several steps towards fashionable developments outside the monastic sphere, thereby presenting a changing view of the relationship between monastic and secular, as well as the monk, his ascetic practice and his habit. In this ornamental development of the small mantles, they again speak to worldly society but do not lose their traditional formal association with ascetic authority within the monastic context.

31

32

33

Other papers presented at the ‘Living for eternity’ conference at the University of Minnesota in 2003 provided archaeological evidence that monks in Egypt were not always living the ‘simple life’ that historians associated with monasticism. Fine pottery ware, for example, runs counter to modern expectations of ancient monasticism. This evidence again cautions us not to rely too heavily on monastic rules to create a historically accurate picture of life in the monastery, as noted by Brooks Hedstrom (2003) and in Ivančica Dvoržak Schrunk (2003). Karel Innemée (1992, e.g., at 162–63), noted the ubiquity of elaborate vestments in contemporaneous paintings of Nubian monks and ecclesiastics, remarking as well upon institutional connections between the Coptic and Nubian churches. Or might it allude to wise monks of old, such as the Syrian Bahira, who recognized Muḥammad as the prophet? Maguire

Conclusion How should we interpret these displays of ornament? They reflect the institutional role of the monastery for the education of priests and, increasingly, as the incubator for future patriarchs, who were now charged with leadership of the Egyptian Christian population, monastic and secular, in the dominant Islamic culture. A turning point may have been reached in the tenth century with the collection and intensification of clothing regulations in the Covenant of Umar (Mikhail 2014, 171). By this time, assimilation had reached a point where it worried some in the Christian community as well (Mikhail 2014, 172). High-ranking clerics (along with Egyptian civil servants) mediated with the caliph and his representatives on behalf of their Church, while negotiating the political and the social settings of the caliphal court (Swanson 2010, e.g., 43–48). This visual discourse seems to reflect Christians’ views of their high social status rather than the discriminatory dress codes for non-Muslims that were introduced sporadically from the seventh century on (e.g., Stillman 2000, 52–53; Swanson 2010, esp. 8). In addition to foregrounding ascetic monastic perspectives (e.g., Swanson 2010, 92–93), it would be interesting to focus on the portrayal of ornate Egyptian monastic habits through the lens of a fuller range of Eastern Christian monastic encounters, much as Karel Innemée did in his 1992 study of monastic and ecclesiastical dress.32 Within the context of the wider Islamic world, might such dress have brought to mind a shared scriptural tradition of expressive mantles of prophetic authority?33

(2003, 54–55, 60) makes connections to images on ceramics that are relevant to the latter point. On the odes, in which Bahira plays a key role, see Stetkevych 2010; on Bahira, see Roggema 2007 (2009) and Szilágyi 2008. Or might this portrayal be seen in the broader context of Arabic diyarat literature, which in its Medieval form celebrated monasteries as places of leisure and licence, wine drinking and carousing, that were outside the moral strictures of Islam and beyond the constraints of urban courtly life? See, e.g., Fowden (2007), whose narrative begins with Bahira. Perhaps, the ‘cloth associations’ of the Qur’ānic story of Joseph’s coat or shirt (S. 12) to his prophetic character and the gifting of this garment across generations are involved in the network of associations linking ancient and Medieval Jewish and Christian with Islamic monastic mantles of authority? See Lawson 2012.

FASHIONING ASCETIC LEADERSHIP

Within the changing image of monastic authority in Late Antique and Medieval Egypt, dress may be seen as a field for the display of group membership, distinction within the group and expression of belonging to a social world beyond the group. Surviving painted portraits, archaeological remains of garments and a wide range of textual attestations build the foundation for a broader survey of early trends emerging from the Graeco-Roman system of dress current in Late Antique Egypt. As in the secular sphere, mantles were a primary expression of authority and status. From the fourth century, the earliest symbolism of monastic mantles drew upon ascetic dress traditions, associating the Greek himation with a philosophical way of life, and the fleece or leather melote with Hebrew prophets in the Christian Old Testament. As early as the sixth century, the image of Egyptian monastic authority began to change as some painted portraits represented monastic leaders wearing fine, ornate mantles (previously known only from texts). By the ninth century, monastic leaders could be portrayed in highly decorated mantles. Portraits of the historical figure of Apa Shenoute, archimandrite of a powerful federation of monasteries, are particularly revealing as they allow for comparisons to portraits of other monastic leaders, suggesting that the striking shift in the appearance of these mantles is tied to different ways of characterizing authority, including authority recognized in non-monastic spheres. In the Medieval era, the ornate mantle also identified a monastic leader as the possessor of worldly authority. Further enquiry should address the historical context of increasing monastic leadership of the Church and

34

This enquiry would address other vestimentary signs of authority, such as the mismatched imperial red shoe and monastic sandal said to have been worn by the seventh-century Melkite patriarch Cyril (Davis 2004, 233 n. 180). Parallel developments in, for example, the adoption of the Arabic language should also be considered as discussed, e.g., in Davis (2008, 203–4): During the tenth century, young Copts learned Arabic as part of their education, a linguistic training that facilitated their roles in the Arab administration of the country. With this scholastic background, Sawirus began his career as a scribe in the government administration. However, he eventually left his post to become a monk, and later he was elected bishop of al-Ashmunayn (ancient Hermopolis Magna) under the name of Sawirus (the Arabic form of Severus). In that capacity, he attended the

271

engagement in governance, including during the Medieval era, as monk-patriarchs came to represent their community to the Islamic ruling elite.34 The visible display of monastic status developed from the adaptation of items in regular use in Mediterranean culture, including the ubiquitous tunic and the mantle always worn in public, which, in the monastic as in the secular sphere, displayed signs of authority and social status, group membership and individual identity. The habit spoke to the world in an established, living language of dress. The habit served monks’ negotiation of a renunciation of life in the world that was not an absolute rejection of the world. Thomas Merton, who thought deeply about the habit he wore, did not say but likely knew how deeply entangled was his display of himself as a monk with the traditions of the Desert Fathers, as well as with his own complicating engagement with the spirituality of Zen Buddhism and other Eastern religious traditions.35 In the tradition of the Antonian anchoritic ideal that he aspired to and that he shared with Shenoute, the cenobitic leader, in his habit turned within himself and towards his community, still concerned with the effect of his existence in the world. Acknowledgements I thank Elisabeth R. O’Connell and the participants in the conference for their comments as well as my generous colleagues Jennifer Ball, Elizabeth Bolman and Ellen Poteet.

35

court of the Fatimid caliph al-Muizz (972–75) and is said to have represented the Coptic community in public debates with leaders of other churches (such as the Syrian bishop Yunis ibnal-Shama), as well as with Jewish rabbis and Muslim imams. While Merton wore the habit, he too used elements of it along with everyday clothing, as in a posed portrait by his friend the photographer Eugene Meatyard, in which he wears his scapular (related to the melote-apron through the writings of John Cassian) and a baseball cap, an example of the flexible exterior framework he built for structuring his multifaceted life with God and the world. For the portraits, see Meatyard, Davenport and Magid 1991. For examples of his thoughts on the habit, see Merton 1973 and 2012.

272

T. K. THOMAS

Bibliography Bell, D. N. (trans.). 1983. Besa: The Life of Shenoute. Cistercian Studies Series 73. Kalamazoo, MI. Bolman, E. S. (ed.). 2016. The Red Monastery Church: Beauty and asceticism in Upper Egypt. New Haven, CT. Bolman, E. S., S. J. Davis and G. Pyke. 2010. Shenoute and a recently discovered tomb chapel at the White Monastery. Journal of Early Christian Studies 18 (3): 453–62. Bolman, E. S. and A. Szymańska. 2016. Ascetic ancestors: Identity and genealogy. In Bolman 2016, 165–74. New Haven, CT. Boud’hors, A. 2009. Le ‘scapulaire’ et la melote: Nouvelles attestations dans les textes coptes? In Études coptes XI: Treizième journée d’études (Marseille, 7–9 juin 2007), A. Boud’hors and C. Louis (eds), 65–79. Cahiers de la Bibliothèque copte 17. Paris. Brennan, T. C. 2008. Tertullian’s De pallio and Roman dress in North Africa. In Roman dress and the fabrics of Roman culture, J. Edmondson and A. Keith (eds), 258– 70. Toronto. Brooks Hedstrom, D. L. 2003. Tableware and monastic practice, 600–1400: New questions from the ceramic corpus at John the Little’s monastery. In Living for eternity: The White Monastery and its neighbourhood. Proceedings of a symposium at the University of Minnesota, Minneapolis, March 6–9, 2003, P. Sellew (ed.), 133. http://egypt.umn.edu/Egypt/1-pb%20pdfs/Brooks.pdf (last accessed 29 July, 2019). Carroll, R. P. 1969. The Elijah–Elisha sagas: Some remarks on prophetic succession in ancient Israel. Vetus Testamentum 19 (4): 400–15. Casiday, A. M. 2006. Evagrius Ponticus. The Early Church Fathers. New York. Castel, G. 1979. Étude d’une momie copte. In Hommages à la mémoire de Serge Sauneron, 1927–1976. Vol. 2: Égypte post-pharaonique, J. Vercoutter (ed.), 121–43. Bibliothèque d’étude 82. Cairo. Coquin, R.-G. 1992. À propos des vêtements de moines égyptiens. Bulletin de la Société d’archéologie copte 31: 3–23. Craik, J. 1993. The face of fashion: Cultural studies in fashion. London; New York. ———. 2003. The cultural politics of the uniform. Fashion Theory 7 (2): 127–48. Crosby, H. L. (trans.). 1951. Dio Chrysostom. 5 vols. Loeb Classical Library 385. Cambridge, MA; London. Davis, S. J. 2004. The early Coptic papacy: The Egyptian Church and its leadership in Late Antiquity. Popes of Egypt 1. Cairo. ———. 2008. Coptic Christology in practice: Incarnation and divine participation in Late Antique and Medieval Egypt. Oxford; New York.

Dijkstra, J. and M. van Dijk (eds). 2006. The encroaching desert: Egyptian hagiography and the Medieval West. Leiden; Boston, MA. Dilley, P. C. 2008. Dipinti in Late Antiquity and Shenoute’s monastic federation: Text and image in the paintings of the Red Monastery. Zeitschrift für Papyrologie und Epigraphik 165: 111–28. ———. 2016. Textual aesthetics: Dipinti and the early Byzantine epigraphic habit. In The Red Monastery Church: Beauty and asceticism in Upper Egypt, E. S. Bolman (ed.), 175–82. New Haven, CT. Emmel, S. 2008. Shenoute’s place in the history of monasticism. In Christianity and monasticism in Upper Egypt 1: Akhmim and Sohag, G. Gabra and H. N. Takla (eds), 31–46, with bibliography 321–50. Cairo; New York. Fowden, E. K. 2007. The lamp and the wine flask: Early Muslim interest in Christian monasticism. In Islamic crosspollinations: Interactions in the Medieval Middle East, A. Akasoy, J. E. Montgomery and P. E. Pormann (eds), 1–28. Cambridge. Fussell, P. 2002. Uniforms: Why we are what we wear. Boston, MA. Gehn, U. 2012. Ehrenstatuen in der spätantike: Chlamydati und Togati. Spätantike-frühes Christentum-Byzanz. Reihe B, Studien und Perspektiven 34. Wiesbaden. Giorda, M. 2007. L’abito non fa il monaco? La veste nei primi secoli del monachesimo egiziano. Annali di studi religiosi 8: 67–88. ———. 2009. Bishops-monks in the monasteries: Presence and role. Journal of Juristic Papyrology 34: 49–82. ———. 2011. Does the cowl make the monk? A monastic accessory during the first centuries of Egyptian monasticism. In Dress accessories of the 1st millennium AD from Egypt: Proceedings of the 6th conference of the research group ‘Textiles from the Nile valley’, Antwerp, 2–3 October 2009, A. De Moor and C. Fluck (eds), 182–87. Tielt. ———. 2012. Shenouda e la Chiesa copta ortodossa: L’abito fa il Monaco. La Rivista il Mulino (30 March), http:// www.rivistailmulino.it/item/1494 (last accessed 7 July 2016). Goehring, J. E. 1992 (1999). The origins of monasticism. In Ascetics, society, and the desert: Studies in early Egyptian monasticism, J. E. Goehring, 13–35. Harrisburg, PA. ———. 2006. Remembering Abraham of Farshut: History, hagiography and the fate of the Pachomian tradition. Journal of Early Christian Studies 14 (1): 1–26. Goffmann, E. 1961 (2007). On the characteristics of total institutions. In Asylums: Essays on the social situation of mental patients and other inmates, E. Goffmann, 1–124. New York. Gregg, R. C. (ed.). 1980. Athanasius: The Life of Antony and the Letter to Marcellinus. New York.

FASHIONING ASCETIC LEADERSHIP

Grossmann, P., D. L. Brooks Hedstrom, M. Abdal-Rassul and E. S. Bolman. 2004. The excavation in the monastery of Apa Shenute (Dayr Anba Shinuda) at Suhag, with an appendix on documentary photography at the monasteries of Anba Shinuda and Anba Bishoi, Suhag. Dumbarton Oaks Papers 58: 371–82. Hanfmann, G. M. A. 1951. Socrates and Christ. Harvard Studies in Classical Philology 60: 205–23. Hunink, V. (ed. and trans.). 2005. Tertullian: De pallio (On the mantle). Amsterdam. Innemée, K. C. 1992. Ecclesiastical dress in the Medieval Near East. Studies in Textile and Costume History 1. Leiden; New York. ———. 2005. On the necessity of dress: Should a hermit wear clothes? Khilꜥa 1: 69–78. Joseph, N. 1986. Uniforms and nonuniforms: Communication through clothing. Contributions in Sociology 61. New York. Kalamara, P. 1995. Le système vestimentaire à Byzance du IVe jusqu’à la fin du XIe siècle. 2 vols. PhD dissertation, École des hautes études en sciences sociales. Kampen, N. 2009. Family fictions in Roman art: Essays on the representation of powerful people. Cambridge; New York. Kiilerich, B. 2015. Visual dynamics: Reflections on Late Antique images. Bergen. Krawiec, R. 2002. Shenoute and the women of the White Monastery: Egyptian monasticism in Late Antiquity. Oxford; New York. ———. 2009. Garments of salvation: Representations of monastic clothing in Late Antiquity. Journal of Early Christian Studies 17 (1): 125–50. ———. 2014. The holy habit and the teachings of the elders: Clothing and social memory in Late Antique monasticism. In Dressing Judeans and Christians in Antiquity, K. Upson-Saia, C. Daniel-Hughes and A. J. Batten (eds), 55–73. Farnham; Burlington, VT. Lawson, T. 2012. Typological figuration and the meaning of ‘spiritual’: The Qur’anic story of Joseph. Journal of the American Oriental Society 132 (2): 221–44. Layton, B. 2002. Social structure and food consumption in an early Christian monastery: The evidence of Shenoute’s Canons and the White Monastery federation, A.D. 385–465. Le Muséon 115 (1–2): 25–55. ———. 2007. Rules, patterns, and the exercise of power in Shenoute’s monastery: The problem of world replacement and identity maintenance. Journal of Early Christian Studies 15 (1): 45–73. ———. 2014. The Canons of our fathers: Monastic rules of Shenoute. Oxford Early Christian Studies. Oxford; New York. Leroy, J. 1974. Les manuscrits coptes et coptes-arabes illustrés. Paris.

273

Linscheid, P. 1999. Kapuzen im spätantiken und koptischen Ägypten. In Ägypten und Nubien im spätantiken und christlicher Zeit. Akten des 6. Internationalen Koptologenkongresses, Münster 20.–26. Juli 1996, S. Emmel et al. (eds), 1: 238–48. Sprachen und Kulturen des christlichen Orients 6.1–2. Wiesbaden. Lopez, A. G. 2013. Shenoute of Atripe and the uses of poverty: Rural patronage, religious conflict and monasticism in Late Antique Egypt. Berkeley, CA. MacMullen, R. 1964. Some pictures in Ammianus Marcellinus. Art Bulletin 46 (4): 435–55. Maguire, E. D. 2003. Dressed for eternity: A prelude. In Living for eternity: The White Monastery and its neighbourhood. Proceedings of a symposium at the University of Minnesota, Minneapolis, March 6–9, 2003, P. Sellew (ed.), 33–69. http://egypt.umn.edu/Egypt/1pb%20pdfs/maguire.pdf (last accessed 28 July, 2019). Mann, V. 2000. Jewish texts on the visual arts. Cambridge; New York. Marsengill, K. 2013. Portraits and icons: Between reality and spirituality in Byzantine art. Turnhout. Maspero, J. 1931. Fouilles exécutées à Baouit: Notes mises en ordre et éditées par Étienne Drioton. Mémoires publiés par l’Institut français d’archéologie orientale du Caire 59. Cairo. Mauss, M. 1935 (1973). Les techniques du corps. Journal de psychologie 32: 271–93. (Translated as: Body techniques. Ben Brewster [trans.]. Economy and Society 2 [1]: 70–88). Meatyard, R. E., G. Davenport and B. Magid. 1991. Father Louie: Photographs of Thomas Merton. New York. Merton, T. 1960. The wisdom of the desert: Sayings from the Desert Fathers of the fourth century. New York. ———. 1973. The Asian journal of Thomas Merton, N. Burton, P. Hart, J. McLaughlin and A. Chakravarty (eds). New York. ———. 2012. The life of vows: Initiation into the monastic tradition 6, P. F. O’Connell (ed.). Monastic Wisdom Series 30. Trappist, KY. Mikhail, M. S. A. 2014. From Byzantine to Islamic Egypt: Religion, identity and politics after the Arab conquest. New York. Miller, P. C. 2009. The corporeal imagination: Signifying the holy in Late Ancient Christianity. Philadelphia, PA. Morfin-Gourdier, N. 1991. Costume of the religious. In The Coptic encyclopedia, A. S. Atiya (ed.), 2: 650b–655a. New York. Quibell, J. E. 1908. Excavations at Saqqara (1906–1907). Cairo. Rapp, C. 2005. Holy bishops in Late Antiquity: The nature of Christian leadership in an age of transition. The Transformation of the Classical Heritage 37. Berkeley, CA.

274

T. K. THOMAS

Rassart-Debergh, M. 1981. La décoration picturale du monastère de Saqqara: Essai de reconstitution. Acta ad archaeologiam et artium historiam pertinentia 9: 9–124. Rassart-Debergh, M. and J. Debergh. 1981. À propos de trois peintures de Saqqara. Acta ad archaeologiam et artium historiam pertinentia 9: 187–206. Riegl, A. 1889. Die Ägyptischen Textilfunde im K. K. Österreich Museum: Allgemeine Charakteristik und Katalog. Vienna. Roggema, B. 2007 (2009). The legend of Sergius Bahira: Eastern Christian apologetics and apocalyptic in response to Islam. Leiden; Boston, MA. Rollason, N. K. 2016. Gifts of clothing in Late Antique literature. London; New York. Rubenson, S. 1995. The letters of St. Antony: Monasticism and the making of a saint. Studies in Antiquity and Christianity. Minneapolis, MN. Russell, N. (trans.). 1981. The Lives of the Desert Fathers. Cistercian Studies Series 34. London; Oxford; Kalamazoo, MI. Rutschowscaya, M.-H. 1990. Coptic fabrics. Paris. Schiffhorst, G. J. 2011. Thomas Merton’s desert spirituality. Cithara 50 (2): 3–13. Schrenk, S. 2000. Textilien—Malerei: Sind Parallelen zwischen verschiedenen Gattungen auswertbar? In Byzantinische Malerei, Bildprogramme—Ikonographie—Stil: Symposion in Marburg vom 25.–29.6.1997, G. Koch (ed.), 293–306. Spätantike, frühes Christentum, Byzanz; Reihe B: Studien und Perspektiven 7. Wiesbaden. Schrenk, S. 2004. Textilien des Mittelmeerraumes aus spätantiker bis frühislamischer Zeit. Textilsammlung der Abegg-Stiftung 4. Riggisberg. Schroeder, C. T. 2001. Purity and pollution in the asceticism of Shenute of Atripe. In Ascetica, Gnostica, Liturgica, Orientalia, M. F. Wiles and E. J. Yarnold (eds), 142–47. Studia Patristica 35. Leuven. ———. 2004. ‘A suitable abode for Christ’: The church building as symbol of ascetic renunciation in early monasticism. Church History 73 (3): 472–521. ———. 2007. Monastic bodies: Discipline and salvation in Shenoute of Atripe. Divinations: Rereading Late Ancient Religion. Philadelphia, PA. Schrunk, I. D. 2003. Spiritual economy and spiritual craft: Monastic pottery production and trade. In Living for eternity: The White Monastery and its neighbourhood. Proceedings of a symposium at the University of Minnesota, Minneapolis, March 6–9, 2003, P. Sellew (ed.), 83–96. http://egypt.umn.edu/Egypt/1-pb%20pdfs/ schrunk.pdf (last accessed 28 July 28, 2019) Smith, R. R. R. 1999. Late antique portraits in a public context: Honorific statuary at Aphrodisias in Caria, A.D. 300–600. The Journal of Roman Studies 89: 155–89.

Sterk, A. 2004. Renouncing the world yet leading the Church: The monk-bishop in Late Antiquity. Cambridge, MA. Stetkevych, S. P. 2010. The mantle odes: Arabic praise poems to the prophet Muhammmad. Bloomington, IN. Stewart, P. 2003. Statues in Roman society: Representation and response. Oxford; New York. Stillman, Y. K. 2000. Arab dress from the dawn of Islam to modern times: A short history. Leiden. Swanson, M. N. 2010. The Coptic papacy in Islamic Egypt (641–1517). Popes of Egypt 2. Cairo; New York. Szilágyi, K. 2008. Muhammad and the monk: The making of the Christian Bahīra legend. Jerusalem Studies in Arabic and Islam 34: 169–214. Tellenbach, M., R. Schulz and A. Wieczorek (eds). 2013. Die Macht der Toga: DressCode im Römischen Weltreich. Begleitband zur Sonderausstellung ‘Die Macht der Toga—Mode im Römischen Weltreich’ im Roemerund Pelizaeus-Museum Hildesheim … 20. April 2013 bis 8. September 2013. Regensburg. Thomas, T. K. 2012. Mimetic devotion and dress in some monastic portraits from the monastery of Apa Apollo at Bawit. Coptica 11: 37–79. ———. 2018. Portraits of Apa Jeremias at his monastery in Memphis (Saqqara): Holy man, philosopher, prophet, and monastic father. In Across the Mediterranean – Along the Nile. Studies in Egypotology, Nubiology, and Late Antiquity dedicated to Lászlo Török on the occasion of his 75th birthday, T. Bács, Á. Bollók and T. Vida (eds), 2: 659–83. Budapest. ———. 2019. The honorific mantle as furnishing for the household memory theater in Late Antiquity: A case study from the monastery of Apa Apollo at Bawit. Dumbarton Oaks Papers 73: 355–88. ———. In progress. Dressing souls, making monks: The habit and the portrait in the figuration of the earliest monks of Egypt. Torallas Tovar, S. 2007. The terminology of Egyptian monastic garments. In Material culture and well-being in Byzantium (400–1453): Proceedings of the international conference (Cambridge, 8–10 September 2001), M. Grünbart, E. Kislinger, A. Muthesius and D. Ch. Stathakopoulos (eds), 219–24. Denkschriften (Österreichische Akademie der Wissenschaften, PhilosophischHistorische Klasse) 356; Veröffentlichungen der Kommission für Byzantinistik 11. Vienna. Tsirpanlis, C. 1986. The origin, nature and spirit of Christian monasticism. Orthodox Thought and Life 3: 81–95. Urbano, A. 2013. ‘Dressing a Christian’: The philosopher’s mantle as signifier of pedagogical and moral authority. In Papers presented at the Sixteenth International Conference on Patristic Studies held in Oxford, 2011, Volume 10, M. Vinzent (ed.), 213–29. Studia Patristica 62. Leuven.

FASHIONING ASCETIC LEADERSHIP

———. 2014. Sizing up the philosopher’s cloak: Christian verbal and visual representations of the tribōn. In Dressing Judeans and Christians in Antiquity, K. Upson-Saia, C. Daniel-Hughes and A. J. Batten (eds), 175–94. Farnham; Burlington, VT. ———. 2016. The philosopher type in Late Roman art: Problematizing cultural appropriation in light of cultural competition. In Religious competition in the GrecoRoman world, C. Vuong and N. P. Desrosies (eds), 27–40. Atlanta, GA. Veilleux, A. (ed.). 1981. Pachomian koinonia 2: Pachomian Chronicles and Rules. Kalamazoo, MI. van der Vliet, J. 2006. ‘In a robe of gold’: Status, magic and politics on inscribed Christian textiles from Egypt. In Textile messages: Inscribed fabrics from Roman to

275

Abbasid Egypt, C. Fluck and G. Helmecke (eds), 23–67. Studies in Textile and Costume History 4. Leiden; Boston, MA. Ward, B. (trans.). 1975. The sayings of the Desert Fathers: The alphabetical collection. London; Kalamazoo, MI. Winlock, H. E. and W. E. Crum. 1926. The Monastery of Epiphanius at Thebes 1: The archaeological material. Publications of the Metropolitan Museum of Art Egyptian Expedition. New York. Wipszycka, E. 2009. Moines et communautés monastiques en Égypte (IVe–VIIIe siècles). Journal of Juristic Papyrology Supplement 11. Warsaw. Zanker, P. 1996. The mask of Socrates: The image of the intellectual in Antiquity, translated by A. Shapiro. Sather Classical Lecture Series 59. Berkeley, CA.

V LANGUAGES AND SCRIPTS

GREEK AND EGYPTIAN: DID LINGUISTIC POLICIES EXIST IN GRAECO-ROMAN EGYPT? Sofía TORALLAS TOVAR

For the Castilian Antonio de Nebrija, writing the first Spanish grammar in the fateful year of 1492, it was obvious that ‘language has always been the companion of empire’. He might have added religion, too, given that he wrote in the same year that his monarchs expelled Iberia’s Jews, and conquered its last Muslim territories. But what seemed obvious and eternal to this early European theorist of empire and religion is far from apparent to the student of Graeco-Roman Egypt. There we not only find diglossia and the religious use of Egyptian for many centuries, but we can scarcely find explicit evidence of a linguistic policy, nor of any effort to impose a language on a population in all their activities. The use and interaction of Greek and demotic/Coptic in Egypt during the Graeco-Roman period is attested by a large number of papyri. This documentary treasure attests to ups and downs in the use of the languages: the continued use of demotic in temples and certain parts of scribal culture during the first centuries of Hellenistic rule; the decline and eventual disappearance of demotic from the bureaucracy at some point around the reign of Augustus; the revival of the Egyptian language, particularly in ritual and religious contexts, around the third century through the use of an alphabet adopted and modified to its needs from the Greek alphabet, that is, Coptic; and its blooming, from the arrival of the Arabs in the land of the Nile, in both documentary texts and Christian literature. We should not doubt that empire and religion were of crucial importance in this history of language change. But if there was any kind of policy governing this centurieslong interaction (not to say competition) between these two languages, it is very difficult to discern.1

In this chapter I will not attempt to determine reasons (imperial or religious) for the relative position of languages at any given point in time. Rather, given the opacity of our sources for linguistic policy in GraecoRoman Egypt, I have set out to distinguish specific policies that may have shaped or arbitrated the use of languages in public life, and that may have had some role in the demise of demotic (at least in bureaucracy), and in the re-emergence of Egyptian in its Coptic garb some time later. I do not mean to suggest that linguistic policy drove or caused these changes. Such an argument would be difficult in the best of cases, and impossible in this one, given that we know so little about what linguistic policies there were. My goal here is to explore what evidence we have that might increase that knowledge of the dynamics between law, language use, language preference—often bound by religious affiliation—in Egypt, especially in imperial times. As is well known, Latin and Greek prevailed over the other languages of the Roman empire at least until Arabic superseded Greek and Latin in areas conquered by the Arabs in the seventh and eighth centuries of the common era. Some of the vernacular languages of the Roman empire disappeared at a pace that cannot be reconstructed from the written evidence, as is the case with Gallic or Iberian languages (Clackson 2012 and 2015, 68).2 Some of these languages partly disappeared from the written record for a while to later recover and flourish into the Middle Ages, as Egyptian did. Does the emergence of Coptic around the third century AD have anything to do with the Constitutio Antoniniana in 212, after which we have evidence that linguistic policy was less strict in certain domains? We might want to think so, but it is very difficult to make a causal

1

2

This is a wide topic, on which much work has been done since the 1960s. I refer to Rochette 1995; 1996; 1998; Torallas Tovar 2005; 2010a; 2010b; Thompson 2009; Fournet 2009; Vierros 2012; 2014; Choat 2012; Keenan 2014; all with ample bibliography on earlier work on the topic.

On the ‘forgetting’ of culture more generally in the western provinces of the Roman empire, see Woolf 1996.

280

S. TORALLAS TOVAR

argument about the effects of imperial policies such as this one on linguistic practice in the empire, and even less at the local level of Egypt. Indeed, so far as we know there was no attempt to impose Greek and Latin on conquered populations, nor do we have any direct evidence of central policy or planning on these questions.3 Yet there was awareness of the linguistic situation and measures were taken, as I will try to argue below, to address specific cases in order to avoid or prevent misunderstandings and make sure that language use would not hinder the progress of administrative, financial or legal issues. These measures may provide some indirect evidence for an awareness of the importance of linguistic issues and policies.

Linguistic policy, broadly speaking, can be defined as any intervention by the authorities, whether by recommendation, ban or restriction, on a community’s linguistic use. More narrowly speaking it refers to the explicit set of rules and laws with which the authorities (in a broad sense) intervene in the use of languages and dialects in a community. Different policies take effect at the municipal level, within educational institutions, libraries, the judiciary, bureaucracy and other bodies, such as the Church or unions, in different ways. Today’s Europe is a good example of a multilingual situation and the application of linguistic policies. Although we cannot try to construct a model for historical situations based on modern linguistic contexts, scholarship on those contexts can provide suggestive analogies with which to approach ancient evidence.4 In the current European context (with the coexistence and conflict of a large diversity of languages in all spheres: public and private, political, social or educational) the development of linguistic regulations has reached extraordinary levels of refinement. Whole governmental departments on minority languages deal with everyday questions and conflicts rising from the identity issues originating in the use of languages in the same territories, and their presence in the streets,

schools, courts and parliaments. Each of these environments has a different regulatory approach towards the interaction of two or more languages in the same space: more conservative in court, for example, less so in the street. But many of today’s linguistic issues, or regulation of situations involving some kind of linguistic circumstance, tend to be addressed by separate and relatively uncoordinated parts of a complex administrative structure, rather than governed by an overarching policy with a clear agenda behind it. Spolsky (2004, 1) makes this point using the example of a specific situation at a German hospital. A Turkish woman was denied treatment in Germany because the doctors were afraid she would not understand their instructions and would fail to take the appropriate medication and cooperate fully in her treatment. This situation elicited a specific response from the Ministry of Health, tailormade for the specific situation. Most questions like this one have all kinds of components, not just the linguistic. The favourable attitude of citizens towards multilingualism and their acceptance of the ‘language of the other’ are directly connected to their acceptance of the presence of ‘the other’ in their territory, to the extent of their conception of their language as representative of their national, ethnic, religious or other identity. These ‘social’ issues often drive language evolution, as we are witnessing in the twenty-first century, and often have an effect on policies (Spolsky 2004, 6). The tools being developed to help us understand these shifts in contemporary Europe can serve us, not as a general paradigm, but as an invitation to think in new ways about the various issues that could have affected a linguistic environment such as Ptolemaic Egypt, and later, the Roman empire. Unfortunately the evidence is sparse, but a careful analysis of our fragmentary pieces of information in comparison with current situations can perhaps provide comparative insight into something approaching the linguistic policies of the ancient world. What can we identify as ‘linguistic policy’ in antiquity? Although the concept of linguistic policy is difficult to typify and describe, and is a debated matter

3

4

Linguistic policies: Today and in antiquity

Indeed, we even have instances of cities such as Cumae asking permission to use Latin in public affairs (Livy 40.42.13), evidence that Rome was not imposing Latin, even as it also suggests that the use of Latin was understood to have symbolic importance.

Although it is undoubtedly true that the national identity and its connection to language is a modern development that cannot be found in ancient situations, there are details that can help us understand the development of a conflict between languages.

GREEK AND EGYPTIAN

even in modern situations where there is much more information to hand (Schiffman 1996, 26–40),5 we can, for the sake of convenience, talk of policy when there is a clear statement in an official document, or in a copy of or reference to it, that attests to an attempt to regulate or define the use of a language in a particular context. I will suggest that, to the extent we find such evidence of linguistic policy in Ptolemaic Egypt and later the Roman empire, such policy was, in Kloss’ (1977) terminology, promotive, in that it promoted the use of a language in specific domains, and at the same time tolerant, in that it tolerated the use of languages other than Greek and Latin in other domains; and that this language tolerance was closely connected with tolerance of alien legal systems and alien religions. The sources The sources at our disposal for this inquiry are the legal texts, the documents on papyrus and the literary and historical evidence. Of these, the verbatim legal regulations, albeit often devoid of a context, are the most reliable evidence for a linguistic policy, but in the form of details that often do not convey a bigger picture. More generally, the surviving documentation in toto provides a corpus of information about language use. I will discuss later three examples: the use of demotic in specific documents in the first centuries of Roman rule; the preference for Greek in police reports; and the use of Latin or Greek in documents concerning citizens (in this case connected to a specific linguistic requirement). Finally, the historical sources provide partial glimpses into situations involving linguistic events, offering a sense of context, but seldom providing us with a clear and defined view or description of a specific policy regulating that event. I will comment now on two of these historical texts in order to give a sense of the problems that attend their use, and suggest why such sources are not particularly useful for the purpose of reconstructing policy. Both are from the first century, and involve the concept of sermo patrius. Their interpretation is controversial among modern scholars, both because the modern concept of national language pride is

5

Modern linguistic policy is also hard to assess as a whole. The attempts at typology or at definition never reach the extent of the casuistic. However, the scholarship, which is wide, provides

281

anachronistic in antiquity, and because the purpose and motivations behind the scenes described in the texts are far from clear. The first text deals with a speaker of a Celtiberian language in Hispania Citerior (Eastern Spain). Tacitus (Annals 4.45) describes a scene of torture of a man from Termes who was accused in AD 25 of murdering a provincial official. Since the murder of the senator was considered a collective act, he was questioned under torture about the names of his accomplices. But according to Tacitus or his informants, he replied magna voce sermone patrio that he would not give up and confess the names, no matter how terrible the torture was: et repertus cum tormentis edere conscios adigeretur, voce magna sermone patrio frustra se interrogari clamitavit: adsisterent socii ac spectarent; nullam vim tantam doloris fore ut veritatem eliceret. When he was discovered and forced with torments to give away his accomplices, with a loud voice he cried in his native tongue that questioning was useless: his partners might stand by and watch—for there would be no violent pain sufficient to extract the truth from him (Tacitus, Annals 4.45).

The only certain linguistic information that we can derive from this passage is that a vernacular Celtiberian language continued to be spoken in the first century AD. The circumstances of its use instead of Latin are debated. Adams (2003a, 280 and 752) interprets this scene as a proof of cultural resistance to Romanization, also evidenced in the murder of the oppressive senator. He considers the use of the native language as a mark of identity. Beltrán Lloris (2011, 19–20) does not agree: there is a good deal of evidence of Romanization in the area of Termes, both textual and archaeological. This suggests to Beltrán Lloris that the Termesine was addressing his accomplices, present at the scene, in their own language in order not to be understood by the torturers. Both are possible interpretations. I need not take sides in order to suggest that we should not deduce much about language use from such evidence.

a neat terminology that I will try to apply, in order to create a link to similar situations.

282

S. TORALLAS TOVAR

As a second example: Quintus Curtius Rufus narrates an episode of the Life of Alexander the Great. His general Philotas, accused of treason, is asked to defend himself. The language that he chooses to speak will be interpreted by Alexander as a sign of disrespect to Macedonian customs, and thus confirm his treason (cf. Kapetanopoulos 1999): Iamque rex intuens eum, ‘Macedones’, inquit, ‘de te iudicaturi sunt: quaero, an patrio sermone sis apud eos usurus’. Tum Philotas, ‘Praeter Macedonas’, inquit, ‘plerique adsunt, quos facilius, quae dicam, percepturos arbitror, si eadem lingua fuero usus, qua tu egisti, non ob aliud, credo, quam ut oratio tua intellegi posset a pluribus’. Tum rex, ‘Ecquid videtis adeo etiam sermonis patrii Philotan taedere? solus quippe fastidiit eum discere. Sed dicat sane, utcumque ei cordi est, dum memineritis, aeque illum a nostro more quam a sermone abhorrere’. And now the king, staring at him, said: ‘The Macedonians are about to pass judgment on you; I inquire whether you will use their native tongue in addressing them.’ Then Philotas replied: ‘Besides the Macedonians there are many present who, I suppose, will more easily understand what I say if I use the same language which you just employed, for no other reason, I think, than in order that your speech might be understood by most of the people.’ Then the king said: ‘Indeed you see to what extent Philotas is sickened by the language of his fatherland? For he alone disdains to learn it. But certainly let him speak in whatever way he desires, provided that you remember that he abhors our customs as much as our language’ (Hist. Alex. 6.9. 34–36).

This is the first-century view of a historian who is perhaps thinking in anachronistic terms. But in any case, it attests to language as a marker of identity that can meaningfully be understood (at least in the first century) as part of an accusation of treason, and to the connection of language, custom (more nostro) and identity in the mind of this historian. Without going into the question of the real status of Macedonian or Greek, or whether Rufus’ information was accurate or not, we can speak of a sense of what sermo patrius is

6

The inscription RDGE 52 = I.Milet. 1.2 no. 3 = I.Priene 106, is a famous letter of a Roman governor to the cities of the koinon of Asia in 56–50 BC. He ends by saying, l. 55: τὴν δὲ αἰτίαν δι’ ἣν ἑλλη[νι]κοῖς ἔγραψα, μὴ ἐπιζητήσητε· κατὰ νοῦν γὰρ

and its correlation to ethnicity in the first century (for more cases in antiquity, see Clackson 2015, 66). In this case, as in Tacitus’ text too, the context for the scene is judiciary: both the Termesine and Philotas are being accused and questioned. Their language choice seems significant for the historians, perhaps to give the scene a dramatic flavour and the protagonist a rebellious image, as he refuses to comply with the prescribed language of the army or the court. However, there are interesting linguistic phenomena present, especially in Rufus’ text. The motive adduced by Philotas for his language choice, that of koiné Greek versus Macedonian, is that more people will understand what he is saying. This appeal to employ the more widely understood language is strikingly similar to arguments found in modern ‘dominant language ideologies’ that are important contributors to the ‘minorization’ of languages in the contemporary world (Patrick 2010).6 The general shape of the argument is reinforced later by Rufus himself, when he explains that Macedonian has fallen into disuse owing to the contact with other peoples (Hist. Alex 6.10.23): Iam pridem nativus ille sermo commercio aliarum gentium exolevit: tam victoribus, quam victis peregrina lingua discenda est. The text even includes the paradox of the dominant language being left behind and the ‘victors’ adopting the language of the vanquished. But we should not, therefore, go so far as to see in Rufus a defense of Latin or Greek in the linguistic situation of his day, just because more people would understand them. As discussed here, these passages, and many others, are extremely interesting in that they provide evidence for ancient linguistic situations, which are open to discussion. The question is whether they can tell us much about linguistic policy in the Graeco-Roman world, rather than just linguistic attitudes. Since it is difficult to answer this question, I will in the remainder of this paper focus on explicit instances of linguistic policy, according to the definition provided above: not excluding the support of historical texts like these, but remaining cautious in their use.

[ἔσ]χον, μή τι παρὰ τὴν ἑρμηνείαν ἔλασσον τὰ [γεγραμ][μ] ένα νοῆσαι δύνησθε· (‘Do not ask why I wrote in Greek, since it was my intention that nothing contrary to the correct interpretation of my letter could possibly be in your mind’).

GREEK AND EGYPTIAN

Ptolemaic policies In the Ptolemaic period the extant documents display linguistic diversity. With a population of Egyptian origin in the majority, the evidence in demotic is extensive, especially in the first century of Ptolemaic rule (Thompson 1992a; 1992b), much more so than it will be later, under Roman rule (see below). From the reign of Ptolemy II on, the intensification of bureaucracy brought about an increase in the use of Greek. As Thompson (1992b) describes it, it is not a case of the imposition of Greek, but rather the adoption over time of the language of the new rulers by a collaborative priestly and scribal class. It appears that the Ptolemies issued a number of laws on the use of koiné for official and administrative communication (rapidly eliminating any other Greek dialects, and contributing to the uniformity of the Greek official language that we have in the papyri).7 This created a situation of diglossia. Some evidence of Ptolemaic linguistic policies does survive in a few papyri that provide a glimpse into some policies that attempted to regulate the use of language in specific domains: notarial deeds, in bookkeeping and in courts of justice (Crespo 2007). a. The Revenue Laws of Ptolemy Philadelphos (P.Rev. col. 9, ll. 4–5; Bingen 2007, 157–88) stated in 259 BC that the proclamation had to be issued in both the Greek and the Egyptian languages (col. 9, ll. 4–5: γράψαντες γράμμασιν ἑλληνικοῖς τε καὶ ἐγχωρίοις). This suggests an awareness of the linguistic situation and recognition of the need to address subjects in their own language, especially in a matter as crucial as their tax obligations. b. As far as the registration of public contracts is concerned, the creation of the grapheia and the requirement of registration of all contracts, both in Greek and in demotic (register and collection of tax on demotic and Greek contracts written outside the grapheia) probably dated back to Ptolemy II (Muhs 2010, 587). Although the policy is not preserved in an official document, P.Paris 65 (Sel.Pap. II 415 = UPZ I, pp. 596–97, from Thebes, 145 BC) may provide evidence of a revision of it. This document contains a letter between two professional scribes, Paniskos, an

7

Peremans 1964; 1983; Thompson 1994; Clarysse 1998; Vierros 2014. See also Remondon 1964.

283

official in charge of registers in Thebes, and Ptolemaios, probably a subordinate. The letter describes the procedure for registration of Egyptian documents in the grapheion as stipulated in a circular, even mentioning the dates of issue and enforcement (Depauw 2011), including names and patronymics, and the details and validity of the agreement and legal transaction. It stresses the importance of ‘knowing the correct procedure’, and this might well mean that it had been recently revised. The revisions had, in this case, a linguistic profile: the Greek registration subscription was now made mandatory, and a separate abstract was necessary. c. A decree from Tebtunis issued by Ptolemy VIII in 118 BC established the competence of Greek and native courts of law for lawsuits between Greeks and Egyptians, depending on the language used in the contract that was in conflict (Mélèze-Modrzejewski 1975, 1995; Pestman 1985; Allam 1991; Capponi 2010). προστετάχασι δὲ καὶ περὶ τῶν κρινομένων Α[ἰ]γυπτίων| πρὸς Ἕλληνας καὶ περὶ τῶν Ἑλλήνων τῶν [π]ρὸς τοὺς | Αἰγυπτίους ἢ Αἰγυ(πτίων) πρὸς Ἕλληνας γενῶν πάντων | πλὴν τῶν γεω(ργούντων) βα(σιλικὴν) γῆν καὶ τῶν ὑποτελῶν καὶ τῶν | ἄλλων τῶν ἐπι\πε/πλεγμένων ταῖς προσόδοις τοὺς | μὲν καθʼ Ἑλληνικὰ σύμβολα συνηλλαχότας | Ἕλλησιν Αἰγυπτίους ὑπέχειν καὶ λαμβάνειν | τὸ δίκαιον ἐπὶ τῶν χρηματιστῶν. ὅσοι δὲ Ἕλληνες | ὄντες συνγράφονται κατʼ αἰγύ(πτια) συναλλάγματα | ὑπέχειν τὸ δίκαιον ἐπὶ τῶν λαοκριτῶν κατὰ τοὺς |τῆς χώρας νόμους. τὰς δὲ τῶν Αἰγυ(πτίων) πρὸς τοὺς |αὐτοὺς Γυ(πτίους) κρίσεις μὴ ἐπισπᾶσθαι τοὺς χρημα(τιστὰς) |ἀλλʼ ἐᾶν ⟦κριν⟧ διεξάγεσθαι ἐπὶ τῶν λαοκριτῶν κατὰ τοὺς | τῆς χώρας νόμους. And they have decreed in cases of Egyptians who bring actions against Greeks and in cases of Greeks who bring actions against Egyptians, or of Egyptians against Egyptians, and Greeks against Greeks, regarding all classes except the cultivators of Crown land and the tax-payers and all others connected with the revenues, that Egyptians who reach an agreement with Greeks with contracts in Greek shall have the right of suing and being sued before the chrematistai; but those who are Greek who sign agreements with contracts in Egyptian they shall be sued before the native judges according to the local laws; and

284

S. TORALLAS TOVAR

that suits of Egyptians against Egyptians shall not be brought by the chrematistai, but they shall allow them to be administered before the native judges according to the local laws (P.Tebt. 1.5.207).

