Muslims and Others in Early Islamic Society [Reprint ed.] 0860787133, 9780860787136

Citation preview

THE FORMATION OF THE CLASSICAL ISLAMIC WORLD

General Editor: Lawrence I. Conrad Volume 18

Muslims and Others in Early Islamic Society

THE FORMATION OF THE CLASSICAL ISLAMIC WORLD General Editor: Lawrence I . Conrad 1 2 3 4 5

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

B yzantium before th e R ise of Islam T h e Sasanian East before the Rise of Islam T h e Arabs and Arabia on the Eve of Islam T h e Life of M uham m ad T h e Expansion of th e Early Islamic S tate T h e A rticulation of Islamic S tate Structures P roblem s of Political Cohesion in Early Islam A rab -B yzan tin e R elations in Early Islam ic T im es T h e Turks in th e Early Islamic World P attern s of Everyday Life Production and th e Exploitation of R esources M anufacturing and Labour Trade and Exchange in Early Islam P roperty and Consum ption in Early Islam ic Society C ities in the Early Islamic World N om ads and the D esert in th e Early Islam ic World Society and the Individual in Early Islam M uslim s and Others in Early Islamic Society T h e Christian C om m unities in th e Early Islam ic W orld T h e Jewish C om m unities of the Early Islam ic World A rchaeology and Early Islam Early Islamic N um ism atics and M onetary H istory Early Islamic A rt and A rchitecture T h e Qur’an: Style and C ontents T h e Qur’an: Form ative Interpretation T h e D evelopm ent of Islamic Ritual T h e Formation of Islamic Law Hadîth: Origins and D evelopm ent Early Islamic Historiographical Traditions Early Islamic Theology Eschatology and A pocalyptic in Early Islam Early Islamic V isions of C om m unity Shl^ism: Origins and Early D evelopm ent K härijite M ovem ents in Early Islam T h e Em ergence of Islam ic M ysticism T h e Islamic Philological Tradition Early Arabic Poetry and P oetics Early Arabic Prose Literature T h e Rise of Islamic Philosophy T h e Rise of A rab-Islam ic M edicine T h e Exact Sciences in Early Islam M agic and D ivination in Early Islam Education and Learning in th e Early Islam ic W orld T h e Early Islamic M anuscript Tradition Early Islamic N orth Africa T h e Formation of al-A ndalus I T h e Formation of al-A ndalus II T h e M odern Study of Early Islam

Averil Cameron Shaul Shaked Frank E. Peters Uri Rubin Fred M. Donner Fred M. Donner R. Stephen Humphreys Michael Bonner C.E. Bosworth David Waines Michael G. Morony Michael G. Morony A.L. Udovitch Baber Johansen Hugh Kennedy Hugh Kennedy to be announced Robert E. Hoyland Sidney H. Griffith David Wasserstein Donald Whitcomb Michael Bates Jonathan Bloom Andrew Rippin Andrew Rippin G.R. Hawting Wael B, Hallaq Harald Motzki Lawrence I. Conrad Josef van Ess Wilferd Madelung Wadãd al-Qãdt Etan Kohlberg Ridwan al-Saiid Bernd Radtke Ramzi Baalbaki Suzanne Stetkevych Fedwa Malti-Douglas Everett Rows on Lawrence I. Conrad Jamil Ragep Emilie Savage-Smith Claude Gilliot Jan Just Witkam Elizabeth Savage Manuela Marxn M. Fierro/J. Samsó Lawrence I. Conrad

THE FORMATION OF THE CLASSICAL ISLAMIC WORLD

General Editor: Lawrence I. Conrad Volume 18

Muslims and Others in Early Islamic Society edited by Robert Hoyland

First published 2004 in the series The Formation of the Classical Islamic World by Ashgate Publishing Published 2017 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN 711 Third Avenue, New York, NY 10017, USA

Routledge is an imprint of the Taylor & Francis Group, an informa business This edition copyright© 2004 by Taylor & Francis, and Introduction by Robert Hoyland. For copyright ofindividual articles refer to the Acknowledgements. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library CIP Data Muslims and Others in Early Islamic Society- (The Formation of the Classical Islamic World) 1. Islam - History 2. Islam - Relations I. Hoyland, Robert 297'.09 US Library of Congress CIP Data Muslims and Others in Early Islamic Society/ edited by Robert Hoyland p. cm. - (The Formation of the Classical Islamic World; v. 18) Includes bibliographical references. 1. Islamic Empire - History- 622-661- Historiography. 2. Islamic Empire-History- 661-750-Historiography 3. Islamic Empire - History- 750-1258 - Historiography. 4. Historiography- Islamic Empire. 5. Islamic Empire-Ethnic Relations. 6. Dhimmis. I. Hoyland, Robert G., 1966- II. Series. DS38. l.M876 2002 909'.097671-dc21 2001027829

ISBN 13: 978-0-86078-713-6 (hbk) THE FORMATION OF THE CLASSICAL ISLAMIC WORLD- 18

CONTENTS Acknowledgements

vii

G eneral E d ito r’s Preface

xi

Introduction

1.