This regulation did not impose the use of any language, but showed awareness of the legal situation entailed by the use of one language or the other, and consequently the use of one code of law or the other, and organized the access to court depending on the language in which contracts were written. Language use is not the main issue at stake here, but rather the laws written in that language. The danger of affecting the letter of the law by translating or interpreting into a second language seems to prompt an effort at regulation. The impulse presumably did not derive from ignorance of these languages, since bilingualism was very high in Egypt in this period, at least among the scribal class. Through bureaucratic requirements that affected everyday economic life, Ptolemaic authorities influenced linguistic usage. Knowledge of Greek was not imposed, but a certain familiarity was required for important aspects of civil life, such as recourse to Greek-speaking scribes in order to have agreements registered in the prescribed way. The policy is thus promotive, while still tolerant of the use of the other language (and the other law) as long as the bureaucratic use complied with the minimum requirement of registration in Greek. d. Another type of linguistic policy, which Spolsky (2004) calls covert, favoured the use of Greek through indirect paths. The tax exemptions offered to Greek schoolteachers, and even Egyptian teachers of Greek as a second language (obol tax in CPR 13.2; salt tax in P.Hal. 1.260; Thompson 1992a, 325; Clarysse and Thompson 2006, 125–33) discouraged the use of Egyptian and incentivized the extensive spread of Greek.8 The promotion of literacy in Greek through the path of education supported a widespread bureaucracy in Greek.9

8

9

Bresciani 1983, 181–84: P.Rain.Cent. 5, AD 95/96, Soknopaiou Nesos, establishes the required conditions for a temple scribe. On demotic education emanating from temples, see Maehler 1983, 191–203, 192–93, 196–97. Thompson (1992a, 326) discusses a Ghoran papyrus (P. Lille Dem. 3.101 [244/3 BC]) in which apparently an Egyptian with a Greek name enjoyed exemption from a tax although his brothers, with Egyptian names, did not. Perhaps the social

This specific instance of language promotion through tax exemption can be compared to a few historical texts that attest to the promotion of schooling in Latin in the provinces. As I stated above, these historical texts provide contexts, but no policy, and their interpretation is the subject of debate. There seems to have been no legislated effort to promote schooling in Latin, and some of the schools we know of, like the one in Augustodunum in Gallia, attested in Suetonius (Caligula 45, 2; Balsdon 1979, 122), were probably an initiative of the local elites in an effort to make a Roman education available to their own children for their social advancement. Two examples, perhaps taken as isolated cases, attest to a promotive foundation of schools and teaching of Latin: Tacitus refers to the efforts of Agricola to educate the young sons of the local noble families, with the result that those who once disdained the language of Rome, now coveted its eloquence (Agr. 21).10 Plutarch (Sert. 14) informs us about Sertorius who opened a school in Osca, as an excuse to take the Oscans as ‘hostages’ to his Latinization. In general, it can be said that schools and the learning of Latin were established by the local elites in order to provide their children with instruments of social advancement, the Latin language and rhetoric. However, the exemption from tax for teachers attested in the papyri is a rare instance of actual policy, presumably intended to have an effect on the use of a language, as discussed above. Another example of a covert policy in Ptolemaic Egypt may be the establishment of Greek-speaking authorities and police forces in towns and villages. This must have brought about an extensive use of Greek for certain practical matters. But for this policy we only have the resulting documents as witness: the fact that all petitions to the primarily Greek-speaking law enforcement administrative class of Ptolemaic Egypt were written in Greek shows the effect of these indirect policies. Few traces of demotic appeals survive. The

10

advancement that the knowledge of Greek provided, such as a career in administration, qualified him to be considered Greek. Tacitus Agr. 21: ‘Moreover he began to train the sons of the chieftains in a liberal education, and to give a preference to the native talents of the Briton as against the trained abilities of the Gaul. As a result, the nation which used to reject the Latin language began to aspire to rhetoric’ (Hutton and Peterson 1914, 66–67).

GREEK AND EGYPTIAN

Greek and demotic elements of the archive of Hor (O.Hor; Ray 1976) cannot be considered fragments of a proper petition, as many of the texts that make up the archive are dream texts and oracles and thus of questionable value as witnesses to administrative petition practices. A fragmentary demotic version of a Greek petition that survives elsewhere in two copies demonstrates only that an Egyptian petitioner may have wished to keep a private copy of a complaint to the authorities, written in his own tongue (UPZ I 6a, Memphis, 163 BC). Aside from these few texts, we occasionally see Egyptians writing in Egyptian to Egyptian police officials, but only to arrange for bail payments.11 We should deduce that the agents of the law were Greek-speaking and that petitions, in order to receive attention, had to be written in Greek. Egyptians from all segments of society knew how to construct appeals to law enforcement or to hire those who did, and they did so when necessary (Bauschatz 2013, 188–89). But, as stated above, this hypothesis is based on the lack of documents, an argument ex silentio. Roman Egypt, Roman empire After almost three centuries of intermixing and the progressive predominance of Greek, Philo of Alexandria still found it meaningful to say in his Embassy to Gaius (147) that Augustus had ‘hellenized’ Egypt and the Eastern Empire: οὗτος ὁ τὰς πόλεις ἁπάσας εἰς ἐλευθερίαν ἐξελόμενος, ὁ τὴν ἀταξίαν εἰς τάξιν ἀγαγών, ὁ τὰ ἄμικτα ἔθνη καὶ θηριώδη πάντα ἡμερώσας καὶ ἁρμοσάμενος, ὁ τὴν μὲν Ἑλλάδα Ἑλλάσι πολλαῖς παραυξήσας, τὴν δὲ βάρβαρον ἐν τοῖς ἀναγκαιοτάτοις τμήμασιν ἀφελληνίσας. And he was able to claim every state to freedom, he who led disorder into order and tamed and and harmonised all unsocial and savage peoples, he who enlarged Hellas with many Hellenic states and hellenized the barbarians in the most important regions (Philo, Legatio, 147).

It is difficult to know what Philo might mean by ‘hellenization’. But the use of demotic and other

11

12

P.Bürgsch 16 (Memphis, 159 BC); 22 (Ghoran, 243 BC); P.Lille Dem. 1.4 (Ghoran, 247 BC). I do not mean there was a period when no Egyptian at all was written. It just disappeared from the domain of bureaucracy. Consequently, no more scribes were trained for the production

285

vernacular languages decreased dramatically in the written records in the first century of the common era, and eventually disappeared in some domains (Depauw 2003, 89–90; Muhs 2005). Is this an instance of the application of Roman linguistic policy? It is generally agreed that this is not exactly the case. Bagnall (1988) has offered the most convincing explanation for this. He relates the demise of demotic to the progressive loss of power and influence of Egyptian religion. The temples, together with their staff, festivals, belongings and scripts, suffered increasing decline from the first to the third century AD. Although the reasons for the demise of the demotic script are still under discussion (see Lewis 1993; Stadler 2008), it is clear that the Egyptian language had no easily accessible written medium for everyday purposes during the beginning of the Roman period. This seems a clear case of ‘minorization’ of a language in some specific domains: some, but not all, since we know that the language continued to be spoken, eventually re-emerging in all its glory to become the language of the Coptic Church.12 It remains to be discussed here whether the disappearance of Egyptian from our written record, i.e., from the domain of bureaucracy, is the result of a specific policy or not. Roman linguistic policy belongs among a larger set of legal tactics by which Rome dealt with the foreign populations of its annexed territories. Roman authorities seem to have believed that respecting local legal traditions and barring non-citizens from access to the use of Roman law would be more profitable for Rome, both economically and in terms of social order, than imposing their law relentlessly. The annexed peoples functioned under their own legal systems, and the challenge for Rome was to administer an ever-growing realm, heterogeneous in all possible ways: ethnically, religiously, and of course, juridically and linguistically (Ando 2011; 2015). A similar respect for and integration of ‘foreign’ legal systems in Ptolemaic Egypt (in this case the ‘native’ legal system) has been discussed above. In the Roman period regulations come from outside Egypt, and often apply to other areas of the empire as

of documents and the demotic script was relegated to temple use. The eventual use of the Greek alphabet for Egyptian by bilingual scribes is connected to the development of magic, and later on to the first Coptic versions of the Bible.

286

S. TORALLAS TOVAR

well as Egypt.13 Moreover another language—Latin— appears on the playground and alters the balance. I will not discuss the tension between Latin and Greek, since it is beyond the scope of this paper, and there is plenty of bibliography on this, amply discussing the question.14 For this period the position of the vernacular languages, in our case Egyptian, is overshadowed by the friction between these two languages of power. It is quite clear that there was no imposition of the Latin language, no aggressive linguistic dirigisme, just as there had not been in the Ptolemaic period. Rather, we largely find indifference towards languages other than Latin and Greek. Although it is true that Claudius is said to have removed immediately a Lycian from the privilege of Roman citizenship when he heard his bad Latin (Latini sermonis ignarum), both sources for this ‘anecdote’ highlight the emperor’s inconstant and capricious character.15 It was in fact possible to be a Roman citizen without knowing the Latin language, as the Constitutio Antoniniana would prove in AD 212, when it granted citizenship to all free inhabitants of the empire, for many of whom Latin cannot have been the first language and might have been very imperfectly mastered, if spoken at all. Most of the available Roman legal sources for what we might see as linguistic policy16 refer only to Latin

13

14 15

16

17

For a possible comparison to the linguistic situation at the same time (thus, under similar linguistic influences), one might turn to the Latinization of Hispania (cf. Tovar 1968; 1974; 1980; De Hoz 2010; Beltrán Lloris 2011). The situation is quite different: in the Augustan period the Romans had been in Hispania already for almost 200 years, and the pre-Roman civilizations and cultures did not have a developed intellectual life, scribal practice and bureaucracy, as Egypt had, and Latinization often arrived together with literacy in some cases (see e.g., Lusitanian inscriptions, all in Latin script). Kaimio 1979; Dubuisson 1982; Adams 2003a; Clackson 2015. Suetonius, Claud. 16.4; perhaps the same case in Cassius Dio 40.17.4. See Kaimio 1979, 134–35; Dubuisson 1982, 208. The interpretation instead might be that the deprivation of citizenship was due to Claudius’ being furious and agitated because of a revolt. For the sample of texts, Wacke 1993; 1999. I thank José Domingo Rodríguez Martín for his invaluable assistance in this matter. Digest 2.13.7.1 (Paulus, lib. III ad Edictum): Haec vox ‘iterum’ duas res significat: alteram, qua demonstraretur tempus secundum, quod Graeci δεύτερον dicunt; alteram, quae ad insequentia quoque tempora pertinet, quae Graece dicitur πάλιν, quod ita accipitur ‘quotiens opus erit’. Νam potest fieri ut bis editam sibi rationem quis perdiderit: ut verbum ‘iterum’ pro ‘saepius’ accipiatur; Digest 50.16.5.1 (Paulus, lib. II ad Edictum): ‘Opere

and Greek, displaying no interest in vernacular languages. The new languages (and legal systems) that came in contact with Latin in the growing empire required flexibility: the old contracts of the Roman ius civile, based on strict Latin formulas, accommodated to the tools of the ius gentium, adapted to trade between peoples of diverse origin who operated within the limits of Roman law in a multilingual environment (Gaius 3.93 and 179, on sponsio and stipulatio). This new situation also brought about debates on the correct interpretation and translation of juridical terms (Digest 2.13.7.1; Coll. 2.5.1).17 Interestingly, these texts often give, after definitions of legal Latin terms, ‘quod Graeci appellant...’, the corresponding term in Greek, even also in other languages.18 Most of these texts refer exclusively to Greek as the other possible legal language (e.g., the possibility of formulating a stipulatio in Greek),19 as long as there was coherence and understanding (Gaius 3.39, cf. 179): Sed haec quidem verborum obligatio ‘Dari spondes? Spondeo’ propria civium romanorum est; ceterae vero iuris gentium sunt, itaque inter omnes homines, sive cives romanos sive peregrinos, valent. Et quamvis ad graecam vocem expressae fuerint, velut hoc modo Δώσεις; Δώσω. Ὁμολογεῖς; Ὁμολογῶ. Πίστει κελεύεις; Πίστει κελεύω. Ποιήσεις; Ποιήσω, etiam hae tamen inter cives romanos valent, si modo graeci sermonis intellectum habeant; et

18

19

locato conducto’: his verbis Labeo significari ait id opus, quod Graeci ἀποτέλεσμα vocant, non ἔργον, id est ex opere facto corpus aliquod perfectum; Coll. 2.5.1: Generaliter dicitur iniuria omne, quod non iure fit: specialiter alia est contumelia, quam Graeci ὕβριν appellant, alia culpa quam Graeci ἀδίκημα dicunt, sicut in Lege Aquilia damnum iniuriae accipitur, alia iniquitas et iniustitia, quam Graeci ἀδικίαν vocant. Nam cum praetor non iure adversum nos pronuntiat, iniuriam nos accepisse dicimus: unde apparet non est verum, quod Labeo putabat, apud praetorem iniuriam ὕβριν dumtaxat significare. Digest 23.3.9.3 (Ulpianus, lib. XXXI ad Sabinum): Ceterum si res dentur in ea, quae Graeci παράφερνα dicunt quaeque Galli (alii, Mo.) ‘peculium’ appellant, videamus, an statim efficiuntur mariti. Stipulatio was a form of contract, which gave rise to a unilateral obligation: a person was bound by returning an affirmative answer to an oral question of the stipulator (Cicero, pro Caec. 3.7). Originally there was a strict requirement to use that exchange of question and answer. Moreover, only Roman citizens could use this form of contract, while peregrini could not (Gai. 3.93, 179), until other words were sanctioned to make available the form of contract to aliens. By the time any language was accepted between question and answer (Cod. 8.38.10; Inst. 3.15.1), the need was felt to have contracts put down in writing, instead of just having an oral agreement in the presence of witnesses.

GREEK AND EGYPTIAN

e contrario quamvis latine enuntientur, tamen etiam inter peregrinos valent si modo latini sermonis intellectum habeant. At illa verborum obligatio ‘Dari spondes? Spondeo’ adeo propria civium romanorum est, ut ne quidem in graecum sermonem per interpretationem proprie transferri possit, quamvis dicantur a graeca voce figurata esse.

Or according to a frequently cited passage of Ulpian, the fideicommissum could be written even in vernacular languages:

But while the verbal obligation contracted by the expressions, ‘Do you solemnly agree to give?’ ‘I do solemnly agree to give,’ is peculiar to Roman citizens, the other (obligations) belong to the Law of Nations, and therefore they are valid among all men, whether they are Roman citizens or foreigners. And even if they are uttered in the Greek language: either Δώσεις; Δώσω (will you give? I will give) Ὁμολογεῖς; Ὁμολογῶ (do you agree? I agree) Πίστει κελεύεις; Πίστει κελεύω (do you guarantee? I guarantee) Ποιήσεις; Ποιήσω (will you fulfil this? I will fulfil) they are still valid, as long as Roman citizens are concerned, if they understand Greek; and on the contrary, although they may be stated in Latin, they will nonetheless be valid among foreigners, provided they understand the Latin language. The obligation contracted by the words, ‘Do you solemnly agree to give?’ ‘I do solemnly agree to give,’ is peculiar to Roman citizens to such an extent that it cannot properly be translated into the Greek language, although it is said to have been modeled on the Greek (Gaius 3.93).

Fideicommissa may be left in any language, not only Latin or Greek but Punic, Gallic or that of any other nation (Ulpianus, Digest 32.11 pr).

The fact that only one phrase seems to be untranslatable allows the inference that there was a broad tolerance for translation. Wills had to be written in Latin, while fideicommissa—the provisions made in connection to wills for those who were not direct heirs—could be written in Greek): Item legata graece scripta non valent; fideicommissa vero valent. Legacies bequeathed in the Greek language are not valid; fideicommissa (in Greek) however are valid (Gaius 2.281; cf. Ulpianus Reg. 25.9).20

20

21

287

See also BGU 5.1210, a mid-second-century AD copy of the Gnomon of the Idioslogos, §8, ll. 35–37, which clearly states that using Greek in a Latin testament voids the document. Institutes 3.15.1: The paraphrasis of Theophilus Antecessor includes, as an example, the Syrian and Egyptian languages: Inst. Teoph. 3.15.1 Καὶ ἀδιάφορον ἦν, πρότερον Ῥωμαϊστί,

Fideicommissa quocumque sermone relinqui possunt, non solum Latina vel Graeca, sed etiam Punica vel Gallicana vel alterius cuiuscumque gentis

Only three texts in the Digest, like the one just above (Digest 32.11 pr.), mention vernacular languages (see Digest 45.1.1.6 and tangentially Digest 23.3.9.3), with reference to Punic, Syrian and Gallic. None mentions Egyptian explicitly, except for the much later Greek paraphrasis by Theodoros Antecessor21 (see below). These languages were not legally relevant and thus, they appear only when a practical issue needed to be addressed in relation to them. Daily transactions probably required ‘legal’ attention to these languages in the provinces, but not enough to require specific regulation, at least not regulation emanating from Rome. Moreover, no linguistic policy could cover the enormous variety of linguistic situations in the empire. Thus, the few instances we have seem to show some flexibility in the use of languages—any language including the vernacular—as long as there is mutual comprehension. Documents concerning citizens, including birth certificates (professiones in albo when the child was legitimate, testatio for others), wills and documents concerning inheritance (e.g., cretio hereditatis and agnitio bonorum possessionis), and requests for appointing a guardian had to be produced in Latin, at least until the Constitutio Antoniniana.22 We have evidence of a constitution of Severus Alexander, attested only in a papyrus document dated to AD 235, allowing the use of Greek for wills:

22

Ἑλληνιστὶ ἢ οἱᾳδήποτε ἑτερᾳ διαλέκτῳ, οἷον τῇ τῶν Σύρων ἢ τῶν Αἰγυπτίων, ἡ ἐπερώτεσις γέγονε, μόνον μέντοι εἰ ἑκάτερος τῶν συναλλαττόντων ἐνόει τὰ ἐκφωνούμενα. Amelotti 1966, 112–15; Wacke 1999, 133–39; Halla-aho 2013. On the chronology of language use in Latin testaments, see Rochette 2000.

288

S. TORALLAS TOVAR

τὴν διαθήκην ἐποίησα γράμμασιν] Ἑλληνικοῖς ἀκο[λού]θως τῇ θείᾳ κ[ελε]ύσ[ει][ τοῦ κυρίου ἡμῶν Αὐτοκράτορος Μάρκου Αὐρηλίου] Σεουήρου Ἀλεξάνδρ[ο]υ. I have produced the will in Greek letters, according to a divine constitution of our Lord, the emperor Marcus Aurelius Severus Alexander (SPP 20.35; SB 1.5294.12–14).

Even before this constitution, soldiers and veterans were (exceptionally) allowed to write their wills in Greek.23 Also in the field of wills, according to Digest (28.2.20) no knowledge of Latin was required for witnesses to wills. The linguistic question is solved the moment the witness knows (or it is explained to him) what his presence is for: nam si vel sensu percipiat quis, cui rei adhibitus sit, sufficere. In a different field, that of contracts, there is one more example I know of concerning language regulation, this one mentioning even the vernacular languages, referring to verbal contracts: Eadem an alia lingua respondeatur, nihil interest. Proinde si quis Latine interrogaverit, respondeatur ei Graece, dummodo congruenter respondeatur, obligatio constituta est: idem per contrarium. Sed utrum hoc usque ad Graecum sermonem tantum protrahimus an vero et ad alium, Poenum forte vel Assyrium vel cuius alterius linguae, dubitari potest. Et scriptura Sabini, sed et verum patitur, ut omnis sermo contineat verborum obligationem, ita tamen, ut uterque alterius linguam intellegat sive per se sive per verum interpretem. It makes no difference whether the reply is made in one language or the other. Therefore, if someone asks in Latin but receives a reply in Greek, as long as the reply is congruent, the obligation is settled. Similarly in the reverse case. Whether we can extend this rule to the Greek language only, or also to another, such as Punic or Assyrian, or some other tongue, is a matter for doubt. The writings of Sabinus however permit that all tongues can produce a verbal obligation, provided that each of both parties understand each other’s language, either personally or by means of a truthful translator (Dig. 45.1.1.6; Ulpianus, lib. XLVIII ad Sabinum).

The requirement was that of ‘congruenter respondere’ and there was no required language to be used as long as the verbal agreement was reached. Equally, any

23

See Stein 1915, 143; Taubenschlag 1955, 193, 199; Rochette 1997, 112–13 and 115; 2000. For a list of documents, see Halla-aho 2013.

language was good for declarations as long as there was no doubt about it: Gaius 2.114, quomodo velint vel quomodo possint. Finally I will mention a passage of Ulpian, where, referring to a public notice about administrators, he insists on intelligibility: the notice has to be placed visibly and has to be written in the language that everybody will understand. Proscribere palam sic accipimus claris litteris, unde de plano recte legi possit, ante tabernam scilicet vel ante eum locum in quo negotiatio exercetur, non in loco remoto, sed in evidenti. Litteris utrum Graecis an Latinis? Puto secundum loci condicionem, ne quis causari possit ignorantiam litterarum. Certe si quis dicat ignorasse se litteras vel non observasse quod propositum erat, cum multi legerent cumque palam esset propositum, non audietur. We understand by public notice one that is clearly written, which is visible and can be easily read, for example, in front of the shop or the place of business, not in a remote place but on display. Should the notice be written in Greek or Latin? I think it depends on the locality; in order that no one can claim ignorance of the written notice. Certainly, if the notice was posted publicly and was read by many, no one will be heard to say that he did not understand the text nor see it was on display (Digest 14.3.11.3; Ulpianus, lib. ΧXVIII ad Edictum).

The drive for intelligibility might seem here to be motivated primarily by the desire to facilitate trade, but the concern is more broadly with legal legitimacy. If people are to be held accountable to or liable for the content of law, they must have reasonable access to that content. The drive for legal legitimacy, therefore, results in various strategies of accommodation to local linguistic practice. As we can see, legislators seem quite aware of such local practices: veterans and soldiers, on their deathbed, were allowed to write or dictate their last will in any language; the witnesses for some transactions were not required to know Latin, as long as they understood what they were acting as witness for; verbal agreements could be performed in any language as long as there was mutual comprehension; some public notices needed to be visible and written

GREEK AND EGYPTIAN

in a language understood by most. These few instances suggest that Roman linguistic regulation was flexible enough to facilitate some aspects of life in the provinces, and even recognized the existence of the vernacular languages and their position in society, though it seldom found it necessary to address these explicitly. Apart from these few instances it is clear that the study of the legal role of these languages depends almost exclusively on documentary evidence, rather than legal sources. The problem in the interpretation of these legal sources and their relationship to the actual linguistic situation in Egypt is that the distance is almost impossible to bridge. The papyri may provide a ‘statistical’ idea of tendencies, but that can only be used with care. Conclusion The surviving evidence for linguistic policies in the Graeco-Roman period is problematic. While literary and historical sources loom large, their rhetoric seldom gives us purchase on policy ‘realities’. The general corpus of papyri, whether viewed statistically or in detail, are useful for understanding the result of whatever policies there may have been, but they rarely make any such policies explicit. Few are the passages referring to languages in the Corpus of Roman Law, even fewer those referring to vernacular languages. It is hard to tell how these decrees affected the use of Coptic in Egypt. As much as the Egyptian language evidently did not disappear, some vernacular languages in other parts of the empire certainly did.24 The maintenance and disappearance of vernacular languages in the Roman empire may have had less to do with any linguistic policy imposed by the Romans, and more with the individual situation of each language and the different levels of collective pressure from a language of power that each of them could endure. Collective pressure can produce a spectrum of attitudes, from embarrassment about the use of the vernacular language to language loyalty and pride. The embarrassment of Septimius Severus at his sister’s

24

Egypt presents a special case, since the production and preservation of documents throughout this period gives at least a glimpse into linguistic use, while other areas of the empire present a much more difficult situation owing to lower

289

precarious Latin serves as an example of one extreme, very different from the proud attitude of the Hispanus from Termes attested by Tacitus, mentioned previously. Very few sources provide direct evidence for linguistic policies that might underlie linguistic evolution in Graeco-Roman Egypt, although evolution certainly took place. In the Ptolemaic period the use of Greek increasingly occupied the domain of bureaucracy, noticeable in the drastic decrease of certain forms of documentation in demotic, relegating the Egyptian language to temple use, and ultimately to literary and private use. The Augustan ‘hellenization’ of Egypt, according to Philo, coincides with the gradual ‘minorization’ of the Egyptian language that excluded it from the written record. This coincides with a similar phenomenon affecting other languages in the Mediterranean. Some of these languages were ‘minorized’ by the use of Latin or Greek even in the private domain, so that they ultimately vanished. The Roman legal texts that illustrate some kind of linguistic policy during the first Roman centuries seem to regulate law and trade rather than linguistic use, the latter becoming relevant primarily as a vehicle of law. The Constitutio Antoniniana in 212 seems to be a turning point in these policies. The extension of Roman citizenship to subjects who did not speak Latin, at least as a first language, brought flexibility in the use of languages in domains where previously there was protectionism. Perhaps this seemingly new flexibility reflects a pre-existing reality evident, for example, in the use of Greek in wills: even before the date of the constitution of Severus Alexander half of Roman wills preserved on papyrus were written in Greek. It is possible that these were really translations of Latin originals, but that is difficult to determine (P.Oxy. 38.2857, AD 134, pretends to be the model of a Latin will, while P.Oxy. 20.2348, AD 234, is a translation into Greek of a Latin testament that was perhaps never produced). After AD 212, a point of inflection, we find the return of the Egyptian language into the written record. The sources permit us to watch how, after a period of experimentation, a new Coptic writing system took over from the waning demotic, as

production and no preservation of organic material. The fate of other languages is, therefore, more difficult to determine. See Millar 1968; Clackson 2015.

290

S. TORALLAS TOVAR

a way to record a language that had evidently been widely spoken, even as (in demotic form) it had been more and more rarely written. Clackson (2015, 69) criticizes McMullen (1966) in his reconstruction of the linguistic situation in the empire at this point, for not having taken into account the level of bilingualism of the provinces. The vernacular languages had not ceased to be spoken, but had simply disappeared from the written record, and their return into it is not the result of a revival or a language

choice, but the recovery of the written practice, especially in ritual and religious contexts.

Bibliography

Capponi, L. 2010. Spaces of justice in Roman Egypt. In Spaces of justice in the Roman world, F. De Angelis (ed.), 251–76. Leiden. Choat, M. 2012. Coptic. In The Oxford handbook of Roman Egypt, C. Riggs (ed.), 581–93. Oxford. Clackson, J. 2012. Language maintenance and language shift in the Mediterranean. In Multilingualism in the GraecoRoman worlds, A. Mullen and P. James (eds), 36–57. Cambridge. ———. 2015. Language and society in the Greek and Roman worlds. Cambridge. Clarysse, W. 1993. Egyptian scribes writing in Greek. Chronique d’Égypte 68: 186–201. ———. 1998. Ethnic diversity and dialect among the Greeks of Hellenistic Egypt. In The two faces of Graeco-Roman Egypt: Greek and demotic and Greek-demotic texts and studies presented to P. W. Pestman, A. M. Verhoogt and S. P. Vleeming (eds), 1–13. Papyrological LugdunoBatava 30. Leiden; Boston; Cologne. Clarysse, W. and D. J. Thompson. 2006. Counting the people in Hellenistic Egypt 1: Population registers (P. Count); 2: Historical studies. 2 vols. Cambridge. Crespo, E. 2007. The linguistic policy of the Ptolemaic kingdom. In ΦΩΝΗΣ ΧΑΡΑΚΤΗΡ ΕΘΝΙΚΟΣ: Actes du Ve congrés international de dialectologie grecque (Athènes, 28–30 septembre 2006), M. Hatzopoulos (ed.), 35–49. ΜΕΛΕΤΗΜΑΤΑ 52. Athens. De Hoz, J. 2010. Historia lingüística de la Península Ibérica en la antigüedad I: Preliminares y mundo meridional prerromano. Madrid. Depauw, M. 2003. Autograph confirmations in demotic private contracts. Chronique d’Égypte 78: 66–111. ———. 2011. Physical descriptions, registration and εἰκονίζειν with new interpretations for P.Par. 65 and P.Oxy. I 34. Zeitschrift für Papyrologie und Epigraphik 176: 189–99.

All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed 8 July 2018). Adams, J. N. 2003a. Bilingualism and the Latin language. Cambridge. ———. 2003b. Romanitas and the Latin language. Classical Quarterly 53: 184–205. Allam, S. 1991. Egyptian law courts in pharaonic and Hellenistic times. Journal of Egyptian Archaeology 77: 109–27. Amelotti, M. 1966. Il testamento romano. Florence. Ando, C. 2011. Law, language and empire in the Roman tradition. Philadelphia, PA. ———. 2015. Roman social imaginaries: Language and thought in contexts of empire. Robson Classical Lectures. Toronto. Bagnall, R. S. 1988. Combat ou vide: Christianisme et paganisme dans l’Égypte romaine tardive. Ktema 13: 285– 96. Balsdon, J. P. V. D. 1979. Romans and aliens. London. Bauschatz, J. 2013. Law and enforcement in Ptolemaic Egypt. Cambridge. Beltrán Lloris, F. 2011. Lengua e identidad en la Hispania Romana. Palaeohispanica 11: 19–59. Bingen, J. 2007. The Revenue Laws Papyrus: Greek tradition and Hellenistic adaptation. In Hellenistic Egypt: Monarchy, society, economy, culture, R. S. Bagnall (ed.), 157–88. Hellenistic Culture and Society 49. Berkeley; Los Angeles, CA. Bresciani, E. 1983. P.Rain.Cent. 5. In Festschrift zum 100-jährigen bestehen der Papyrussammlung der Österreichischen Nationalbibliothek, H. Loebenstein (ed.), 181–84. Vienna.

Acknowledgements I am grateful for the comments and corrections of Cesar Montoliu (Barcelona), José Domingo Rodríguez Martín (Madrid), Clifford Ando (Chicago) and David Nirenberg (Chicago), which helped me to improve this paper. All translations are my own.

GREEK AND EGYPTIAN

Dubuisson, M. 1982. Y a-t-il une politique linguistique romaine? Ktema 7: 197–210. Fournet, J. L. 2009. The multilingual environment of Late Antique Egypt: Greek, Latin, Coptic and Persian documentation. In The Oxford handbook of papyrology, R. S. Bagnall (ed.), 418–51. Oxford; New York. Halla-aho, H. 2013. Bilingualism in action: Observations on document type, language choice and Greek interference in Latin documents and letters on papyri. In Bilinguisme et digraphisme dans le monde gréco-romain: L’apport des papyrus latins. Actes de la Table Ronde internationale (Liège, 12–13 mai 2011), M.-H. Marganne and B. Rochette (eds), 169–81. Papyrologica Leodiensia 2. Liège. Hutton, M. and W. Peterson. 1914. Tacitus: Agricola. Germania. Dialogue on Oratory. Loeb Classical Library 35. Cambridge. Kaimio, J. 1979. The Romans and the Greek language. Helsinki. Kapetanopoulos, E. 1999. Alexander’s Patrius Sermo in the Philotas affair. Ancient World 30 (2): 117–28. Keenan, J. 2014. Languages of law. In Law and legal practice in Egypt from Alexander to the Arab conquest, J. Keenan (ed.), 96–144. Cambridge. Kloss, H. 1977. The American bilingual tradition. Rowley, MA. Lewis, N. 1993. The demise of the demotic document: When and why. Journal of Egyptian Archaeology 79: 276–81. Maehler, H. 1983. Die griechische Schule im ptolemäischen Ägypten. In Egypt and the Hellenistic world: Proceedings of the international colloquium, Leuven, 24–26 May 1982, E. Van’t Dack, P. Van Dessel and W. Van Gucht (eds), 191–203. Studia Hellenistica 27. Leuven. McMullen, R. 1966. Provincial languages in the Roman empire. American Journal of Philology 87 (1): 1–17. Mélèze-Modrzejewski, J. 1975. Chrematistes et laocrites. In Le monde grec: Hommages à Claire Preaux, J. Bingen, G. Cambier and G. Nachtergael (eds), 699– 708. Brussels. ———. 1995. Law and justice in Ptolemaic Egypt. In Legal documents of the Hellenistic world, M. J. Geller and H. Maehler (eds), 1–19. London. Millar, F. 1968. Local cultures in the Roman empire: Libyan, Punic and Latin in Roman Africa. Journal of Roman Studies 58: 126–34. Muhs, B. P. 2005. The grapheion and the disappearance of demotic contracts in early Roman Tebtynis and Soknopaiou Nesos. In Tebtynis und Soknopaiou Nesos: Leben im römerzeitlichen Fajum. Akten des Internationalen Symposions vom 11. bis 13. Dezember 2003 in Sommerhausen bei Würzburg, S. Lippert and M. Schentuleit (eds), 93–104. Wiesbaden. ———. 2010. A late Ptolemaic grapheion archive in Berkeley. In Proceedings of the 25th International Congress

291

of Papyrology, Ann Arbor, July 29–August 4, 2007, T. Gagos (ed.), 581–88. American Studies in Papyrology. Ann Arbor, MI. Patrick, D. 2010. Language dominance and minorization. In Society and language use, J. Jaspers, J.-O. Östman and J. Verschueren (eds), 176–91. Amsterdam. Peremans, W. 1964. Über die Zweisprachigkeit im ptolemäischen Ägypten. In Studien zur Papyrologie und antiken Wirtschaftsgeschichte: F. Oertel zum achtzigsten Geburtstag gewidmet, H. Braunert (ed.), 49–60. Bonn. ———. 1983. Le bilinguisme dans les relations gréco-égyptiennes sous les Lagides. In Egypt and the Hellenistic world: Proceedings of the international colloquium, Leuven, 24–26 May 1982, E. Van’t Dack, P. Van Dessel and W. Van Gucht (eds), 253–80. Studia Hellenistica 27. Leuven. Pestman, P. W. 1985. The competence of Greek and Egyptian tribunals according to the decree of 118 BC. Bulletin of the American Society of Papyrologists 22: 265–69. Ray, J. D. 1976. The archive of Hor. London. Rémondon, R. 1964. Problèmes du bilinguisme dans l’Égypte lagide (UPZ I 148). Chronique d’Égypte 39: 126–46. Rochette, B. 1995. Grecs et Latins face aux langues étrangères: Contribution à l’étude de la diversité linguistique dans l’antiquité classique. Revue Belge de Philologie et d’Histoire 73 (1): 5–16. ———. 1996. Sur le bilinguisme dans l’Égypte grécoromaine. Chronique d’Égypte 71: 153–68. ———. 1997. Le latin dans le monde grec. Recherches sur la diffusion de la langue et des lettres latines dans les provinces hellénophones de l’empire romain. Brussels. ———. 1998. Le bilinguisme gréco-latin et la question des langues dans le monde gréco-romain: Chronique bibliografique. Revue Belge de Philologie et d’Histoire 76 (1): 177–96. ———. 2000. La langue des testaments dans l’Égypte du IIIe s. ap. J.-C. Revue internationale des droits de l’antiquité, 3e série 47: 449–61. Schiffman, H. 1996. Linguistic culture and language policy. London; New York. Spolsky, B. 2004. Language policy. Cambridge. Stadler, M. A. 2008. On the demise of Egyptian writing: Working with a problematic source basis. In The disappearance of writing systems: Perspectives on literacy and communication, J. Baines, J. Bennet and S. D. Houston (eds), 157–81. London; Oakville, CT. Stein, A. 1915. Untersuchungen zur Geschichte und Verwaltung Aegyptens unter römischer Herrschaft. Stuttgart. Taubenschlag, R. 1955. The law of Greco-Roman Egypt in the light of the papyri: 332 BC–640 AD. 2nd ed. Warsaw. Thompson, D. J. 1992a. Literacy and the administration in early Ptolemaic Egypt. In Life in a multicultural society: Egypt from Cambyses to Constantine and beyond, J. Johnson (ed.), 323–26. Chicago IL.

292

S. TORALLAS TOVAR

———. 1992b. Literacy in early Ptolemaic Egypt. In Proceedings of the 19th International Congress of Papyrology (Cairo 2–9 September 1989), A. H. S. el-Mosalamy (ed.), 2: 77–90. Cairo. ———. 1994. Literacy and power in Ptolemaic Egypt. In Literacy and power in the ancient world, A. Bowman and G. Woolf (eds), 67–83. Cambridge. ———. 2009. The multilingual environment of Persian and Ptolemaic Egypt: Egyptian, Aramaic and Greek documentation. In The Oxford handbook of papyrology, R. S. Bagnall (ed.), 395–417. Oxford. Torallas Tovar, S. 2005. Identidad lingüística e identidad religiosa en el Egipto Grecorromano. Barcelona. ———. 2010a. Greek in Egypt. In Blackwell’s companion to the ancient Greek language, E. Bakker (ed.), 253–66. Oxford. ———. 2010b. Linguistic identity in Graeco-Roman Egypt. In The multilingual experience in Egypt: From the Ptolemies to the Abbasids, A. Papaconstantinou (ed.), 17–43. Farnham; Burlington, VT. Tovar, A. 1968. Latín de Hispania: Aspectos léxicos de la romanización. Madrid. ———. 1974. Implantación y desarrollo del latín en Hispania: Los ejemplos de Varrón y Columela. In Colloquio

italo-spagnolo sul tema Hispania Romana (Roma, 15–16 maggio 1972), 95–107. Rome. ———. 1980. Das Vulgärlatein in der Provinzen. In Die Sprachen im Römischen Reich der Kaiserzeit, G. Neumann and J. Untermann (eds), 331–42. Bonner Jahrbücher, Beiheft 40. Pulheim. Vierros, M. 2012. Bilingual notaries in Hellenistic Egypt: A study of Greek as a second language. Collectanea Hellenistica 5. Brussels. ———. 2014. Bilingualism in Hellenistic Egypt. In Encyclopedia of ancient Greek language and linguistics, G. K. Giannakis (ed.), 234–38. Leiden. Wacke, A. 1993. Gallisch, Punisch, Syrisch oder Griechisch statt Latein? Zur schrittweisen Gleichberechtigung der Geschäftssprachen im römischen Reich. Zeitschrift der Savigny-Stiftung für Rechtsgeschichte 110: 14–59. ———. 1999. El pluralismo lingüístico y su reglamentación jurídica en el Imperio Romano. Seminarios Complutenses de Derecho Romano 9: 117–46. Woolf, G. 1996. The uses of forgetfulness in Roman Gaul. In Vergangenheit und Lebenswelt: Soziale Kommunikation, Traditionsbildung und historisches Bewußtsein, H.-J. Gehrke and A. Möller (eds), 361–81. Tübingen

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT Arietta PAPACONSTANTINOU

Like Greek, Arabic was introduced to Egypt through conquest and regime change. The linguistic situation it encountered, however, was very different. The Greekspeaking conquerors, and the Ptolemaic kingdom that succeeded them, arrived and established themselves in a monolingual country where the language of power was the high variety of the local language. The Arabicspeakers who arrived almost a millennium later, in AD 641/642, encountered a bilingual country, where the language of power was Greek, also the administrative and learned lingua franca of the eastern Mediterranean, and the imperial language of their rival empire. In the intervening time the Roman conquest had also resulted in the introduction of Latin as a language of power, but, as in the rest of the eastern provinces, its presence had remained limited and Greek had prevailed as the main language of the administration and the urban elites. Yet far from remaining a simple imperial superstructure, Greek had become naturalized and was spoken widely also in a low variety alongside Egyptian. As for the latter, it had itself undergone a series of changes, in script as in linguistic form: it had adopted the Greek script, had integrated a large amount of loanwords from Greek, and, crucially, had all but lost its high variety—even though in the few decades before the conquest a legal and administrative idiom had gradually started developing again (Richter 2008; Fournet 2010; 2016; 2018a; 2018b). Just like the Romans before them, on arrival the Arabs left much of the administration of the new provinces to be run in Greek. Thus for at least seventy years after the mid-seventh-century conquest, Greek continued to be used in Egypt as a language of empire—that is, used both by the central chancery and by the Egyptian provincial elites who relayed power locally. It also continued to be used in institutional and legal contexts until the end of the seventh century, especially (but not only) in the northern half of the country. The low variety of Greek, however, all but disappeared, at least in writing, within a very short time. This indicates that it was only used as a second language, which it was no longer felt necessary to know. At the same time, in contrast to Latin, Arabic became more and more prominent as a bureaucratic language, causing the use of

Greek to recede in the administration as well, so that by the mid-eighth century it was used only in a few institutional and ecclesiastical contexts. The advent of the ‘Abbāsids quickly saw Arabic become the only imperial language used in Egypt. During the almost century-long period of bilingual administration, Egyptian—or Coptic—gained ground again as the only vernacular, but also continued developing a high legal and administrative idiom which was invariably used for legal transactions in the villages and towns of the Nile valley, as well as for exchanges between villages and their local administrative nodes. Literary production in Coptic also took off during that period. Texts from that time show little linguistic contact with Arabic, which is not surprising given the limited presence of Arabic-speakers in the valley. Indeed, until the early eighth century, the newcomers were mostly settled in garrison cities and did not interact much with the local population. This changed in the course of the eighth century, when we have increasing evidence of Arabic-speaking Muslims settling in the countryside. From the ninth century Coptic seems to have steadily receded, at least as an everyday written language, and with time as a literary one too. It was still used in the late eleventh century in some areas concurrently with Arabic, and in monasteries Coptic texts were copied until well into the fourteenth century. The written version of the language was still mastered by some thirteenth-century Christians sufficiently to compose the Life of John of Phanijoit, recounting the story of a martyr under Saladin. This unique text, however, was an ideological identity statement, which involved an antiquarian attempt to revive what had by then become a fossil liturgical and literary language (Zaborowski 2005). The difference in the linguistic balance between the language of Egypt and the two languages of empire that entered the country by conquest and served for centuries as languages of power is very striking: while Egyptian was used for a millennium alongside Greek, being redefined and reinforced in the process (Fournet 2018a; 2020), within half that time it gave way to Arabic, having shown only a relatively low and specialized degree of linguistic interaction with it (Richter 2001;

294

A. PAPACONSTANTINOU

2004; 2006; 2015a; 2015b; Legendre 2014). To be sure, the relation of Egyptian to Greek was not stable over the long millennium of their coexistence. The role, function and prestige of Egyptian fluctuated depending on context and circumstances (Papaconstantinou 2008; Fournet 2020). Yet at no point in time did it seem to be about to stop being used altogether as it did in the Middle Ages. The linguistic evolution of Egypt has been studied at some length, and with the invaluable evidence offered by the papyrological sources it has been possible to make very fine and detailed observations on the chronology and distribution of linguistic change. Although in recent decades academic discussion of the subject has been entirely reframed, its image outside specialist circles remains beset with the shadow of its Orientalist origins, and the concomitant underlying assumptions about the relative value of languages. It was in such terms that in 1950 the Bollandist Paul Peeters, in his book Le tréfonds oriental de l’hagiographie byzantine, first commented on the difference in the linguistic relation that Egyptian maintained with Greek and with Arabic respectively. Writing about the local elites who ran the country on behalf of the Romans, he wrote: But that minority of Hellenised Egyptians could only fulfil its role as mediating agents by preserving, under its veneer of Greek culture, the language and the thoughtworld of the servile masses that it had to lead. The latter had only learnt the little Greek that they needed for the occasional quarrel with the tax-man and other representatives of the tyrannical administration. And yet it was not the Egyptian race as such that showed itself to be resistant to assimilation: it took the Arabs less than three centuries to disfigure and then abolish its national character. It is rather Hellenism that did not deign to raise the Egyptian peasantry to its level (Peeters 1950, 14).

In Peeters’ view, the reason why Egyptian elites were only superficially Hellenized—and therefore maintained their indigenous language—was that Hellenism was unattainable for them. This was presumably not the case for Arab culture, which the Egyptians adopted relatively quickly. It is clear that for Peeters, this accessibility of Arabic and Arab culture was a sign of its lower cultural status—and that degree of accessibility was indeed very high, since even what he considered as the uncultured and functionalistic ‘Egyptian masses’ were able to acquire it. Much more recently, in a seminal article entitled ‘Why did Arabic succeed where Greek failed?’, David Wasserstein (2003) tackled the same paradox, focusing

mainly on the Levant. A Semiticist less attached to the classical tradition than Peeters, Wasserstein reversed that perspective on Arabic: he saw the inability of Greek to penetrate into all levels of society despite a very long presence in the area as its failure rather than as a sign of its superiority. Conversely, the success of Arabic lay precisely in its ability to be at the same time a language of power and culture, and the first language of everyday speech throughout the Near East and Egypt, to the point that local languages were abandoned or marginalized. Thus, according to Wasserstein, Arabic developed both high and low varieties, while Greek was an exclusively high language, leaving the low variety to the local tongues. Unsurprisingly, neither scholar saw the shift from local languages to Arabic as the result of a deliberate policy on the part of the Arabs. Indeed, as Sofía Torallas shows in this volume, the very notion of linguistic policy is not adapted to the ancient and Medieval worlds. If anything came close to it, it was the combination of incentive and positive reinforcement naturally provided by the possibilities that official languages offered: allowing better careers and direct access to law and administration. Never did it involve, as has been the case in nation states, forceful imposition or prohibition, and this has generally been acknowledged by scholars. Yet on the whole, scholarship has tended to see the language shift in Medieval Egypt as a top-down process, one that saw the higher, more prestigious, more common, more powerful language prevail over the weaker and less significant one. I shall begin with a discussion of this top-down approach, discuss its antecedents and the assumptions that underpin it, and show how it started evolving under the impetus of Leslie MacCoull’s work. I shall then suggest, on the basis of some core sociolinguistic insights into language shift and recent research on early Medieval Egypt, a more balanced view of the progressive establishment of Arabic as the country’s only language. Although this was ultimately the result of the arrival on Egyptian soil of a new empire with a different religion, Arabic won over the population irrespective of its religious affiliation, and it is on the implications of this complex phenomenon that I would like to focus here. Imperial contexts, ancient and modern The top-down approach is linked—not always explicitly—to underlying assumptions regarding the hierarchy of languages and the cultures they carry. It

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

accepts implicitly that imperial languages are in some way superior, acting as vectors of civilization and culture—a phrase most commonly describing some version of classical culture or its derivations. Local languages, on the other hand, are typically constructed as vectors of identity—that of the subjected cultures. The tenacious notion that some languages are superior to others has a long imperial background which is deeply rooted in the colonial past and in Western classical education. Some of its earliest and strongest expressions come from the Spanish New World, where the encounter with cultures using non-alphabetic scripts fostered many assumptions about them. The absence of proper ‘letters’ was given as the explanation for their backwardness and lack of achievement in learning technology, and religion, prompting a concerted effort not only to teach them Spanish, but also to reinterpret the local languages in terms familiar to the Europeans (Mignolo 2003, 29–67). That imposition was justified in the name of progress on all those fronts—especially culture—and was inscribed in the venerable tradition of the classical world: There has never been a Cultured Nation in the World, that when it extended its Conquests, did not attempt the same with its Language: the Greeks took for barbarians those other Nations that remained ignorant of its speech; the Romans, after conquering the Greeks, required them to accept their Latin Language—that of Latium, the Roman countryside—with such rigor that they did not allow anyone who spoke another, foreign language to conduct business in the Senate (Lorenzana 1770, 92; transl. Mignolo 2003, 59–60).

This statement by Francisco Antonio Lorenzana y Buitrón, archbishop of Mexico in the second half of the eighteenth century, is part of his fifth Pastoral Letter, in which he complained that two centuries after the conquest, the Spaniards had not yet taught the Mexicans proper Castilian so that they still needed interpreters to communicate. He most certainly knew that his historical example of conquest was slightly specious, considering how little effort the Romans had put into imposing Latin—a fact well known in the eighteenth century. Accuracy mattered little, however, as the claim

295

was above all intended to give gravitas and legitimacy to his argument for a more forceful linguistic policy. It is ironical that the fate of Spanish in the New World ultimately resembled that of Arabic rather than that of Greek in the conquered territories, a fact that Archbishop Lorenzana may well have found somewhat unnerving. A couple of decades after Lorenzana’s diatribe, in the wake of the French Revolution, the politicization of language was taken one step further. Similarly to imperial contexts, the high French language was seen as superior because it was a tool of knowledge, culture, and, crucially in this case, political communication. The promoters of une nation, une langue had the ambition of reversing the tables, and turning French from a tool of domination to one of empowerment: where the monarchy and aristocracy had monopolized the language of power, the revolutionaries wanted the people to speak it so that they could participate fully in the new political project. Yet however noble the intention, it continued to rest on the idea that there was a natural hierarchy of languages, those at the top being, unsurprisingly, the languages of power and learning that prevailed in Europe. At the lower end, in this case, were the patois, the countless local languages and dialects spoken in the territory of the French state at the end of the eighteenth century. Their elimination in favour of French became one of the central aims of the new establishment, in what was probably the first systematic case of a national linguistic policy, based on the new notion of ‘public instruction’. The imposition from above was often ruthless, negating local identities in favour of national unity (Balibar and Laporte 1974; de Certeau, Julia and Revel 1975; 2002; Bourdieu 1982, 28–31; Schlieben-Lange 1996). The process that followed the French Revolution soon became a model for many emerging nation states, and its underlying conceptions still colour our understanding of language and linguistic shift to this day. The very notions of ‘national language’, ‘linguistic identity’, ‘linguistic policy’, and even our definition of what constitutes a language as opposed to a dialect,1 are deeply rooted in the idea of une nation, une langue and its implications, and they have heavily influenced

1

See the discussion on the notion of dialect in Adams 2007, 8–12, with further references.

296

A. PAPACONSTANTINOU

the way modern scholars have thought about language in historical settings. Working from the vantage point of the Western tradition, imbued as it is with classical rhetorical culture, they have unwittingly tended to reproduce such imperial tropes. This is especially striking in some early judgements on Coptic texts. We have seen what the great Orientalist Bollandist Paul Peeters thought of the Egyptian ‘masses’. In 1922, in his classic article on translations and translators in oriental hagiography, he had already given his view on Coptic as a language: Coptic, which was still experiencing its early babblings, did not even fight against that monopoly [of Greek]. It suffered the domination of a superior language and literature (Peeters 1922, 243).

Even leaving aside the classic imperial trope of Coptic as a baby in relation to Greek, the passage could not be more candid. It reflects Peeters’ overall approach to Egyptian Christian culture. Hellenism found in Egypt ‘a barbaric mass’ that was ‘amorphous’ and ‘passive’ (Peeters 1950, 15); the Egyptian people had to ‘borrow from Greek literature the makeshift processes that made possible their few delayed and gauche attempts at popular education’ (1950, 14). Even though they undeniably invented monasticism, they were only able to make ‘mediocre’ use of that ‘principle’, which only fulfilled its potential after ‘Hellenism lent it a superior vitality and transmitted it to the Latin spirit, which enabled its full development’ (1950, 47–48). All this had an effect on the levels of literacy of the Egyptians—nay, their very capacity for literacy—which for Peeters was mediated through Greek: ‘In general, those who knew only the vernacular did not write’ (1922, 243). In the same 1922 issue of Analecta Bollandiana, Peeters’ fellow Bollandist Hippolyte Delehaye published a long article of over 120 pages entitled ‘Les martyrs d’Égypte’, presenting and classifying the evidence on Egyptian martyrs, and containing a large section on Coptic martyrological literature. The two scholars were in absolute agreement in their appraisal of the value of Coptic culture, and Delehaye’s most famous— some would say infamous—passages are worth citing again in this context: The artificial character of that miserable literature immediately catches the eye. It proceeds through platitudes and often through borrowings. … This disdain for history that the authors seem to affect goes hand in hand with a lack of measure and a taste for exaggeration that does not fear

the absurd. … Those extravagances testify to an inferior level of culture and to a profound indigence of thought. … Those repulsive acts of butchery, interrupted by scenes of fantasy, recounted in gauche compositions that smack of barbarism, seem to be the natural products of the Coptic mind, and nothing could be further from the measure and the elegance of the Greek spirit (Delehaye 1922, 148–49; my emphasis).

Delehaye had previously written at length on Greek hagiography, and more specifically on what he had defined as ‘epic’ martyr stories on the basis of their high degree of absurdity and implausibility (Delehaye 1921). Yet he found the same characteristic in the Coptic texts much more off-putting. The reasons for this are complex, but in a nutshell his reaction reveals his expectation of finding the hallmarks of Western prose composition, and of reading texts structured according to rhetorical rules that go back to Greek practice. His understanding of texts that did not follow those rules as reflecting an ‘inferior level of culture’ and an inability to think straight are characteristic of his time and of the imperial tradition within which his work is inscribed. Evidently, if the two Bollandists persisted in studying Coptic literature, it was because it preserved memories of early saints, whose lives they had dedicated themselves to reconstructing, and not because of its intrinsic literary or religious value. Within such a framework, it is perhaps natural that texts in Coptic conforming better to the classical tradition should be found puzzling, and therefore considered to be translations from the Greek. A common reflex among Copticists, this attitude has been taken to its extreme expression by Enzo Lucchesi, who maintains that Egyptians simply never wrote anything directly in their language, even in the late sixth or early seventh century, but continued to write literature in Greek and then translate it into Coptic. Commenting on an edition of the homilies of the late-sixth-century bishop Rufus of Shotep by Mark Sheridan, Lucchesi—whose outstanding knowledge of Coptic texts must be noted— stated that he had ‘long been convinced that, until proof is brought to the contrary, every Coptic literary text presupposes a Greek original’ (Lucchesi 2000, on Sheridan 1998). Putting the burden of proof on those who claim that authors were writing in their own language will seem curious to most modern sociolinguists. The cultural domination of Christian Egypt by Hellenism has been internalized to such an extent by scholars that Coptic has often been discussed as if it were a dialect rather

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

than a language—in other words, a spoken idiom in which there was no instruction, in which it was impossible to conceive of a work of literature, and into which one could only translate for the illiterate. As Peeters had put it, if one knew only Coptic, one did not write. However extreme this position may sound, it nonetheless seems to have surreptitiously crept into the way most Copticists approach their subject. As late as 2008, Stephen Emmel was wondering: ‘As a Coptologist, I feel compelled to pose this question—why in Coptic?—not only with regard to the Coptic Gnostic literature, but with regard to much other Coptic literature as well’ (2008, 43). After a discussion of Lucchesi’s positions, Emmel continued: ‘The question “Why in Coptic?” has two aspects to it: first, why was anyone motivated to translate this literature into Coptic in the first place? and second, once it was available in Coptic, why was anyone motivated to read it in that language?’ (2008, 45). The reason why this seems so improbable, to Lucchesi as to Emmel, is that if any readers were reading those texts, they were necessarily bilingual, and therefore not in need of a translation. This series of questions not only assumes the superiority and primacy of one language over another, but it also seems to take for granted that a bilingual society has no reason to translate texts, be they technical, religious or literary. This in turn presupposes another underlying assumption, which would also surprise modern sociolinguists, namely that translations have as their unique function to transfer meaning from one language to another for those who do not understand one of the two—that is, that they always and only arise out of communicative necessity. Needless to say, this line of reasoning would put into question practically all modern translations from English into Gaelic, Welsh or Scandinavian languages, for instance. Language is a vehicle of value, social positioning and emotion well beyond its primary function as a tool of communication, and language choice in a bilingual setting has far-reaching implications that are overwhelmingly extra-linguistic. This brings us back to Arabic. Peeters did indeed note that it replaced Coptic in ‘less than three centuries’, but was not interested in the process through which that

297

happened. The inferiority of the Egyptian language seems to have provided him with a sufficient reason. Only in the 1980s did the language shift become an object of study in itself. The first to bring up the topic was Leslie MacCoull, in an article entitled ‘Three cultures under Arab rule: the fate of Coptic’. MacCoull asked the same question as Peeters, and like him, she framed the process as one of assault: ‘what Persians, Greeks, and Romans (not to mention Ottomans) could not do, the Arabic language … did: Coptic died’ (MacCoull 1985, 64). Being the language ‘of the conquered’, Coptic was unable to resist the power of Arabic as it had that of Greek. To understand the language shift, she rightly insisted that the evidence from documentary sources was directly relevant and went on to explore some of that material. One important phenomenon she noted was the use of alternative Arabic names by Egyptian Christians in some of their documents, citing for example lists of names from a late manuscript of the grammatical work of the fourteenthcentury bishop Athanasius of Qūṣ. The examples she gave were Makarios paired with Sa‘id, Kirikos with Bishara, Johanna with Sitt al-Fadl, Agathos with Abu‘l-Kheir, Theophanes with Abu-‘l-Farag and Nicodemus where the first part (Nico-) had been rendered as Manṣūr (MacCoull 1985, 65; see also Bauer 1972, 17). The Arabic names are chosen to reflect the meaning of the Greek ones, a practice that had prevailed during the Graeco-Roman centuries, when Egyptian names were routinely paired with Greek ones in the standard form [Greek name] ὁ καὶ (who is also) [Egyptian name], more often than not with a semantic or phonetic correspondence between the two names.2 However, instead of seeing a mirror of the Graeco-Roman practice, MacCoull interpreted the double names as a form of ‘camouflage’, in other words a way for Egyptian Christians to hide their identity, in line with her adopted paradigm of linguistic and cultural aggression. The title of MacCoull’s article also described her central tenet, namely that language and culture are synonymous. The ‘three cultures’ were, in her words, carried by languages, that is, ‘the Greek-speaking, Coptic-speaking and Arabic-speaking spheres of thought and action’ (1985, 61). ‘Language was the

2

Similar double-naming was also common among Jews in both the Graeco-Roman and the Islamic periods.

298

A. PAPACONSTANTINOU

carrier of each of the three cultures’ (1985, 66), and because of that, she equated the fate of the language and that of the culture. She entitled a subsequent article ‘The strange death of Coptic culture’, and in its introduction asked why it was that: The Coptic culture of Christian Egypt, which had flourished in high originality and creativity in the sixth and early seventh centuries, was more thoroughly submerged than was any other of the high cultures of the Christian Orient by the effects of the Arab/Islamic conquest. … I am investigating the phenomenon of cultural death through that of language death, as language … has been par excellence the carrier of culture and of identity (MacCoull 1989, 35).

Such an approach marked a clear departure from the unequivocal condemnations of Coptic writings, presenting the culture as original and creative. MacCoull also implied, however, that maintaining a ‘high’ culture was essential to linguistic or cultural survival: ‘The consequences of the Coptic devaluation of learning were to be tragic even up to the present day’ (MacCoull 1989, 35). Like Wasserstein, she suggested that the existence and production of learned texts had a direct influence on the maintenance of a spoken version of a language—and one assumes that by ‘learning’ she had in mind the same Western classical understanding of the term as her predecessors. Displaying a mix of nostalgia, regret and ultimately incomprehension, MacCoull’s analysis remains more descriptive than explanatory. Yet despite leaning overall towards a process of destruction by external forces, she also made some very important observations that are much more in line with current sociolinguistic thinking. Firstly, even though her equation of language with culture cannot be sustained,3 not only but mainly because it is static and essentialist, she also expressed very clearly a notion that has become central in current debates about linguistic diversity and the cognitive importance of language: ‘The cognitive style of the Coptic mind, its ways of classifying and conceptualising, are embedded in the Coptic language. The death of that language is the loss of a whole way of being human (1989, 35).’

3

See also Myers-Scotton 2006, 102–3 on the separation between language and culture.

For its International Year of Indigenous Languages (2019), UNESCO described indigenous languages as ‘unique systems of knowledge and understanding of the world’, as well as repositories of unique individual identities, cultural histories, traditions and memories. Because of its political dimension, the question of endangered indigenous languages has become highly contested ground, and has spawned innumerable studies in several interrelated fields. On the whole, this large body of research agrees with MacCoull’s statement. Secondly, in describing the language shift, MacCoull put the Egyptian Christians at the centre of the process, considering them as the main agents of that change— even if in her narrative this is judged negatively. She laments that Egyptian Christians did not value their language enough to continue speaking it, and, in a sense, that they abandoned ship when Arabic attacked. Still, as Sebastian Richter has already pointed out, she was right to make the speakers themselves responsible for the choice of not maintaining and transmitting their language (Richter 2009, 428). That external factors had an important influence on that choice is almost a truism, but there was no single direct cause for it and certainly not any form of coercion. Learning from modern language shift After MacCoull’s articles, the subject was taken up again by a number of scholars. Rather than give a linear overview of subsequent research here, however, I would like to follow up on MacCoull’s insights, weaving the more recent work on early Islamic Egypt, as well as broader advances in the study of language shift, into that exploration. Research on language shift has grown at a very quick pace in recent decades, largely because of the realization that several languages become extinct every year. Like the biological metaphor that describes it, ‘becoming extinct’ means that a language, like a species, is no longer reproduced—in other words, it is no longer transmitted to the next generation by the parents. To happen, a shift presupposes a bilingual family context, where the traditional language of the family

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

ceases to be used by one generation, which prefers to use the second language exclusively (Potowski 2013). This can be the generation of the parents themselves, who thus do not transmit the language to the children, or that of the children, who will understand the language but choose not to use it, maintaining a passiveonly knowledge of it. For the language to cease being that of a speech community, this family pattern needs to be replicated socially, that is, it has to be a group phenomenon and not only an individual one. The shortest period for such a shift to occur within an individual family is three generations, where the first is monolingual in the first language, the second bilingual, and the third monolingual in the second language. This is a pattern typical in immigrant communities, but in other situations—especially when the first language is the local language and the second an imported lingua franca—the bilingual phase can remain stable for a very long time. In other words, there can be many bilingual generations before one of them stops transmitting the first language in sufficient numbers for the shift to occur at the level of the speech community, and not of a single family. Such stable bilingual situations exist when there is a homogeneous speech community speaking the first language, and, crucially, when that language is spoken exclusively in the home and in most informal situations between native speakers, while the second language is reserved for official purposes and in more formal economic transactions. The ability to maintain clearly compartmentalized linguistic domains, and especially keeping the domestic sphere linguistically intact, is a key factor in the preservation of the first language (Romaine 2000, 49, 54). A large number of factors can combine to cause bilingual stability to evolve. One of the concepts developed by modern linguists to assess and classify the variety of those factors is that of ethnolinguistic vitality. Although it is mainly a model aimed at predicting where languages are most at risk today, it contains several observations that are also relevant for the ancient and Medieval worlds. There are two categories of variables, subjective and objective. The former are

4

See for example Romaine 1995, 43, who gives the example of English in Ireland.

299

essentially in-group factors and individual factors, related to the perception of the language by its speakers and its role in various aspects of community life. However, even though positive attitudes towards a language have been shown to be important, they do not always guarantee its maintenance, and vice versa.4 It is also much more difficult to assess subjective variables when working with historical sources, as they often represent partial or fragmented views. The table reproduced in Fig. 1 presents the principal objective variables affecting the balance of languages in bilingual speech communities, which are much easier to relate to historical societies. When a language has high amounts of several of these, and therefore high ethnolinguistic vitality, it is less likely that a shift will occur; conversely, when those variables are lacking or not strongly present the likeliness of a shift is higher (Myers-Scotton 2006, 90; Mullen 2012, 26–27). Even in this table, not all the variables have relevance for early Islamic Egypt, or indeed for pre-modern societies more generally. What it does show is the complexity of the phenomenon and the number of factors involved in language shift, even if we take only the ‘objective’ or external ones. In addition to these, a number of other factors have been highlighted, most notably the degree of isolation of a speech community, which favours the maintenance of the first language, while increased mobility favours a shift to the second language. On the whole, different combinations of all the variables above can yield different results, and it is often proving very difficult to predict when a language shift will occur, especially in the case of indigenous languages (see Romaine 1995, 40–46; Myers-Scotton 2006, 105–6). As ultimately the question of language maintenance is one of generational transmission within the household, the linguistic choices made by women are essential in determining the fate of a language (Romaine 2000, 53; Langslow 2002, 28; Potowski 2013, 325– 26). This can go in different directions depending on a number of factors. It is common to consider women as the transmitters of tradition and culture, and therefore as more conservative in their choices—something

300

A. PAPACONSTANTINOU

Fig. 1: Indicators of high ethnolinguistic vitality (Table from Mullen 2012, 27).

that is well reflected in the expression ‘mother tongue’. The model certainly exists, but it is far from being the norm: many women hope to find in the use of the dominant or high language a path to social mobility through marriage for their daughters, or through education for their sons, and are therefore more likely to use it (Romaine 2000, 78–82). With the observations above in mind we can now come back to Coptic and its relationship with Greek and Arabic. Although the nature of the evidence for Late Antiquity and the early Middle Ages is very different from the data gathered and treated by sociolinguists, this framework of analysis has been successfully applied to ancient evidence, for instance by Alex Mullen in her work on Southern Gaul (Mullen 2013).5 For Egypt, one could posit that even though Greek had both a high (formal) and a low (informal) variety, it was generally kept outside the domestic domain, and the

5

More generally, on the difficulties raised by ancient source materials for sociolinguistic analysis, see for instance Adams 2003, 29–84; Langslow 2002; Marotta, Putzu and Donati 2017.

overall linguistic situation was one of extended diglossia. The compartmentalization of the different domains was maintained for at least three centuries after the Arab conquest, but at some point—certainly not the same in different communities throughout the country—the household was no longer an intact linguistic domain. This is a key factor that makes it possible for one of the generations to choose the dominant language as its exclusive language. By that time, a low variety of Arabic was certainly common in the country, and easy to adopt. Because of the nature of our sources, which are overwhelmingly bureaucratic and legal, Arabic—and in the first century after the conquest also Greek—appear to have been most prominent in situations that mobilize the formal version of the language, especially in the early years. Arabic entered the country as an official language, and several very early documents are extant

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

today that pertain to the administration and organization of various levies (Ragheb 2013). The new rulers also produced documents in Greek in the chancery of their new capital, Fusṭāṭ, while the local administrators initially produced documentation in Greek. This was because the administrative pyramid was composed of speakers of the two languages, and presumably also of a number of bilingual individuals who connected the two linguistic groups. Indeed, the locals held most of the lower positions, and their working language was Greek—even though many would have spoken Coptic in a domestic context. The documentation clearly shows that they understood it (Richter 2010a; 2013). As this was a pyramid, the largest number of documents came from the bottom, and they were, therefore, in Greek in the early years. With time, however, the ratio of Greek to Arabic in the administration changed, as Arabic descended more and more towards the bottom of the pyramid. By the second half of the eighth century this process was practically complete. We do not know whether this means that the locals were converting and adopting Arabic or that more Arabic-speakers were populating the lower administrative positions, as converts would often adopt Arab names, but it is likely there was a mix of the two. From the linguistic point of view the two categories were different, as one would have been bilingual, using Coptic as their informal language and Arabic as the formal one (much as they would have done before with Greek), while the other would have been monolingual and was very likely to have used both a formal and an informal version of Arabic. As a result of this, the knowledge of Arabic as a second language will have increased among the local population, at least at a functional level. Village communities would have needed at least a small group of people who could mediate linguistically, especially if the administrators were monolingual Arabic-speakers. At a higher level, local urban elites with the ambition to rise in the administration were required to learn Arabic, as they had learnt Greek in earlier centuries. Learning the official language has always been a timeless tool for bureaucratic success among the local elites of

301

multilingual empires, and the caliphate was no exception. As it had been under Rome, the integration of educated urban elites into the government structures was very successful. Egyptians with a good Arabic education could rise in the service of the government. Several Egyptian Christians are known to have worked for the administration from the early years in Fusṭāṭ (Bouderbala 2014). From the first half of the tenth century onwards this was even more common, including several well-known figures, such as the theologian Sawīrus ibn al-Muqaffa‘, who, before becoming bishop of al-Ashmūnayn, bore the name of Abū al-Bishr and worked as a secretary for the government (Griffith 1996, 17), or the vizier of the Fāṭimids, ‘Īsā ibn Nasṭūrus (994–996). The theological writings of Sawīrus show an excellent and nuanced mastery of high Arabic, no doubt acquired during early education and refined in the context of his administrative duties. That such education was widely available to the local, Christian populations, and not only in Egypt, is implied in the tenth-century geographer al-Muqaddasi’s comment that the administration in Syria and Egypt was full of Christians because of the quality of their Arabic. The Arabs, he said, made less effort with the literary version of their language because they contented themselves with their native knowledge of it (de Goeje 1906, 183; Collins 2001, 153). Arabic did not remain limited to the administration for long, however. Formal legal transactions of an economic nature among the locals gradually adopted it too. In the late seventh and eighth centuries, private legal documents among locals were drawn up overwhelmingly in Coptic, while documents in Greek, which had dominated the record until then, dwindled after the conquest. By the tenth century, however, most surviving contracts drawn up between two local parties were in Arabic. Scholars have noted that they often contain the provision that they would be read aloud to the parties in their language,6 something that was also common in the Roman period. This certainly reflects the difference in language, but it does not mean that the parties did not understand any Arabic at all—only that they would not have been able to follow the formal

6

Abbott 1937, 7–14 (text 1, lines 19–20; date: 946). FrantzMurphy 1981, 209–12 (text 1, lines 14–15; date: 962) and 216– 18 (text 2, lines 13–14; date: 963).

302

A. PAPACONSTANTINOU

register of the legal documents. It must also be noted that in the majority of contracts in Arabic between locals this clause is not present. At the same time, we have evidence that Coptic continued to be used for legal transactions until the eleventh century (Green 1983; Richter 2000). The preference for Arabic in legal documents can be explained in the same way as the earlier preference for Greek, namely that documents in that language could be more easily produced before an official court. There is clear evidence that by the ninth century, the qāḍī’s court ruled on cases that involved not only Muslims, but also Christians and Jews (Tillier 2017; Tillier and Vanthiegem 2018). In a series of fascinating publications, Sebastian Richter has shown in detail the interest shown by Egyptian scholars in Arab science, most notably alchemical knowledge.7 Indeed, the earliest lexical borrowings from Arabic into Coptic were in the scientific domain, indicating that Arabic enjoyed great prestige as a language of learning. The other semantic fields where Arabic made early inroads into the vocabulary were luxury items and commerce. Arabic, says Richter, was perceived as ‘the language of science, the language of advanced civilisation and the language of material wealth’ (Richter 2009, 431, author’s emphasis). Thus far from being the result of a devaluation of learning, as MacCoull suggested, the adoption of Arabic by the Egyptian Christian intelligentsia rather seems to have been due to a thirst for learning and a wish to exchange with the world beyond the confines of Egypt. In all these areas we are generally dealing with a formal and/or learned register of Arabic, and thus not one that would have interfered with everyday speech. What was more significant was the presence of a less formal register of the language which would have been used in oral interactions. This could only happen at a large scale after a sufficient number of monolingual Arabic-speakers had settled in the country. Yet even without extensive settlement of Arabic-speakers, the foundation of Fusṭāṭ, which was the country’s capital and functioned largely in Arabic, had created from the very start a pole of everyday Arabic with which many

7

On Coptic alchemy see Richter 2001; 2004; 2006; 2010b; 2015a. For an overview with supplementary material, see Richter 2009.

inhabitants of the country were bound to have to interact, if only for commercial reasons (Vorderstrasse 2015). The locals who had converted to Islam and were bilingual often served as interpreters and more generally mediators between the speech communities. The hope of participating fully in the possibilities offered by urban life, as well as the greater density of the population and hence increased contact, are factors that would have played in favour of Arabic as the sole language of communication (van Bladel 2018, 168–69). For the rest of the country the situation is much more difficult to assess. The importance of settlement has recently been highlighted by Kevin van Bladel as a key element in Arabicization, and indeed it is a variable that cannot be overestimated (van Bladel 2018). As the table on ethnolinguistic vitality (see Fig. 1) shows, and as many researchers have found, however, demography is a tricky variable, and often quality (who speaks) proves more important than quantity (how many speakers) when it comes to the maintenance of the local language (Romaine 1995, 40). And demography itself is not about mere numbers. Our evidence, for example, tells us little about mixed marriages, often identified as one of the most influential factors of language shift. It is also mute on the ethnicity, and consequently also the first language, of the many migrants into Egypt from the ‘Abbāsid period onwards, and especially under the Fāṭimids (den Heijer, Lev and Swanson 2015, 331–37). Their settlement in the country would have accelerated the need to speak a lingua franca for very practical purposes. Arabic-speakers had already started settling in the countryside in the first half of the eighth century. A letter in Arabic from a travelling Muslim landowner to his estate manager in the Fayyūm shows that he employed local Christians, whose names he had Arabicized (Sijpesteijn 2004). The estate manager must have been bilingual and served as an interpreter, especially if the Christian wage labourers were monolingual Coptic-speakers. Over time, however, this sort of context would favour the acquisition of at least functional Arabic by those seeking seasonal work in agriculture.

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

303

The majority of the Egyptian countryside remained Coptic-speaking for another couple of centuries, but here again the balance certainly shifted with time. By the tenth century, even though the Nile valley was still largely inhabited by Christians, Arabic seems to have been a common vernacular. The revival of the upriver trade route towards the Red Sea, South Arabia and the Indian Ocean under the Fāṭimids made Arabic an important commercial lingua franca as well, and certainly gave the language a boost in the valley from the late tenth century.8 That Arabic was a Christian vernacular at that time, at least in public life, is also indicated by the fact that a tenth-century bishop of al-Ashmūnayn could write a letter to his congregation on matters of discipline in Arabic (Reinhardt 1897). As I mentioned above, the isolation of speech communities favours the maintenance of their language. Conversely, mobility endangers it, because the need to communicate leads to the adoption of the prevailing lingua franca. The revival of commercial relations along the valley in the tenth century is certainly a case in point. Rural Egyptians had moved out of their local environments long before that time, however. Very early, from the 660s onwards, men were requisitioned for corvée labour in the caliph’s shipyards at Klysma/ al-Qulzūm. This intensified in the early eighth century, because to the shipyards that continued to function were added several caliphal building projects in Fusṭāṭ, Jerusalem and Damascus. Corvée labour was presumably not only levied in Egypt, and the building sites in question would have been very multicultural spaces, where being able to function in Arabic would have been imperative. The same is true of the conscription of locals for the caliphal fleet’s raids against Byzantium in the early eighth century. Conscripts would have found themselves with speakers of different languages, and under the orders of an Arabic-speaker. In all these cases, the presence of interpreters does little to mitigate the need for a common language among the individuals present, however elementary its practice. The Egyptian villagers who had been requisitioned or conscripted

most certainly returned with some basic knowledge of informal Arabic. As we can see, there were a number of factors that contributed to the symbolic and practical value of Arabic and that facilitated its acquisition. None of these factors, however—except perhaps the scale of the corvée—was so different from what prevailed under Rome, or from the role played by Greek. The domains mentioned until now are indeed both formal and informal, meaning that depending on their activity and social level, Egyptians would have been able to acquire both a high variety and a low variety of Arabic. Again, this was also true of Greek. What made the difference ultimately was that Arabic not only dominated the political and economic domains, but it eventually also prevailed in the rural Egyptian domestic domain. It is difficult to apprehend an early version of domestic Arabic in use in Egypt, as it would have been predominantly oral. Until very recently, interest in late Coptic documentary material was very limited, and even more limited the interest in Christian Arabic papyri. This is now beginning to change, and a picture is starting to emerge of bilingual individuals who could communicate in both languages in writing. In the eighth century, two converts to Islam wrote to each other in Coptic (Boud’hors 2016). From the ninth to the eleventh century, letters could begin in Arabic and continue in Coptic.9 An interesting case of intra-communal bilingualism is that reflected in the so-called Teshlot archive, which contains the business papers of a certain Raphael, son of Mina, from the village of Dashlūt near al-Ashmūnayn. The documents date between 1022 and 1063, and constitute the latest extant group of Coptic legal documents. The archive also contains some texts in Arabic from some correspondents, indicating that this was a bilingual network, the members of which chose their preferred language for communication, at least in writing. Unfortunately we do not know the ages of the correspondents, so we cannot know if the choice was generational or of a different nature.

8

9

This is more fully argued in Papaconstantinou 2012.

Delattre et al. (2012) edit two such letters, and also give a brief overview of late documentary texts produced by Egyptian Christians.

304

A. PAPACONSTANTINOU

Religious institutions and local identities Such bilingual practices are typical of a critical moment in the process of linguistic shift, and other evidence also points to the eleventh century as the time when it took place. This evidence comes from the religious domain, and more specifically from institutional religion. In the first half of the eleventh century, the Egyptian church, after a period of hesitation, adopted Arabic as its official language. This came about at the same time as the Patriarchate moved its headquarters from Alexandria to Cairo under the Fāṭimids. The reasons for this move were political in the broader sense, allowing the heads of the Christian community to be near the court where, apart from the rulers, there was a network of influential Christians and Jews. Opening up to the other religious communities was especially important for the Fāṭimids, who were Ismāꜥīli in a country whose Muslims were mostly Sunnī, and sought both legitimacy and balance by co-opting the local elites (Garcin 1987, 133; den Heijer, Lev and Swanson 2015, 325). For several decades, patriarchs had spent more time in Cairo than Alexandria, and eventually, in the second half of the eleventh century, Cyril II (1078– 1092), under pressure from the Fāṭimid vizier Badr al-Jamālī, made this state of affairs official and gave the Patriarchate new headquarters in the capital (den Heijer 1999; Papaconstantinou 2007). In such a context, it was impossible for the Egyptian church not to function in Arabic. Many members of the church had already been using the language, for example the tenth-century bishop and author Sawīrus ibn al-Muqaffa‘ already mentioned above, or the clergy of Cairo that was close to the court, but the official language of the institution was still Coptic, and its documents were still produced in that language. A number of histories of the church existed in Coptic, as well as the foundational stories of the martyrs of the Great Persecution, who were so important to the Egyptian church that it had its own era, known as the Era of the Martyrs. In a quite remarkable move, as the church adopted Arabic, it took with it, so to speak, virtually its entire literary heritage. It initiated an unprecedented venture of ‘translating the tradition’, as Samuel Rubenson has so aptly put it (Rubenson 1996). This included historical, hagiographical and theological works—in short, everything except the liturgy. New works were henceforth composed in Arabic, including the institutional history of the Patriarchate itself, which took the form of biographies of the successive holders of the office. The first

to be composed in Arabic, by Mawhūb ibn Manṣūr ibn Mufarrij, who was also the co-ordinator of the collection and translation of the older biographies, were those of the patriarch Christodoulos (1047–1077) and his successor Cyril II (den Heijer 1989; 1991; 2015). Thus the shift in language did not leave the past behind, but recast it in the new linguistic medium. Not only that, but those translations were integrated with the new compositions seamlessly, as if there had been no rupture whatsoever (den Heijer 2015). Among other things, this indicates that, for many clerics, their heritage was perceived as religious and cultural rather than linguistic, and that contrary to MacCoull’s assumption, they did not equate their language and their culture. This move of the church was significant because it gave Arabic and Islam religious legitimacy: if the bishop and the patriarch spoke Arabic and mingled with the Muslims, no one could or should feel guilty for doing the same. Even though that decision mainly concerned domains that functioned within the high variety of the language, its symbolic importance cannot be overestimated. I have argued elsewhere that the gradual Arabicization of the church hierarchy was not well received in some monastic circles, which were generally more isolated from the urban centres and strongly attached to the traditional language (Papaconstantinou 2007). These monastic institutions were much closer to the rural population than the urban clergy, and would have had an audience among the Coptic-speakers in their area. As we have seen, in the eleventh century there were still villages where Coptic was the dominant language, but their inhabitants were bilingual. The terrain was therefore prepared for the shift made by the church. Whatever the resistance, ultimately the combination of symbolic capital and outright power of the Patriarchate won the day, and eventually monasteries were themselves Arabicized. The weight of ecclesiastical institutions in the life of the Christian population at large was enormous by the tenth century. Arguably then, the most crucial element in tipping the balance from a stable bilingual situation to a shift towards Arabic may have been the arguments put forth by the different religious actors to obtain the adhesion of the population. Certainly disseminated orally to a large extent, many of them can be found in textual compositions as well. The position of the traditionalists, so to speak, finds its most accomplished expression in the now famous —if misnamed—Apocalypse of Samuel of Qalamūn, a text purporting to be a prophecy from the mouth of the

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

seventh-century founder of a monastery in Qalamūn, on the southwestern fringe of the Fayyūm. The monastery had produced a Coptic Life of its founder which seems to have been a very important text for the community, and which the author of the Apocalypse knew well (Zaborowski 2003). We know that the monastery remained active throughout the Middle Ages, and the Apocalypse has been dated from that period, variously between the eighth and the fourteenth century.10 The text is part of a series of such ‘apocalypses’, which contain eschatological ‘prophecies’ about the fate of Christians in the Muslim empire, and are filled with admonitions for said Christians as to the correct behaviour to adopt.11 They tend to promote a conflictual vision and a strict separation of the religious groups. The reason the Apocalypse of (pseudo-)Samuel has been more popular is that it alone discusses the loss of the Coptic language, presenting it as the essence of Egyptian Christianity. Adoption of Arabic came with assimilation to the ways of the Arabs, according to the author, and was therefore to be avoided. He castigates priests and monks for having already adopted the language, which points to a time when this was beginning to happen. I quote a short passage from the text, which gives a taste of its style:12 They will also do something else, that if I were to tell you of it your hearts would be greatly pained: they will abandon the beautiful Coptic language, in which the Holy Spirit has often spoken through the mouths of our spiritual fathers; they will teach their children from an early age to speak the language of the Bedouin, and will take pride in it. Even the priests and monks will—they too!— dare to speak in Arabic and to take pride in it, and that within the sanctuary! Woe upon woe!! Oh my dear children! What shall I say? At that time the readers in the church will no longer understand what they read nor what they say because they will have forgotten their language, and they will truly be miserable and deserving to be wept over because they will have forgotten their language and will speak the language of the Hagarenes. But woe to

10

Décobert (1992) was the first to suggest that the text was later than its frame story claims, and suggested the eighth century; for van Lent (2000, 667), it could be anywhere between the ninth and the fourteenth century; Zaborowski (2008) suggests the fourteenth century. Iskander (1998) argues for the eleventh century, and Papaconstantinou (2007) for a late tenth-/early eleventh-century date.

305

every Christian who teaches his son the language of the Hagarenes from an early age, causing him to forget the language of his fathers!

The late tenth century, and even more the eleventhcentury context, as we have seen, fits the content of this passage admirably. The clergy was proud of their Arabic, but unable to understand what was said in church— most certainly referring to the text of the liturgy, which remained in Coptic throughout. The shift is happening because the transmission of Coptic from the parents to the children is no longer taking place. The author is very conscious of the key importance of this factor, and repeats it several times in the text, even in this short extract. He also laments elsewhere that books will fall into disuse, and the loss of communal memory that goes with not understanding the stories of one’s own past. Adopting another language went, for the author, with adopting that language’s culture, which was foreign and the carrier of a different religion. Conclusion: A new indigeneity Much like recent research on indigenous languages, and like Leslie MacCoull’s approach, the Apocalypse insists strongly on language as a vector of unique and irreplaceable knowledge—more specifically, in this case, religious and cultural knowledge. The church, on the other hand, without making a constructed argument for its position, offered a much more practical approach, as well as a relatable new vision and place for the Christians, which found an especially favourable context in the Fāṭimid caliphate. I have suggested in an earlier publication that the importance of Fāṭimid cosmopolitanism and integration of the Jews and Christians into the system of power was essential in giving the local population a sense of belonging that was intimately connected to the use of Arabic. This included their participation in the newly revived Indian Ocean trade, which once again turned the Nile valley into a

11

12

See now Palombo 2019, who discusses a corpus of monastic texts from the ninth and tenth centuries that promote a similar approach. Ziadeh 1915–1917, 379; the passage is translated in Papaconstantinou 2007, 275, but an improved translation will appear in Papaconstantinou in press (in particular regarding the term Hagarenes, for which see also Iskander 1998 and Samir 2014).

306

A. PAPACONSTANTINOU

major trade route, something it had not been since the fourth century. Contrary to the Roman system, however, which had its centre outside Egypt and simply used the Nile as a pipeline for sealed goods, under the Fāṭimids the wealth remained in Egypt (Papaconstantinou 2012). The reversal of the social importance and symbolic role of the merchant between Rome and the caliphate also meant that even without being a landowner, one could hope to participate in the new prosperity through assimilation. Arabic became the means for such social mobility and participation. A short study by Jean-Claude Garcin (1987), which had escaped me at the time, points out some additional aspects of the Fāṭimid context that offer an essential complement to my argument. The first of these, to which I have already alluded, is the important role of the Christians and Jews at court and in the administration, because of the rulers’ reluctance to recruit only among the local Muslim grandees who were of a different persuasion. The influx of locals into the circles of power, whether Christian or converted, had an important effect in the development of local histories in Arabic which reimagined the country’s past and integrated its mythology. Stories such as that of Mary the Copt, a female slave of the Prophet Muḥammad, who became his wife after she gave him his only son, Ibrāhīm, first appeared in the sources in the ninth century and became increasingly popular, both in Muslim and in Christian circles (Hidayatullah 2010). A parallel development can be seen in the Jewish community, as Muḥammad had also married the Jewish Ṣafiyya. This allowed the development of stories and discursive strategies that highlighted the common ground between Islam and the two protected religions, Judaism and Christianity (Frenkel 2015; Fierro 2015, 522–23). As

13

For an overview of Medieval Arabic histories of Egypt see Cook 1983; more specifically on Ibn ꜤAbd al-Ḥakam and the beginnings of that movement, see Zychowicz-Coghill 2017, and ongoing work by the author on the Egyptian Arabic historical tradition. Den Heijer (2015) suggests that the History of the patriarchs of Alexandria was part of that broader movement.

Egyptian Christian stories had done from the fourth century onwards (Papaconstantinou 2001), these histories also highlighted the biblical connections of Egypt: Abraham’s visit and his Egyptian wife Hagar, the mother of Ismāꜥīl, Joseph’s stay and Moses’ departure. Narratives of the visit of the Holy Family also saw something of a revival.13 The fact that Egypt was the centre of the Fāṭimid caliphate accommodated this nativist discourse, as it could only legitimize the rulers and reinforce their hold on the country. In many ways then, the eleventh-century context addressed the very cultural anxiety that is evident in the Apocalypse. By participating in the construction of a new form of indigeneity and a new imagined community, the church and the Christian elites of Cairo made a more appealing case for the younger generations than did the author of the Apocalypse and those who thought like him. Outside the country, Egyptian Christian authors started being read again as they had when they wrote in Greek. Contrary to pseudo-Samuel’s fears, communal memory did not die with the language, and the thirst for knowledge was reinvigorated. Inside, Egyptians could participate more fully in a polity that they saw as their own again, even if there were tensions and religious differences.14 This state of affairs did not last very long, but it was enough for Arabic, initially only a language of empire, eventually to become the language of Egypt. Acknowledgements I would like to express my heartfelt thanks to Elisabeth R. O’Connell for her patience and understanding, and to the anonymous reader for their discussion of double names.

14

On late versions of the story of the holy family’s flight to Egypt, see Boud’hors and Boutros 2000; 2001. See also Sijpesteijn 2011 on the integration between ancient Egyptian and biblical history. See Lev 2015 on the ups and downs of Fāṭimid–Christian relations.

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

Bibliography Abbott, N. 1937. The monasteries of the Fayyūm. Chicago, IL. Adams, J. N. 2003. Bilingualism and the Latin language. Cambridge. ———. 2007. The regional diversification of Latin. Cambridge. Balibar, R. and D. Laporte. 1974. Le français national: Politique et pratique de la langue nationale sous la Révolution. Paris. Bauer, G. 1972. Athanasius von Qūṣ, Qilādat at-taḥrīr fī ꜥilm at-tafsīr: Eine koptische Grammatik in arabischer Sprache aus dem 13./14. Jahrhundert. Freiburg im Breisgau. Bouderbala, S. 2014. Les mawālī à Fusṭāṭ aux deux premiers siècles de l’islam et leur intégration sociale. In Les dynamiques de l’islamisation en Méditerranée centrale et en Sicile: Nouvelles propositions et découvertes récentes = Le dinamiche dell’islamizzazione nel Mediterraneo centrale e in Sicilia: Nuove proposte e scoperte recenti, A. Nef and F. Ardizzone (eds), 141–51. Bari; Rome. Boud’hors, A. 2016. Degrés d’arabisation dans l’Égypte du VIIIe siècle: CPR II 228 revisité. In Mélanges Jean Gascou: Textes et études papyrologiques (P.Gascou), J.-L. Fournet and A. Papaconstantinou (eds), 71–89. Paris. Boud’hors, A. and R. Boutros. 2000. La sainte famille à Gabal al-Tayr et l’homélie du Rocher. In Études coptes VII: Neuvième journée d’études (Montpellier, 3–4 juin 1999), N. Bosson (ed.), 59–79. Leuven. ———. 2001. L’homélie sur l’église du Rocher, attribuée à Timothée Aelure: Texte copte et traduction. Deux textes arabes et traduction. Turnhout. Bourdieu, P. 1982. Ce que parler veut dire: L’économie des échanges linguistiques. Paris. de Certeau, M., D. Julia and J. Revel. 1975. Une ethnographie de la langue: L’enquête de Grégoire sur les patois. Annales: Economies, sociétés, civilisations 30: 3–41. de Certeau, M., D. Julia and J. Revel. 2002. Une politique de la langue: La Révolution française et les patois. 2nd ed. Paris. Collins, Basil A. 2001. The best divisions for knowledge of the regions/Aḥsan al-taqāsīm fī maꜥrifat al-aqālīm. 2nd ed. Reading. Cook, M. 1983. Pharaonic history in Medieval Egypt. Studia Islamica 57: 67–103. Décobert, C. 1992. Sur l’arabisation et l’islamisation de l’Égypte médiévale. In Itinéraires d’Égypte: Mélanges offerts au père Maurice Martin, S.J., C. Décobert (ed.), 273–300. Cairo. de Goeje, M. J. (ed.). 1906. Al-Muqaddasi, Kitāb aḥsan al-taqāsīm fī maꜥrifat al-āqālīm. Leiden.

307

Delattre A., B. Liebrenz, T. S. Richter and N. Vanthieghem. 2012. Écrire en arabe et en copte: Le cas de deux lettres bilingues. Chronique d’Égypte 87: 170–88. Delehaye, H. 1921. Les passions des martyrs et les genres littéraires. Brussels. ———. 1922. Les martyrs d’Égypte. Analecta Bollandiana 40: 5–154, 299–364. den Heijer, J. 1989. Mawhūb ibn Mansūr ibn Mufarriğ et l’historiographie copto-arabe: Étude sur la composition de l’Histoire des patriarches d’Alexandrie. Leuven. ———. 1991. History of the Patriarchs of Alexandria. In The Coptic encyclopedia, A. S. Atiya (ed.), 4: 1238–42. New York. ———. 1999. Considérations sur les communautés chrétiennes en Égypte fatimide: L’État et l’Église sous le vizirat de Badr al-Jamālī (1074–1094). In L’Égypte fatimide, son art et son histoire, M. Barrucand (ed.), 569–78. Paris. ———. 2015. The martyrdom of Bifām ibn Baqūra al-Ṣawwāf by Mawhūb ibn Manṣūr ibn Mufarrij and its Fatimid background. Medieval Encounters 21: 452–84. den Heijer, J., Y. Lev and M. Swanson. 2015. The Fatimid empire and its population. Medieval Encounters 21: 323–44. Emmel, S. 2008. The Coptic Gnostic texts as witnesses to the production and transmission of Gnostic (and other) traditions. In Das Thomasevangelium: Entstehung— Rezeption—Theologie, J. Frey, E. Edzard Popkes and J. Schröter (eds), 33–49. Berlin. Fierro, M. 2015. Dhimmīs in Fatimid Egypt: A view from the Islamic West. Medieval Encounters 21: 516–23. Fournet, J.-L. 2010. Sur les premiers documents juridiques coptes. In Études coptes XI: Troisième journée d’études (Marseille, 7–9 juin 2007), A. Boud’hors and C. Louis (eds), 125–37. Paris. ———. 2016. Sur les premiers documents juridiques coptes (2): Les archives de Phoibammôn et de Kollouthos. In Études coptes XIV: Seizième journée d’études (Genève, 19–21 juin 2013), A. Boud’hors and C. Louis (eds), 115–41. Paris. ———. 2018a. Sur les premiers documents juridiques coptes (3): Les ‘archives’ d’Apa Abraham. In Études coptes XV: Dix-septième journée d’études (Lisbonne, 18–20 juin 2015), A. Boud’hors and C. Louis (eds), 199–226. Paris. ———. 2018b. Les documents bilingues gréco-coptes dans l’Égypte byzantine: Essai de typologie. In Written sources about Africa and their study. Le fonti scritte sull’Africa e i loro studi, M. Lafkioui and V. Brugnatelli (eds), 59–83. Milan. ———. 2020. The rise of Coptic: Egyptian versus Greek in Late Antiquity. Princeton, NJ. Frantz-Murphy, G. 1981. A comparison of the Arabic and earlier Egyptian contract formularies, Part I: The Arabic

308

A. PAPACONSTANTINOU

contracts from Egypt (3rd/9th–5th/11th centuries). Journal of Near Eastern Studies 40: 203–25. Frenkel, M. 2015. Adaptive tactics: The Jewish communities facing new reality. Medieval Encounters 21: 364–89. Garcin, J.-C. 1987. L’arabisation de l’Égypte. Revue de l’Occident musulman et de la Méditerranée 43: 130–37. Green, M. 1983. A private archive of Coptic letters and documents from Teshlot. Oudheidkundige Mededelingen uit het Rijksmuseum van Oudheden 64: 61–121. Griffith, S. 1996. The Kitāb Misbāh al ꜥAql of Severus Ibn al-Muqaffaꜥ: A profile of the Christian creed in Arabic in tenth-century Egypt. Medieval Encounters 2: 15–42. Hidayatullah, A. 2010. Mariyya the Copt: Gender, sex and heritage in the legacy of Muhammad’s umm walad. Islam and Christian-Muslim Relations 21: 221–43. Iskander, J. 1998. Islamization in medieval Egypt: The Copto-Arabic ‘Apocalypse of Samuel’ as a source for the social and religious history of medieval Copts. Medieval Encounters 4: 219–27. Langslow, D. R. 2002. Approaching bilingualism in corpus languages. In Bilingualism in ancient society: Language contact and the written text, J. N. Adams, M. Janse and S. Swain (eds), 24–51. Oxford. Legendre, M. 2014. Perméabilité linguistique et anthroponymique entre copte et arabe: Exemple de comptes en caractères coptes du Fayoum fatimide. In Coptica Argentoratensia: Textes et documents de la troisième université d’été de papyrologie copte (Strasbourg, 18–25 juillet 2010), A. Boud’hors, A. Delattre, C. Louis and T. S. Richter (eds), 325–440. Paris. Lev, Y. 2015. The Fatimid caliphs, the Copts and the Coptic church. Medieval Encounters 21: 390–410. Lorenzana, F. A. 1770. Cartas pastorales y edictos del illustrissimo señor D. Francisco Antonio Lorenzana, y Buitrón, arzobispo de México. Mexico City. Lucchesi, E. 2000. La langue originale des commentaires sur les Évangiles de Rufus de Shotep. Orientalia 69: 86–87. MacCoull, L. S. B. 1985. Three cultures under Arab rule: The fate of Coptic. Bulletin de la Société d’archéologie copte 27: 61–70; repr. in her Coptic perspectives on Late Antiquity, Variorum CS (Aldershot 1993), no. XXV. ———. 1989. The strange death of Coptic culture. Coptic Church Review 10: 34–45; repr. in her Coptic perspectives on Late Antiquity, Variorum CS (Aldershot 1993), no. XXVI. Marotta, G., I. Putzu and M. Donati. 2017. Data from ancient languages and sociolinguistic analysis. In Language and identity in multilingual Mediterranean settings: Challenges for historical sociolinguistics, P. Molinelli (ed.), 201–21. Berlin. Mignolo, W. D. 2003. The darker side of the Renaissance: Literacy, territoriality, colonization. 2nd ed. Ann Arbor, MI.

Mullen, A. 2012. Introduction: Multiple languages, multiple identities. In Multilingualism in the Graeco-Roman worlds, A. Mullen and P. James (eds), 1–35. Cambridge. ———. 2013. Southern Gaul and the Mediterranean: Multilingualism and multiple identities in the Iron Age and Roman periods. Cambridge. Myers-Scotton, C. 2006. Multiple voices: An introduction to bilingualism. Malden, MA. Palombo, C. 2019. The view from the monasteries: Taxes, Muslims and converts in the ‘Pseudepigrapha’ from Middle Egypt. Medieval Encounters 25: 297–344. Papaconstantinou, A. 2001. ‘Là où le péché abondait, la grâce a surabondé’: Sur les lieux de culte dédiés aux saints dans l’Égypte des Ve–VIIIe siècles. In Le sacré et son inscription dans l’espace à Byzance et en Occident: Études comparées, M. Kaplan (ed.), 235–49. Paris. ———. 2007. ‘They shall speak the Arabic language and take pride in it’: Reconsidering the fate of Coptic after the Arab conquest. Le Muséon 120: 273–99. ———. 2008. Dioscore et le bilinguisme dans l’Égypte du VIe siècle. In Les archives de Dioscore d’Aphroditè cent ans après leur découverte. Histoire et culture dans l’Égypte byzantine, J.-L. Fournet (ed.), 77–88. Paris. ———. 2012. Why did Coptic fail where Aramaic succeeded? Linguistic developments in Egypt and the Near East after the Arab conquest. In Multilingualism in the Graeco-Roman worlds, A. Mullen and P. James (eds), 58–76. Cambridge. ———. In press. A monk deploring the assimilation of the Christians to the Hagarenes. In Conversion to Islam in the premodern age: A sourcebook, N. Hurvitz, C. Sahner, U. Simonsohn and L. Yarbrough (eds). Berkeley, CA. Peeters, P. 1922. Traductions et traducteurs dans l’hagiographie orientale. Analecta Bollandiana 40: 241–98. ———. 1950. Le tréfonds oriental de l’hagiographie byzantine. Brussels. Potowski, K. 2013. Language maintenance and shift. In The Oxford handbook of sociolinguistics, R. Bayley, R. Cameron and C. Lucas (eds), 321–39. Oxford. Ragheb, Y. 2013. Les premiers documents arabes de l’ère musulmane. In Constructing the seventh century, C. Zuckerman (ed.), 679–729. Paris. Reinhardt, K. 1897. Eine arabisch-koptische KirchenbannUrkunde. In Aegyptiaca: Festschrift für Georg Ebers zum 1. März 1897, 89–91. Leipzig. Richter, T. S. 2000. Spätkoptische Rechtsurkunden neu bearbeitet (II): Die Rechtsurkunden des Teschlot-Archivs. Journal of Juristic Papyrology 30: 95–148. ———. 2001. Arabische Lehnworte und Formeln in koptischen Rechtsurkunden. Journal of Juristic Papyrology 31: 75–89.

ARABIC: LANGUAGE OF EMPIRE, LANGUAGE OF EGYPT

———. 2004. O.Crum Ad. 15 and the emergence of Arabic words in Coptic legal documents. In Papyrology and the history of early Islamic Egypt, L. Sundelin and P. Sijpesteijn (eds), 97–114. Leiden. ———. 2006. Coptic [Arabic loanwords in]. In Encyclopedia of Arabic language and linguistics I, K. Versteegh (ed.), 595–601. Leiden. ———. 2008. Rechtssemantik und forensische Rhetorik. Untersuchungen zu Wortschatz, Stil und Grammatik der Sprache koptischer Rechtsurkunden. 2nd ed. Wiesbaden. ———. 2009. Greek, Coptic and the ‘language of the Hijra’: The rise and decline of the Coptic language in Late Antique and Medieval Egypt. In From Hellenism to Islam: Cultural and linguistic change in the Roman Near East, H. Cotton, R. Hoyland, J. Price and D. Wasserstein (eds), 398–443. Cambridge. ———. 2010a. Language choice in the Qurra dossier. In The multilingual experience in Egypt from the Ptolemies to the ‘Abbāsids, A. Papaconstantinou (ed.), 189–219. Farnham. ———. 2010b. Naturoffenbarung und Erkenntnisritual: Diskurs und Praxis spätantiker Naturwissenschaft am Beispiel der Alchemie. In Honi soit qui mal y pense. Studien zum pharaonischen, griechisch-römischen und spätantiken Ägypten zu Ehren von Heinz-Josef Thissen, H. Knuf, C. Leitz and D. von Recklinghausen (eds), 585–605. Leuven. ———. 2013. ‘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 Ägypten und sein Umfeld in der Spätantike. Vom Regierungsantritt Diokletians (284/5) bis zur arabischen Eroberung des Vorderen Orients um 635–646. Akten der Tagung vom 7.–9. 7. 2011 in Münster, A. Lohwasser and F. Feder (eds), 121–38. Wiesbaden. ———. 2015a. The master spoke: ‘Take one of the “sun” and one unit of almulgam’: Hitherto unnoticed Coptic papyrological evidence for early Arabic alchemy. In Documents and the history of the early Islamic world: Third Conference of the International Society for Arabic Papyrology, Alexandria, 23–26 March 2006, A. T. Schubert and P. M. Sijpesteijn (eds), 194–233. Leiden. ———. 2015b. On the fringes of Egyptian language and linguistics: Verb borrowing from Arabic into Coptic. In Fuzzy boundaries: Festschrift für Antonio Loprieno, H. Amstutz, A. Dorn, M. Müller, M. Ronsdorf and S. Uljas (eds), 1: 227–42. Hamburg. Romaine, S. 1995. Bilingualism. 2nd ed. Oxford. ———. 2000. Language in society: An introduction to bilingualism. 2nd ed. Oxford.

309

Rubenson, S. 1996. Translating the tradition: Some remarks on the Arabization of the patristic heritage in Egypt. Medieval Encounters 2: 4–14. Samir, K. S. 2014. L’Apocalypse de Samuel de Qalamūn et la domination des Hagaréens. In ‘Guerra santa’ e conquiste islamiche nel Mediterraneo (VII–XI secolo), M. Di Branco and K. Wolf (eds), 17–63. Rome. Schlieben-Lange, B. 1996. Idéologie, révolution et uniformité de la langue. Liège. Sheridan, M. 1998. Rufus of Shotep, Homilies on the Gospels of Matthew and Luke: Introduction, text, translation and commentary. Rome. Sijpesteijn, P. M. 2004. Travel and trade on the river. In Papyrology and the history of early Islamic Egypt, P. M. Sijpesteijn and L. Sundelin (eds), 115–52. Leiden. ———. 2011. Building and Egyptian identity. In The Islamic scholarly tradition: Studies in history, law and thought in honor of Professor Michael Allan Cook, A. Q. Ahmed, M. Bonner and B. Sadeghi (eds), 85–105. Leiden. Tillier, M. 2017. L’invention du cadi: La justice des musulmans, des juifs et des chrétiens aux premiers siècles de l’Islam. Paris. Tillier, M. and N. Vanthieghem. 2018. Un registre carcéral de la Fusṭāṭ abbasside. Islamic Law and Society 25: 1–40. van Bladel, K. 2018. Arabisering, islamisering en de kolonies van de veroveraars. In Mohammed en de Late Oudheid, J. van den Bent, F. van den Eijnde and J. Weststeijn (eds), 159–87. Hilversum. van Lent, J. M. 2000. The nineteen Muslim kings in Coptic apocalypses. Parole de l’Orient 25: 643–93, Vorderstrasse, T. 2015. Linguistic diversity at Fusṭāṭ. In A cosmopolitan city: Muslims, Christians and Jews in Old Cairo, T. Treptow and T. Vorderstrasse (eds), 53–59. Chicago, IL. Wasserstein, D. J. 2003. Why did Arabic succeed where Greek failed? Language change in the Near East after Muhammad. Scripta Classica Israelica 22: 257–72. Zaborowski, J. R. 2003. Egyptian Christians implicating Chalcedonians in the Arab takeover of Egypt: The Arabic Apocalypse of Samuel of Qalamūn. Oriens Christianus 87: 100–15. ———. 2005. The Coptic martyrdom of John of Phanijoit: Assimilation and conversion to Islam in thirteenthcentury Egypt. Leiden. ———. 2008. From Coptic to Arabic in Medieval Egypt. Medieval Encounters 14: 15–40. Ziadeh, J. 1915–1917. L’Apocalypse de Samuel, supérieur de Deir-el-Qalamoun. Revue de l’Orient chrétien 20: 374– 407. Zychowicz-Coghill, E. 2017. Conquests of Egypt: Making history in ‘Abbāsid Egypt. D.Phil. thesis, University of Oxford.

VI MINORITIES AND MAJORITIES

EGYPT, EMPIRE AND JUDAISM, 650 BC–AD 650 David NIRENBERG

The very title of this volume reminds us that in Egypt (and not only there) religious identities were linked to and shaped by political formations and imperial projects, not separate from them. Nor is this a linkage confined to Roman and post-Roman times, as anyone can attest who wandered among the treasures of the British Museum’s ‘Egypt: Faith after the Pharaohs’ exhibition that inspired this volume, many of which precede by centuries any nesting of Roman eagles on the Nile, and which include precious papyri from periods of Persian and Hellenistic rule over pharaonic lands. This linkage suggests something we might find disconcerting had we not imbibed, nolens volens, so much Foucault in our youth: that the representations of any religion produced at any given point in time are not, or not only, the product of any essential attribute of that religion, but rather (or also) the product of political processes of power and resistance. Many historians today are inclined to accept such a view— despite its implication that, since historians only have access to representations produced in the past, and are moreover also subject to the politics of their own present, every history they write would be doubly structured by power. Epistemology is above my pay grade. But what I would like to do in this essay is push towards its more radical implications this linkage between empire and religious identity, power and representation. If we take this linkage with imperial power seriously, what are the consequences for what we thought we knew about ‘religious identity’ in Egypt? (I put the term in scarequotes because it is not one I would choose, in part for the reason just mentioned, in part for others best left to a different essay.) In pursuit of that question I will focus on one case, and summarize how Judaism, across its very long

history in the lands of the Nile, provided Egyptians of many different periods, polities and faiths—among them pharaonic, Hellenistic, Roman, Christian and Muslim—with ways of thinking about themselves and their relationship to power. (To this list I should of course add ‘Jewish’ as well, but my emphasis here will be more on the work that non-Jews did with ‘Judaism’, than what Judaism might have meant to a self-identified Jew.) I will suggest that this ‘thinking with Judaism’ in the context of imperial power was so important that it has, in fact, shaped much of what we think we know about Judaism in Egypt. The ways in which Egypt’s inhabitants put ‘Judaism’ to work in their thinking about power and piety affected the possibilities of existence for Judaism in Egypt to such a degree that even in cases where there had for long been no living Jews to be found—as in my concluding examples—Egyptians were capable of creating figures of Judaism from, so to speak, the entrails of their own traditions. Already in some of our earliest sources we can see Egyptians and Jews (both words are anachronistic) making sense of themselves in terms of each other. Starting sometime around 650 BC, we find an Israelite garrison and an Egyptian one defending the frontier side by side on the island of Elephantine, at the southern limits of ancient Egypt (Grelot 1986; Porten 2003; Kahn 2007). There they worshipped in close proximity, with a ‘Temple of the Jewish God YHW’ near that of the local Egyptian deity, the ram-god Khnum.1 Though their collaboration began earlier, it comes most sharply into our documentary field of vision after the Persian king Cambyses conquered Egypt in 525 BC.2 It is this period of Persian imperial rule that is illuminated by an extraordinary ‘archive’ of documents written in Aramaic by or for the Jews of Elephantine, roughly from the years 495 until 400 BC.

1

2

On religious practices in Elephantine see Alyoueny 1981; Bolin 1995; Dion 2002; Kottsieper 2003.

On Cambyses’ conquest, see Sternberg-El Hotabi 2003 and Yamauchi 1996. For discussion of the chronology see Devauchelle 1998.

314

D. NIRENBERG

Among the more famous of those documents is one of the very few attesting to friction between Egyptians and Jews: a letter (included in the exhibition) that a man called Hananiah wrote to the Jewish garrison of Elephantine in the fifth year of King Darius’ rule (419 BC). The letter is known as the ‘Passover letter’ because in it King Darius apparently instructs his governor or satrap, Arsames, to tell the Egyptians of Elephantine to stay away from their Jewish neighbours during Passover: ‘it has been sent from the king to Arsames the prince, saying: keep away from the Jewish garrison’ (C21/B13/TAD A4.1).3 From two other papyri we learn that nine years later, when Arsames was absent from Egypt, the Egyptians destroyed the Jewish Temple: In the month of Tammuz, year 14 of Darius the king, when Arsames had departed and gone to the king, the priests of Khnum the god…, in agreement with Vidranga, who was chief here, (said), saying: ‘The Temple of YHW the God which is in Elephantine the fortress let them remove from there.’ … [They] broke into that Temple, demolished it to the ground, … But the basins of gold and silver and the (other) things which were in that Temple— all (of these) they took and made their own … (C30/B19/ TAD A4.7, ll. 14–43; see also C 31/B20/A4.8).4

The same document tells of how order was restored, and of the diplomatic missions sent by the Jews of Elephantine to the High Priest in Jerusalem, and to the governors of Judah and Samaria, in order to gain permission to rebuild the Temple. Permission was granted in 407 BC, but without the privilege of carrying out the animal sacrifices that the earlier Temple had enjoyed (Lindberger 2001). In 399 BC the Jewish garrison at Elephantine vanishes from the historical record. Why did the Egyptians destroy the Temple of the God YHW? We can only hypothesize, but one answer might be that Egyptians resented the Jews as allies of the Persians, whom they viewed as their oppressors. In this interpretation, the Jews are attacked because they represent something else—in this case, imperial Persian power: a process we could already call ‘Egyptians

3

C21/B13/TAD A4.1 refers to the three editions and translations of this text: C=Cowley 1923, no. 21; B=Porten 2011, no. 13; TAD=Porten and Yardeni 1986–1989, no. A4.1. On the status of this text, see Gass 1999. The reconstruction of line 3 given here is from Porten 1968, 129 and 311–12, following suggestions in Grelot 1954 and Galling 1964. The phrase was an attempt to fill in a lacuna, and Porten later abandoned this

thinking with Judaism’. Or we could relate the destruction to the ‘Passover letter’ King Darius had sent nine years before. Why Passover? Perhaps the traditional Jewish sacrifice of a paschal lamb offended the Egyptian priests who were acolytes of the ram-god Khnum, making the festival a special point of friction. For those attracted to biblical echoes, this explanation has the virtue of evoking Moses’ objection in Exodus 8.26 to Pharaoh’s request that the Hebrews remain in Egypt rather than go out into the desert to make their sacrifice: ‘Lo, if we shall sacrifice the abomination [sacred animal] of the Egyptians, will they not stone us?’ But we could also imagine that the Egyptians were offended as much by the ‘history’ the sacrifice commemorated as by the species sacrificed: offended, that is, by the festival’s ritual re-enactment of the defeat of Egypt and the drowning of its armies as described in Exodus. Perhaps the priests of Khnum, becoming aware (some two centuries before the Septuagint translations) of the negative roles played by figures of Egypt in the work of self-definition done by the Jews’ Passover, began to develop their own commemorations interpreting ‘the Exodus’ in counter-dialogue with the Hebrews’ version. From this perspective, the Passover attacks at Elephantine could be seen as symptoms not only of indirect protests against imperial power, but also of a developing Egyptian sense of history in which the Jews were beginning to be understood as figures inimical to Egyptian piety, sovereignty and prosperity. The priests of Khnum have left us no papyri, no version of the events in their own words. But in the following century, after Alexander the Great’s conquest of Egypt in 332 BC, we do start to find evidence of precisely such Egyptian traditions about the Jews. Of course we should not naively treat these traditions as simply continuous with what came before. Greek empire brought with it new ways of thinking about governance and politics, new questions about the world, new tools with which to work on those questions, and even a new language (Greek) in which to ask them. These new questions and tools

4

reconstruction (Porten 2011, 127, n. 13). This does not, to my mind, remove the possibility that the ensuing conflict had ritual roots. See also Schäfer 1997, 124–26. Briant 2002, 603–7, on the other hand, sees here simply a land dispute with the priests of Khnum. My thanks to Bezalel Porten for his personal communication on this subject. For abbreviations, see n. 3 above.

EGYPT, EMPIRE AND JUDAISM

transformed the kinds of thinking that Egyptians could do with Judaism. But neither should we be in too much of a hurry to treat ideas about Jews that we discover in Hellenistic Egypt as entirely discontinuous from those of previous centuries. Certainly Egypt’s own inhabitants did not do so. Indeed, history was itself an important tool of Hellenistic empire: the many peoples that Alexander’s arms brought under Greek rule needed to be intertwined, not just by conquest, but also by a shared sense of origins and destiny capable of supporting newly cosmopolitan visions of the polity.5 This need to redeploy the past in order to produce the present generated histories of Egypt such as the one written c. 320 BC by a Greek historian called Hecataeus of Abdera, which survives only as excerpts quoted in the works of later historians. One of these excerpts preserves the earliest non-biblical version of the Exodus story we possess. In it Hecataeus described a time in the long distant past, before the Greeks had even been dreamed of, when Egypt was afflicted by a terrible plague: The common people ascribed their troubles to the workings of a divine agency; for indeed, with many strangers of all sorts dwelling in their midst and practicing different rites… their own traditional observances in honor of the gods had fallen into disuse. … At once, therefore, the aliens were driven from the country. The most outstanding and active among them banded together and, as some say, were cast ashore in Greece and certain other regions…. But the greater number were driven into what is now called Judaea, which is not far distant from Egypt and was at that time utterly uninhabited. The colony was headed by a man called Moses, outstanding both for his wisdom and for his courage (Hecataeus of Abdera, Aegyptiaca, excerpts in Diodorus, Library of History, 40.3.1–3; Walton and Geer 1967, 279–83; cf. Bar-Kochva 2010, 100–3).

5

6

7

For multicultural history, see Hartog 1996, 73–75. On Hecataeus specifically, see also Redford 1986, 281–82. On the complexities of this fragment see Mendels 1983; Albertz 2001; Grabbe 2008; Bar-Kochva 2010, 90–135. Hecataeus dismissed the validity of Egyptian traditions about the origins of the Athenians, but he seems to have accepted their traditions concerning the Babylonians, the Colchi, and the Jews (Bar-Kochva 2010, 113, 115–16). He also drew on biblical material, although he wrote before the translation of the Torah into Greek. There is a debate about the valence of Hecataeus’ representation of Jews and Judaism. J. Gager argues that Hecataeus’ ‘non condemnatory’ rhetoric was transmuted into antiJudaism by subsequent Greek writers (Gager 1983, esp. 39–40, 69–76). Against this, P. Schäfer contends that the basic elements of anti-Judaic rhetoric can already be found in Hecataeus and Manetho (Schäfer 1997, 15–39). See also Berthelot 2008.

315

Moses, Hecataeus tells us, also founded a new religion, distinguished by its ban on the worship of images, and especially noteworthy for the misanthropy or a-sociability of its adherents, ‘for as a result of their own expulsion from Egypt he [Moses] introduced an unsocial [separate from men] and intolerant mode of life (ἀπάνθρωπόν τινα καὶ μισόξενον βίον)’.6 This brief account provides a beautiful example of how Hellenistic historians turned to Egyptian history in order to provide an account of the origins of their own world: in this case, an account of the origins of the Greeks (and elsewhere the Babylonians), as well as of the Jews. But given that Hecataeus tells us that he drew on some version of Egyptian sources in crafting this account (Bar-Kochva 2010, 109–10, and see below), it also provides the earliest window we have into how Egyptians could draw upon their past in order to create historical representations of Jews and Judaism with which to make sense of their present place in the world.7 Roughly a generation after Hecataeus we find a native Egyptian version of precisely such a history. Its author was Manetho, a priest of Heliopolis during the reign of Ptolemy II Philadelphus (282–246 BC).8 As an ethnic Egyptian and a priest, Manetho had greater access than Hecataeus to the physical and textual remnants of Egyptian history. But from both accounts it becomes clear that these Egyptian histories of the origins of the Jews drew upon a number of traditional themes that were much older still, themes that had long shaped the ways in which Egyptians thought about the ups and downs of their history. One of these themes was that of foreign invasion by a people known as the Shepherds or Hyksos, who swept into Egypt and dominated it for more than a hundred years circa the seventeenth century BC.9 After

8

9

Bar-Kochva contrasts Hecataeus’ critical comments about the Jews and their way of life with his ‘determination to justify or explain away every deplorable or strange aspect of Egyptian life’ (Bar-Kochva 2010, 132–35). Quotes from Manetho are given by fragment number from Waddell 1948, followed by the passage of Josephus that cites them. On Josephus’ Manetho see Raspe 1998 and Pucci ben Zeev 1993. On Manetho as ethnic historian in an imperial age see Moyer 2011, 84–141, and Dillery 2015, passim. On Manetho’s sources, and his adaptation of ‘proto-apocalyptic’ narratives, including the Oracle of the Lamb, see Dillery 2015, 301–3. On the Hyksos, see Goharghi 1999. The association of Hyksos (‘rulers of foreign lands’) with ‘shepherds’ is a false etymology. See Bietak 2001. The etymology is Manetho’s, fr. 42; Josephus, Against Apion 1.82–83.

316

D. NIRENBERG

their defeat and expulsion (c. 1555 BC), retrospective stories about the Hyksos became an important motif in Egyptian political and historical thought.10 (A famous example of this type of ‘historical propaganda’ comes from an inscription known as the Tutankhamun Restoration Stela today in the Cairo Museum.) The point of such stories was to demonstrate how, through careful attention to the proper maintenance of temples and their cult, the good king regains divine favour, and catastrophe becomes prosperity.11 Many centuries later, as we have just seen in the examples of Hecataeus and Manetho, Egyptians could draw upon these traditions to cast the Jews as the degenerate descendants of defeated invaders from the distant past.12 But we should not forget that this strategy towards the Jews was itself part of a broader Egyptian discourse of historical theodicy that redeployed ancient traditions in order to address a series of foreign imperial conquerors: Assyrian, Persian, Greek, Roman. The stories were used to emphasize the dangers ‘aliens’ posed to Egypt, to rally resistance to them, and even to cast contemporary invaders as descendants of those expelled by Egypt long ago, and therefore doomed to a similar fate.13 The histories produced by Hecataeus, Manetho, Chaeremon, and others (including Jews, such as Artapanus Judaeus14) could be understood as part of this process, in which the traditional

10

11

12

13

The Hyksos retreated to Palestine c. 1555 BC, but remained a threat for another half-century or more. The New Kingdom in Egypt (Dynasties 18–20) lasted from roughly 1569 BC (shortly before the expulsion of the Hyksos) to around 1076 BC (Spalinger 2001; Murnane 2001). Redford 1986, 260–61, 264–65. A very early example of such a lamentation is found in the fragments from the palace at Tod, celebrating the achievements of Senwosret I, of Dynasty 12 (1971–1928 or 1958–1913 BC). J. J. Collins (2000) explores this association of the Jews with the Hyksos, drawing out the political implications of the retelling of exodus narratives by both Jews and Egyptians. See also Davies 2001. In the Graeco-Roman period, for example, we hear echoes of these analogies in works of prophetic literature such as the Oracle of the Lamb, works that ‘foretold’ invasion and promised liberation. The prophecy of the Lamb is edited by K. Th. Zauzich (1983). It is a composite text and difficult to date, but the extant demotic version was put together at least two centuries after Manetho’s floruit. For a survey of the scholarship, see Ritner 2003, 445–49; Gozzoli 2006, 293–97. On the genre of prophecy in Egyptian/demotic literature, see Koenen 1968. See also Fragment 1 of the first-century Egyptian priest Chaeremon (Van der Horst 1984, 8–9, 49–50). In this tradition the conflict

tools with which Egyptians had fashioned images of enemies now long forgotten (such as the Shepherds) were applied in the service of new struggles taking place in a multicultural Hellenistic court in which native Egyptians deployed their history in competition with Jews, Greeks, and others for prestige. The Jewish translation of the Hebrew Bible into Greek—the Septuagint—can itself be understood as a product of this same competition. It was undertaken, according to ancient tradition, in Alexandria at the behest of Ptolemy II Philadelphus (282–246 BC): that is, at more or less the same time as Manetho was writing his summation of the glories of Egyptian history (Hermann and Baumgärtel 1923, 48–50; CPJ 1.1, pp. 32, 42).15 We might see both projects as (among other things) products of a competition among the elites of subject peoples for prestige in the eyes of their historically minded overlords. The Jews, like the Egyptians, were translating the chronicles of their heroic past into a language that their Greek sovereigns could understand. The process of translation may even have given new edge to the competition. The increasing availability of Jewish scriptures in Greek may have made Egyptians more aware of (and more offended by) the Hebrew Bible’s negative representations of their country and their ancestors, further inspiring the production of Egyptian narratives about Jews.16

14

15

16

was understood as an enduring confrontation between the followers of Isis and Osiris, and worshippers of Seth (= Baal). On this tradition’s place in Egyptian anti-Judaism see van Henten and Abusch 1996. Like Hecataeus and Manetho, we know of Artapanus only because fragments of his work are preserved in later histories, in Artapanus’ case those of the Greek historian Alexander Polyhistor. For the fragments, see Jacoby 1954, no. 726, pp. 680–86. On Artapanus, see Koskenniemi 2002; Kugler 2005; Jacobson 2006. CPJ = Tcherikova and Fuks 1957. An Egyptian Jewish account of the tradition is provided by the second-century BC Letter of Aristeas (trans. Hadas 1973; ed. Pelletier 1962), but scholarship is sharply divided about its reliability. A. van der Kooij (2008), for example, preserves the hope that the letter can tell us something about the Septuagint and its Alexandrian context. Against that hope, see Carbonaro 2008. T. Rajak (2008) attempts a mediation. See also in the same volume Gruen 2008, and, more generally, Wasserstein and Wasserstein 2006. But see now especially Wright 2015. Kasher (1985) argued, somewhat implausibly, that Manetho was himself motivated by the Septuagint to counter the humiliating picture of Egypt in the Hebrew Bible. Cf. Schäfer 1997; Gruen 1998; and especially Borgeaud 2007.

EGYPT, EMPIRE AND JUDAISM

Over the course of the next three centuries of Greek and then Roman rule in Egypt, the ‘Manethine’ negative image of Moses and the Jews was put to many uses. Rather than making sense of their vicissitudes by telling stories about the Hyksos or about general neglect of the gods, Egyptians could use Jews to do the work of these earlier motifs, as when a prophetic papyrus from the Roman period warns Egyptians of impending disaster, when ‘impious people will destroy your temples’, ‘your largest temple will become sand for the horses’ and Jews will inhabit the sacred city of Helios. The solution? ‘Attack the Jews’, ‘lawbreakers’ who have already been ‘once expelled from Egypt by the wrath of Isis’ (CPJ 3.520, late second–early third century AD, as quoted in Frankfurter 1993, 189–91).17 I will not attempt to explain how and why this shift occurred—that is, how it began to make sense to many Egyptians, both native and Greek, to understand their past history and their present circumstances under Greek or Roman rule in terms of Judaism. Instead, let me simply offer an example of how powerful that sense could be, and how separated it could become from what we might want to call ‘reality’. The ‘Acts of the Alexandrine Martyrs’ are stories of heroic citizens risking death to defend their city’s freedoms from the tyrannical power of Rome and its malicious agents, ‘usually the Jewish community resident in Alexandria’ (Harker 2008, 1). Isidorus, for example, came before the emperor Claudius (r. AD 41–54) to complain of ‘my native city’s sufferings’ and proclaimed ‘I am… the gymnasiarch of the glorious city of Alexandria. But you [Claudius] are the cast off son of the Jewess Salome!’ The emperor was not amused: Isidorus was executed (Musurillo 1954, 18–26).18 A delegate in a later embassy sometime between AD 105 and 112 deployed similar terms to the emperor Trajan:

17

18

19

On this papyrus, known as the Oracle of the Potter, see Bohak 1995 and Koenen 2002. Other versions of the oracle mention other enemies, such as the Alexandrians and Greeks. The dating is debated, but almost certainly falls between AD 41 and 53. If after 43/44, the case was against Agrippa II. If earlier, it was against Agrippa I, whose visit to Alexandria had sparked the riots of AD 38. On the political context of this story and the difficulties of its dating, see Harker 2008, 10–24. Hermaiscus was saved from death when the bust of Serapis that he was carrying began to sweat, an event interpreted by the

317

Hermaiscus: ‘Why, it grieves us to see your Privy Council filled with impious Jews.’ Caesar said: ‘… I am telling you Hermaiscus: you are answering me insolently, taking advantage of your birth.’ Hermaiscus said: ‘What do you mean, I answer you insolently, greatest Emperor? Explain this to me.’ Caesar said: ‘Pretending that my Privy Council is filled with Jews.’ Hermaiscus said: ‘So then, the word ‘Jew’ is offensive to you? In that case you ought rather to help your own people and not play the advocate for the impious Jews.’19

Some historians have wondered which Salome could have been Claudius’ mother (Salome I would be too old, while Salome II, famous from the story of John the Baptist, was born a generation after Claudius), or whether the Roman Senate of Trajan’s day might indeed have been full of Jews. (It was not.) (Musurillo 1954, 168–72). But the search for a Roman reality behind the charges misses the point of these texts: they are the product of a discourse that represented the struggle against a foreign and imperial tyranny in terms of a struggle against the Jews. Representation and reality are here mutually constitutive. The earliest texts of the Alexandrian martyr genre presented themselves as accounts of embassies sent to negotiate with the emperor the consequences of the lethal Alexandrian riots against the Jews in the year AD 38, events made famous to posterity through the writings of Philo (who lived through the riots and participated in the embassies) and later Josephus.20 Those riots were themselves to some extent the product of existing Graeco-Egyptian discourses about Jews as enemies of the polity. The accounts of Alexandrian martyrdom produced in the wake of those riots and their attendant embassies became in turn powerful vehicles for discourses representing the Jews as enemies of Egypt, the Romans (if they protected or supported the Jews) as ‘Jewish’ tyrants, and the

20

emperor as an omen. Jewish privy council: Acta Hermaisci 3.41–53, in Musurillo 1954, 45. Sweating statues were a standard type of marvel in Antiquity. See, for example, Plutarch, Coriolanus 38.1 and Alexander, 14.8–9. Some centuries later John Lydus states that weeping or sweating statues portend internal disorder and civil strife; see his De ostentis, proem. 8 (Wachsmuth 1897, 16). Although a number of Philo’s and Josephus’ writings are relevant here, a good starting place is the introduction, commentary and appendices to Barclay 2007.

318

D. NIRENBERG

Egyptians as martyrs to both. We know from the surviving papyrological evidence that these texts circulated widely, and continued for centuries to be ‘read in Egypt wherever people could read’ (Harker 2008, 2), thereby presumably shaping future possibilities of coexistence. For example, in AD 66 when the citizen body of Alexandria met to elect delegates for a mission to Rome, the crowd spotted three Jews, assailed them with cries of ‘spies’ and ‘enemies’, and burned them to death. During the ensuing riot Roman legions and Alexandrines entered the Jewish quarter and killed thousands. Josephus says 50,000 (Josephus, War 2.490–97; see also Feldman 1993, esp. 117–18). In AD 115 the citizens of Alexandria again attacked the Jews and called in the Roman legions, claiming that ‘impious Jews’ were planning to invade the city (CPJ 2.158a, col. VI; Fuks 1953; Pucci Ben Zeev 1981, 1982 and 1983; Smallwood 1981; Barnes 1989). The Jewish population of Alexandria was virtually eliminated (CPJ 2.158a, col. I; Eusebius, Church History 4.2–3). In the rest of Egypt the Jews revolted. According to our sources, Jews slaughtered their Egyptian neighbours, tore down Egyptian temples and destroyed statues of the gods: a list of horrors suspiciously akin to those attributed to Moses and his followers by Manetho and the Egyptian Exodus tradition.21 Papyri preserve mothers’ prayers to the gods, begging them to protect their children from being defeated by the Jews (P.Giss. 24 = CPJ 2.437).22 Native Egyptian priests mobilized peasant armies, Greek citizens formed militias, and an expeditionary army reinforced the regular Roman legions. By AD 117 the Jews of Egypt, city and countryside, were destroyed. With these victories ‘real’ Jews more or less disappear from Roman Egypt. From the following three centuries scholars have identified no more than forty-four papyri that may refer to Jews, and in most of those the identification seems improbable. But these papyri do show that even in the absence of Jews, their memory

could still be put to work. A petition from the town of Oxyrhynchus in 199/200 invoked the goodwill the city had shown to the Romans during the ‘war against the Jews’, and added that ‘even now they [are] keeping the day of victory as a festival every year’. In other words, the Oxyrhynchites (and perhaps other towns as well) were celebrating the defeat of the Jews some eighty years after the fact, much as we today might celebrate the end of the First or the Second World War (CPJ 2.450; see also Modrzejewski 1995, 224). If we are to believe a critique of theatrical and athletic spectacles written at about the same time, some places staged representations of Jews in some sort of subjection: a funny sight, judging from the papyrus’ description of a mime imitating a ‘man bearing a Jewish burden (Ἰουδαϊκὸν φορτίον)’. ‘Why do you laugh? Why are some of you disgusted at what was said or at the man you see?’ (What this Jewish burden was we do not know: the most likely options are either a religious object such as Torah or tefillin, or a representation of the special Jewish tax burden that had been paid by the Jews before their destruction [CPJ 3.519, late second–third century AD].)23 The Acta Alexandrinorum hit the height of their popularity (or at least, of their survival in the papyrological record) in this same period as well (Harker 2008, 2). And as late as the third century we still find Egyptian oracles and prophetic texts, such as a version of the ‘Oracle of the Potter’, that explain the miseries of Egypt in terms of Isis’ anger at impurity, and urge readers to ‘go against the Jews’, though it does not seem that there were many (or indeed any) Jews to go against. The point bears stressing: Jews could remain useful, even as they became impossible to find. Egypt provides a particularly stark example of this curious phenomenon, because the (near?) elimination of the Jews there in the early second century coincided with a new religion’s rapid rise along the Nile. Christianity placed figures of Judaism at the centre of its imagination more than any cult of Isis or Osiris ever had. We need

21

23

22

Appian, Civil Wars 2.90; Historia Augusta, Life of Hadrian 5; Dio Cassius 68.32, 69.8; and Eusebius, Church History, 4.2–3. Cf. the papyri in CPJ 2.435–3.500, as well as Tcherikover’s comments in CPJ 1.86–93, and Barclay 1996, 79, n. 69. The edition has ‘roasted’ rather than ‘defeated’, but see Bagnall and Cribiore 2006, 157–58.

A reading of Jewish texts such as Lamentations Rabbah might yield a rabbinic perspective on this ritual representation of defeat. Allen Kerkeslager attempts to assign the text a much earlier date (c. AD 40), in order to interpret it as referring to the ridicule suffered by a naked Jewish athlete in a Greek gymnasium, the burden being a circumcised penis (Kerkeslager 1997).

EGYPT, EMPIRE AND JUDAISM

not rehearse here what everyone already knows: that Christianity emerged from Judaism, a New Israel (as Christians would come to understand it) from an Old. Indeed the earliest writings of the followers of Jesus—some of which were later gathered into what came to be called the canonical New Testament, others not—are full of concern about the relationship between their movement and Judaism, and about how to distinguish between the two. Already in their earliest writings, such as the letters of Paul, we can see that concern expressed in terms of the danger of falling from the one to the other: a danger that Paul distilled very early into the concept of ‘Judaizing’, that is, of Gentile converts to Christ acting or believing in an inappropriately ‘Jewish’ way (Gal. 2.14).24 As a result of the importance of Judaism and Judaizing as a term of criticism in the thought and writings of the earliest followers of Jesus, the terms remained central to the Christian imagination long after Christianity had separated from Judaism, grown far larger than its ‘parent’, and indeed become the official religion of the Roman empire, capable of mobilizing the full force of imperial law and power against its rivals, whether pagan or Jewish. Volumes have been and continue to be penned about the place of Jews and Judaism in the thought of theologians such as Sts Augustine, Chrysostom and Jerome. And while it is clear that this place was a large one, there is an ongoing debate about whether or not that importance reflects a concomitant importance of ‘real’, living Jews in the society of those authors. Some scholars continue to insist (implausibly to my mind), that St John Chrysostom’s dialogues against the Jews were motivated by the real threat that the Jews of Damascus posed to the Christian community there. Of Augustine, we can say that in all of his voluminous writings he speaks only once of having met a living Jew: a plaintiff who came to his court to complain of having been illegally dispossessed by another bishop. In the particularly well-documented case of Egypt, we should expect that if there were an important Jewish presence there it would have left some trace in the plentiful ostraca or papyri produced by the rapacious

24

For a treatment of the meaning of Galatians 2.14 in its Pauline context, and of the long future of the logic of Judaizing, see Nirenberg 2013, 59–60, and passim.

319

Roman administrative apparatus. It does not. Where it frequently appears is in the preaching of those who, like the aforementioned Chrysostom or Augustine, criticized the behaviour of their fellow Christians in terms of Judaism. I very much doubt that the famed Egyptian monk Shenoute (AD 347–465) ever met a Jew, but he certainly understood the utility of ‘Judaizing’ as an accusation, and (like many other early Christians) did not hesitate to cast Christians who resisted his teachings as ‘Jews’. In the sermon known today as A26, for example, Shenoute likened his (Christian ‘crypto-pagan’) opponent Gesios to the Jews, ‘so that the destruction that came upon those people will come doubled upon the crown of his head’. And in the sermon I am amazed, he did the same to those Christians who did not pray as he advocated: ‘O you faithless Jewish hearts and other heretics of this sort, who are like you and you like them in this same deceptive spirit?’ (Brakke and Crislip 2015, 248 and 81 respectively; see also Brakke 2016). The utility of this concept of Judaizing, combined with the widespread tendency to interpret the Old Testament as a figuration of the New, meant that Christians could fill Egypt with figurative Jews even in the total absence of living ones. (Just as, for example, English writing of the fourteenth through seventeenth centuries overflows with figures of Judaism—think of Shylock—and accusations of Judaism and Judaizing, despite the absence of any living Jews in the British Isles.) Put in other words, imperial Christian languages of power were just as capable of producing figures of Judaism as the earlier Persian, Hellenistic, and Roman ones had been. This poses difficulties for historians seeking to assess the survival of ‘real’ Jews in Christian Egypt, for there is always the danger of confusing figures of thought and figures of flesh. How, for example, to interpret the so-called Exodus Chapel in the Bagawat Necropolis at the Kharga Oasis, near the ancient town of Hibis, the capital of the Great Oasis in Antiquity? The decorative programme of this late fourth- or early fifth- century Christian chapel is rich with Old Testament images, such as the ‘Parting of the Red Sea’, or

320

D. NIRENBERG

‘Three Hebrews in the Oven’. It might be tempting to move from the figural to the real, and hypothesize that the chapel’s congregation was made up of a Late Antique Egyptian Jewish community that converted to Christianity. But such a deduction would be as improbable as it is unnecessary. We have virtually no evidence of any Jews in that area for the previous 300 years, and these Old Testament themes can all be found in both pre-and post-Constantinian Christian art from ‘Gentile’ congregations.25 Many centuries would pass before the reappearance of documentary evidence for living Jews in Egypt, now under Muslim, rather than Roman or Christian, rule. (On that evidence see Petra Sijpesteijn’s important contribution to this volume.) We might want to assume that this later community implies a continuous settlement of Jews along the Nile, a population submerged and silent from the second century that suddenly appears once more on the surface of our documentary record in the ninth century AD. That explanation is possible, although 700 years is a long time for a community to hold its breath, so to speak. It is also possible that, as happened in other areas (such as the Iberian peninsula), the first centuries of Muslim rule produced widespread migration of Jewish populations to Egypt, as they adapted to and were transformed by the new possibilities potentiated by Islamic empire, much as had occurred under previous empires, and would in those yet to come. We should not forget that those possibilities remained just as complex as they had been before. Islam, like Christianity, claimed to share a common history and scripture with Judaism and therefore had to spend a great deal of time thinking about and differentiating itself from ‘Judaism’. Relations between power and representation, and between the figural and the real, were no simpler under Islamic empire than they had been under Christian, Roman, or Greek, though of course there is much that was highly distinctive about the spaces available for Judaism in Egypt under Islam. Let me conclude with just one example of the relationship between representation and what we can know

about ‘religious identity’ in Islamic Egypt, an example that reverses the optic insofar as it is not so much about how representation shapes what we think we know about ‘Judaism’, but rather about ‘Islam’. It is a curious irony that much of our knowledge about Islamic Egypt comes from ‘Jewish’ rather than ‘Islamic’ sources, that is, from the hands of a subordinate community rather than from those of the dominant religion. The thousands of manuscripts discovered in the book cemeteries of Cairo’s synagogues (the Cairo Genizah) remain a vital source for large swathes of ‘everyday life’ under Islam. Historians of Medieval Islamic trade, for example, can look to Islamic law books and notarial manuals for the formal rules of business and trade, but to trace its actual practice they must often turn to the Genizah. As a result we often find historians asking themselves whether a business practice documented in ‘Jewish’ sources is Islamic (that is, borrowed from Islam), and conversely, reconstructing ‘Islamic’ business practices on the basis of ‘Jewish sources’, and similarly for other spheres of Medieval life on the Nile, in a kind of historical co-production of Judaism and Islam.26 Co-production: perhaps that inelegant compound provides a serviceable conclusion for this essay. Egyptians of whatever religion (pagan, Jewish, Christian, Muslim), and living under a series of diverse empires (Persian, Ptolemaic, Roman, Islamic) made sense of their worlds and their histories by thinking about each other. I dare say that they are still doing so today. One result of this co-production is that it is both very important and very difficult for the historian to separate the figures of thought produced by that process in the distant past from whatever ‘religious identity’ any adherents of one of these communities might have recognized as their own. How do we distinguish between ‘Jewish identity’ in Egypt in the third century BC or the first century AD, or the first century after the hijra; and the images of ‘Judaism’ produced in those periods by pagan and later Christian Egyptians struggling to make space for themselves under Ptolemaic or Roman or Muslim rule?

25

26

On the Bagawat chapel, see Martin 2006.

For a critique of this tendency, see Ackerman-Lieberman 2014.

EGYPT, EMPIRE AND JUDAISM

321

Translations into examples from our contemporary world are sometimes suggestive, though also of course misleading. Today, in Egypt, Muslim views of Judaism and Christianity, as well as of their own Islamic ‘identities’, are very much shaped by their experience of Western imperial, colonial and post-colonial domination. A critical historian would not want to confuse them with some essential attribute of the Jewish or Christian ‘identities’. And yet, precisely because they have such a tremendous impact on the possibilities of life for Muslims, Christians and Jews alike, these figures of thought cannot be separated from the identities they claim to represent. The creation of an image of one group by another transforms the possibilities of existence for both, in different ways and depending upon countless variables, including asymmetries of power. Returning to the distant past: already in Antiquity the historian Josephus noted that the figures of Judaism generated by Egyptians in their engagements with empire had been and would continue to be of great consequence for the future of ‘Judaism’. Shortly after

Roman armies had obliterated many of his people and all of their political hopes, he wrote—with Manetho very much in mind—that ‘the Egyptians are the originators of the calumnies against us’. I am not endorsing that claim. But I do hope to have convinced you that in the Egyptian case, at least, the images of ‘Judaism’ that Egyptians generated through the tools with which they thought about power and empire had such a profound effect on the possibilities of existence for what we might call real Jews that it becomes difficult for the student of the past to separate figures of thought from figures of flesh. I doubt it is much easier for the student of the present.

Bibliography

Barnes, T. D. 1989. Trajan and the Jews. Journal of Jewish Studies 40: 145–62. Berthelot, K. 2008. Hecataeus of Abdera and Jewish ‘misanthropy’. Bulletin du Centre de recherche français à Jérusalem 19. https://bcrfj.revues.org/5402 (last accessed 28 Nov. 2016). Bietak, M. 2001. Hyksos. In The Oxford encyclopedia of ancient Egypt 2, D. B. Redford (ed.), 136–42. Oxford. Bohak, G. 1995. Corpus Papyrorum Judaicarum III, 520: The Egyptian reaction to Onias’ temple. Journal for the Study of Judaism in the Persian, Hellenistic and Roman Period 26: 32–41. Bolin, T. 1995. The Temple of YHW at Elephantine and Persian religious policy. In The triumph of Elohim: From Yahwisms to Judaisms, D. Edelman (ed.), 127–42. Kampen. Borgeaud, P. 2007. Moïse, son âne et les Typhoniens: Esquisse pour une remise en perspective. In La construction de la figure de Moise, T. Römer (ed.), 121–30. Transeuphratène Suppl. 13. Paris. Brakke, D. 2016. Shenoute and the Jews. In Coptic society, literature and religion from Late Antiquity to modern times: Proceedings of the Tenth International Congress of Coptic Studies, Rome, September 17th–22nd, 2012, P. Buzi, A. Camplani and F. Contardi (eds), 945–52. Leuven.

All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed 8 July 2018). Ackerman-Lieberman, P. I. 2014. The business of identity: Jews, Muslims and economic life in Medieval Egypt. Stanford, CA. Albertz, R. 2001. An end to the confusion? Why the Old Testament cannot be a Hellenistic book! In Did Moses speak Attic? Jewish historiography and scripture in the Hellenistic period, L. Grabbe (ed.), 30–46. Sheffield. Alyoueny, M. 1981. The religion of the Elephantine Jews. Beit Mikra 26: 217–30. Bagnall, R. S. and R. Cribiore. 2006. Women’s letters from ancient Egypt, 300 BC–AD 800. Ann Arbor, MI. Barclay, J. M. G. 1996. Jews in the Mediterranean diaspora: From Alexander to Trajan (323 BCE–117 CE). Edinburgh; Berkeley, CA. ———. 2007. Against Apion: Translation and commentary. Boston, MA; Leiden. Bar-Kochva, B. 2010. The image of the Jews in Greek literature: The Hellenistic period. Berkeley, CA.

Acknowledgements The author would like to thank the anonymous reader for the press for their careful reading; Bezalel Porten for a helpful suggestion; Elisabeth R. O’Connell for her generous care with this essay and this volume; and Sofía Torallas Tovar for her inspiration, learning, and correction.

322

D. NIRENBERG

Brakke, D. and A. Crislip. 2015. Selected discourses of Shenoute the Great: Community, theology and social conflict in Late Antique Egypt. Cambridge. Briant, P. 2002. From Cyrus to Alexander: A history of the Persian empire. Translated by P. T. Daniels. Winona Lake, IN. Carbonaro, P. 2008. Aristobule et Hécatée d’Abdère. Ephemerides Theologicae Lovanienses 84 (1): 181–93. Collins, J. J. 2000. Reinventing Exodus: Exegesis and legend in Hellenistic Egypt. In For a later generation: The transformation of tradition in Israel, early Judaism and early Christianity, R. A. Argall, B. A. Bow and R. A. Werline (eds), 52–62. Harrisburg, PA. Cowley, A. E. 1923. Aramaic papyri of the 5th century B.C. Oxford. Davies, P. 2001. Judeans in Egypt: Hebrew and Greek stories. In Did Moses speak Attic? Jewish historiography and scripture in the Hellenistic period, L. Grabbe (ed.), 108–28. Sheffield. Devauchelle, D. 1998. Un problème de chronologie sous Cambyse. Transeuphratène 15: 9–17. Dillery, J. 2015. Clio’s other sons: Berossus & Manetho. Ann Arbor, MI. Dion, P.-E. 2002. La religion des papyrus d’Éléphantine: Un reflet du Juda d’avant l’exil. In Kein Land für sich allein: Studien zum Kulturkontakt in Kanaan, Israel/ Palästina und Ebirnari für Manfred Weippert zum 65. Geburtstag, U. Hübner and E. A. Knauf (eds), 243–54. Göttingen. Feldman, L. 1993. Jew and Gentile in the ancient world. Princeton, NJ. Frankfurter, D. 1993. Elijah in Upper Egypt: The apocalypse of Elijah and early Christianity. Minneapolis, MN. Fuks, A. 1953. The Jewish revolt in Egypt (AD 115–117) in the light of the papyri. Aegyptus 33: 131–58. Gager, J. 1983. The origins of anti-Semitism: Attitudes toward Judaism in pagan and Christian Antiquity. Oxford. Galling, K. 1964. Studien zur Geschichte Israels im persischen Zeitalter. Tübingen. Gass, E. 1999. Der Passa-Papyrus—Mythos oder Realität? Biblische Notizen 99: 55–68. Goharghi, R. 1999. The land of Goshen in Egypt: The Hyksos. Bulletin on Biblical Studies 18 (1): 99–131. Gozzoli, R. 2006. The writing of history in ancient Egypt during the first millennium BCE: Trends and perspectives. London. Grabbe, L. 2008. Hecataeus of Abdera and the Jewish law: The question of authenticity. In Berührungspunkte: Studien zur Sozial- und Religionsgeschichte Israels und seiner Umwelt, I. Kottsieper, R. Schmitt and J. Wöhrle (eds), 613–26. Münster. Grelot, P. 1954. Études sur le ‘papyrus pascal’ d’Éléphantine. Vetus Testamentum 4: 349–84.

———. 1986. Elephantine: Araméens et Juifs en Égypte. Monde de la Bible 45: 32–35. Gruen, E. S. 1998. The use and abuse of the Exodus story. Jewish History 12 (1): 93–122. ———. 2008. The letter of Aristeas and the cultural context of the Septuagint. In Die Septuaginta: Texte, Kontexte, Lebenswelten, M. Karrer and W. Kraus (eds), 134–56. Tübingen. Hadas, M. (trans.) 1973. Aristeas to Philocrates: The letter of Aristeas. New York. Harker, A. 2008. Loyalty and dissidence in Roman Egypt: The case of the Acta Alexandrinorum. Cambridge. Hartog, F. 1996. Mémoire d’Ulisse: Récits sur la frontière en Grèce ancienne. Paris. Hermann, J. and F. Baumgärtel. 1923. Beiträge zur Entstehungsgeschichte der Septuaginta. Berlin. Jacobson, H. 2006. Artapanus Judaeus. Journal of Jewish Studies 57 (2): 210–21. Jacoby, F. 1954. Die Fragmente der griechischen Historiker 3: Geschichte von Städten und Völkern (Horographie und Ethnographie) C: Autoren über einzelne Länder. Leiden. Kahn, D. 2007. Judean auxiliaries in Egypt’s wars against Kush. Journal of the American Oriental Society 127: 507–16. Kasher, A. 1985. The Jews in Hellenistic and Roman Egypt: The struggle for equal rights. Tübingen. Kerkeslager, A. 1997. Maintaining Jewish identity in the Greek gymnasium: A ‘Jewish load’ in Corpus Papyrorum Judaicarum 3.519 (= P. SCHUB. 37 = P. BEROL. 13406). Journal for the Study of Judaism in the Persian, Hellenistic and Roman Period 28: 12–32. Koenen, L. 1968. Die Prophezeiungen des Töpfers. Zeitschrift für Papyrologie und Epigraphik 2: 178–209. ———. 2002. Die Apologie des Töpfers an König Amenophis oder das Töpferorakel. In Apokalyptik und Ägypten: Eine kritishce Analyse der relevanten Texte aus dem griechish-römischen Ägypten, A. Blasius and B. U. Schipper (eds), 139–87. Orientalia Lovaniensia Analecta 107. Leuven. Koskenniemi, E. 2002. Greeks, Egyptians and Jews in the fragments of Artapanus. Journal for the Study of the Pseudepigrapha 13: 17–31. Kottsieper, I. 2003. Die Religionspolitik der Achämeniden und die Juden von Elephantine. In Religion und Religionskontakte in Zeitalter der Achämeniden, R. Kratz (ed.), 150–78. Gütersloh. Kugler, R. 2005. Hearing the story of Moses in Ptolemaic Egypt: Artapanus accommodates the tradition. In The wisdom of Egypt: Jewish, early Christian and Gnostic essays in honour of Gerard P. Luttikhuizen, A. Hilhorst and G. H. van Kooten (eds), 67–80. Leiden. Lindberger, J. 2001. What ever happened to Vidranga? A Jewish liturgy of cursing from Elephantine. In The

EGYPT, EMPIRE AND JUDAISM

world of the Aramaeans 3: Studies in language and literature in honour of Paul-Eugène Dion, P. M. M. Daviau, J. Wevers and M. Weigl (eds), 3: 134–57. Sheffield. Martin, M. 2006. Observations on the paintings of the Exodus Chapel, Bagawat Necropolis, Kharga Oasis, Egypt. In Byzantine narrative: Papers in honour of Roger Scott, J. Burke et al. (eds), 233–57. Melbourne. Mendels, D. 1983. Hecataeus of Abdera and a Jewish ‘Patrios Politeia’ of the Persian period (Diodorus Siculus XL 3). Zeitschrift für die alttestamentliche Wissenschaft 95: 96–110. Modrzejewski, J. M. 1995. The Jews of Egypt: From Rameses II to Emperor Hadrian. Translated by R. Cornman. Edinburgh. Moyer, I. S. 2011. Egypt and the limits of Hellenism. Cambridge. Murnane, W. J. 2001. New Kingdom: Overview. In The Oxford encyclopedia of ancient Egypt 2, D. B. Redford (ed.), 519–25. Oxford. Musurillo, H. A. (ed.) 1954. Acts of the pagan martyrs 1. Oxford. Nirenberg, D. 2013. Anti-Judaism: The western tradition. New York. Pelletier, A. (ed.) 1962. La lettre d’Aristée à Philocrate. Paris. Porten, B. 1968. Archives from Elephantine: The life of an ancient Jewish military colony. Berkeley, CA. ———. 2003. Settlement of the Jews at Elephantine and the Arameans at Syene. In Judah and the Judeans in the Neo-Babylonian period, O. Lipschits and J. Blenkinsopp (eds), 451–70. Winona Lake, IN. ———. 2011. The Elephantine Papyri in English: Three millennia of cross-cultural continuity and change. Atlanta, GA. Porten, B. and A. Yardeni. 1986–89. Textbook of Aramaic documents from ancient Egypt. Winona Lake, IN; Jerusalem. Pucci Ben Zeev, M. 1981. La rivolta ebraica al tempo di Traiano. Pisa. ———. 1982. La rivolta ebraica in Egitto (115–117 d.C.) nella storiografia antica. Aegyptus 62: 195–217. ———. 1983. Corpus Papyrorum Judaicarum II 158, 435 e la rivolta ebraica al tempo di Traiano. Zeitschrift für Papyrologie und Epigraphik 51: 95–103. ———. 1993. The reliability of Josephus Flavius: The case of Hecataeus’ and Manetho’s accounts of Jews and Judaism: Fifteen years of contemporary research (1974– 1990). Journal for the Study of Judaism in the Persian, Hellenistic and Roman Period 24 (2): 215–34. Rajak, T. 2008. Translating the Septuagint for Ptolemy’s library: Myth or history? In Die Septuaginta: Texte, Kontexte, Lebenswelten, M. Karrer and W. Kraus (eds), 176–93. Tübingen.

323

Raspe, L. 1998. Manetho on the Exodus: A reappraisal. Jewish Studies Quarterly 5: 124–55. Redford, D. 1986. Pharaonic king-lists, annals and daybooks: A contribution to the study of the Egyptian sense of history. Mississauga, ON. Ritner, R. K. 2003. The Prophecy of the Lamb. In The literature of ancient Egypt, W. K. Simpson (ed.), 445–49. 3rd ed. New Haven, CT. Schäfer, P. 1997. Judeophobia: Attitudes toward the Jews in the ancient world. Cambridge. Smallwood, E. M. 1981. Jews under Roman rule from Pompey to Diocletian: A study in political relations. Leiden. Spalinger, A. J. 2001. Chronology and periodization. In The Oxford encyclopedia of ancient Egypt 2, D. B. Redford (ed.), 264–68. Oxford. Sternberg-El Hotabi, H. 2003. Die persische Herrschaft in Ägypten. In Religion und Religionskontakte in Zeitalter der Achämeniden, R. Kratz (ed.), 111–49. Gütersloh. Tcherikover, A. and A. Fuks. 1957. Corpus Papyrorum Judaicarum. Cambridge. Van der Horst, P. W. (ed. and trans.). 1984. Chaeremon: Egyptian priest and Stoic philosopher. Leiden. van der Kooij, A. 2008. The promulgation of the Pentateuch in Greek according to the letter of Aristeas. In Scripture in transition: Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in honour of Raija Sollamo, A.Voitila and J. Jokiranta (eds), 179–91. Leiden. van Henten, J. W. and Abusch, R. 1996. The Jews as Typhonians and Josephus’ strategy of refutation in Contra Apionem. In Josephus’ Contra Apionem: Studies in its character and context with a Latin concordance to the portion missing in Greek, L. H. Feldman and J. R. Levison (eds), 271–309. Leiden; New York; Cologne. Wachsmuth, C. (ed.). 1897. Ioannis Laurentii Lydi Liber de ostentis: Et calendaria Graeca omnia. 2nd ed. Lipsiae. Waddell, W. G. 1948. Manetho with an English translation. Loeb Classical Library. Cambridge, MA. Walton, F. R. and R. M. Geer. 1967. Diodorus of Sicily in twelve volumes. Loeb Classical Library 423. London. Wasserstein, A. and D. J. Wasserstein. 2006. The legend of the Septuagint, from Classical Antiquity to today. Cambridge. Wright III, B. G. 2015. The Letter of Aristeas: ‘Aristeas to Philocrates’ or ‘On the Translation of the Law of the Jews’. Berlin; Boston, MA. Yamauchi, E. 1996. Cambyses in Egypt. In ‘Go to the land I will show you’: Studies in honor of Dwight W. Young, J. Coleson and V. Matthews (eds), 371–92. Winona Lake, IN. Zauzich, K.-Th. (ed.). 1983. Das Lamm des Bokchoris. In Papyrus Erzherzog Rainer. Festschrift zum 100-jährigen Bestehen der Papyrussammlung der Österreichischen Nationalbibliothek. 2 vols. 1: 165–74. Vienna.

VISIBLE IDENTITIES: IN SEARCH OF EGYPT’S JEWS IN EARLY ISLAMIC EGYPT1 Petra SIJPESTEIJN

When the Arabs arrived in Egypt in the mid-seventh century AD, they reportedly encountered substantial Jewish communities in Alexandria and at the Roman fortress of Babylon at the point where the Nile enters the Delta. Descriptions of synagogues, courts and religious schools suggest a well-organized and functioning religious community. References to Jewish Egyptians in documentary sources, however—Greek, Coptic and Arabic papyri and inscriptions—remain scanty in the first two centuries of Muslim rule in Egypt. To quote Maged Mikhail’s recent historical study of the period, ‘Richly documented throughout Late Antiquity and especially from the late tenth century on, Egyptian Jews are marginally attested in the sources of the centuries under investigation [i.e. the early post-conquest centuries]. This prevents even a cursory survey’ (Mikhail 2014, 9). Only from the ninth century AD do Jewish Egyptians become more visible in the documents. How to explain the absence of Jews in the documentary record from the first two centuries of Arab rule in Egypt? What caused the increase in attestations from the ninth century onwards? Does this signify an increase in the Jewish Egyptian population or are other factors at play? In this paper I will argue that the introduction of ethnic/religious identity markers in the written record in the ninth century relates to demographic changes and consequent socio-political reorientations rather than a numerical increase only. Sources Arabic literary sources reporting on Egypt under the first centuries of Muslim rule abound. Chronicles, geographical dictionaries and local histories provide information on historical events and on the human and natural environment, starting with the conquest and even

1

This paper will also be published under the same title in Israel in Egypt, ed. S. Pearce and A. Salvasen, Ancient Judaism and Early Christianity (Leiden). I would like to thank Eve Krakowski,

pre-dating it. The earliest and doubtlessly richest source is Ibn ꜥAbd al-Ḥakam’s (d. 870) Conquest of Egypt (Futūḥ Miṣr). As Hugh Kennedy (1998) has pointed out, this text is written to record the achievements of the Arabs who had conquered and subsequently ruled Egypt, but who had lost their privileged position to Persian-Turkish Muslims migrants to Egypt from the eastern part of the empire from the 830s onwards. In this sense it is a valuable source not only for the earliest period under examination, but especially for the demographic and consequent social changes taking place in the ninth century. From the tenth century comes another rich source, al-Kindī’s (d. 961) Book of governors and judges (Kitāb al-wulāt wa kitāb al-quḍāt), which was also produced in Egypt. Later Egyptian sources, such as al-Maqrīzī’s (d. 1442) account of the topography and history of Egypt (al-Mawā’iz wa-l-iꜥtibār fī dhikr al-khiṭaṭ wa-l-āthār) and Ibn Duqmāq’s (d. 1407) description of Cairo (Kitāb al-intiṣār li-wāsiṭat ꜥiqd al-amṣāri), also preserve useful earlier material despite their late date of composition. Geographical and historical texts written outside the province contain useful information as well. Documentary sources such as papyri and inscriptions from early Islamic Egypt are especially bountiful and useful sources for examining issues of self-identification amongst Egypt’s population. There are, however, some limitations to this material. Most of the papyri from the Islamic period are found in a limited number of rural find-spots outside the main places of residence. A substantial number of Arabic papyri and ostraca dating to the ninth century were discovered in Fusṭāṭ (Denoix 1986; Gayraud, Björnesjö and Denoix 1986, 25). Most papyri do not have an absolute date and can only be roughly dated on the basis of their palaeography. Moreover, because most papyri are found on

Susanna de Vries, Nicholas de Lange, Adam Silverstein and Jelle Bruning for their help in providing references and feedback. Any remaining mistakes are entirely my own.

326

P. SIJPESTEIJN

disposal heaps, they lack an archaeological context. The individuals mentioned in the documents mostly cannot be identified. Owing to the haphazard nature of these papyrological remains, moreover, documents appear mostly in isolation without other texts related to the same events or individuals, offering only one side of a story. Documents pertaining to Egypt’s Jewish community were partially produced by (and for) Jewish people themselves, but documents produced by non-Jews also purposely or unintentionally record information. The most obvious are ethnic/religious markers (Gr. hebre(a)ios, ioudaios; Ar. yahūdī, isrā’īlī) which can be voluntarily used or imposed by others to denote individuals or groups. References to religious or community buildings and rituals, celebrations or any object or transaction related to the organization thereof are attested as well. Religious and other literary and semiliterary texts such as amulets might be considered at first sight to be community-specific, but can of course have been consumed by other groups as well. Names are notoriously difficult identifiers (on all these issues, cf. Ilan 2016). As for the languages in use in Islamic Egypt, some changes occurred under the Arabs, while continuity prevailed in other ways. When the Arabs arrived in Egypt in 639, Arabic was immediately added to the languages already in use, namely Greek and Coptic.2 Greek remained the language most commonly used in the administration into the ninth century. The use of Coptic increased under the Arabs, as it was also used to write administrative documents for the first time. Coptic and Syriac had already become more important in the Near East, probably as a form of anti-Chalcedonian resistance, but no Christian Syriac or Palestinian Aramaic papyri are attested from Byzantine Egypt.3 The use of Hebrew (and Aramaic) in Egypt increased from the fifth century onwards, but the impact of the supposed introduction of Hebrew as a liturgical language in the sixth century has left few traces in the documentary record (Sirat 1985; Stroumsa 2012;

Smelik 2012; Ilan 2016). Both Syriac and Judaeo-Arabic papyri are attested from the ninth century onwards (see below).

2

3

From Late Roman Egypt date also a small number of Hebrew and Aramaic papyri (Sirat 1985; Ilan 2016) as well as a cache of very important but particular Syriac Manichaean texts found in the Dakhla oasis (Gardner 1996, 161–76).

The Jewish population of Egypt at the time of the Arab conquest Let us begin with an overview of what the historical, literary and documentary sources say about Egypt’s Jewish population on the eve of the Arab conquest. Documentary and literary sources suggest that Byzantine Egypt was home to a significant, religiously active and transregionally connected Jewish community residing both in towns and in the countryside (Ilan 2016). In spite of the reports of repeated expulsions of and attacks on the Jewish population of Alexandria by the Romans, a continuing Jewish presence in the city up to the Arab conquest is confirmed by documentary and literary sources (Starr 1935, 281, 286; Stroumsa 2012). Presumably such expulsions were actually not applied to the whole Jewish population of the city, but rather imposed symbolically only on the Jewish leadership. There are suggestions that the Jewish population in fact increased in the early seventh century when refugees from Palestine, fleeing the Persian armies, settled in Alexandria (St John the Almsgiver 1980–81; Ibn al-Biṭrīq 1985). Other sources point to a continued Jewish presence in the Egyptian countryside. The Babylonian Talmud contains a discussion on the permissibility of reading the book of Esther in Coptic (Megillah 18a). Probably related to the sixth-century debate within the Jewish community over whether scriptures should be read in public in Hebrew only, the reference points to the presence of a Jewish community in the Egyptian countryside which was in touch with the Babylonian religious authorities (Smelik 2012). This Jewish community left its traces in Greek papyri, where members are identified as ioudaios or hebre(a)ios. Their community buildings, such as a ‘communal synagogue’ in an account of one of the Appion estates in the Fayyūm dated after AD 566

For the small group of Syriac Manichaean papyri found in Dakhla, see above n. 2.

VISIBLE IDENTITIES

(P.Oxy. 55.3805), are mentioned as well. Hebrew and Aramaic papyri record Jewish marriage contracts, biblical, liturgical and magical texts (Sirat 1985; Ilan 2016). All in all, the documentary and literary evidence points to an intensified Jewish presence in Egypt from the fifth century AD onwards (Ilan 2016, 216–17). The presence of a significant Jewish population in Egypt is confirmed by descriptions of the Arab conquest. These accounts concentrate on the cities, where the Arabs encountered a substantial and active Jewish population. It is reported that when the Arabs entered Alexandria in AD 642, there were forty or seventy thousand Jews living there (Ibn Duqmāq 1893; Ibn ꜥAbd al-Ḥakam 1922, 81; Ibn al-Biṭrīq 1985, §285, p. 147). Egypt’s conqueror and first governor, ꜥAmr b. ꜥĀṣ (d. 664), subsequently determined in the treaty concluded in 641 with the city’s patriarch that the Jews could remain in the city (John of Nikiu 1916, 194). Several synagogues were located in and around the Roman fortress of Babylon when the Arabs founded Fusṭāṭ on the same spot, at the entry to the Nile Delta (Stiefel 2014, 27; Dridi 2015). Amongst the eleven functioning synagogues in the Egyptian capital that al-Maqrīzī (d. 1442) mentions, he identifies four explicitly as dating back to the first or second century AD (al-Maqrīzī 2002–2003, 4, 939). While such a date for continuously functioning synagogues might seem unlikely, it does suggest that these religious buildings were considered to be very old by the time al-Maqrīzī described them. In fact, religious buildings, mosques as well as churches and synagogues, had to be built and expanded to serve the growing population of Fusṭāṭ. Immigrants flowed into the capital from Egypt and beyond (Dridi 2015). That this growing population residing in the capital contained Jews is clear from other sources as well. The chief judge of Fusṭāṭ, Khayr b. Nu‘aym (in office 737–44), announced in 738–39 that from then on evidence of Jews and Christians in cases against their co-religionists was acceptable in court (al-Kindī 1912, 351). While there is thus a strong indication that there was a significant Jewish population in Egypt during the first two centuries of Muslim rule, their presence in the documentary sources is very slim indeed. Greek papyri attest individuals identified as hebre(a)ios and, to a much lesser extent, ioudaios (Ilan 2016, 218–19), all but one produced by the Arab-Muslim administration.

327

A Greek ostracon records contributions for the military provisions (annona) by different population groups in Edfu, amongst whom the hebraioi are mentioned. This text is dated to the very end of the Byzantine or early Arab period (Gascou 1978). A poll-tax receipt for ‘Jakob hebraios’ from the Fayyūm is dated 645 (SPP 8.741). Two fiscal lists dating to the eighth century from the Fayyūm include men identified as hebraios (Levi and Abraham in P.Sijp. 36; Moses, Jacob and Orbi in SPP 10.182). The only non-administrative text is a seventh-century receipt for rent made out to the lady ‘Senufe hebraia’ (P.Brook 15). Only one Arabic document from the second half of the eighth century uses the qualifier ‘the Jew’ or ‘Jewish’. Three generations of Jewish men, Yaꜥqūb, his son Yūsuf and his grandson, were all identified as al-yahūdī in an Arabic sale-contract for a mule dated 761–62 (Hanafi 2004). While not many in number, the documents show a rather wide geographical spread throughout Egypt’s countryside. Interestingly, this period also yields only one Arabic document that uses the descriptive ‘the Christian’ or ‘Christian’. In an eighth-century list of goods shipped for different individuals a Yasūꜥ al-naṣrānī and a Mūsā al-naṣrānī appear (Khan 1992, no. 7). The adjective ‘Christian’ appears occasionally in seventh- and eighthcentury Greek and Coptic legal documents referring to Christian institutions such as monasteries, a Christian legal system or rulings (cf. Papaconstantinou 2009). In a Greek letter dating to the second half of the seventh century ‘Christian slaves’ are requested to be sent from Edfu to the Arab authorities in Fusṭāṭ (P.Apoll. 51). The description of the governor’s retinue as ‘Arab and Christian officials’ in an early eighth-century Greek papyrus is probably a reference to the linguistic–administrative domains that these two groups of administrators occupied. The officials are the recipients of the food (such as dates, sheep, vinegar, oil, boiled wine and poultry) and firewood for which the maintenance costs (Ar. rizq; Gr. dapanē) are demanded from the community of Aphroditō/Ishqūh (P.Lond. 4.1375). Famously, the Arab-Muslim conquerors are not religiously identified, but rather referred to in Greek as moagaritai, probably a rendering of the Arabic word muhajirūn, migrants. Greek and Coptic papyri attest the term moumin, a rendering of Arabic mu’minūn, ‘believers’, but only in the combination amirmoumin, which refers to the Arab administration as a whole and should not be understood as ‘prince of believers’, its

328

P. SIJPESTEIJN

literal translation. Sarakēnos is also occasionally used continuing pre-Islamic practice, as is arabos, both geographic markers.4 Two further documents, one in Coptic and the other in Arabic, contain names that unambiguously refer to Jews, but without identifying them explicitly as Jewish. A seventh–eighth-century Coptic document concerning the leasing of land from Bawīṭ mentions Eliezer son of Toual (?) as the lessee (Delattre, Pilette and Vanthieghem 2015, no. 3). Finally, in a letter dating to the last quarter of the eighth century an individual named Judah (Yahūdā) appears (Jahn 1937, no. 10 republished as Diem 2018, no. 2). Many more Jewish Egyptians are presumably hidden in the documents behind less unambiguous names and markers. The small number of explicit attestations does not prove that there were hardly any Jews in Egypt at this time, but rather that identifying them in the written record as such, both as a group and as individuals, was not deemed necessary during the first two centuries of Muslim rule. To understand when and why this changed in subsequent centuries, we should examine the relation between the Arab ruling authorities and their Egyptian subjects. From people of the land to Copts In the period immediately following the Arab conquest the distinction that mattered was that between conquerors and conquered, Arabs and Egyptians, soldiers and peasants. The Egyptians who had remained in possession of their properties after the Arab takeover worked the land to raise the taxes that paid the Arab soldiers’ stipends and provisions. Arabs and Egyptians occupied two distinctive hierarchical categories, and they also remained separated spatially, with the Arabs confined to the garrisons of Fusṭāṭ and Alexandria. The distinction is reflected in the terms used to describe the two groups: soldiers (jund) for the Arabs versus ‘people of the land’ (ahl al-arḍ) for the Egyptians. The governor Qurra b. Sharīk (in office 709–15) uses these terms in his letters to the pagarch Basileios of the district of Aphroditō/Ishqūh in Upper Egypt in the years 709–10 (e.g. Grohmann 1938a, nos 148, 159; Becker

4

See for example the expression kata sarakēnous or kat’arabas to indicate the hijra calendar in Greek and Coptic papyri (Bagnall and Worp 2004, 300). For arabes, see P.Lond. IV 1375.7, dated 710, provenance Aphroditō/Ishqūh. For arabikos,

1906, nos 1–3; Abbott 1938, no. 2). Another term used to describe the Arab rulers is ‘people of Egypt’ (ahl miṣr) (Rāġib 1981, no. 1; Diem 1984, no. 1). The name miṣr, introduced by the Arabs, was the current identification for the province in Arabia. It appears in Greek (masr) on Umayyad copper coins. This expression is also used to qualify the chief judge of Fusṭāṭ (qāḍī ahl miṣr) in a papyrus letter from the Egyptian governor to the Nubian king dated 758. This same text refers to the Christian Nubians as a religious community (ahl millatika) (Hinds and Sakkout 1981). A new phase in Arab–Egyptian relations, which had an impact on the use of descriptive terms attested in the papyri, starts in the mid-eighth century. This is when Arab-Muslim expansion into Egypt’s countryside (with Arab-Muslims taking possession of agricultural lands and settling outside the garrisons) started in earnest (Sijpesteijn 2013, 163–72). Arabicization and Islamicization, albeit still limited at this time, increased as a result. This in turn impacted the relationship between rulers and ruled and how the two groups were identified in the written record. The changed vocabulary first occurs in documents related to taxation. A document dated c. 735 enjoins Muslims (ahl al-Islām) in the Fayyūm to pay the alms-tax (ṣadaqa, zakāt) (Sijpesteijn 2013, no. 8), while the same term appears in a letter from the governor of Egypt to the king of Nubia dating to 758 (Hinds and Sakkout 1981). Later in the eighth century, documents attest to the use of the two categories ‘Muslims’ and ‘people of the covenant of protection’, i.e., non-Muslims (ahl aldhimma) as administrative entities (dated 789 by Grohmann 1952a, 132–33; dated 790 by Diem 1984, no. 7). Copts (qibṭ) are opposed to Muslims in another eighth-century papyrus related to agricultural taxes (Sijpesteijn 2007, no. 1). Fiscal categories are also specified as ‘Coptic’ (baqṭ al-qibṭ; fuḍūl al-qibṭ) when referring to dues imposed on the indigenous population (Grohmann 1938b, nos 2 and 4). Starting in the late ninth century, Arabic fiscal and legal documents make frequent use of the expression ‘months of the Copts’ (shuhūr al-qibṭ) in the date or ‘calculation of the Copts’ (ꜥadad al-qibṭ) for the kind of payment requested. These terms are, however, used to indicate population

see P.Lond. IV 1447.140 and 190, dating between 685 and 705; 1434.229 and 1435.84, both dated 716, provenance of all is Aphroditō/Ishqūh.

VISIBLE IDENTITIES

groups as a whole, not individuals, and seem to be used to identify non-Muslim Egyptian tax-paying peasants as opposed to Muslim ones, or even indigenous Egyptian practice as opposed to newly introduced (Arab) customs, without distinguishing between Christian and Jewish Egyptians. In fact the term qibṭ does not seem to have had a religious connotation at all. Copt, the Arab vulgarization of the Greek Aegyptus was, after all, a generic term applicable to all indigenous inhabitants of the province at that juncture. Changes in the ninth century Processes of Islamicization, Arabicization and migration in the ninth century altered Egypt’s population composition in ethnic and religious terms. While the rate of conversion seems to have continued to be negligible, the impact of the movement of populations across the empire and within Egypt led to the rise of communal–religious designations. The late eighth and early ninth century witnessed great changes in settlement patterns within Egypt and migration into the province from the eastern part of the empire. The breakdown of central control gave new population groups a chance to expand their presence in the province, dramatically changing its socio-ethnic organization. Greater interaction led on the one hand to a levelling of differences, but on the other, and partially in reaction to it, to greater diversification and sharper demarcations between groups with a direct impact on how members of those communities self-identified and were classified. The documentary and literary record shows how the application of identity markers used by and for individuals changed as a result. From the late eighth century onwards troops composed of Persian–Turkish soldiers originating in the eastern Islamic empire had found their way to Egypt. This development introduced by governors and military leaders sent to rule the province, was reinforced by political developments at the end of the eighth and early ninth centuries. Representatives of the ruling military elite from the caliphal capital in Iraq, who had been settling in Egypt since the late eighth century, had taken over land holdings and offices at the expense of Egypt’s wujūh, the Arab families who had conquered and first settled Egypt in the seventh century. From the last decade of the eighth century the country had effectively been in the control of a Persian–Turkish military elite initially based in Iraq. Order was restored by the Abbasid general ῾Abd Allāh b. Tāhir (d. 845), but the

329

measures he took during his short presence in the province (825 or 826–27) decisively reduced the power of the old Arab leaders. Eastern appointees replaced local Arabs in such crucial positions as the police and heads of the fiscal office (al-Kindī 1912, 183–84). With the administrators arrived personnel who introduced new administrative terminology and practices (Sijpesteijn 2011). In other words, support and co-operation from the local wujūh was no longer necessary, while a new class of administrators and bureaucrats was put in control of the province (Kennedy 1998). The wujūh turned to other ways to make their presence felt in the province. It is exactly when their role in the administration diminishes that Arab-Muslim Egyptians appear more prominently in mediation processes of informal conflict resolution, as can be observed in the papyri. In this arena, which fell outside the official legal and administrative institutional structures, the Arabs still wielded substantial power based on their standing amongst the local population (Sijpesteijn 2020). In other areas too, the wujūh’s attempts at self-legitimization can be observed. The first local history of Egypt, the Futūḥ Miṣr, is a conscious attempt to create the history and emphasize the importance of the Arabs who had settled in and ruled Egypt from the seventh century (Kennedy 1998; Omar 2013). The writing of a history for the wujūh connecting them to the Egyptian land coincided with the development of an Egyptian Muslim identity under the influence of the processes of Islamicization and Arabicization (Sijpesteijn 2011; Omar 2013). Arab-Muslim settlement in the Egyptian countryside starting in the mid-eighth century had indeed led to increased Arabicization and, to a lesser extent, Islamicization (Sijpesteijn 2013, 163–72). The use of Arabic had been gradually increasing to the point that no securely dated Greek papyri can be identified after the first half of the ninth century, while the use of Coptic diminished as well. The number of Arabic papyri, on the other hand, dramatically increases in the ninth century, suggesting that more Arabic-speakers were settled in the Egyptian countryside where the papyri were found and that more Egyptians in the countryside used Arabic for their written communications (Sijpesteijn 2013, 235–36). While Egypt’s Muslim population would continue to form a numerical minority for several centuries, intensified contact resulted in Islamicization in cultural and religious terms as well (Brett 2005; Mikhail 2014).

330

P. SIJPESTEIJN

Increased interaction between Arabs and Egyptians lay at the base of these developments and was stimulated by it. The result was cultural and social convergence. The opposing categories of Arab-Muslim ruling authorities and local Egyptians could even give way to a shared socio-economic and political agenda. In the eighth century Egyptians and Arabs together rose up against an increasingly intrusive central state (Lev 2012). Commercial transactions in which Muslims, Christians and Jews work together are recorded in Arabic, Coptic and Judaeo-Arabic letters. The loss of distinctive communal identities also led to disaffection and protest. When the ninth/tenth-century author of the Apocalypse of Samuel of Qalamun warned against the use of the Arabic language and other forms of Arabicization amongst the Egyptian Christians, he was fighting a lost battle.5 He was, however, expressing the discomfort of Egyptian Christians about the increasingly close interaction between the Christians and Muslims of Egypt (Papaconstantinou 2007). The author called for a strict adherence to Coptic cultural expressions and for a redefining of group identity on the basis of religious and ethnic terms. Indeed, the first occurrence of the ‘era of the martyrs’ in Egypt is found in a Coptic dating clause from 861/62. This name probably arrived in Egypt from Nubia, where it appears some seventy-five years earlier, and became the dating method of choice in Egyptian Christian texts (Bagnall and Worp 2004, 67). It is also in the ninth century that the restrictive measures concerning clothing, hair styles and other external markers distinguishing Muslims from non-Muslims were most probably systematized as the shurūṭ ꜥUmar (Cohen 1999; Yarbrough 2016; pace Levy-Rubin 2016). In other words, at a time characterized by social unrest and political turmoil, the call for clear demarcations and boundaries, ostensibly to preserve existing identities, created a new norm of religious identification expressed in the literary and documentary record. The results of these changes are obvious in the documentary record. Individuals appear in the papyri more

frequently with a distinctively ethnic or religious nisba or name. In some cases inter-confessional distinctions might have been expressed as well. This can be seen for example in the different terms—al-yahūdī (David-Weill 1951–52; Khan 1992, no. 33; Vanthieghem 2014, no. 2); and al-isrā’īlī (Grohmann, 1952b, no. 248)—used to identify Jewish Egyptians.6 These documents all originate in an administrative– fiscal context. A ninth-century quittance records the payment of ten dinars by Sha‘ūn al-Yahūdī to a certain Yazīd b. Yazīd (the unpublished Mich.Pap.B 111 from the Cambridge University Library).7 In a ninthcentury letter an individual is identified as ‘the son of the Christian woman’ (ibn al-masīḥiyya) (Diem 1997, no. 61). While the previous texts can be considered the result of centralized directives, other papyri were produced by members of the community itself. The name Isrā’īl (e.g. Grohmann 1943, no. 71; Grohmann 1952b, no. 248) occurs for the first time in ninth-century Arabic papyri. An unedited ninth-century letter mentions a certain Rabbī Yahūdā, the likes of whom the sender of the letter asks God to increase amongst the ‘people of Israel’ (banū Isrā’īl).8 The name Yahūdā also appears twice in a list of quittances by individuals whose names suggest they were all Jewish (the unpublished AP 11325 from the Austrian National Library).9 In addition a small number of Judaeo-Arabic papyri is preserved (Sirat 1985; Blau and Hopkins 1987; Vanthieghem 2014, no. 2). The documents do not bear dates, but on the basis of palaeography and the expressions used in them can be dated to the ninth century. Their origin, as far as it is known, is representative for the papyrological record in general, showing that Jewish communities were present throughout Egypt. Written in Arabic but in Hebrew characters, these documents clearly originate in a Jewish milieu. They can be considered to continue the Hebrew and Aramaic writings of earlier centuries (Sirat 1985; Ilan 2016), but the large number of Judaeo-Arabic commercial letters shows a shift from a prevalent legal and religious to the private and

5

8

6

7

For the arguments that this text is a product of ninth/tenth-century developments, see Hoyland 1997, 285–89. The different references to al-rūmī all refer to Byzantine slaves, rather than Melkite Christians (Diem 1997, no. 41). I would like to thank Jelle Bruning for bringing this text to my attention.

9

I am preparing the text for publication. The name is also attested in an eighth-century papyrus discussed above (Jahn 1937, no. 10 = Diem 2018, no. 2), which seems to be related I would like to thank Jelle Bruning for bringing this text to my attention.

VISIBLE IDENTITIES

commercial sphere. Besides the Hebrew script the documents do not generally contain specific references to a ‘Jewish context’, using the same religious blessings and formulae as contemporary non-Jewish letters. One letter, however, opens with the expression, ‘In Thy Name, O Merciful One’, instead of the usual ‘In the name of God, the Compassionate, the Merciful’ (Blau and Hopkins 1987, 111). The Judeo-Arabic letters do represent, however, features of early phonetic JudeoArabic spelling which can, in a rudimentary form, also be observed in a group of three eighth-century Arabic papyri produced in a Jewish context (Diem 2018, 13–15 and the unpublished Arabic papyrus mentioned above n. 9). The use of a distinctive language associated with a particular religious group can also be observed in the ninth-century Syrian orthodox community in Egypt. The ninth century witnessed important migration of Syrian orthodox Christians from Iraq and Syria to Egypt (Fiey 1963; Hunt 2003; Shaked 2010). Forming a successful mercantile community in the capital Fusṭāṭ, they also fostered Syriac as a written language, albeit seemingly only in the literary domain. The library of the Dayr al-Sūryān monastery in the Wadī al-Naṭrūn was founded with Syriac manuscripts brought from Iraq and Syria. While there had always been Syrian orthodox Christians temporarily and permanently settled in Egypt, only from the ninth century onwards were new Syriac texts composed in monasteries throughout Egypt (Van Rompay 2009). The trend to make religious affiliations explicit through identity markers in writing continued in the tenth century. The appearance of the name ꜥAbd al-Masīḥ in large numbers in the Arabic papyri shows the high level of acculturation amongst Egyptian Christians, who took over Arab-Muslim naming patterns (ꜥAbd Allāh, etc.), but added a Christian twist (Grohmann 1938a, no. 195; Diem 2008, nos 18 and 21). Jewish-named authors are known from this period for the first time since the Hellenistic period (Rustow 2008, 36). The famous theologian and Judaeo-Arabic translator Saꜥadya b. Yoseph Gaon (d. 942) and the physician and philosopher Isaac Israeli (d. 955), although flourishing outside Egypt, in Baghdad and Qayrawan respectively, both received their basic Jewish theological education in the province (Fenton 1995). Clearly, a Jewish intellectual and religious infrastructure had been in place in Egypt to train scholars before, but only now were Gaon and Israeli brought to the forefront as representatives of the Jewish community. The desire to draw boundaries between

331

religious communities also gave rise to the polemic and apologetic texts that flourished in the tenth century. Larger in numbers or more visible? When the Persian geographer Ibn Khurdādhbih (d. 912) described the position of the Jews of Alexandria he wrote that the community had had an inferior status in the pre-Islamic period. Indeed, Egyptian Jews had suffered regular persecutions and expulsions under the Romans, with some of the worst maltreatment after the Byzantine restoration of power in 628 as a punishment for Jewish collaboration with the Persians (Stillman 1997, 199). Orders to leave, however, were probably only seriously imposed on the Jewish leadership and a Jewish population continued to educate and practise their religion in Egypt into the Islamic period. While the numbers of Egyptian Jews encountered by the Arabs according to the chronicles might have to be taken with a grain of salt, references to synagogues and educational institutions indicate that the community was active and numerous. Documentary references in Hebrew, Greek and Arabic also show that a Jewish population existed throughout Egypt. Egypt’s demographic composition changed drastically in the ninth and tenth centuries under the influence of Islamicization and Arabicization, as well as changed settlement patterns and immigration from the eastern Islamic empire. This brought about deep changes in Egypt’s social organization, resulting in greater diversification between ethnic and religious groups. In the changed social landscape, it became common to express group and individual identities religiously, ethnically and linguistically. An increased awareness of, and necessity to document, religious and ethnic demarcations affected all sectarian groups in Egypt—Muslims, Christians and Jews—and manifested itself in new identity markers following similar patterns used in contemporary documents. Scripts (Hebrew) and languages (Syriac, Arabic) were applied in new contexts and on a larger scale by Egyptian communities to demarcate communal boundaries. Nisbas and names referring to the ethnic and religious identity of individuals were used more frequently or appeared for the first time. Egypt’s Jews participated in these processes. JudaeoArabic commercial letters and other documents appear for the first time in the ninth century. Nisbas and other identity markers appear more frequently from the ninth/

332

P. SIJPESTEIJN

tenth century onwards in the documents. The increased visibility of Egypt’s Jews in the documentary record thus reflects a changed attitude towards the use of religious and ethnic markers in written texts in ninth–tenth century Egypt. While no substantial numerical change seems to have occurred in Egypt’s Jewish population before the substantial influx of Jews (and Christians) under the Fatimids, migrations from Iraq in this period might have had an impact nevertheless (Ashtor 1972; Rustow 2008). As discussed above, Syrian orthodox populations from Takrit in Iraq moved to Egypt, while soldiers and administrative personnel also moved from Baghdad and Khurasan to Egypt. In 882 a group of recently arrived Iraqi immigrés acquired a building from the Coptic patriarch for use as a synagogue (Stiefel 2014, 27). Amongst the earliest documents preserved in the Genizah of the Ben Ezra synagogue in Fusṭāṭ are indeed some marriage and divorce contracts drawn up in Iraq and northern Syria, presumably brought to Egypt by Jewish immigrants (OlszowySchlanger 2006). Egypt’s Jewish population continued to utilize Hebrew and Aramaic, used and compiled biblical and

liturgical texts, and educated its youth in religious matters in the late Roman and early Islamic period. While occasionally using or being identified by religious and ethnic markers, the documentary record does not identify Egypt’s Jews clearly. The ninth/tenth-century expressions of their religious identity, on the other hand, increase and operate in new domains. When Ibn Khurdādhbih described the position of Egypt’s Jews in his lifetime positively compared to earlier times, he might have been influenced similarly by the augmented visibility and consequently increased presence of Jewish Egyptians.

Bibliography

Brett, M. 2005. Population and conversion to Islam in Egypt in the Medieval period. In Egypt and Syria in the Fatimid, Ayyubid and Mamluk eras IV, U. Vermeulen and J. Van Steenbergen (eds), 1–32. Leuven. Cohen, M. R. 1999. What was the pact of ῾Umar? A literaryhistorical study. Jerusalem Studies in Arabic and Islam 23: 100–57. David-Weill, J. 1951–52. Un papyrus inédit du Musée du Louvre. Semitica 4: 67–71. Delattre, A., P. Pilette and N. Vanthieghem. 2015. Papyrus coptes de la Pierpont Morgan Library I: Cinq documents du monastère de Baouît. Journal of Coptic Studies 17: 33–53. Denoix, S. 1986. Les ostraca de Isṭabl ‘Antar, 1985. Annales Islamologiques 22: 27–33. Diem, W. 1984. Einige frühe amtliche Urkunden aus der Sammlung Papyrus Erzherzog Rainer (Wien). Le Muséon 97: 109–158. ———. 1997. Arabische Briefe des 7. bis 13. Jahrhunderts aus den Staatlichen Museen Berlin. Wiesbaden. ———. 2008. Arabische Steuerquittungen des 8. bis 11. Jahrhunderts aus der Heidelberger Papyrussammlung und anderen Sammlungen. Wiesbaden.

All papyrological abbreviations are cited throughout the chapter in accordance with the conventions established in J. F. Oates, R. S. Bagnall, S. J. Clackson, A. A. O’Brien, J. D. Sosin, T. G. Wilfong and K. A. Worp, Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed 13 Nov. 2018) and P. M. Sijpesteijn, J. F. Oates, A. Kaplony, E. M. Youssef-Grob and D. Potthast, Checklist of Arabic documents, http://www.naherosten.lmu.de/isapchecklist (last accessed 13 Nov. 2018). Abbott, N. 1938. The Kurrah papyri from Aphrodito in the Oriental Institute. Chicago, IL. Ashtor, E. 1972. Un mouvement migratoire au haut Moyen Âge: Migration de l’Irak vers les pays méditerranées. Annales 27: 185–214. Bagnall, R. S. and K. A. Worp. 2004. Chronological systems of Byzantine Egypt. 2nd ed. Leiden; Boston, MA. Becker, C. H. 1906. Papyri Schott-Reinhardt I. Heidelberg. Blau, J. and S. Hopkins. 1987. Judaeo-Arabic papyri: Collected, edited, translated and analysed. Jerusalem Studies in Arabic and Islam 9: 87–160.

Acknowledgements I would like to thank Eve Krakowski, Nicholas de Lange, Adam Silverstein and Susanna Wolfert-de Vries for their helpful references and discussions. I discussed this paper at the workshop ‘Egypt Incorporated’ at Leiden in December of 2016 and I would like to thank the participants for their useful comments. Any remaining mistakes are, of course, my own.

VISIBLE IDENTITIES

———. 2018. Vier Studien zu arabischen Dokumenten des 8.–14. Jahrhunderts. Wiesbaden. Dridi, A. 2015. Christian and Jewish communities in Fusṭāṭ: Non-Muslim topography and legal controversies in the pre-Fatimid period. In The Late Antique world of early Islam: Muslims among Christians and Jews in the Eastern Mediterranean, R. Hoyland (ed.), 107–132. Princeton, NJ. Fenton, P.-B. 1995. Saꜥdyā ben Yōsēf. In Encyclopaedia of Islam 2, P. Bearman, T. Bianquis, C. E. Bosworth, E. van Donzel and W. P. Heinrichs (eds), 8: 661–62. Leiden. Fiey, J. M. 1963. Tagrit: Esquisse d’histoire chrétienne. L’Orient Syrien 8: 289–342. Gardner, I. 1996. Kellis literary texts 1. Oxford. Gascou, J. 1978. Ostrakon grec tardif de l’IFAO. Bulletin de l’Institut français d’archéologie orientale 78: 227–30. Gayraud, R.-P., S. Björnesjö and S. Denoix. 1986. Isṭabl ꜥAntar (Fostat) 1985: Rapport de fouilles. Annales Islamologiques 22: 1–26. Grohmann, A. 1938a. Arabic papyri from the Egyptian library III. Cairo. ———. 1938b. Ein arabischer Steuerpapyrus aus der Sammlung der Papyrus Erzherzog Rainer. Zeitschrift für die neutestamentliche Wissenschaft 37: 52–53. ———. 1943. Arabische Papyri aus der Sammlung Carl Wessely im Orientalischen Institut (Orietální Ústav) zu Prag. Archiv Orientální 14: 161–260. ———. 1952a. From the world of Arabic papyri. Cairo. ———. 1952b. Arabic papyri from the Egyptian Library IV. Cairo. Hanafi, A. 2004. Two paper documents and a papyrus. In Papyrology and the history of early Islamic Egypt, P. M. Sijpesteijn and L. Sundelin (eds), 45–62. Leiden. Hinds, M. and H. Sakkout. 1981. A letter from the governor of Egypt to the king of Nubia and Muqurra concerning Egyptian–Nubian relations in 141/758. In Studia Arabica et Islamica: Festschrift for Ihsan Abbas, W. al-Qāḍī (ed.), 209–29. Beirut. Hoyland, R. G. 1997. Seeing Islam as others saw it. A survey and evaluation of Christian, Jewish and Zoroastrian writings on early Islam. Studies in Late Antiquity and Early Islam 13. Princeton, NJ. Hunt, L.-A. 2003. Stuccowork at the monastery of the Syrians in the Wadi al-Natrun: Iraqi-Egyptian artistic contact in the ‘Abbasid period. In Christians in the heart of Islamic rule: Church life and scholarship in ‘Abbasid Iraq, D. Thomas (ed.), 93–127. Leiden. Ibn ꜥAbd al-Ḥakam. 1922. Futūḥ Miṣr: The history of the conquests of Egypt, North Africa and Spain, C. Torrey (ed.). New Haven, CT. Ibn al-Biṭrīq. 1985. Das Annalenwerk des Eutychios von Alexandrien: Ausgewählte Geschichten und Legenden kompiliert von SaꜤīd ibn Baṭrīq um 935 A.D., M. Breydy (ed. and trans.). Leuven.

333

Ibn Duqmāq. 1893. Kitāb al-Intiṣār li-wāsiṭat ꜥiqd al-amṣār: Description de l’Égypte, K. Vollers (ed.). Cairo. Ilan, T. 2016. The Jewish community in Egypt before and after 117 CE in light of old and new papyri. In Jewish and Christian communal identities in the Roman world, Y. Furstenberg (ed.), 203–24. Leiden. Jahn, K. 1937. Vom frühislamischen Briefwesen: Studien zur islamischen Epistolographie der ersten drei Jahrhunderte der Hiǧra auf Grund der arabischen Papyri. Archiv Orientální 9: 153–200. John of Nikiu. 1916. The Chronicle of John, Bishop of Nikiu. Translated from Zotenberg’s Ethiopic Text by R. H. Charles. London. St John the Almsgiver. 1980–81. The Life of Saint John the Almsgiver, K. Urwin (ed.). London. Kennedy, H. 1998. Egypt as a province in the Islamic caliphate. In The Cambridge history of Egypt I: Islamic Egypt, 640–1517, C. F. Petry (ed.), 62–85. Cambridge. Khan, G. 1992. Arabic material: Selected material from the Khalili collection. Oxford. al-Kindī. 1912. Kitāb al-wulāt wa-kitāb al-quḍāt, R. Guest (ed.). Leiden. Lev, Y. 2012. Coptic rebellions and the Islamization of Medieval Egypt (8th–10th century): Medieval and modern perceptions. Jerusalem Studies in Arabic and Islam 39: 303–44. Levy-Rubin, M. 2016. ῾Umar II’s ghiyār edict: Between ideology and practice. In Christians and others in the Umayyad state, A. Borrut and F. M. Donner (eds), 157– 72. Chicago, IL. al-Maqrīzī. 2002–2003. Al-Mawā’iz wa-l-iꜥtibār fī dhikr al-khiṭaṭ wa-l-āthār, A. F. Sayyid (ed.). 6 vols. London. Mikhail, M. 2014. From Byzantine to Islamic Egypt: Religion, identity and politics after the Arab conquest. London. Olszowy-Schlanger, J. 2006. Les plus anciens documents datés de la Guenizah du Caire: Lectures et relectures. Livret-Annuaire de l’EPHE 20: 47–50. Omar, H. 2013. The crinkly-haired people of the black earth: Examing Egyptian identitied in Ibn ῾Abd al-Ḥakam’s Futūḥ. In History and identity in the Late Antique Near East, P. Wood (ed.), 1499–67. Oxford. Papaconstantinou, A. 2007. ‘They shall speak the Arabic language and take pride in it’: Reconsidering the fate of Coptic after the Arab conquest. Le Muséon 120: 273–99. ———. 2009. What remains behind: Hellenism and Romanitas in Christian Egypt after the Arab conquest. In From Hellenism to Islam: Cultural and linguistic change in the Roman Near East, H. M. Cotton et al. (eds), 447–66. Cambridge. Rāġib, Y. 1981. Lettres nouvelles de Qurra b. Šarīk. Journal of Near Eastern Studies 40: 173–87. Rustow, M. 2008. Heresy and the politics of community: The Jews of the Fatimid caliphate. Ithaca, NY.

334

P. SIJPESTEIJN

Shaked, S. 2010. Persian–Arabic bilingualism in the Cairo Genizah documents. In ‘From a sacred source’: Genizah studies in honour of Stefan C. Reif, S. Bhayro and B. Outhwaite (eds), 319–30. Leiden. Sijpesteijn, P. M. 2007. The archival mind in early Islamic Egypt: Two Arabic papyri. In From al-Andalus to Khurasan: Documents from the medieval Muslim world, P. M. Sijpesteijn, L. Sundelin, S. Torallas Tovar and A. Zomeño (eds), 163–86. Leiden. ———. 2011. Building an Egyptian identity. In The Islamic scholarly tradition: Studies in history, law and thought in honor of professor Michael Allan Cook, A. Q. Ahmed, M. Bonner and B. Sadeghi (eds), 85–106. Leiden. ———. 2013. Shaping a Muslim state: The world of a mideighth-century Egyptian official. Oxford. ———. 2020. Establishing local elite authority in Egypt through arbitration and mediation. In Transregional and regional elites: Connecting the early Islamic empire, H.-L. Hagemann and S. Heidemann (eds), 387–406. Berlin. Sirat, C. 1985. Les papyrus en caractères hébraïques trouvés en Égypte. Paris. Smelik, W. F. 2012. Justinian’s Novella 146 and contemporary Judaism. In Greek scripture and the rabbis, T. M. Law and A. Salvesen (eds), 141–63. Leuven.

Starr, J. 1935. Byzantine Jewry on the eve of the Arab conquest (565–638). Journal of the Palestine Oriental Society 15: 280–93. Stiefel, B. L. 2014. Jews and the renaissance of synagogue architecture, 1450–1730. London. Stillman, N. A. 1997. The non-Muslim communities: The Jewish community. In Cambridge history of Egypt I, C. F. Petry (ed.), 198–210. Cambridge. Stroumsa, G. G. 2012. Jewish survival in Late Antique Alexandria. In Jews in Byzantium: Dialectics of minority and majority cultures, R. Bonfil, O. Irshai, G. G. Stroumsa and R. Talgam (eds), 257–70. Leiden. Van Rompay, L. 2009. Syriac and related scripts in Egypt. In Journey of writing in Egypt, K. Azab and A. Masour (eds), 150–63. Cairo. Vanthieghem, N. 2014. Contrats de location arabes II. Chronique d’Égypte 89: 210–15. Yarbrough, L. 2016. Did ῾Umar b. ῾Abd al-῾Azīz issue an edict concerning non-Muslim officials? In Christians and others in the Umayyad state, A. Borrut and F. M. Donner (eds), 173–206. Chicago, IL.

RETHINKING PERSECUTIONS: P.RYL. III 469 AND THE MANICHAEANS IN EGYPT Roberta MAZZA

How did different religious systems negotiate their coexistence in the multicultural environment of Egypt? Did the spread and success of monotheistic universal religions, like Christianity, which will be the focus of this article, make the negotiation more difficult than before? What kind of impact did political imperial powers have on the terms of the negotiation between different groups? Finally, who ‘persecuted’ whom in this context? The exhibition that gave occasion to this volume was centred on questions of great interest for ancient historians and in general also for the wider public, in view of their relevance for the modern world. In the Late Roman and Byzantine period, religious violence, physically enacted and verbally expressed, was widespread in Egypt and elsewhere. But it seems to have become particularly acute in those moments when ruling elites took a stance on religious freedom, targeting specific groups or implementing wider policies on state religion; as well as when religious leaders tried to affirm their beliefs and practices as normative, and their organization as the only one that was allowed to establish and control texts, rituals, creeds, norms and other aspects of people’s lives. Papyri offer a privileged point of view for observing these dynamics and trying to answer the questions at stake for the period at the core of the present contribution, the second half of the third and the fourth century AD, for they often offer a different perspective on themes and issues from those presented in literary or legal sources. In this chapter, we will observe how in fourth-century Egypt a Christian, possibly a bishop, reacted to the spread of Manichaeism, a religious system so rooted in the Jewish and Christian traditions that it was at times considered a heretical form of Christianity. We will undertake a close analysis of a script left behind by this Christian, a unique fragmentary papyrus letter held in the John Rylands Library of Manchester: P.Ryl. III 469. An extraordinary manuscript for its content and material aspects, it reports a violent attack against the Manichaeans, and communicates warnings to an unnamed Christian community not to be led astray by Manichaean preachers and teachings. As will be shown, the text offers insights into religious coexistence and its

challenges in Late Antiquity: in my interpretation, the papyrus sheds light on the competition between different universal religions, in particular Christianity in its multiple forms, in Egypt and beyond. While inhabitants of the Byzantine empire were encountering different experiences, even adhering to different religions in the course of their lives, some of the parties involved in such competition wanted to persuade, defeat and even eliminate the others. Coexistence for them was simply impossible: in the fight to prevail, alliances with the imperial political power would become the key to success and lead to different forms of persecution. Although we still tend to identify persecution with the imperial persecutions against Christians, violent oppositions within Christianity make the study of persecution and martyrdom much more complex and fascinating. Rather than simple opposition between fading Graeco-Roman polytheism and rising Christian monotheism, Late Antique religious violence is to be interpreted as a more pervasive phenomenon of resistance against diversity, often within the same religious and philosophical movements. In order to understand better the nature and content of the Rylands letter, it is necessary to recall briefly what Manichaeism was. This universal religion, which shared some scriptures with Christianity, was named after Mani (or Manichaios), a teacher, mystic and prophet who flourished in the early Sassanian empire, at first supported by the emperor Shapur I, but later persecuted and condemned to death by Vaharam I and his entourage (c. AD 216–276/77). In the complex religious panorama of the third century, the peculiarity of Mani seems to have been that he consciously conceived his religion as based on a precise set of scriptures, teachings and a church (in the sense of an organized structure) with a universal mission. From Persia, Manichaeism spread towards both the East and the West, where it developed as an expression of the Hellenistic culture typical of the Syrian-speaking belt, stretching from Roman Syria into Persia (Gardner and Lieu 1996, 146); in the Romano-Byzantine empire it became at some point both a religious and a political problem because of its Persian ties (Brown 1969; Coyle 2009,

336

R. MAZZA

3–23). As already mentioned, Mani’s belief system was heavily indebted to the Jewish and Christian traditions; according to the Chinese sources on his life, Mani’s father joined the Baptist sect of Elchasaios after a divine voice ordered him to abstain from meat, wine and sex, and Mani was raised in this environment which he left around the age of twenty-four or twentyfive (Gardner and Lieu 2004, 3–4). Despite these roots, Manichaeism also assumed elements from other religions coexisting in the Persian empire and in the areas into which it had spread from the third century onwards. Because Manichaeism incorporated aspects of other religions, it could more easily be presented as a heresy within them and in fact attracted accusations of this kind. Since Late Antiquity, in the Roman empire Manichaeism has been tied to other heretical groups, first and foremost the Gnostics, because of the similar dualism that informs these philosophical systems. According to both systems, two principles, one good and divine, the second evil and opposed to God, governed the world, and humankind had to combat the latter in order for the first to prevail; in human beings, the soul contains the divine principle while the flesh contains the evil. However, there were also substantial differences, for instance in what concerns the creation principle.1 In order to make the spirit overcome matter, followers of Mani had to respect a number of ethical and practical norms often seen as contrary to common sense by their rivals, who, as will be shown, transformed Manichaean diversity into deviance through repeated attacks against the sect. When Christianity later became the empire’s religion, persecution was expressed not only through libels and discourses, but also, and more dangerously, through the passing of laws against Manichaeans, along with Jews, heretics, Hellenes (= Greeks, i.e., pagans)2 and basically any other non-Christians. For a long time modern knowledge of Manichaeism relied on anti-Manichaean writings, in some cases entire treatises, produced by Christian and also nonChristian authors, until new original sources produced

1

Connections between different heretical groups were created by Church heresiologists, and should be carefully handled. In Late Antiquity, the term ‘gnostic’ was indiscriminately applied to different groups and systems of thought; as a result, the term Gnosticism has been rightly criticized as an inappropriate collective

by Manichaeans began to emerge in different areas of the Mediterranean and the East, and were published especially over the course of the last century (Gardner and Lieu 1996). As expected, in Manichaean texts Mani is presented as the true interpreter of the Christian revelation and the apostle of Jesus, while in antiManichaean scripts he appears as the deviant founder of a sect. It should be remembered that the term Manichaean (manichaios) seems to have been rarely if ever used by Mani or his followers, but rather employed by external observers, most often opponents, of the group (Coyle 2009, 3–4). As will be shown, a great amount of the new Manichaean evidence has been found in Egypt: we now know that communities of Manichaeans were probably present in Oxyrhynchus (modern el-Bahnasa), certainly in the Fayum, the Dakhla Oasis and Lykopolis (modern Asyut), which mainstream scholarship acknowledges as a main centre for the diffusion of this religion in Egypt (Gardner and Lieu 1996, 168; Römer 2009, 637; Minale 2013, 1–13). According to ninth- and tenthcentury Manichaean texts from Central Asia, which collect more ancient traditions, the first missions to the West were sent by Mani himself, and Adda, one of the disciples, went as far as Alexandria (Gardner and Lieu 2004, 111–12). There is little doubt that the cosmopolitan environment of the city must have been one of the main hubs for Manichaean missionaries. In the early fourth century, the first Church historian, Eusebius of Caesarea, paid little attention to Mani (Hist. Eccl. 7.31.1–2; Oulton 1932), but in the fifth century, Socrates of Constantinople devoted more space to him and his biography (Hist. Eccl. 1.22; Hansen 1995). Socrates explicitly mentions that he relied on the antiManichaean tradition of the Acta Archelai, a work attributed to Hegemonius, an otherwise unknown author, who documented the fight of Archaelaus, a Christian bishop of a Mesopotamian city, against the Manichaeans (Vermes 2001). The Acta established a connection between three figures: Scythianus, a Saracen based in Egypt who married a woman from the

2

noun, which did not correspond to any real, uniform religious and philosophical system. On the question, see King 2003. In the period with which we are concerned, the term ‘Hellenes’ was used in the Greek language to indicate what we now define as ‘pagan’.

RETHINKING PERSECUTIONS

Upper Thebaid; his disciple Terebinthus (or Buddha), who then moved to Babylon; and lastly Cubricus, a slave-boy raised and later freed by the widow who was hosting Terebinthus in the city when he died as a result of his unsuccessful magic performances. The widow left everything to Cubricus, including Terebinthus’ property and books, which were carried by Cubricus into Persia, where he changed his name to Manes (i.e., Mani) and passed off the books as his own writings. After some success at the Persian court, Manes fled but was found by the king and killed. The legendary nature of this account is evident, but the structure of the narrative is revealing because it establishes an interesting link between Egypt, Babylon and Persia as Manichaean territories, so to speak, and discredits not only the Manichaeans’ founding figure but also their scriptures. There is another detail contained at the very beginning of Socrates’ chapter on Mani that is important for the discussion that follows. Socrates explains that, ‘A little while before the time of Constantine, a Hellenizing Christianity grew aside the true Christianity, just as false prophets grow aside the prophets, and false apostles aside the apostles. For at that time the dogma of Empedocles, philosopher among the Hellenes, deceived Christianity through Manichaios’ (Hist. Eccl.1.22.2; author’s translation). The passage is relevant in two ways: first it locates the spreading of Manichaean Christianity shortly before the Constantinian period, i.e., c. 306–37. In my opinion the sentence relates to the flourishing of this religion and its influence on other forms of Christianity, and not to the first phase of the Manichaean mission to Egypt that some scholarship locates earlier, in the second half of the third century, even before Mani died. This is a point to which I will return later. Secondly Manichaeism is defined as a form of Hellenizing Christianity, influenced in particular by Empedocles, to whom later in the text Socrates adds also Pythagoras. Now Mani was viewed as the leader of a sect of Christianity, itself an underdeveloped and simplified system of thought, by a Hellene Neoplatonic observer, Alexander of Lykopolis (Contra man. 1–2; Van der Horst and

3

All these laws found their source in Constantine’s provisions against division in the Church and the definition of heresy that originated in the years immediately preceding and following the

337

Mandfeld 1974, 51–52). Little is known about this philosopher, except what is transmitted in a short and not fully reliable notice by Photius, and what he writes in his own treatise on the Manichaeans, which scholars date between 277 and 297 (Villey 1985, 20–22). Alexander’s origin, from Lykopolis, renders him a particularly interesting source; in his work, he states he learned the Manichaean system of thought from Mani’s own pupils (Contra man. 2; Van der Horst and Mandfeld 1974, 52). After a summary of the main principles of Manichaeism, Alexander says that despite their philosophical weaknesses, they were able to convert some of his fellow-philosophers (Contra man. 5; Van der Horst and Mandfeld 1974, 58). This statement, taken with the abovementioned lines of Socrates, sheds light on a rich and variegated philosophical environment, where Christian and Hellenic thinkers intermingled and easily passed from one philosophical system to another (DePalma Digeser 2012, esp. 1–23). Christian anti-Manichaean scripts and treatises flourished especially from the fourth century onwards; as is widely known, one of the most important Christian figures of Late Antiquity, Augustine, wrote repeatedly against the sect, which he knew intimately as he had been a Manichaean himself (Brown 2000, 35–49; Coyle 2009, 209–19). So Manichaeans were confronted by both Christian leaders and Neoplatonic philosophers, and also by the imperial power. Anti-Manichaean measures were taken first by Diocletian at the end of the third or beginning of the fourth century AD (Corcoran 1983, 136–37: the rescript is most probably to be dated 31 March 302, but 287, 297 and 307 are also possible), and later by Valentinian and Valens (C.Theod. XVI.5.3; AD 372) and other emperors (Gardner and Lieu 2004, 145–50).3 As pointed out by Stroumsa (1982; 1986), Manichaeans escaped from persecution by hiding in monasteries where their way of living could be well adapted into monastic asceticism. The threat of Manichaeans coming to Rome disguised as African monks was denounced by popes Gregory I and Gregory II as late as the late sixth and early seventh centuries AD (Stroumsa 1986, 312). In other words,

Council of Nicaea in AD 325 and are only partially preserved (see Drake 2000, 346–50 and most recently Minale 2013, which is focused on the Manichaeans).

338

R. MAZZA

despite Christian opposition and imperial legislation, Manichaeism is attested well into the Byzantine era. Let us now consider our papyrus. P.Ryl. III 469 was purchased on the antiquities market, on behalf of the John Rylands Library, by Oxford-based papyrologist Bernhard P. Grenfell during his last trip to Egypt of 1920; as a result, we do not know anything about the find-spot or archaeological context of the papyrus. It was published by Colin Roberts in 1938, and to my knowledge the dating, reading and general interpretation of the text has never been challenged since. According to Roberts, and successive scholarship, this is a pastoral epistle sent by a bishop of Alexandria to one or more communities of the Egyptian chora (the territory of Egypt beyond Alexandria) towards the end of the third century AD (P.Ryl. III 469, introduction; followed by, e.g., Gardner and Lieu 2004, 114; Römer 2009, 637; Blumell 2012, 204 n. 184). It should be stressed that P.Ryl. III 469 is formatted as an actual letter, similar to others preserved on papyrus, and it is important material proof of what we know about the mechanics of Church communication in Late Antiquity, according to a tradition rooted in the epistles of Paul, the apostles, bishops and other early Christian leaders. The material aspect of the letter shows that it was carefully laid out by a trained scribe, written in correct Greek (except for common forms of iotacism, and confusion among letters due to phonetics), and enhanced by adding punctuation and even breaths and accents in some cases (Fig. 1). Then it was folded in the typical way letters were, as demonstrated by the nine still visible folds. The text probably extended for three if not more columns, of which we have only the remains of the last two; considering the height of the extant text (c. 35 cm) and the average height of papyrus rolls, not many top lines have been lost (Johnson 2004, 141–43). The address, if any, should have been written in the missing portion. It is worth mentioning that the papyrus is of light colour and good quality: the surface is smooth, and the roll is unusually thick. Roberts already noted in his edition that P.Ryl. III 469 shares similarities with Festal Letters, official letters that bishops of Alexandria used to send throughout Egypt in order to announce the date of the following Easter while addressing theological and other themes. Some of these letters have been transmitted in later edited collections; few fragmentary Greek and Coptic copies are still extant. As is the case with our papyrus, these are usually written along the fibres of a papyrus

roll, the back of which has no address and tends to be empty (Camplani 2003, 20–23; Stroppa 2013, 351). While the material aspect and topics may recall such epistles, our papyrus does not conclude by giving the Easter date as the others do, meaning that this was probably an official letter of another kind. My argument is that this letter dates not to the end of the third, but to the fourth century AD for palaeographical and historical reasons, and its author, as also pointed out by previous scholarship, is likely a bishop of Alexandria, although I would not exclude other Christian leaders, such as intellectuals from the Alexandrian catechetical school who participated in the anti-Manichaean polemics, or other bishops also involved in it. Let us consider palaeographical dating first. Methodological weaknesses in attributing overly precise dates to Christian papyri on the basis of palaeography have been brought to attention in recent scholarship. Handwriting styles were in use for longer periods than previously thought, and it is basically impossible to attribute a date more precise than a century to papyri which not only lack a date, but also an archaeological context or some internal elements that may help identify when they were written (Bagnall 2009, 1–24). As with other examples, including the edition of the famous fragment of the Gospel of John, P.Ryl. III 457 (=P52), Roberts’ dating of this text has suffered from an old palaeographical approach that was overconfident in attributing precise dates to this category of texts (Nongbri 2005). In his palaeographical analysis of P.Ryl. III 469, the papyrologist refers only to one securely dated parallel, P.Flor. II 254, a letter from the Heroninus archive of AD 259. In the introduction to the papyrus edition, P.Ryl. III 469 handwriting is defined as: regular and official in style, though not as stereotyped as chancery hands, and common near the end of the third century; to this period (approximately AD 275–300) I would assign the text. A date in the early fourth century cannot be excluded, but the hand seems to me less formal and more free than fourth century hands.

Roberts then identifies the author with bishop Theonas of Alexandria, by establishing a firm, although unproven, link between our anti-Manichaean epistle and Diocletian’s rescript to the prefect of Africa regarding the sect, which he dates to 297. Interestingly this identification has been taken as certain by subsequent and recent scholarship on Manichaeism, which labels our papyrus as the most ancient extant Christian

RETHINKING PERSECUTIONS

339

Fig. 1: P.Ryl. III 469 recto (Courtesy of the John Rylands Library. Copyright of the University of Manchester).

document on the sect (see, e.g., Dignas and Winter 2007, 28; Coyle 2009, 195–96; Kristionat 2013, 193– 94). To sum up, the date proposed by Roberts was merely based on palaeographical observations linked to Diocletian’s rescript; everyone dealing with the papyrus after him has taken Roberts’ opinion for granted. In order to give a stronger basis to the palaeographical analysis of the papyrus, I compared the handwriting to that of official correspondence and documents. Similarities are to be found, for instance, with P.Ryl. IV 617,4 a draft of a petition to the emperor belonging to the early-fourth-century archive of Theophanes and possibly dating to around 317, and P.Oxy. LXVII 4602,5 a letter to a strategos, the official in charge of the nome, of 361. An interesting parallel is also offered by P.Lond. VI 1924 (with pl. IV), a letter written by a professional scribe on behalf of Ausonius, probably the homonymous praeses of Augustamnica of 341–42. Our epistle, although written in cursive handwriting of the

same type as those here listed, presents higher levels of tidiness and regularity, and almost ornamental features (e.g., the loopy ductus of many letters) due to its official ecclesiastical nature and degree of literary construction. Other elements, such as the diacritical signs, typical of literary writings appear as well. In other words, this kind of handwriting was in use for a period longer than that noted by the first editor. Moreover, while Roberts supported his late-third-century date with the issuing of anti-Manichaean measures by Diocletian, I believe that these were not necessarily a drive for bishops to intensify their opposition to the sect, especially considering that these years were difficult for Christianity as a whole (DePalma Digeser 2012). If we accept the common date of 31 March 302 for the anti-Manichaean rescript, the measure was followed just one year later by the anti-Christian edict and the beginning of the so-called Great Persecution.

4

5

Digital image available at: https://luna.manchester.ac.uk/luna/ servlet/s/l31v82.

Digital image available through a link available via http:// papyri.info/ddbdp/p.oxy;67;4602.

340

R. MAZZA

Let us move now to the other reason that makes me inclined to argue for the fourth century: the content of the letter. Although we cannot be sure of what preceded it, the extant part of the epistle is aimed at warning the addressees about the deviant beliefs and debased practices of the Manichaeans. The letter seems to show some direct knowledge of the sect and its scripts, and worries about it spreading. The first accusation against Manichaeans concerns marriage (ll. 12–16; the preceding lines are too broken to make any sense): Again the same Manichaeans speak falsely against marriage saying that he does well who does not marry. Paul says that the man who does not marry does better, but that the adulterer and fornicator are evil is clear from the Holy Scriptures, from which we learn that marriage is honoured, while God abominates fornicators and adulterers.

The last sentence (marriage is honoured, while God abominates fornicators and adulterers) is a quote from Hebrews 13.4 that has been elaborated on (‘Marriage should be honoured by all, and the marriage bed kept pure, for God will judge the adulterer and all the sexually immoral’) and contains a variation (otherwise unattested) since it employs the verb misein (to hate) instead of the standard krinein (to judge), which is in fact added above the line as a note or correction, probably by a second scribe. The point under discussion in the lines in question is the opposition of Manichaeans to marriage, and their inclination towards celibacy, a common idea and practice in some normative Christian environments too, first and foremost among monks and ascetics. Marriage and abstinence from sex were at the centre of a debate that started in the third century, but reached its peak in the fourth. In fact, it seems that Manichaeans recommended abstinence from marriage and sexual intercourse only to some categories of believers, namely the Elects, the highest level of their Church, but not to the Auditors or catechumens, the great majority of the people adhering to the creed (Gardner and Lieu 2004, 23–24). Accusations of anti-marriage positions are also contained, and counter-argued, in other anti-Manichaean Christian writings, which confirms that this was regarded as a key issue. Within the Egyptian context, a good parallel is furnished by Didymus the Blind’s Treatise against the Manichaeans (Bennett 1997; Layton 2004, 15–34). The author, a student of Origen and a Christian teacher close to Athanasius the Great, who, according to some scholars, made him

head of the catechetical school of Alexandria, puts forward the argument that for those who live according to the Gospel of the Saviour it is possible to contract an honourable marriage and have undefiled intercourse; to strengthen the point, he interestingly recalls the same passage from Paul’s Letter to the Hebrews found in P.Ryl. III 469 (Heb. 13.4; Bennett 1997, 312–13). Didymus’ broader line of reasoning was that the principle of evil was not innate, as Manichaean philosophy supposed, but derived from intention (proairesis): so there was nothing evil in the flesh as opposed to the spirit, but rather in the intention of human beings regarding both the flesh and the spirit (Bennett 1997, 135–40). Didymus also deals with Manichaean antimarriage positions in his commentary on Ecclesiastes, following a similar argument and stressing the point of procreation as the main element of Christian unions. What is also interesting about this passage, which unfortunately is marred by textual problems and lacunas, is the fact that the author refers to conversations he had with a Manichaean on the point (Comm.Eccl. T. 274.18–275.2, cited by Bennett 2001, 49–50). In the end of the first and the beginning of the second column of the Rylands papyrus, the argument against the Manichaeans moves on to accusations of idolatry, referring to other quotations from the Bible and also, interestingly, citing a direct (although fragmentary) quotation of what the author defines as an ‘apology’ over the bread (ll. 25–26). This ‘apology’ is a blessing formula recited over the bread during the rite of almsgiving, in which the Auditors provided food to the Elects, and is transmitted by other Christian sources, the Acta Archelai and Cyril of Jerusalem’s sixth catechetical lecture. It is worth noticing that the final redaction of the Acta Archelai, probably systematizing precirculating materials, is dated by scholars to the first half of the fourth century (BeDuhn and Mirecki 2007, 8–9), and Cyril’s lectures were probably delivered around 348–50, so later than the Rylands letter as dated by Roberts. BeDuhn thinks that our papyrus was in fact a source for the other two scripts, but I wonder if that is really the case (BeDuhn 2000, 131–32). The author of the Rylands papyrus mentions a direct reading from a Manichaean script (ll. 29–30: ‘as above mentioned, I have presented these things in brief from a written document of the madness of the Manichaeans which fell into my hands’): although clearly reporting the same formula as the other two texts, I do not see any solid reason why the the Acta and Cyril of Jerusalem should be necessarily relying on the letter in question

RETHINKING PERSECUTIONS

rather than other Manichaean or anti-Manichaean sources that may have reported it. It appears, then, that the writer of the Rylands letter had first-hand knowledge of Manichaean rites and beliefs, and was aware of Manichaean activities of proselytism among orthodox Christian communities, as will be shown later. This is unsurprising not only for what we know from the literary accounts already mentioned, but also from the many papyrological sources from Egypt bearing Manichaean scriptures and other, including everyday, writings, which have started coming to light since the last century. Among these archaeological finds, all dating no earlier than the fourth century AD, the most prominent are: the socalled Köln Mani Codex; the codices from Medinet Madi (ancient Narmouthis); and the texts and other archaeological evidence from Kellis in the Dakhla Oasis (Gardner and Lieu 1996). Acquired on the antiquities market by the University of Köln in 1969, the Mani codex is an extraordinary object: a ‘miniature’ luxury book, consisting of 192 pages of fine parchment, each containing twentythree lines of writing. The codex was written in Greek in a regular, calligraphic handwriting by possibly two scribes who used a magnifier (probably a glass with water in it) in order to carefully preserve the teachings and life of Mani as reported by his disciples. The book has been dated to the fifth century on palaeographical grounds, although there is some debate on this point (a later dating, however, seems less probable). Its ancient function is also debated: it could have been a phylactery, but also—and maybe at the same time—a text to be read on some private or public occasion (Koenen 1970; Koenen and Römer 1988; Römer 2009, 633–37). Starting in the late thirties and early forties of the last century, papyrus codices from Medinet Madi, the ancient town of Narmouthis in the Fayum, also began to emerge from the antiquities market and are now dispersed in different collections (Gardner and Lieu 1996, 148–54; Robinson 2013). The texts are translations in Coptic, from either the Syriac originals or their Greek translations, of Manichaean literature, including the Teaching of Mani, his Epistles, the Psalms and

341

Homilies (Römer 2009, 638). They have been dated to the fifth or even sixth century AD. It is interesting to notice that the Coptic dialect of the texts in question was not used in the Fayum, but is typical of the Lykopolitan area. This and other details have led specialists to formulate the hypothesis that these books were possibly part of the library of some members of the Manichaean Church who were fleeing their residence to escape persecution (Gardner and Lieu 1996, 149). This impression of communities under threat has also been observed in some of the Manichaean papyrus letters from Kellis (Gardner 1997, 170–71). The Roman site of Kellis (modern Ismant el-Kharab) is located at the centre of the Dakhla Oasis in the Western Desert; since 1978, together with the rest of the Oasis, it has been at the centre of intensive archaeological and environmental research by a pool of international institutions.6 Excavation finds have proven that Kellis was inhabited from the first to the end of the fourth or beginning of the fifth century AD; the reasons for its abandonment are still obscure, and are possibly due to climate changes that made living conditions in the Oasis more difficult. At the centre of the town there was a temple dedicated to the Egyptian deity Tithoes, which seems to have lost its primary cultic function in the middle of the fourth century, while excavations of residential areas have brought to light the establishment of Christianity (three churches and a cemetery have been found) alongside the presence of Manichaeans, attested by both fragments of Manichaean books and everyday writings bearing traces of their owners’ creeds (P.Kellis II, V, VI and VII). Editors of the texts have emphasized that the archaeological evidence as a whole indicates that the religious context of the site was indeed varied (Gardner 1997, 162). The Manichaean texts, on papyrus and wooden tablets, have for the most part been found in Area A of the site, mainly in House 3, but there have also been scattered finds in other houses in this area and elsewhere. It is interesting to notice that in fourth-century Kellis inhabitants were using both Coptic and Greek, while some writings in Latin and Syriac are also attested. In letters, Coptic seems to have been reserved especially for private communication, while Greek was used for business

6

See the project’s website at http://dakhlehoasisproject.com/.

342

R. MAZZA

(Gardner 1997, 167; letters are edited mostly in P.Kellis V and VII). Returning to our Rylands papyrus, it is clear then that it fits well within this evidence, especially that from Kellis, which dates to the fourth century AD and attests a Manichaean community flourishing in an area that was to some extent peripheral. Kellis was located far away from the Nile, and at the same time not too far away from Lykopolis (modern Asyut), which, as we have already said, was a main centre of Manichaean teaching and proselytism; the ancient route linking the Oasis to the Nile valley began from Asyut. In other words, it may be possible that the Kellis community stemmed from Lykopolitan groups, who moved to the Oasis to escape tensions with other groups and even imperial persecution. This hypothesis remains unproven; however, the letters found at Kellis demonstrate that its Manichaean inhabitants were in contact with other communities. Some women, in particular, are the addressees of a Coptic letter sent by an unnamed ‘Father’ who is in Egypt (i.e., the Nile valley) and invites them to continue supporting him and his group by sending oil (P.Kellis V 31, with the more fragmentary 32 and 33 in the same volume). The editor of this text connects its content to the almsgiving practice, which sustained the Manichaean Church and was also at the basis of its rituals, as we have seen in the case of the Apology of the Bread. The conclusion of the letter, l. 54 (‘Do not let [the letter] stay with you, write to somebody!’), might indicate that this was a sort of circular epistle, while at l. 34 the writer explains that they needed the oil since they were afflicted, a sentence that unfortunately remains obscure, as often happens in ancient letters, but may indicate some sort of persecution or threat. In the Kellis documentary evidence, women often are addressees of letters and seem busy with many activities, including textile production. Manichaean women are increasingly emerging from literary and documentary sources with important roles as both catechumens and Elects, as highlighted by recent studies (Coyle 2009, 187–205; Kristionat 2013, esp. 132–42). The mention of women leads us back to our Rylands letter and the warnings relating to Manichaean women, following the mention of the Manichaean script already recalled (ll. 29–35, Roberts’ translation slightly revised): As I said before, I have cited this in brief from the script of the madness of the Manicheans (eggraphos tes manias ton Manicheon) that fell into my hands; that we may watch out against those entering in our houses with

deceits and false words, and especially those women whom they call Elects and whom they have in great esteem manifestly because they need their menstrual blood for the defiled rituals of their foolishness (mania).

This section of the letter is important in many ways. It confirms the presence of women in this religious group, in particular among the Elects, but also presents them as the negative alter egos of those mentioned by the ‘Father’ of the Kellis letter. Then it introduces a typical topos of the Christian anti-heretical discourse: the accusation of deviant, especially sexual, behaviours and the contamination of rituals through the use of bodily fluids, which in fact convert those rituals into the exact opposite of normative Christian ones. The rhetoric deployed here is particularly interesting. The writer maintains that as the women preachers defile Christian houses with deceit and false words, similarly their menstrual blood defiles the already foolish Manichaean rites. Manichaeans esteem these women ‘manifestly’ because they need their menstrual blood. But this need is far from being ‘manifest’, at least to the modern, and I suspect even the ancient, reader. The author of the Rylands letter leaves the nature and modality of such rituals underdeveloped, titillating the audience’s imagination; he does not add any description or explanation about the use of the menstrual blood, but rapidly moves on to exhort the addressees to abstain from evil, and to be pure both in soul and body, through a series of biblical quotations. In order to make more sense of the lines in question, we should take into consideration similar accusations of using sperm and menstrual blood in rituals raised by other Christian authors against Manichaeans and other heterodox groups. Augustine is enlightening in this respect. In his treatise on the Manichaeans, written around 388, he reports allegations of ritual eating of semen among the Elects, right after a section where he attacks their position on marriage. According to Augustine, Manichaeans’ chastity is inherently wrong because they do not forbid sexual intercourse, but the begetting of children, which on the contrary is the aim of marriage. The core of the accusation is expressed through two direct questions to his ideal Manichaean opponent: Is it not you who used to counsel us to observe as much as possible the time when a woman, after her purification, is most likely to conceive, and to abstain from cohabitation at that time, lest the soul should be entangled in flesh? This proves that you approve of having a wife, not for the procreation of children, but for the gratification of

RETHINKING PERSECUTIONS

passion. In marriage, as the marriage law declares, the man and woman come together for the procreation of children (De mor. man. 18.65–66; Schaff 1956, 86–87).

In so doing, Augustine asserts that Manichaeans made women not wives, but mistresses, since there is no marriage without motherhood. At the basis of this allegation is the idea that for Mani’s followers the world contained living principles of light which should be carefully handled by following a set of behaviours and norms: for instance, the Elects should not eat meat since dead animals have lost their vital, divine principle, and have become decayed matter to be avoided. The same reasoning, according to Augustine, regulated their continence: they abstained from conjugal intercourse to avoid confining the soul in the flesh. This opened the possibility that accusations of eating human semen, which Augustine heard about, were actually true: For what you say of the flesh of a dead animal, that it is unclean because there is no soul in it, cannot be said of the seed of the animal; for you hold that it keeps confined the soul which will appear in the offspring, and you avow that the soul of Manichaeus himself is thus confined. And as your followers cannot bring these seeds to you for purification, who will not suspect that you make this purification secretly among yourselves, and hide it from your followers, in case they should leave you? If you do not these things, as it is to be hoped you do not, still you see how open to suspicion your superstition is [...] (De mor. man. 18.66; Schaff 1956, 87).

It is clear that our letter differs from Augustine’s longer composition as they belong to two different genres. However, there are striking similarities between the two writings. First of all, in both cases the topic of marriage is bound to that of inappropriate use of bodily fluids; then both authors represent Manichaean women in a disquieting manner, as mistresses instead of mothers and wives in the case of Augustine, as wandering preachers and the source of tainted fluids in the case of the Rylands epistle’s author. Finally, both writers do not offer any proof that misuse of bodily fluids did actually take place, but in any case spread the hearsay, according to a technique that has been aptly described in the case of Augustine as ‘a process of marginalising the followers of Mani’s teachings through the deployment of rumour and insinuation. In specific terms, Augustine is engaged in transforming Manichaean Christians into the ‘Other’ through the imputation of a revolting rite to the religion’s hierarchical class’ (Baker-Brian 2013, 507). Manichaeans are against

343

marriage because they are against procreation, which in fact is the core of Christian marriage: intercourse is only allowed for such an aim, without which it becomes pure lust. Christian food norms and ritual meals are equally transformed into abominable rituals through the suspicion of the use of bodily fluids. In both examples Manichaeism is represented as the opposite of normative Christianity. Augustine tackled the use of seminal fluids by Manichaean men and women in his later treatise on heresies, this time adding more details (Haer. 46.9; Gardner and Lieu 2004, 144). Here he says that ‘their Elect are forced to consume a sort of Eucharist sprinkled with human seed in order that the divine substance may be freed even from that, just as it is from other foods of which they partake’. Augustine goes on to say that although Manichaeans denied partaking in such practices and attributed them to other groups usurping their name, a trial took place in Carthage in which two Manichaean women were involved who reported a ritual where ‘flour is sprinkled beneath a couple in sexual intercourse to receive and commingle with their seed’, letting the reader infer that the flour had then been used to bake the ritual bread mentioned above. An attestation of a Manichaean practice involving the eating of semen has been also found in a passage of Cyril of Jerusalem; however, the details are again vague (van Oort 2016). Besides the improper use of bodily fluids, the specific mention of menstrual blood is found in Epiphanius of Salamis’ Panarion, a sort of encyclopaedia of heresies, completed around 375. In it, the Nicolaites and other Gnostic groups, which according to Epiphanius’ reconstruction all stemmed from the main heresy of Nicolaus, are accused of deviant sexual practices, and in particular of employing semen and menses in their rituals. Similarly to Augustine, Epiphanius explains that these rites result from the belief that seeds, including those of humans, contain particles of light, the divine vital principle that Gnostics wanted to preserve from decaying (Pan. 1.25.3:2; 1.26.1:9; 1.26.4:7–8; Holl 1915; Williams 1987). In the passage regarding the Gnostics or Borborites, the female menstrual blood is consumed in a communal meal clearly reversing the Christian Eucharist (Pan. 1.26.4:8; Williams 1987, 86): ‘And so with the woman’s emission when she happens to be having her period (en rhysei tou aimatos)—they likewise take the unclean menstrual blood they gather from her, and eat it in common. And “This,” they say, “is the blood of Christ”’.

344

R. MAZZA

There has been extensive discussion among specialists about how to evaluate Christian anti-Manichaean, and in general anti-heretical, writings and their reliability. In the specific cases involving the misuse of body fluids, some Gnostic sources explicitly condemn such practices. In book four of the Pistis Sophia, a text preserved in a Coptic manuscript dated to the fifth century AD, but probably composed in the third or fourth century AD, the apostle Thomas poses the following question to Jesus: ‘We have heard that there are some upon earth who take male sperm and female menstrual blood and make a dish of lentils and eat it, saying: ‘We believe in Esau and Jacob. Is this then a seemly thing or not?’ Jesus’ following answer is of eternal condemnation: ‘Truly I say that this sin surpasses every sin and every iniquity. [Men] of this kind will be taken immediately to the outer darkness, and will not be returned again into the sphere’ (Pistis Sophia, IV 147; Schmidt and MacDermot 1978, 381). This presents the possibility that practices involving such rituals were actually taking place in some types of Christian communities; interestingly, similar accusations had been made by Mandaeans about other Christian groups (Lidzsbarski 1925, 229). Accusations of deviant behaviour, especially of a sexual nature, had also been typical of Christian polemics against the Jews from the second century onwards, in an attempt to isolate the new religion from its Judaic roots (Drake 2013). Lust and sexual promiscuity had been also part of the GraecoRoman repertory of accusations against the barbarians, who were broadly conceived as the ‘Others’ and included Christians (see e.g., Minucius Felix, Octavius 9). In other words, this kind of blame reflects a cultural attitude embedded in the wider society of the time. Women behaving against the norms, like the Elects of our papyrus, in particular, are part of this same repertory. In any case, what is really important for the broader theme of this volume, the impact of empire on Egypt, is not so much to establish if the accusations made in P.Ryl. III 469 were based on real facts, but rather to observe through such an object how people living in the Romano-Byzantine world were in contact with, and often experimenting with, different religious environments and practices, while powerful members of the elite, bishops, intellectuals and even emperors, tried to normalize culture, including religion, and keep diversity at bay through moral discourse, legal actions and even physical violence. The famous case of Augustine, who was for about a decade a member of the Manichaean sect that he later

violently attacked, is symptomatic of an incredibly interesting and fluid Late Antique cultural environment. The same vitality is observed throughout the papyrological evidence from Kellis, where, as we have seen, Manichaeans were living among Christians, and it is not always easy to find the borders that divided the two groups. To what extent Christian communities of various kinds were intermingled is also reflected in the Oxyrhynchus papyri: some private letters, dated to the fourth century, have been attributed either to ‘orthodox’ Christians or to Manichaeans, owing to the common terminology the two groups shared (Gardner, Nobbs and Choat 2000; Blumell 2012, 72–74). Along with these letters, a theological hymn previously attributed to bishop Melito of Sardis preserved in a fifthcentury papyrus, P.Oxy. XVII 2074, has recently been re-edited and identified as a Manichaean hymn closely resembling the Psalms found at Medinet Madi and Kellis (Smith 2016). To sum up on the Rylands letter’s author and date of redaction: as I have explained in the previous pages, Roberts’ dating of P.Ryl. III 469 has serious methodological flaws. Palaeographical analysis locates this type of handwriting in a period that spans from the second half of the third into the second half of the fourth century AD. Moreover, there is no reason to believe that the Diocletian rescript of 302 and the wider political climate of those years immediately preceding the Great Persecution must have been the occasion for the Christian in question to write the letter, as argued by Roberts. On the contrary, I believe that the edict could have been a warning for Christians too, since after all, as well attested by Alexander of Lykopolis, Manichaeans were seen by the ‘Hellenes’ as part of Christianity. Egypt was certainly one of the first regions to be reached by Manichaean missionaries; Egyptian intellectuals have left writings against the sect, and Manichaean sources have come to light. Now, it should be emphasized that, as we have shown in the previous pages, all the extant Egyptian material concerning the Manichaeans dates to the fourth century AD onwards, with the apparent exception of the treatise of the rather elusive Alexander of Lykopolis, which is not only an attack on Manichaeism, but also against Christianity as a whole in a moderate, although firm, tone. The contents of the Rylands letter presume both a wellestablished and organized Manichaean presence, and a proselytizing campaign so successful as to constitute a serious threat in the eyes of a bishop or other Church authority. Some of the topoi used in the letter were

RETHINKING PERSECUTIONS

systematized and circulated in the fourth century: the pun on the name of Mani and the Greek word mania (= folly, madness); the accusations of deviant practices involving bloody female fluids; and also the defence of the possibility of a pure Christian marriage. In my view there are two figures active in the fourth century who might suit well as authors or inspirers of our letter. The abovementioned Didymus the Blind (c. 309–394; Bennett 1997, 13–14) and his circle deserve deeper investigation, also considering that the citation of Paul’s Hebrews 13.4 on marriage is absent from other Christian sources that address this question.7 The other possible author might be Athanasius the Great, bishop of Alexandria from 328 to 373. In this respect, confirmation of the circulation of letters of Athanasius against the heretics, including the Manichaeans, comes from Shenoute of Atripe, who, right at the end of a section of his work Who speaks through the prophet where he discusses Manichaean positions on the resurrection of the body, reveals his source: ‘So again we have understood the error of many heretics, from his [i.e., Athanasius’] letters’ (Gardner and Lieu 2004, 230). Overall, the mid-fourth century of Athanasius the Great emerges as the period to which the Rylands letter most probably belongs. These are the same years in which Serapion, bishop of Thmuis from about AD 339 to 360, also wrote against the sect (Contra Man.; Casey 1931), and Egyptian monastic sources started reporting clashes between Christians and Manichaeans and anti-Manichaean sentiment (e.g., Hist. Monach. X, 30–35; Festugière 1971, 87–88; Vita Antonii 68; see Stroumsa 1986 for further examples), not to mention that the composition and circulation of the Acta Archelai discussed above, which is considered an archetype for many of them, also dates to this period. While Manichaean preachers could have reached Egypt very early, from the mid-third century onwards, I find it more probable that their presence started to be felt as a serious, distinctive threat by ‘orthodox’ Christians only from the fourth century onwards. It is in the first half of that century that the Christian anti-Manichaean polemic evolved and became a common topic. This process, embedded into a longer historical trajectory,

7

I plan to expand this part of my argument further in a separate study. It should be noted that the two men were very close and

345

started in the mid-third century AD but came to maturation only with Constantine’s policies against heresy during the years immediately preceding and especially following the Council of Nicaea in AD 325, and was carried on by his successors, with the obvious exception of Julian. To conclude: the Rylands letter against the Manichaeans is an enlightening source for the study of intra-religious violence in Late Antique Egypt, and more broadly in the early Byzantine empire. Together with the rest of the fourth-century evidence, it signals the development of an increasing intolerance towards religious diversity inside the intellectual and political ruling class in the course of that century. In this climate, minorities, external and internal to normative Christianity, found increasingly less space to express their own beliefs and practices, and (as had happened with other groups, including Christians, before) they became the target not only of harsh cultural polemics, but also of laws condemning them to conversion or death. Nonetheless, they clearly found some ways to resist normalization, as the Manichaean archaeological evidence emerging from Egypt and elsewhere demonstrates. When imperial condemnation struck through rescripts and edicts, resistance became particularly tough; however, it was possible, especially in periods and regions which saw political infrastructures working less efficiently on the ground, to survive by fleeing to more secure places or hiding, as Christian history of the first three centuries after the death of Christ demonstrates. The Manichaean community of Kellis offers some perspective on resistance from the ground: its proximity to and intermingling with the rest of the town’s inhabitants, as attested by the archaeology of the settlement, is a testimony to the mostly peaceful cohabitation of different groups of Christians and nonChristians at a local level, which greatly alarmed Christian polemicists. This preoccupation of normalizing society seems to have been more an obsession of the elites, rather than a real issue for people probably occupied with solving more pressing problems of survival in a world where poverty, illness and other life-threatening issues were widespread.

for some scholars Didymus was put at the head of the catechetical school of Alexandria by Athanasius himself.

346

R. MAZZA

Bibliography All papyrological abbreviations are cited throughout this volume according to J. Oates et al. Checklist of Greek, Latin, Demotic and Coptic papyri, ostraca and tablets, http://papyri.info/docs/checklist (last accessed 8 July 2018). Bagnall, R. S. 2009. Early Christian books in Egypt. Princeton, NJ; Oxford. Baker-Brian, N. J. 2013. Women in Augustine’s antiManichaean writings: Rumour, rhetoric and ritual. In St Augustine and his opponents, M. Vinzent (ed.), 499–520. Studia Patristica 70. Leuven; Paris; Walpole, MA. BeDuhn, J. D. 2000. The Manichaean body in discipline and ritual. Baltimore, MD. BeDuhn, J. D. and P. Mirecki. 2007. Placing the Acts of Archelaus. In Frontiers of faith: The Christian encounter with Manichaeism in the Acts of Archelaus, J. D. BeDuhn and P. Mirecki (eds), 1–22. Leiden. Bennett, B. 1997. The origin of evil: Didymus the Blind’s Contra Manichaeos and its debt to Origen’s theology and exegesis. Unpublished PhD thesis, Toronto School of Theology. Bennett, B. 2001. Didymus the Blind’s knowledge of Manichaeism. In The light and the darkness: Studies in Manichaeism and its world, P. Mirecki and J. BeDuhn (eds), 38–67. Leiden; Boston, MA; Cologne. Blumell, L. H. 2012. Lettered Christians: Christians, letters and Late Antique Oxyrhynchus. Leiden. Brown, P. 1969. The diffusion of Manichaeism in the Roman empire. Journal of Roman Studies 59: 146–69. (Reprinted in P. Brown. 1972. Religion and society in the age of Augustine, 94–118. London.) ———. 2000. Augustine of Hippo: A biography. Rev. ed. Berkeley, CA; Los Angeles, CA. Camplani, A. 2003. Atanasio di Alessandria: Lettere Festali. Anonimo: Indice delle Lettere Festali. Milan. Casey, R. P. (ed.). 1931. Serapion of Thmuis: Against the Manichees. Harvard Theological Studies 15. Cambridge, MA. Corcoran, S. 1983. The empire of the Tetrarchs: Imperial pronouncements and government AD 284–324. Oxford. Coyle, J. K. 2009. Manichaeism and its legacy. Leiden. DePalma Digeser, E. 2012. A threat to public piety: Christians, Platonists and the Great Persecution. Ithaca, NY; London. Dignas, B. and E. Winter. 2007. Rome and Persia in Late Antiquity: Neighbours and rivals. Cambridge. Drake, E. 2013. Slandering the Jew: Sexuality and difference in early Christian texts. Philadelphia, PA. Drake, H. A. 2000. Constantine and the bishops: The politics of intolerance. Baltimore, MD; London.

Festugière, A. J. 1971. Historia Monachorum in Aegypto: Édition critique du texte grec et traduction annotée. Subsidia Hagiographica 53. Brussels. Gardner, I. 1997. The Manichaean community at Kellis: A progress report. In Emerging from darkness: Studies in the recovery of Manichaean studies, P. Mirecki and J. BeDuhn (eds), 161–75. Leiden. Gardner, I. and S. Lieu. 1996. From Narmouthis (Medinet Madi) to Kellis (Ismant El-Kharab): Manichaean documents from Roman Egypt. Journal of Roman Studies 86: 146–69. ——— (eds). 2004. Manichaean texts from the Roman empire. Cambridge. Gardner, I., A. Nobbs and M. Choat. 2000. P.Harr. 107: Is this another Greek Manichaean letter? Zeitschrift für Papyrologie und Epigraphik 131: 118–24. Hansen, G. C. 1995. Sokrates Kirchengeschichte. Berlin. Holl, K. 1915. Epiphanius: Anchoratus und Panarion, Haer. 1–33. Leipzig. Johnson, W. A. 2004. Bookrolls and scribes in Oxyrhynchus. Toronto; Buffalo, NY; London. King, K. 2003. What is Gnosticism? Cambridge, MA. Koenen, A. H. L. 1970. Ein griechischer Mani-Codex (P.Colon. inv. nr. 4780). Zeitschrift für Papyrologie und Epigraphik 5: 97–216. Koenen, A. H. L. and C. Römer. 1988. Der Kölner Mani Codex: Über das Werden seines Leibes. Opladen. Kristionat, J. 2013. Zwischen Selbsverständlichkeit und Schweigen: Die Rolle der Frau im frühen Manichäismus. Heidelberg. Layton, R. A. 2004. Didymus the Blind and his circle in Late-Antique Alexandria: Virtue and narrative in biblical scholarship. Urbana, IL; Chicago, IL; Springfield, MO. Lidzsbarski, M. 1925. Der Ginza oder das grosse Buch der Mandaer. Göttingen. Minale, V. M. 2013. Legislazione imperiale e Manicheismo da Diocleziano a Costantino: Genesi di un’eresia. Naples. Nongbri, B. 2005. The use and abuse of P52: Papyrological pitfalls in the dating of the fourth gospel. Harvard Theological Review 98: 23–48. Oulton, J. E. L. 1932. Eusebius: Ecclesiastical history. 2 vols. Cambridge, MA. Robinson, J. M. 2013. The Manichaean codices of Medinet Madi. Eugene, OR. Römer, C. 2009. Manichaeism and Gnosticism in the papyri. In The Oxford handbook of papyrology, R. S. Bagnall (ed.), 623–43. Oxford. Schaff, P. 1956. A Select library of the Nicene and Post-Nicene Fathers of the Christian Church IV: St. Augustine. The writings against the Manichaeans and against the Donatists. Grand Rapids, MI.

RETHINKING PERSECUTIONS

Schmidt, C. (ed.) and V. MacDermot (trans.). 1978. Pistis Sophia. The Coptic Gnostic Library. Leiden. Smith, G. 2016. A Manichaean hymn at Oxyrhynchus: A reevaluation of P.Oxy. 2074. Journal of Early Christian Studies 24 (1): 81–97. Stroppa, M. 2013. L’uso dei rotuli per testi cristiani di carattere letterario. Archiv für Papyrusforschung 59 (2): 347–58. Stroumsa, G. 1982. Monachisme et Marranisme chez les Manichéens d’Égypte. Numen 29 (2): 184–201. ———. 1986. The Manichaean challenge to Egyptian Christianity. In The roots of Egyptian Christianity, B. A. Pearson and J. E. Goehring (eds), 307–19. Philadelphia, PA. Van der Horst, P. W. and J. Mandfeld (trans. and comm.). 1974. An Alexandrian Platonist against dualism: Alex-

347

ander of Lycopolis’ treatise ‘Critique of the doctrines of Manichaeus’. Leiden. van Oort, J. 2016. Another case of human semen Eucharist among the Manichaeans? Notes on the ‘ceremony of the fig’ in Cyril of Jerusalem’s Catechesis VI. Vigiliae Christianae 70: 430–40. Vermes, M. (trans.). 2001. Hegemonius: The Acts of Archelaus. With introduction and commentary by S. N. C. Lieu and K. Kaatz. Turnhout. Villey, A. (trans. and comm.). 1985. Alexandre de Lycopolis: Contre la Doctrine de Mani. Paris. Williams, F. 1987. The Panarion of Epiphanius of Salamis: Book I (Sects 1–46). Leiden; New York; Copenhagen; Cologne.

VII UNIVERSAL AND LOCAL HISTORIES

SEVERAN CHRISTIANS BETWEEN THE ROMAN AND ARAB EMPIRES Phil BOOTH

In the immediate aftermath of the Arab conquest of Roman Egypt (AD 640–642), the victorious general ꜥAmr b. al-ꜥĀṣ is said to have issued a letter bidding Benjamin ‘the patriarch of the Coptic Christians’ —hitherto, according to tradition, in hiding from the persecutions of his Roman enemies—to return to Alexandria, and there ‘manage all his church, and the administration of his sect’. When the patriarch had returned, the same tale continues, he proceeded to an audience with ꜥAmr, in which the Arab general asked him to pray for the successful conquest of Pentapolis and the Maghreb. Then Benjamin pronounced a marvellous moral discourse, and ‘he revealed to him [ꜥAmr] certain things, and he departed from his presence honoured and revered’. ‘And all that the blessed father said to the emir ꜥAmr b. al-ꜥĀṣ’, our tale concludes, ‘proved true, and not a single letter went unfulfilled’ (History of the patriarchs; Evetts 1904–1915, 1: 495–97). This short vignette is embedded in the History of the patriarchs of Alexandria, a Medieval compilation of patriarchal biographies which belongs to the ‘Severan’ tradition, that is, to the tradition of those Egyptian Christians who resisted the doctrinal formulation of the Council of Chalcedon in 451, who adhered to the doctrine of Severus of Antioch (d. 538), and who would come to form the Coptic Orthodox church. Our vignette comes from a section of the compilation which depends upon a text completed in around 715 (den Heijer 1989, 121–24), and seems to encapsulate the Severans’ evolving conception of their church’s place between the Roman and Arab empires. Here, that church is a regional ethnic church, that of al-aqbāṭ or ‘the Copts’; and the persecution of ‘the Romans’ against that church is contrasted with the tolerance, and active patronage, of ꜥAmr. In this context, legitimate rule passes from Roman emperor to Arab governor, and thus the patriarch can perform, even if with some tension, the traditional role of the philosopher-cum-prophet at court, instructing the ruler and predicting the success of his conquests. The themes explored within the History of the patriarchs (the presentation of Severanism as the traditional and authentic faith of local Christians; the

identification of such Christians as ‘Copts’; and the cultural and political antagonism of these indigenous ‘Copts’ towards the foreign, persecuting ‘Romans’) have long influenced conceptions of the pre-Islamic period, although much recent work has now undermined its basic assumptions. Even when it is accepted that the discourse of Severan Christians living under Arab rule did not correspond to complexities on the ground, it is often nevertheless assumed that such discourse was monolithic in its emphases, that is, that all Severan authors sang from the same sheet. But outside of the mainstream of transmitted texts, we catch precious glimpses of the range of contemporaries’ responses both to the Roman past and to the Arab present. The patriarch Benjamin provides a case in point. In the so-called ‘Vulgate’ recension of the History of the patriarchs (as cited above) Benjamin, being in reported exile, is at a safe remove from the conquest, and there is no suggestion that he was complicit in it; but in an alternative recension, we discover the dramatic claim that the conquest occurred upon ꜥAmr’s being apprised of the patriarch’s fate, and a celebration of the Arabs, in far more explicit terms, as liberators of the orthodox from persecution (den Heijer 2000). Indeed, in both the Syrian Orthodox and Islamic traditions, it will later be said that Benjamin ‘surrendered’ Egypt to the Arabs on account of Roman persecution (Dionysius of TelMaḥrē; Palmer 1993, 158; Ibn ꜥAbd al-Ḥakam; Torrey 1922, 58). There is no point in attempting to assess the truth of such claims, and it is impossible to determine when they were first generated or modified. But these competing interpretations of the conquest at once suggest that Severan conceptions of their church’s relation to the successive empires of the Romans and the Arabs were neither static nor monolithic. In this chapter I want to pursue the same point through another marginal text, the extensive but neglected seventh-century Chronicle of John, bishop of Nikiu in the Nile Delta (fl. c. 690). As we shall see, the Chronicle shares with other contemporaneous texts a conception of the Arab conquest as a punishment for Chalcedonian sin, and an ethnic differentiation of anti-Chalcedonian ‘Egyptians’, who represent the

352

P. BOOTH

indigenous population, and Chalcedonian ‘Romans’, who are the representatives of the Roman empire. But in contrast to other contemporaries, John does not limit his political or cultural horizons to his own region, nor to his own church. Rather, he adopts the genre of the universal chronicle, situating Egypt’s recent past within a long narrative starting from creation and embracing all of the eastern Mediterranean. John’s Chronicle is thus caught between two impulses. Janus-like, it looks back, to the ecumenical traditions of the Christian Roman empire, at the same time that it looks forward, to an Arab empire in which some Christian sects were beginning to think of themselves as distinct peoples embedded in distinct regions. From Roman to Arab rule According to an older historiographical tradition, Late Antique Egypt was divided between two ethnic groups: Greeks or Romans and Egyptians or Copts. This ethnic division, in the same tradition, is mapped onto other spatial, social and cultural dichotomies: urban vs rural; northern vs southern; rich vs poor; hellenized vs unhellenized; etc. With the divisive Council of Chalcedon (451), these pervasive divisions were then widened in the doctrinal differentiation between (Roman, urban, northern, rich, hellenized) Chalcedonians and (Coptic, rural, southern, poor, unhellenized) anti-Chalcedonians. This dogmatic conflict, according to this view, further alienated the Copts from a Roman empire perceived to be heretical, encouraged the spread of Coptic ‘nationalism’, and thus eased the transition from Roman to Islamic rule (see e.g., Müller 1964; Orlandi 1986, 74–78). Some of the basic suppositions of this narrative still persist in various, prominent publications. In his stimulating The early Coptic papacy (2004) for example, Stephen Davis has approached post-Chalcedonian Egypt through the lens of postcolonial criticism. Davis’ vision conceives post-Chalcedonian Egyptian society to be polarized into two distinct ‘parties’: on the one hand, the colonialist, alien, pro-Roman, Chalcedonian minority; and, on the other, the colonized, indigenous, Coptic, anti-Chalcedonian majority. Here, anti-Chalcedonism is the natural, unflinching, confession of the indigenous ‘Copts’, and of the majority of Egyptian bishops; while the pro-Chalcedonian patriarchs of Alexandria are isolated accomplices in the political, socio-economic, and cultural colonialism of ‘the Byzantines’, whose policies aim ‘to displace and

disenfranchise the Coptic opposition’. In response to ‘imperialist discourses of power’ and ‘colonial social dynamics’, and through a complex political and cultural ‘resistance’, according to Davis, ‘a type of populist “national culture” was constructed, a shared community defined in terms of religious ethnic solidarity and created through the production of “minority discourse”’ (Davis 2004, esp. 86–88, 119–21). Davis’ perspective depends for the most part on formal written texts, not least those composed in antiChalcedonian circles in the aftermath of Chalcedon, and now extant, for the most part, in Coptic or in Copto-Arabic. This so-called ‘plerophoric’ literature celebrates the heroic resistance of anti-Chalcedonian bishops and ascetics in the face of persecuting emperors, presents anti-Chalcedonism as the natural and legitimate confession of indigenous Christians, and sometimes intimates an ethnic distinction of those ‘Egyptian’ Christians from their ‘Roman’ oppressors (see Johnson 1986). The dating and provenance of individual texts within this ‘plerophoric’ corpus is often far from secure, but we can perhaps be confident that at least some belong to the period before the Arab conquest. Nevertheless it is also certain that these themes intensified in anti-Chalcedonian literature of the Arab period, in which the focus on local heroes, and the ethnic differentiation of ‘Copt’ and ‘Roman’, are far more patent (Papaconstantinou 2006; Palombo 2016). Such texts provide, therefore, the evidential basis for the historiographical narrative described above. But we must, of course, be cautious not to assume that these discursive themes—produced and transmitted within a small religious elite, and performed and consumed in particular contexts—are somehow the prime constituents of the identities of the wider population, nor that religious identities must be coterminous or consistent with wider political, socio-economic and cultural ones. In recent decades various scholars, through turning to the period’s rich documentation (papyri, ostraca, etc.), have assaulted the basic assumptions of this narrative of burgeoning antagonism between ‘Romans’ and ‘Copts’, and the supposed spread of separatist, or ‘nationalist’, sentiment among the latter (e.g., Wipszycka 1992). This revisionism has shown that the existence in the late Roman period of two bounded, distinct, and self-conscious ethnic groups (Greek/Roman vs Egyptian/Copt) is a phantom; that the evident differentiation of Coptic and Greek in texts and documents cannot be mapped onto distinct socio-economic, political, and cultural identities, but is reflective of those

SEVERAN CHRISTIANS BETWEEN THE ROMAN AND ARAB EMPIRES

languages’ function and context; and that there is little to no suggestion that Severan Christians, however antagonistic to the religious policies of Constantinople, ever contemplated political secession from the Roman empire (see further: Clackson 2004; Papaconstantinou 2008; Fournet 2011; Palme 2012). Indeed, it has been demonstrated that various elements of Roman law, administration, and culture long persisted after the Arab conquest, so that the Severan discourse described above was developed in a context of significant continuities with the Roman past (Papaconstantinou 2009; Sijpesteijn 2013). Nevertheless, recent critics of the traditional narrative have tended to acquiesce in the assumption that there existed among the Severans, in the words of one eminent scholar, an ‘emerging totalising discourse of Coptic Christianity’ (Goehring 2012, 43, inter alia), that is, a single, hegemonic discourse which cast the Romans in the role of alien, heretical oppressors. This notion of a singular monolithic discourse seems a priori improbable; and indeed we can disprove it through turning to a contemporaneous text which Coptic tradition has nevertheless forgotten: the Chronicle of John of Nikiu. The Chronicle of John The Chronicle of John of Nikiu is a universal chronicle across 123 chapters, stretching from creation to the Arab conquest. Most of the text is much indebted to the earlier Chronicle of John Malalas, but is interspersed with episodes derived from the Greek ecclesiastical historians and from other, more local sources, one of which seems to be a liturgical calendar. The last recorded event occurs in 646/7 (or perhaps 643/4), but it is probable that our author is identical with the ‘bishop John of Nikiu’ whom two later sources present as active in the 690s (Mena of Nikiu, Life of Isaac [Porcher 1915, 354]; History of the patriarchs [Evetts 1904–1915, 5: 20, 22, 32]). We can perhaps conclude, then, that John decided to end his narrative soon after the conquest, perhaps for fear of offending the Arab regime; or perhaps for the more prosaic reason that he lacked a decent source on subsequent events. The Chronicle survives in three complete manuscripts, and one partial manuscript, in Ge’ez or Classical Ethiopic, all derived (as we discover from a colophon to the text) from a translation produced at the Ethiopian court in 1602. This translation in turn depended on an Arabic translation which appears to be

353

lost. In 1883 Hermann Zotenberg published an edition and French translation of the Ethiopic based upon two manuscripts, but he did not consult Paris BN d’Abbadie 31, which contains significant new readings and even some additional sentences. (Note that the inferior English translation of Charles [1916], whose line numbering I sometimes reproduce below, is also based on Zotenberg’s edition.) On the basis of evident residual words in both Greek and Coptic, Zotenberg suggested that the original text oscillated between passages in both languages, but such a bilingual text would be unparalleled, and since Coptic incorporates large numbers of Greek loanwords, it is probable but not certain that the original text was in Coptic. It can be demonstrated in numerous places that ambiguities or nonsenses within the Ethiopic result from the Arabic translator’s misunderstanding, or simple transcription, of words or constructions in Coptic (see e.g., Crum 1917). A Greek original is still not precluded; but it seems certain that the Arabic, at least, was based upon a Coptic text. The complex process of transmission is but the beginning of the modern interpreter’s problems. Difficult to explain, for example, is a large chronological lacuna between chapters 110 and 111 within the main text, which leaps from 610 and restarts, in medias res, in c. 640, omitting almost all of the reign of Heraclius (610–641). It is probable, I think, that the lacuna is original to John of Nikiu’s text, but it is nevertheless clear that the contents of the extant Ethiopic version are different from those of the original. Prefaced to the Chronicle are a set of chapter rubrics which serve as a table of contents. These sometimes refer to episodes which are absent from the corresponding chapters within the main text, suggesting that that text has at some point been abbreviated. At the same time, a significant dissonance between the rubrics and the main text towards the end of the Chronicle suggests —I have argued in detail elsewhere—that an editor has altered an earlier arrangement so as to account for a new source, a source covering events in Constantinople in the period 641–642 (Booth 2016). It is probable, though not certain, that the editor is John of Nikiu himself. I will return to the significance of this editing below. But suffice it to note here that in approaching the Chronicle we must bear in mind both the levels of translation through which we have the current Ethiopic text and the complex processes of composition and editing through which it was arranged in its earliest existence. Throughout the text, it is difficult to

354

P. BOOTH

disentangle the voice of John from the voices of his sources, which can often seem inconsistent. This should be borne in mind throughout our examination of the text. Interpretations of the Chronicle Although it is the sole historiographical source extant from seventh-century Egypt, the survival of the Chronicle in Ethiopic has entailed a remarkable dearth of reflection upon it—indeed, it has often been excluded from reviews of the period’s Coptic literature (e.g., Orlandi 1986). I would like here, however, to highlight the successive and contrasting contributions of three important critics: Antonio Carile (1981); Ewa Wipszycka (1992); and Jean-Michel Carrié (2003). In one of the few articles dedicated to the Chronicle, Carile contended that John of Nikiu was an advocate of a kind of proto-pharaonism. Carile pointed to those places in the opening part of the Chronicle in which John speaks of the ancient pharaohs, and contended that he had refrained from referencing their paganism, pointing to the existence of now lost Coptic traditions on the pharaonic past. For Carile, this celebration of the pharaohs, as well as a relative dislike for the Roman emperors, is indicative of John’s appreciation for an independent Egyptian culture, which in turn reflects the ‘intransigent anti-Chalcedonism’ and ‘anti-imperial dissent’ which characterize the attitudes of ‘the Copts’ in this period, and which point to the ‘emergence on the political plane of a separatist tension’ (Carile 1981, 119–23). In Carile’s interpretation, therefore, the Chronicle finds a neat place in the older ‘nationalist’ narrative described and critiqued above. Carile’s interpretation was challenged within a celebrated article of Wipszycka, as part of a wider critique of the notion of a late Roman, Coptic ‘nationalism’. In addition to questioning the posited existence of lost Coptic traditions on the pharaonic past (cf. Colin 1995), Wipszycka pointed to a perceived preponderance of local traditions and vignettes, and suggested that this ‘Egyptocentrism’ was the product both of traditional local patriotism, in distinction from ‘nationalism’, and of the ‘narrowing of geographical horizons’ which marked the wider transition from the ancient to the Medieval worlds (1992, 98–99). In a more recent contribution, let us also note, Wipszycka has added important further reflections on the Chronicle (2015, 420–23). Here she notes that the Chronicle is the sole historiographical source from Egypt which extends to

events outside of the church, but also contends that the schism of Chalcedonians and anti-Chalcedonians is ‘completely absent’, and speaks of John’s ‘relative indifference to doctrinal divisions’. ‘It is clear’, she concludes, ‘that the conflicts of such nature are, in his opinion, not important enough to determine the fate of the country’. In 2003, the tenets of Wipszycka’s earlier comments on the Chronicle were also reproduced in an important article of Carrié. Like Wipszycka, Carrié questions John’s investment in the notion of an ‘eternal Egypt’, or of Egypt as the cradle of civilization, even if the Chronicle sometimes departs from its main source (Malalas) to include some anecdotes of local interest. Carrié defines the Chronicle as ‘Egyptian’, not because of its nationalistic perspective on the distant past, but rather for its local focus. Carrié perceives in the progression of the narrative a slippage from universal to regional chronicle, so that in its later sections important, empire-wide events (the Heraclian revolt, the Arab conquest) are treated simply from an Egyptian perspective. Carrié thus speaks of a ‘provincialisation of the Egyptian cultural and mental universe … on the historiographical plane’, mirroring the political and territorial dislocation of the province from the Roman empire. He links this provincialization to a ‘national identity’, constructed not however with reference to a glorious, pharaonic past, or to political separatism, but ‘in a religious conflict which pitched Egyptians against others and in particular imperial power’, that is, the conflict over the Council of Chalcedon. For John, Carrié contends, the Arab conquest is the proof of divine displeasure at the empire’s recognition of Chalcedon, so that ‘[to] describe in detail the events in Egypt was at each instant to prove the error of the Chalcedonian faith and the justice of miaphysitism’. John’s Chronicle is therefore an attempt to understand the fate of his religious community in the transition from Roman to Arab rule. The horizons of the Chronicle How then should we negotiate a path through these competing interpretations? First, we can concur with Wipszycka and Carrié that Carile’s suggestion that John of Nikiu celebrates the pharaonic past is without foundation. It is true that in the first part of the text he sometimes departs from his central source (Malalas), using Herodotus in three chapters on the pharaohs (Chronicle 17–19; Zotenberg 1883, 31–32); and it is

SEVERAN CHRISTIANS BETWEEN THE ROMAN AND ARAB EMPIRES

also true that he includes some otherwise unique passages on local matters (e.g., the foundations of the Busirs and of Nikiu, at Chronicle 13, 31). But it also seems that he omits passages from Malalas on pharaonic Egypt (e.g., Chronicle 28; cf. Malalas 2.7–8; Thurn 2000, 22–24). Indeed, in his précis of the same source contained within the Chronicle’s earlier chapters, John follows no particular regional agenda, but embraces the entire oikoumene. Egypt is afforded no special status or position. John’s overwhelming frame of reference within the text is not the Egyptian but the Graeco-Roman—and, to a lesser extent, biblical—past as described in Malalas. This panoramic vision is indeed characteristic of the text until the Chronicle of Malalas gives out, at the end of the reign of Justinian (r. 527–65). At this point, as Carrié has observed, the text becomes more focused on Egyptian affairs, but it is doubtful that this is somehow representative of an overall narrowing within John’s cultural universe. It should be remembered that his basic instinct is to compose or compile a universal chronicle, and that he has evidently gone to some considerable effort in doing so. This genre is not thrust upon him. It is far more probable that this narrowing of geographical focus, then, reflects a dearth of available sources for wider affairs, as is inevitable for this kind of text, as the narrative approaches the lifetime of the author. That the same narrowing begins with the end of Malalas must be significant. It is moreover false that the Chronicle henceforth adopts a consistent provincial perspective. In the section leading from the end of Malalas up to the aforementioned chronological lacuna (610–640), there is a consistent reference to events in Constantinople and Antioch, even if political affairs in Egypt are dominant. But even after that lacuna, when the narrative focuses on the Arab conquest, Carrié’s argument effaces a large amount of material within chapters 119–20, and focuses on political machinations in Constantinople in the period 641–642. As I have pointed out above, I have elsewhere argued that this material belongs to a single source which has been integrated into a pre-existing scheme of the Chronicle, in order to take account of the new information, and that it is probable, but not certain, that this editor is John of Nikiu (Booth 2016). But if this is correct it demonstrates two important points: first, that the geographical horizons of the later sections of the Chronicle were determined through the available sources; and, second, that John himself (or, perhaps, a later editor) maintained an active interest in narrating

355

past affairs in the (now distant) Roman capital. All of our critics have, then, exaggerated the ‘Egyptocentrism’ of the text. When he began that text, John’s impulse was to compose a universal chronicle. This impulse remained until the end. Doctrine and ethnicity What of the contested status of doctrine in the Chronicle? First let it be said (pace Wipszycka) that Chalcedon, and the deposition of the Alexandrian patriarch Dioscorus, are here important events, presaged in an ominous darkness which descended on the day of Marcian’s accession (Chronicle 87.38). For the period from 451 to the death of Justinian (565), the secular aspects of John’s narrative depend for the most part on Malalas, but he interweaves selections from the latter with various episodes from ecclesiastical history. His sources for such episodes seem to be diverse, and most are unidentified, but one is no doubt a pro-Severan, Alexandrian patriarchal history covering the succession of Alexandrian patriarchs to the exile of Theodosius (r. 536–66). (This source is, let us note, distinct from those used for the biographies contained in the History of the patriarchs.) In the later sections covering the Arab conquest, moreover, one encounters a scattering of clear, anti-Chalcedonian comments. Thus, for example, a prediction of Severus of Antioch that no son would succeed his father on the throne while the emperor adhered to Chalcedon is twice repeated (Chronicle 116, 120); the conquests of the Arabs are several times presented as the product of Chalcedonian persecution (Chronicle 116, 117, 121); and the text at the end celebrates the return of the Severan patriarch Benjamin, with whom we began, from his exile (Chronicle 121). It would be too much, then, to conclude that John is uninterested in doctrine. It is nevertheless true that if, as Carrié suggests, Chalcedon is a pivotal event for John, there are some striking ambiguities. After the patriarch Theodosius, when the Chronicle turns to the post-Justinianic period, there is a remarkable dearth of material on ecclesiastical matters. Thus there is nothing, for example, on the establishment of the Severan episcopate under the patriarch Peter IV (r. 575–77), or the high-profile clashes of his successor Damian (r. 577–c. 606) with his Antiochene counterpart, Peter of Callinicum. This was a formative period for the Egyptian Severan church (Blaudeau 1999; Booth 2017), but neither patriarch is mentioned even in passing. In the sections dealing with

356

P. BOOTH

the Arab conquest, moreover, the Chronicle casts the Chalcedonian patriarch Cyrus in an unexpectedly positive light, while nevertheless calling him, in two places, a persecutor (Chronicle 116, 121), as per his wider reputation within Coptic literature. Both of these problems—that is, the absence of the Severan patriarchs after Theodosius, and the positive presentation of Cyrus—can, however, be mitigated if we once again consider the text’s composition. Since John includes material on earlier Severan patriarchs, we must suppose that the absence of Peter IV and his immediate successors reflects the absence of a decent patriarchal source for that period (however surprising that might seem); while the material on Cyrus seems to be based upon a pre-existing, apologetic source, the general tone of which John has, for one reason or another, not troubled to alter (Booth 2016). There can be little doubt, then, that doctrine is important for John, despite some surprising omissions and inclusions. But we can also detect within the Chronicle, as Carrié has further suggested, hints of the process through which John’s church was assuming, within Severan discourse, an ethnic dimension. Within the later sections of the text there is a pervasive (but hitherto little remarked) differentiation—as in other Coptic texts of this period—between ‘Romans’ (rom) and ‘Egyptians’ (mǝsrāwyān). We must bear in mind that the extant Ethiopic text is the product of a complex transmission, which might have introduced some significant distortions. But it is not impossible that these terms represent an original distinction, in Coptic, of rōmaioi and nrmnkēme (via Arabic rūm and miṣriyūn). What do these terms mean? For the most part of the text ‘Roman’ is used to designate the inhabitants of Rome or of the Roman empire, in particular in the context of foreign campaigns. In the sections covering the Arab conquest, however, it seems instead to designate representatives of the Roman state: administrators and soldiers. ‘Egyptians’, in turn, means for the most part of the text ‘inhabitants of Egypt’, and is used most often in the context of the pharaonic period. With the description of the Arab conquest, however, the usage again intensifies, and appears here to designate the general, civilian, population. Within the text, therefore, the conquest marks the introduction of a stricter ethnic differentiation of ‘Roman’ and ‘Egyptian’, in which the former are combatants, and the latter non-combatants. In at least one place, moreover, this differentiation also assumes a religious dimension. The final chapter

of the text memorializes the patriarch Benjamin, with whom we began: And Abba Benjamin, patriarch of the Egyptians (mǝsrāwyān) [the text records] entered the city of Alexandria, thirteen years after his flight from the Romans (rom). He went to the churches and inspected them all. And everyone began to say, ‘This persecution and the conquest of the Muslims are on account of the oppression of the emperor Heraclius and the affliction of the orthodox through the patriarch Cyrus. For this reason the Romans were destroyed and the Muslims gained dominion over Egypt’ (Zotenberg 1883, 219–20).

Here, then, the religious distinction of anti-Chalcedonians and Chalcedonians also corresponds to an ethnic distinction of Egyptians and Romans. Egypt between empires All this might lend the impression that the Chronicle conceives the Arab conquest as liberation from RomanChalcedonian oppression—but this would be misleading. In the later sections of the text, the Chalcedonian emperors from Justin I (r. 518–27) onwards are for the most part presented with disapproval, although John’s somewhat indiscriminate oscillation between sources again introduces some inconsistencies (see, for example, a passage where he lavishes praise on Justinian for his protection for the poor, his restoration of the empire, and his great faith [Chronicle 90.49–51, 54; Zotenberg 1883, 155]). When the narrative reaches the Chalcedonian Tiberius (r. 578–82), however, it exalts the emperor for his various virtues, but lauds in particular his lack of persecution, and the favour which he showed to the anti-Chalcedonians. This presentation of the reign of Tiberius—whose renunciation of persecution is also recorded in other Severan texts—reflects a real historical situation, in which the emperor’s relative disengagement from doctrinal politics in effect sanctioned the establishment in Egypt of a Severan patriarch and episcopate (Booth 2017). But it suggests that, for John, the emperors at Constantinople are not irredeemable heretics whom God has punished for their simple deviation from the faith. The true spur to divine anger is not adherence to Chalcedon, but the persecution of the orthodox. Indeed, the Chronicle claims that under the tolerant regime of Tiberius, God caused peace to prevail throughout the empire, and the Persians and others to be conquered. In the progression of the narrative, however, this clear equation between religious tolerance

SEVERAN CHRISTIANS BETWEEN THE ROMAN AND ARAB EMPIRES

and imperial success—and thus also between religious persecution and imperial defeat—stands in contrast to the situation under Heraclius where, as we have seen above, the Arab conquest is several times attributed to divine anger at the persecution of the orthodox. In one perspective, then, Arab rule is a scourge sent from God to punish the persecutors; but that rule itself then comes also to be contrasted with the previous regime. In the aforementioned final chapter of the text describing the return of Benjamin—in which, as we have seen, a clear association is made between Roman religious persecution and Arab success—the text continues to refer to ꜤAmr’s burdensome taxation, but nevertheless notes with approval that, ‘he took nothing from the possessions of the churches, and he committed no plunder or act of pillage, but protected them for the entire length of his reign’ (Zotenberg 1883, 220). The Arab regime might, then, be burdensome in terms of taxation, but can nevertheless be celebrated—much like the reign of Tiberius—for its religious tolerance. Can we conclude, then, that the advent of Arab dominion is for John a positive development? This seems doubtful. The Chronicle sets out the terrible violence of the conquest in vivid and undaunted detail; it describes the conqueror, ꜥAmr, as a treacherous barbarian (Chronicle 120.36; Zotenberg 1883, 215); and it twice refers to, and laments, the conversion of some Christians to Islam (Chronicle 114.1, 121.10; Zotenberg 1883, 202, 221). Indeed the overwhelming sense is that the end of Roman (and thus Christian) rule, though attributable to persecution, is in fact something to be regretted, even abhorred. It should be remembered that John belonged to a generation of Egyptians who had, for the first time in centuries, never experienced inclusion in the Roman empire. Arab rule was a fait accompli, and John gives no indication that Roman reoccupation was a realistic expectation. But the Roman empire nevertheless continues to exert a powerful pull on his interest, on his sense of the relevant past. Besides his dependence on a range of Roman sources, and the considerable effort expended on reporting events, both secular and sacred, within the Roman empire, John writes in a Roman idiom, and his geographical vision is, as we have seen, throughout determined by the borders of the former Roman empire. His perspective on the Roman empire, its past and its present, is therefore complex. On the one hand, the Chronicle connects with other contemporaneous Coptic texts in drawing a religious and ethnic distinction between ‘Romans’ and ‘Egyptians’, and in casting the former as persecutors of

357

the latter; but, on the other, it also expends a remarkable effort in recounting for its readership the diverse events which occurred across several centuries of Roman rule, and which continue, for John, to be relevant. This tension should not be dismissed or reduced. It is indicative of the conflicts of perspective which marked the transmission from Roman to Arab rule within Severan circles, and of contemporaries’ attempts to conceptualize their place between the two empires. Conclusion In the period from 690 to 692, according to his extant Coptic Life, the Severan patriarch Isaac used to attend upon the ‘king’ (rro) of the Saracens, the Arab governor ꜥAbd al-ꜥAzīz ibn Marwān. The governor, we are told, had held Isaac’s predecessor John in great affection, and after John’s death continued to court his successor. Having once beheld a miraculous light surrounding Isaac at the altar, ꜥAbd al-ꜥAzīz considered him a prophet, began to call him ‘patriarch’, and included him within his permanent entourage. In a series of anecdotes, the Life describes various miracles of Isaac which the governor witnessed, and which ever deepened both his devotion to the patriarch and his patronage of churches and monasteries. The Life describes, then, the more or less harmonious relationship which existed between the Severan patriarch and the Arab governor, who has here even assumed the Coptic title, rro, once reserved for the Roman emperor. The author of the Life of Isaac is one Mena, bishop of Nikiu, who is no doubt our chronicler’s successor in that office. John of Nikiu himself appears within the text, where he is placed in Isaac’s entourage following his election as patriarch, and described as apotritēs (sc. vicar) for Upper Egypt (Life of Isaac; Porcher 1915, 354). Both John and Mena belong, then, to a circle of literate clerics within the Severan patriarchate, writing in c. 700 and producing original texts, it seems, in Coptic. Within this circle we can also include one George the Archdeacon, who authored (c. 715) a Coptic compilation of patriarchal biographies from Cyril to Simon I (412–700), and which in turn was used as a basis for those sections in the aforementioned History of the patriarchs (den Heijer 1989, 121–24, 142–43). From this we ascertain that he was a personal acquaintance of John of Nikiu (Evetts 1904–1915, 5: 20). Here, the concord said to characterize the relationship of Benjamin and ꜥAmr—as described in the introduction to this chapter—becomes programmatic

358

P. BOOTH

for the subsequent relations of ꜥAbd al-ꜥAzīz with the patriarchs John and Isaac, as also remembered in Mena’s Life of the latter. Besides their shared celebration of Arab patronage over the Severan church, the Life of Isaac and the History of the patriarchs contain several further themes which we have also encountered in the Chronicle of John: for example, the ethnic and confessional distinction of ‘Egyptians’ and ‘Romans’; and the remembrance of the Chalcedonian patriarch Cyrus as a persecutor. Nevertheless, it is important that we also distinguish the preoccupations of John from those of his contemporaries and associates. Both the Life of Isaac and the History of the patriarchs are Egyptocentric texts, in which events outside of Egypt (often garbled and disordered in the latter) are of interest only in so much as they impact upon it. Both, moreover, are ecclesiastical texts, focused on the Severan patriarchate and for the most part disinterested in secular affairs. Neither point is true of the Chronicle. Indeed, those scholars who have emphasized the Chronicle’s ‘Egyptocentrism’, or have seen a sectarian agenda as its guiding principle, risk collapsing it into the texts of John’s contemporaries. Thus it might well seem that Severan churchmen did indeed produce a monolithic ‘totalizing’ or ‘hegemonic’ discourse in this period, in which the effective past is reduced to the level of the local, the Roman empire cast as heretical and oppressive, and Arab rule celebrated for its relative moderation. I have tried in this chapter to counter this impression. It is natural, when assessing the Chronicle, that commentators have tended to focus on its later, less derivative sections, for which the source material is not otherwise extant. But in assessing John’s political and cultural horizons, it is imperative to take account of the entire text. Thus I have argued that the Chronicle embraces the same ecumenical vision as its main source, Malalas, even in the later sections, where the source material is more local and more limited; and that its interests throughout encompass both the secular and the sacred. Within its narrative, the long centuries

of the Roman empire loom large, and the Arabs’ subjugation of Egypt—whose inhabitants John and some contemporaries had come to consider a distinct people, belonging to a distinct ethnic church—was not something to be revelled in so much as regretted, attributed to the intolerance of Chalcedonian persecution. Although the Chronicle is for us a precious source —in particular for the seventh century, for which it often proves an invaluable guide (Booth 2011; 2013; 2016)—it was not to prove as popular as other texts produced within its author’s immediate circle. We know that it was at some point translated into Arabic, and it is not impossible that the Arabic version is still to be discovered. Nevertheless, the Chronicle had no demonstrable impact on later Coptic or Copto-Arabic authors, and if it were not for the commissioning of a further translation at the Ethiopian court in 1602, the text would remain unknown. A recent book has suggested that the text has been forgotten in the Coptic tradition because its vision of the Arab conquest is too conflictual and too violent, and that it therefore proved discordant with the dominant discourse of the Severan church which, as we have seen, emphasized instead the comfortable coexistence of anti-Chalcedonian Christians and their new Arab masters (Mikhail 2014, 32). I would suggest, however, that the divergence between the Chronicle and other contemporaneous texts might be more fundamental. Where others contented themselves with a regional focus centred on the Severan patriarchate, John adopted the Late Antique tradition of the universal chronicle, with all its demands towards chronological, geographical and topical breadth. Although the Chronicle did not, it seem, prove so resonant as to be read or copied to a significant extent in subsequent generations of Coptic historians, the fortunate survival of the text in Ethiopic constitutes a precious witness to a discordant voice within the Severan church of the post-conquest period, and to a pivotal moment in which contemporaries’ understanding of the effective past, of the past that mattered, was transforming.

Bibliography

———. 2013. The Muslim conquest of Egypt reconsidered. In Constructing the seventh century, C. Zuckerman (ed.), 639–70. Travaux et Mémoires 17. Paris. ———. 2016. The last years of Cyrus, patriarch of Alexandria († 642). In Mélanges Jean Gascou, J. Fournet and A. Papaconstantinou (eds), 509–58. Travaux et Mémoires 20.1. Paris.

Blaudeau, P. 1999. Le voyage de Damien d’Alexandrie vers Antioche puis Constantinople (579–580), motivations et objectifs. Orientalia Christiana Periodica 63: 333–61. Booth, P. 2011. Shades of Blues and Greens in the Chronicle of John of Nikiou. Byzantinische Zeitschrift 104: 555–601.

SEVERAN CHRISTIANS BETWEEN THE ROMAN AND ARAB EMPIRES

———. 2017. Towards the Coptic church: The making of the Severan episcopate. Millennium 14: 151–89. Carile, A. 1981. Giovanni di Nikious, cronista BizantinoCopto del VII secolo. Felix Ravenna 121: 103–55. Carrié, J.-M. 2003. Jean de Nikiou et sa Chronique: Une écriture ‘égyptienne’ de l’histoire? In Évenément, écrit, histoire officielle: L’écriture de l’histoire dans les monarchies antiques, N. Grimal and M. Baud (eds), 155–72. Paris. Charles, R. H. 1916. The Chronicle of John, Bishop of Nikiu. London. Clackson, S. 2004. Papyrology and the utilization of Coptic sources. In Papyrology and the history of early Islamic Egypt, P. M. Sijpesteijn and L. Sundelin (eds), 21–44. Leiden. Colin, G. 1995. L’Égypte pharaonique dans la chronique de Jean, évêque de Nikiou. Revue d’égyptologie 46: 43–54. Crum, W. E. 1917. Review of Charles 1916. Journal of Egyptian Archaeology 4: 207–9. Davis, S. J. 2004. The early Coptic papacy: The Egyptian church and its leadership in Late Antiquity. Cairo. den Heijer, J. 1989. Mawhūb ibn Manṣūr ibn Mufarriğ et l’historiographie copto-arabe: Étude sur la composition de l’Histoire des patriarches d’Alexandrie. Corpus Scriptorum Christianorum Orientalium 513. Leuven. ———. 2000. La conquête arabe vue par les historiens coptes. In Valeur et distance: Identités et sociétés en Égypte, C. Décobert (ed.), 227–45. Paris. Evetts, B. (ed.). 1904–1915. History of the patriarchs of the Coptic church of Alexandria. Patrologia Orientalis 1, 5, 10. Paris. Fournet, J.-L. 2011. The multilingual environment of Late Antique Egypt: Greek, Latin, Coptic and Persian documentation. In The Oxford handbook of papyrology, R. S. Bagnall (ed.), 418–51. Oxford. Goehring, J. E. 2012. Politics, monasticism and miracles in sixth century Upper Egypt. Tübingen. Johnson, D. W. 1986. Anti-Chalcedonian polemics in Coptic texts, 451–641. In The roots of Egyptian Christianity, B. A. Pearson and J. E. Goehring (eds), 216–34. Philadelphia, PA. Mikhail, M. 2014. From Byzantine to Islamic Egypt: Religion, identity and politics after the Arab conquest. London. Müller, C. D. G. 1964. Die koptische Kirche zwischen Chalkedon und dem Arabereinmarsch. Zeitschrift für Kirchengeschichte 75: 271–308.

359

Orlandi, T. 1986. Coptic literature. In The roots of Egyptian Christianity, B. A. Pearson and J. E. Goehring (eds), 51–81. Philadelphia, PA. Palme, B. 2012. Political identity versus religious distinction? The case of Egypt in the later Roman empire. In Visions of community in the post-Roman world: The West, Byzantium and the Islamic world, 300–1000, W. Pohl et al. (eds), 81–100. Farnham. Palmer, A. 1993. The seventh century in the West-Syrian chronicles, A. Palmer (ed. and trans.). Liverpool. Palombo, C. 2016. Constantinople and Alexandria between the seventh and eighth centuries: The representation of Byzantium in Christian sources from conquered Egypt. In From Constantinople to the frontier: The city and the cities, N. S. Matheou et al. (eds), 243–59. Leiden. Papaconstantinou, A. 2006. Historiography, hagiography and the making of the Coptic ‘Church of the Martyrs’ in early Islamic Egypt. Dumbarton Oaks Papers 60: 65–86. ———. 2008. Dioscore et le bilinguisme dans l’Égypte du VIe siècle. In Les archives de Dioscore d’Aphrodité cent ans après leur découverte: Histoire et culture dans l’Égypte byzantine, J.-L. Fournet (ed.), 77–88. Paris. ———. 2009. ‘What remains behind’: Hellenism and Romanitas in Christian Egypt after the Arab conquest. In From Hellenism to Islam: Cultural and linguistic change in the Roman Near East, H. M. Cotton et al. (eds), 447–66. Cambridge. Porcher, E. (ed.). 1915. Vie d’Isaac, patriarche d’Alexandrie de 686 à 689, écrite par Mina, évêque de Pchati. Patrologia Orientalis 11: 299–390. Paris. Sijpesteijn, P. 2013. Shaping a Muslim state: The world of a mid-eighth century Muslim official. Oxford. Thurn, I. 2000. Ioannis Malalae Chronographia. Berlin. Torrey, C. (ed.). 1922. The history of the conquest of Egypt, North Africa and Spain, known as the Futūh Miṣr of Ibn Abd al-Ḥakam. New Haven, CT. Wipszycka, E. 1992. Le nationalisme a-t-il existé dans l’Égypte byzantine? Journal of Juristic Papyrology 22: 83–128. ———. 2015. The Alexandrian church: People and institutions. Warsaw. Zotenberg, H. (ed.). 1883. Chronique de Jean, évêque de Nikiou. Paris.

THE MUSLIM ELITE OF EARLY ISLAMIC EGYPT Hugh KENNEDY

The first two centuries of Muslim rule in Egypt, from the Arab Muslim conquest of 641–642 to the assumption of power by Ahmad b. Tulun in 868 and the beginning of the short-lived but important Tulunid dynasty, present something of a challenge to the historian (Kennedy 1981; 1998; Sijpesteijn 2013). Late Antique Egypt offers the opportunity to discuss social structures and religious divisions in a way which cannot be done for any other area of the ancient world at this time. The history of the Apion family, for example, provides a portrait of a late Roman aristocratic family, its properties and its business dealings which has no parallel elsewhere. Equally the emergence of a distinctively Coptic Christianity and the vibrant monastic culture which went with it was a cultural development of major and long-term significance. The ‘political’ history hardly figures in this discourse: the governors of the province are largely unknown and neglected and it never became the power base of ambitious generals or pretenders to imperial power. When the rule of Phokas was challenged and eventually overthrown by Heraclius at the beginning of the seventh century, that challenge was mounted from the poorer and less populated, but more militarized, provinces of Africa rather than Egypt. The Persian invasion and conquest of 614 produced no significant opposition in the province and none of the Persian generals who struggled to take over the Sasanian empire after the murder of Chosroes II in 628 seems to have considered making Egypt his power base. In the aftermath of the rapid Arab Muslim conquest of 641 there follow two centuries when Egypt appears to be no more than a sleepy province of the caliphal empire, ruled successively from Damascus and Baghdad by the Umayyads (661–750) and Abbasids (750–945) and having no local political agency. With the exception of the first phases of the mosque of ῾Amr in Old Cairo, the Muslim conquerors produced no major monuments for art historians and tourists to admire. Compared with the richness of architectural production in the two centuries which followed the Tulunid takeover and the later Fatimid conquest, this is indeed a fairly unimpressive record which in itself may account for much of the neglect of this period in the general history of Egypt.

In this chapter, I will revisit this period and suggest that early Islamic Egypt did have a distinctive political identity and that the local elites negotiated an important role for themselves in the administration of the wider caliphate and there are comparisons to be made between Islamic Egypt’s relationship with the Ottoman empire and that of ancient Egypt with the Roman empire. The governors of Egypt In theory the province, like other areas of the caliphate, was ruled by a governor, wālī, who was appointed and dismissed by the caliph. The system of wālīs had emerged at the time of the first Muslim conquests when the commanders of the victorious armies were put in charge of the administration of the areas they had conquered, including the collection of taxes, the reward and payment of the Muslim warriors and the leading of prayer in the main mosque on Fridays. At times we hear of another top official with the Arabic title of ῾āmil. This title is often synonymous with wālī but it also carries the connotation of tax collector and financial administrator. At certain times when the government of the caliphs, whether in Damascus or Baghdad, was particularly keen to increase the revenues derived from Egypt, a separate ῾āmil would be appointed to ensure that as much as possible of the revenues reached the capital (Kennedy 1981). At first glance the powers of the wālī were pretty impressive. He had an absolute authority to lead prayers, command the local jund (militia) and arrange for the collection of taxes and the forwarding of the profits of taxation to the central government. Imperial control looks like a top-down, authoritarian system, even an ‘oriental despotism’. The reality of the governor’s position was usually very different from this image of unfettered power. The governor had to negotiate between two powers with fundamentally different aims and priorities. The first of these was the government of the caliphs, either the Umayyads in Damascus (or later, during the reign of Hishām [724–743] in Rusafa in northern Syria) and, after 750, the Abbasids in Baghdad. The objective of the caliphs and their administrators was to extract the maximum possible

362

H. KENNEDY

revenue from the province and transport it to the capital where it could be used to finance general caliphal expenditure. The figures that we have suggest that, after Iraq, Egypt was the most important contributor to the budget of the caliphs in Baghdad, a vital resource (el-Ali 1971). The revenues of Egypt were also exported in another direction. It seems that for much of the Umayyad and early Abbasid period, the revenues collected in Egypt were used to support the very large military establishment in the province of Ifriqiya (essentially modern Tunisia with western Libya including Tripoli and parts of western Algeria). Unlike Egypt, this was an area where caliphal control was constantly threatened by the rebellions of warlike Berber tribes and later, in the Abbasid period, by Kharijite ‘heretic’ Muslims, who were largely drawn from these same Berber tribes now converted to Islam. To counter these, the Abbasids were obliged to maintain a jund in the province which was much larger than could be supported by the revenues collected in it. Financial support from the taxation of Egypt was essential. On the other side of the equation that the governors had to balance was the local Egyptian Muslim elite, represented above all by the local militia, the jund, and their leaders, the dozen or so families referred to in the Egyptian sources as the wujūh (the nobles).1 Their interest was to ensure that as much of the revenues of the country as possible were kept in the province, primarily of course to pay their salaries (῾aṭā). This side of the equation was made more complex by the threat, and occasionally the reality, of rebellions by the Coptic Christians, almost certainly the majority of the population until the eleventh century (Lev 2012). In the late eighth and early ninth century these largely rural people raised a number of armed rebellions in protest against the levels of taxation demanded by the government in Fusṭāṭ. The only force the governor had at his disposal to put down the disturbances when they occurred and discourage the Copts from revolting in the first place was the jund. So the governor was squeezed between the caliphal administration, which would put pressure on him and even dismiss him if he

1

In Iraqi and other Eastern sources these notables and tribal leaders are described as ashrāf (sing. sharīf).

did not forward sufficient sums to the capital, and the jund, who resented the export of revenues to which they fundamentally believed they were entitled. When in 784 the governor Mūsā b. Muṣ῾ab, who had been appointed by the Abbasid government specifically to extract more money from the province, failed to look after the interests of the jund they deserted him in the face of a Coptic rebellion and returned to their homes in Fusṭāṭ. Mūsā paid with his life for his failure to respect local interests. Local elites So, who were the local elites (Kennedy 1981, 31–38; Mikhail 2014, 136–59)? The first point to note is that the political elites were composed of Muslims who were, or at least claimed to be, of Arab descent. There were certainly Christian and no doubt Jewish elite families who played very important roles in the administration, but they did not dominate the political scene in Fusṭāṭ in the same sort of way. This was partly because they did not serve in the military and partly because they did not feel that they had a right to enjoy the proceeds of the taxation of the province. In general, the historians to the universal caliphate, notably al-Ṭabarī (d. 923) (de Goeje et al. 1879–1901) in his History of the prophets and kings, have comparatively little to say about Egypt and if we were wholly dependent on them we would have a very limited understanding of the structures and forces which drove local politics. Fortunately we have two local chroniclers who give us a much fuller picture. The earlier of these is ῾Abd al-Raḥmān b. ῾Abd Allah b. ῾Abd al-Ḥakam (d. 871) (Torrey 1922). He came from a prominent wujūh family of jurists and religious scholars, proud of the Arab heritage of Egypt. His book Futūḥ Miṣr (Conquests of Egypt) is essentially a collection of traditions about Egypt, its conquest by the Muslims and its government by the Arab elite. It was written at a time when this elite was rapidly losing its power in the country and his own family were effectively ruined when accused,

THE MUSLIM ELITE OF EARLY ISLAMIC EGYPT

rightly or wrongly, of embezzlement in 851. In many ways his book is both a celebration of and lament for the wujūh and the Arab domination of Egypt. The second author is Muḥammad b. Yūsuf al-Kindī (d. 961) (Guest 1912). Like Ibn ῾Abd al-Ḥakam, he came from an old established Egyptian Arab family who claimed that their ancestors had come to the province at the time of the original Arab conquest. In contrast to the rather rambling work of Ibn ῾Abd al-Ḥakam, his account of the governors and judges of Egypt, Kitāb al-Wulah wa-l-quḍāh, is a meticulous chronological account, short, perhaps, on entertaining anecdotes but full of information not just about governors but also the chiefs of police (aṣḥāb al-shurṭa) and the judges. His work enables us to build up a detailed prosopography of the early Muslim ruling class in the country. The main element in the local elite was the jund, the militia who were the descendants of the army of the first Islamic conquests and those who had followed them in the subsequent three decades. We are told that the number involved in the initial conquest was some 16,000, which had swelled to 40,000 by the reign of the first Umayyad caliph Mu῾āwiya b. Abī Sufyān (661–680) (Butler 1978, 226). It is customary among modern historians to dismiss the numbers quoted in early medieval sources as unreliable, even fanciful, but there are good reasons to treat these figures with respect; firstly because they are by no means improbable and secondly because we know that early Muslim administrations kept records of the numbers of soldiers because they paid them on a regular basis. Even if the figures are only approximate, they suggest that the Arab Muslim soldiers who formed the early elite were very small in number compared with the overall population of the province, which can hardly have been less than 3 million (Kaegi 1998, 34). These conquerors were not, by and large, from Bedouin backgrounds. It is easy to imagine that the Arab invasions saw nomadic tribesmen sweeping through and ravaging the agricultural landscapes in the Nile valley and the Delta. In fact, so far as we can tell, most of the early conquerors came from the Yemen and were members of well-established South Arabian tribes such as Kinda, Himyar and Khawlān. They had mostly lived in small towns and villages and cultivated the land on the terraced slopes of the well-watered mountains of the area. The tribe of Khawlān, recorded in central Yemen in the early Islamic period, can still be found in the same area today, testimony to their settled

363

lifestyle and stability of occupation. The leader of the conquest, ῾Amr b. al-῾Āṣ, came from the merchant aristocracy of the Prophet’s tribe of Quraysh and owned valuable agricultural estates between Mecca and the neighbouring town of Ta’if. It was probably not until the Qaysi Arabs were brought to the eastern fringes of the Delta in the reign of Hishām (r. 724–743) that a significant number of Arabs of Bedouin origin came to Egypt and even then they remained a marginal minority among the Arab community. These South Arabians would have had no difficulty in adapting to city life in Fusṭāṭ (Old Cairo) or Alexandria, or in appreciating and valuing the agricultural resources of the country they had so swiftly overrun. The vast majority of these Arab Muslims settled in Fusṭāṭ (Kubiak 1987; Gayraud 1998). Here, outside the walls of the Trajanic fortress of Babylon, they established themselves in quarters which took the name of their tribal identities. They seem to have lived in areas which had previously been waste or agricultural lands: there is no tradition of them taking or occupying the homes of the existing inhabitants. It was here that they built the first major mosque in the country, known to this day as the mosque of ῾Amr in memory of the leader of the initial conquests. The exact age of the present structure is not entirely clear but it is likely that parts of it date back to the late eighth and early ninth centuries; little, if anything, survives of the work of the first Islamic century (Creswell 1932–40, 1: 96–100, 2: 171–96). From the time of the initial conquest, there was also an Arab garrison in Alexandria. Originally they served on rotation from Fusṭāṭ but by the Umayyad period they were permanently stationed in the seaside city. There seem to have been few, if any, garrisons outside these two cities. As in other newly conquered provinces, the Muslims were settled in towns, newly founded on previously uninhabited sites. These towns were designated by the term miṣr (pl. amṣār). The distinguishing feature of the miṣr, as opposed to other cities, was that it was the centre to which taxes, principally the land tax or kharāj, was brought and where it was paid out in salaries and pensions (῾aṭā᾿) to the soldiers. The names of those to be paid were recorded in lists called dīwāns. In time the miṣr established at Old Cairo gave its name to the whole country and the term Misr is now the generally used Arabic name for Egypt. This system had important and long-lasting consequences. It seems likely that the names of those to be paid were recorded in Arabic from a very early stage: transcribing all these Arabic

364

H. KENNEDY

names into Greek or Coptic would be clumsy and give rise to mistakes and confusions. This in turn encouraged the Arabization of the administration. The system also required large quantities of coined money. In the first decades after the conquest it seems that some taxes were paid in kind (wheat, oil etc.), but by the later Umayyad period, if not before, wages were always paid in cash. This in turn led to the monetization of the economy, especially in the cities where the soldiers spent their money on food, goods and services (Kennedy 2002). Having your name inscribed in the dīwān and being entitled to a salary was a jealously guarded privilege justified by descent from the early conquerors. In the decades after the initial conquest there was certainly continued immigration from Arabia and Syria, but by the end of the seventh century, the dīwān was effectively closed to new arrivals, thus consolidating the rights of this military elite. Far from the authorities using conscription to collect unwilling recruits, as happened for example, in late Ottoman times, the early Muslims strove to maintain their positions in the diwān, even fighting other groups to maintain their monopoly of the revenues if necessary. The leadership of the jund, politically and militarily, lay with a small number of families, no more than a dozen in all, who were mostly descendants of the Yemeni tribal leaders of the time of the initial conquests. They formed the group known as the wujūh referred to above. We can trace the succession in some of these families over a long period. The family of Mu῾āwiya b. Ḥudayj al-Tujībī (the Tujīb were a branch of the great South Arabian tribe of Kinda) can be traced through five generations from the time of the first conquests to the civil strife which followed the death of the caliph Harun al-Rashid in 809, each generation holding office or leadership positions. As far as we can tell, these families were not country landowners or merchants. Their power and wealth came from their access to and control of the payment of salaries. If they were to lose it, they would lose all their power and status. They were also a purely Egyptian elite. As far as we can tell, they never served in the armies of the caliphs in other provinces, nor did they go to Damascus or Baghdad to seek their fortunes. Members of the wujūh were seldom appointed as governors of the province, for these were almost always outsiders appointed by the Umayyad or Abbasid caliphs. Instead they exercised power through the office of chief of police (Ar. ṣāḥib al-shurṭa). This was an

office known from other areas of the Muslim world, including Baghdad itself, but in Egypt it had a unique importance. Governors came and went, often, especially in Abbasid times, with great frequency. Generally, they had never visited Egypt before, at least as far as we know, and they had no contacts in the province. They were obliged to depend on the chief of police to advise them and, especially, to secure the respect and obedience of the jund. The governor could appoint and dismiss the chief of police but he always had to choose a man from one of the elite families. Chiefs of police enjoyed longer spells in office than the governors and might serve a number of governors in turn. This meant that while in theory the governor, as the caliph’s representative was in absolute control, in reality there was a strong local representation in government, a distinctively Arab-Egyptian presence. In this respect, we can make interesting comparisons between Egypt under Abbasid rule in the eighth century and Egypt under Ottoman rule in the eighteenth century. In both cases the country was just one province in a vast empire and governors were sent out from the capital (Baghdad, Istanbul) with theoretically absolute powers. But in both cases, the governors were outsiders with no power base or natural supporters among the Egyptians and they only enjoyed very short terms of office. Just as the Abbasid governors were forced to depend on the jund and the local notables, so the Ottoman governors were forced to cooperate with and rely on the local Mamluk elite. If they failed to respect the rights of the local notables, their terms of office would be short and ignominious. Of course, not all governors were in the same position or enjoyed the same status. The first Muslim governor of the province was naturally ῾Amr b. al-῾Āṣ who had led the first armies of the conquest. At first his position was unassailable and he continued to administer the lands and revenues of which he had taken possession. But even he, with his vast prestige and reputation for competence, was soon challenged by the third caliph ῾Uthmān (r. 644–656) who sought to bring the revenues of the provinces of the still-expanding caliphate firmly under the control of the caliphate in Medina, and ῾Amr was replaced by 645 by ῾Abd Allah b. Sa῾d b. Abī Sarḥ, more a financial administrator than a charismatic leader. It was a foretaste of struggles which were to recur frequently between governors and caliphs. It is worth remarking in this context that ῾Amr put up no resistance to his dismissal. Despite his pivotal role in the conquest and his undoubted prestige

THE MUSLIM ELITE OF EARLY ISLAMIC EGYPT

among the Muslims newly established in Egypt, he seems to have made no attempt to use Egypt as a power base to challenge the caliphal decision or to make a bid for independence. This triumph for central power did not last long and ended disastrously for ῾Uthmān. A group of Arabs from Egypt, along with a similar group from Kufa in Iraq, rose in rebellion against this policy, marched on Medina and killed the aged caliph as he sat in his house reading the Qur’ān in 656. It was the first great trauma in Islamic history and had immediate repercussions in Egypt. In the aftermath of ῾Uthmān’s death, his successor, ῾Ali b. Abī Ṭālib, was challenged by Mu῾āwiya b. Abī Sufyān, the Umayyad based in Syria. This conflict opened up social divisions among the Muslims of Egypt. The rebels against ῾Uthmān came from what one might call the lower classes, Arab Muslims who did not come from the social class of wujūh, nobles or tribal leaders and indeed part of their protest seems to have been against the power and influence wielded by this elite. As in Iraq, these non-elite groups tended to support ῾Ali in his bid to establish himself as caliph while ῾Alī’s Syrian rival, the Umayyad Mu῾āwiya, made contact with the wujūh and won their support. He also sent ῾Amr b. al-῾Āṣ, still active in politics and a close supporter of Mu῾āwiya, to rally support for his cause. In August 658 ῾Amr and his supporters among the wujūh won a decisive battle against their opponents and ῾Amr became governor for the second time. In terms of the wider politics of the caliphate it meant that Egypt came under Umayyad rule, three years before the assassination of ῾Alī and the triumph of the Umayyads in the rest of the caliphate, and ῾Amr remained governor until his death in January 664. In Egypt, it meant the triumph of the established tribal leaders, who were to dominate Muslim Egyptian society for the next two centuries. Mu῾āwiya’s rule in Egypt remained unchallenged for the rest of his reign. As elsewhere in Iraq and Iran, his policy was to work with local Muslim elites and, though we cannot be certain in the absence of firm statistics, it is likely that he demanded little if anything in the way of money from the province. The triumph of Mu῾āwiya had another long-term importance: it destroyed, or at least drastically reduced, support for

365

῾Alī in the province. Whereas many Iraqis looked to the house of the Prophet and the descendants of ῾Alī for political leadership, Egyptian Muslims looked to the Umayyads. Although the divisions between Sunni and Shiꜥi did not become firmly established before the beginning of the tenth century, we can see here the beginnings of a proto-Sunni attitude among Egyptian Muslims which has continued down to the present day and contrasts with the Shi῾i sympathies of the Muslims of southern Iraq, which likewise endure until now. Governors after ῾Amr b. al-῾Āṣ In early Umayyad times, the governors of Egypt were powerful men who often enjoyed long periods in office. The most important and best remembered of these was ῾Abd al-῾Azīz b. Marwān (684–704), the brother of the famous Umayyad caliph ῾Abd al-Malik. He ruled as a virtual viceroy in alliance with the local wujūh. Unlike in Iraq, where the local militias were effectively disbanded and a Syrian garrison was established at Wasit to control the population (and live off the taxes they paid), no Syrian troops were sent to Egypt and the jund kept their effective monopoly of military power. ῾Abd al-῾Azīz was also determined to keep control of Muslim forces then engaged in the conquest of the Maghreb, appointing his own man, Mūsā b. Nuṣayr, to govern the area. It was this same Mūsa who, after ῾Abd al-῾Azīz’s death, took command of the expedition to conquer Spain in 711–712. This period of quasi-autonomy came to an end with the death of the long-serving governor in 704. He had probably hoped that his own son al-Aṣbagh would succeed him, perhaps creating a hereditary appanage in the province as Ibn Tulun was to do later. But this was not to be, for the young man predeceased his father and ῾Abd al-Malik moved quickly to impose his own nominee. This assertion of Syrian control was made clear when, in 706, the order came that Arabic, rather than Coptic or Greek, was to be the language of the governor’s administration, to bring it into line with Syrian practice, and a Syrian bureaucrat was sent to oversee the process.2 Anyone who wished to work for the government, except at the most local level, now had

2

On the use of Greek, Coptic and Arabic, see Mikhail 2014, 79–105.

366

H. KENNEDY

to be able to read and write in Arabic: outside ecclesiastical circles, there were simply no jobs in Greek or Coptic and anyone, Christian, Jewish or Muslim, who wanted a modern education would have had to choose Arabic as the language of instruction. The importance of this for the long-term Arabization of public life can hardly be exaggerated. The surviving letters of Qurra b. Sharīk (709–715), preserved on papyrus, show a governor in full charge of an Arabic-writing administration, and who has a clear understanding of the administration of his province and what he wants to do, even if this administration continued to use Greek to communicate with officials at a local level. It was all part of a growing centralization of caliphal administration and it looks as if the caliph and his bureaucrats were increasingly frustrated at their inability to control the rich revenues of the province because of the effective monopoly the jund had over the public finances and the continuing influence and resistance of the Coptic landholding elites at a local level. Under the last great Umayyad caliph Hishām (724–743), attempts were made to break down this fiscal autonomy. There were two approaches. The first was to appoint determined financial administrators, hard men, trained outsiders who would not bow to local pressures and demands. Such a man was ῾Ubayd Allah b. al-Ḥabḥāb, appointed as head of taxation or ṣāḥib al-kharāj in 724. Until his transfer to North Africa in 734 he, rather than the titular governor, was the real power in the land. He instituted what seems to be the first cadastral survey of Egypt since the conquest, including not only the measurement of land but also the counting of animals and other resources. The other move was to transport Syrian supporters of the Umayyad regime to Egypt so that they could benefit from the revenues of the country directly. ῾Ubayd Allah organized this but the caliph, conscious of the opposition this might arouse among the jund of Fusṭāṭ, forbade his governor to settle them in the capital so they were given lands in the Hawf, on the eastern fringes of the Nile Delta. By the end of Hishām’s reign there were some 5,000 of them. All this activity, and the increase of fiscal pressures, led to the first of the Coptic revolts in 725–726 (Lev 2012). From the time of the conquest for eighty years the Coptic population seems to have accepted Muslim rule, no doubt with grumbles and complaints but essentially peacefully. This revolt was the first of a series which was to last for the next century. We have little information about the leadership or ideology of these revolts. It seems unlikely that they were attempting to

overthrow Muslim rule or drive out the Arabs; rather that they wanted their grievances over taxation to be addressed. The rebellions were always defeated and put down militarily but they increased the military role of the jund, now indispensable for the operation of Islamic government. The rivalries and wars which led to the collapse of Umayyad government in Syria and Iraq from 743 onwards had ripple effects in Egypt as all the various parties attempted to secure their status in these new, disturbed times, but there is no space here to discuss these manoeuvrings. In 750 the last Umayyad caliph, Marwān b. Muḥammad, fled west to Egypt hoping to escape his Abbasid pursuers and perhaps even to organize resistance in the province but he was overtaken and killed at Kharibṭa on the eastern fringes of the Delta. Abbasid governors In Egypt, as in many other areas of the Islamic world, the new Abbasid government looked very much like the old Umayyad one but under new management. On the one hand, governors were appointed by the caliphs but most of the real power lay with the jund and their leaders. The old families continued to provide the main officials, notably of course the chiefs of police. On the other hand, the governors were now chosen from members of the Abbasid family or their military supporters. Two trends which had become apparent in later Umayyad times continued to develop. The first was that the governors were almost always outsiders with, as far as we can see, no previous contacts with the province or experience of it. This weakness was reinforced by the very quick changeover of governors under the early Abbasids. ῾Abd al-῾Azīz the Umayyad had been governor for twenty years, giving him time to build up and consolidate his position in the province. In the twenty-three years of Hārūn al-Rashīd’s reign there were no fewer than twenty-two governors of Egypt. There can be little doubt that this was deliberate government policy, to reduce the power of provincial governors and consolidate power in the capital. Although the documentation is not as full, we can see the same process in the governorates of other important cities of the Muslim world such as Basra and Kufa in Iraq. Along with this reduction in the power of governors, there were repeated attempts to bring the financial administration of the province under direct control from the centre. From 758 to 788 a separate financial

THE MUSLIM ELITE OF EARLY ISLAMIC EGYPT

administrator was appointed from Baghdad to ensure that as much revenue as possible was collected and forwarded to the capital. In 792 a special investigator called ῾Umar b. Mihrān was secretly sent from Baghdad as an agent to find out what was going on. He disguised himself as a poor old man, travelling on a single donkey. He moved unobserved among the government tax collectors and officials in Fusṭāṭ until he had all the information he required. Then he declared himself and revealed his identity, forcing the officials to disgorge their ill-gotten gains. In the end, however, all these devices amounted to very little. As long as the famous Barmakid family of viziers was in charge of the caliphs’ bureaucracy, they worked hard to extract as much money as possible from the province. However, in 803 they were dramatically removed from power, being arrested suddenly and, in one case, executed. After their fall, it is clear that the grip of the central administration was lost and we find the Abbasid authorities resorting to tax farming to try to recover any of the money they felt they were owed. The dominance of Egypt and its wealth by the jund and their wujūh leaders may have been doomed in any case. Certainly their ability to keep so much of the wealth of the province for themselves was increasingly challenged not just by the government of the caliphs but also by other military groups from the East, wanting a share. Turkish domination over Egypt After 809, however, Egypt descended into crisis and chaos. In that year the caliph al-Rashīd died and his sons al-Amīn and al-Ma’mūn and their supporters fought each other for control of the caliphate. Both parties tried to attract support in Egypt and the country was essentially divided between two warlords, one in Fusṭāṭ and the other in Alexandria. The picture was complicated by the arrival of a large group of Andalusi Arabs, expelled from Cordoba by the Umayyad Emir al-Ḥakam I, who sailed to Egypt, took over Alexandria and made it their base for raids in the eastern Mediterranean. The complex details and manoeuvrings do not concern us here but for almost two decades from 809 to 826 Egypt was dominated by the conflicts of the warring parties. The old jund and the wujūh who led them were the main losers; in 816, for example, most of the descendants of Mu῾āwiya b. Ḥudayj were killed when their palace in Alexandria was attacked and burned by the Andalusis and their local allies and in

367

819 a large group of the wujūh were taken out into the Nile on boats and drowned on the orders of a warlord who wanted to destroy their power. Finally in 826 order was restored by al-Ma’mūn’s general ῾Abd Allah b. Ṭāhir, but the political landscape had changed out of all recognition. A new army now dominated Egypt; it was not composed of Egyptian Arabs but of men from the East, especially from northeast Iran, who had been the main supporters of al-Ma’mūn in the civil war. We find Alexandria being governed not by a local man but by the son of the rulers of distant Samarqand. In the years that followed, there were continuous struggles between the Egyptian Arabs and these mostly Turkish soldiers from the East. In 831, for the first time, Copts and Arabs combined together to resist the incomers. Finally, in 832 the caliph al-Ma’mūn came in person to Egypt, the only time a reigning caliph ever visited the province, to consolidate the new regime. Soon the order was given that the Arabs should be dropped from the dīwān, which meant that their salaries were no longer paid: they were now subjects, just like all other Egyptians, whether Muslim, Christian or Jewish, of the new Turkish-dominated regime. The new ruling class brought a new ideology. The Abbasid caliphs of the period, now resident in their new capital at Samarra, 160km north of Baghdad, were wedded to the doctrine of the createdness of the Qur’ān, that is that the Qur’ān had been created by Allah at a certain point in time, the mission of the Prophet Muḥammad, rather than having existed for all eternity. The theological intricacies of the argument may have been lost on most people but the new doctrine, imposed by caliphal decree, was bitterly resented. The authorities set up an inquisition to ensure that all government functionaries subscribed to the new belief. Old and respected religious leaders were imprisoned and tortured and when the Muslims worshipped in the ancient mosque of ῾Amr, a Turk with a whip stood behind them to make sure they did it correctly. This period of humiliation and oppression for the Egyptians was ended not by an uprising of the inhabitants but by the arrival of yet another Turk from the east, Aḥmad b. Ṭūlūn. But Aḥmad was different from his predecessors. It was his political genius to make Egypt his independent power base and, most importantly, to ensure that the taxes raised in the province were spent there on local projects including the magnificent new mosque which still bears his name (on the emergence of Muslim Egyptian identity, see Haarman

368

H. KENNEDY

1980; Sijpesteijn 2013). It is a cliché of historians to say that Ibn Ṭūlūn was the first independent ruler of Egypt since Cleopatra, but there is an element of truth in it. Under his rule, Egypt gained, for the first time, a political power in the Islamic world which reflected its

economic wealth, but that would have been impossible if the foundations had not been laid by the Arab Muslims of Egypt who created a Muslim society in the province.

Bibliography

Kennedy, H. 1981. Central government and provincial elites in the early Abbasid caliphate. Bulletin of the School of Oriental and African Studies 44: 26–38. ———. 1998. Egypt as a province in the Islamic caliphate 641–868. In The Cambridge history of Egypt I, C. F. Petry (ed.), 62–85. Cambridge. ———. 2002. Military pay and the economy of the early Islamic state. Historical Research 75: 155–69. Kubiak, W. 1987. Al-Fusṭāṭ: Its foundation and early urban development. Cairo. Lev, Y. 2012. Coptic rebellion and the Islamization of Medieval Egypt (8th–10th century): Medieval and modern perceptions. Jerusalem Studies in Arabic and Islam 39: 303–44. Mikhail, M. S. A. 2014. From Byzantine to Islamic Egypt: Religion, identity and politics after the Arab conquest. London. Sijpesteijn, P. M. 2013. Shaping a Muslim state: The world of a mid-eighth-century Egyptian official. Oxford. Torrey, C. (ed.). 1922. Ibn ῾Abd al-Ḥakam: Futūḥ miṣr wa-l-maghrib. New Haven, CT.

el-Ali, S. A. 1971. A new version of Ibn al-Mutarrif’s list of revenues in the early times of Hārūn al-Rashīd. Journal of the Economic and Social History of the Orient 14: 303–10. Butler, A. J. 1978. The Arab conquest of Egypt and the last thirty years of the Roman dominion. Edited by P. M. Fraser. Oxford. Creswell, K. A. C. 1932–40. Early Muslim architecture. 2 vols. Oxford. de Goeje, M. J. 1879–1901. Taʼrīkh al-rusul wa-al-mulūk. 15 vols. Leiden. Gayraud, R.-P. 1998. Fostat: Evolution d’une capitale arabe du viie au xiie siècle d’après les fouilles d’Istabl Antar. In Colloque international d’archéologie islamique, R.-P. Gayraud (ed.), 435–60. Paris. Guest, R. (ed.). 1912. The governors and judges of Egypt or Kitâb El ꜥUmarâ (El Wulâh) Wa Kitâb El Quḍâh of El Kindî. London. Haarmann, U. 1980. Regional sentiments in Medieval Islamic Egypt. Bulletin of the School of Oriental and African Studies 43: 62–65. Kaegi, W. 1998. Egypt on the eve of the Muslim conquest. In The Cambridge history of Egypt I, C. F. Petry (ed.), 34–61. Cambridge.