2.

3.

4.

5.

6.

7.

8.

9.

Religious C om m unities in Late Sasanian and Early M uslim Iraq M ichael G. M orony

xiii

1

D him m ah in Q u r’an and H adith M ahm oud Ayoub

25

T he Legislative A utonom y of C hristians in the Islamic W orld Néophyte Edelby

37

How DhimmTs Were Judged in the Islamic W orld A n to in e Fattal

83

Problem s of Differentiation between M uslims and NonM uslims: Re-reading the “O rdinances of ‘U m ar” (A l-Shurüt a l-‘um ariyya) Albrecht Noth

103

“Do not A ssim ilate Yourselves . . La tashabbahü . . . M .J. K ister

125

M inority Selfrule and G overnm ent C ontrol in Islam S.D . G oitein

159

C om parative Religion in the Apologetics of the F irst C hristian A rabic Theologians Sidney H. Griffith

175

Jew ish Polemics against Islam and C hristianity in the Light of Judaeo-A rabic Texts Sarah Stroum sa

201

CONTENTS

vi 10.

11.

12.

13.

14.

15.

M uslim Studies of O ther Religions: T he Medieval Period Jacques Waardenburg

211

C hristian Polemic and the Form ation of Islamic Dogm a C.H. Becker

241

Socio-Economic H istory and Islamic Studies: Problem s of Bias in the A daptation of the Indigenous Population to Islam Claude Cahen

259

Mawlas: Freed Slaves and C onverts in Early Islam Daniel Pipes

277

Conversion in Early Islamic Egypt: T he Economic Factor Gladys Frantz-M urphy

323

Q uestions Concerning the M azdaeans of Muslim Iran Jean de M enasce

331

G eneral Index

343

ACKNOWLEDGEMENTS T he chapters in this volume are taken from the sources listed below. The editor and publishers wish to thank the authors, original publishers or other copyright holders for permission to use their m aterial as follows: C h a p t e r 1: Michael G. Morony, “Religious Communities in Late Sasanian and

Early Muslim Iraq” , Journal of the Economic and Social History of the Orient 17 (Leiden, 1974), pp. 113-35. C h a p t e r 2: Mahmoud Ayoub, “Dhimmah in Qur’an and Hadith” , Arab Stud­

ies Quarterly 5 (Washington, 1983), pp. 172-82. With kind permission from the Association of Arab-American Graduates, Washington, D.C. C h a p t e r 3: Translation of: Neophyte Edelby, “L’autonomie legislative des chré­ tiens en terre d ’islam” , Archives dhistoire du droit oriental 5 (Brussels, 1950-51), pp. 307-51. Translation by Bruce Inksetter; Copyright ©2004 Ashgate Publishing Ltd. C h a p t e r 4: Translation of: Antoine Fattal, “Comment les Dhimmis étaient jugés en terre d ’Islam” , Cahiers dhistoire égyptienne 3 (Cairo, 1951), 321-41. Translation by Susan Pickford; Copyright ©2004 Ashgate Publishing Ltd. C h a p t e r 5: Translation of: Albrecht Noth, “Abgrenzungsprobleme zwischen Mus­ limen und Nicht-Muslimen. Die ‘Bedingungen ‘Umars (as-Surüt al-‘umariyya)’ unter einem anderen Aspekt gelesen” , Jerusalem Studies in Arabic and Islam 9 (Jerusalem, 1987), pp. 290-315. Translation by Mark Muelhaeusler; Copyright ©2004 Ashgate Publishing Ltd. C h a p t e r 6 : M.J. Kister, “ ‘Do not Assimilate Yourselves . . . ’ Lã tashabbahü ..

Jerusalem Studies in Arabic and Islam 12 (Jerusalem, 1989), pp. 321-53. C h a p t e r 7: S.D. Goitein, “Minority Selfrule and Government Control in Islam” ,

Studia Islamica 31 (Paris, 1970), pp. 101-16. C h a p t e r 8 : Sidney H. Griffith, “Comparative Religion in the Apologetics of the First Christian Arabic Theologians” , Proceedings of the Patristic, Medieval and Re­ naissance Conference (Villanova University, Pennsylvania) 4 (Philadelphia, 1979), pp. 63-86. C h a p t e r 9: Sarah Stroumsa, “Jewish Polemics against Islam and Christianity in the Light of Judaeo-Arabic Texts” , in Norman Golb, ed., Judaeo-Arabic Studies: Proceedings of the Founding Conference of the Society for Judaeo-Arabic Studies (Studies in Muslim-Jewish Relations 3; Amsterdam, 1997), pp. 241-50.

viii

ACKNOWLEDGEMENTS

C h a p t e r 10: Jacques Waardenburg, “Muslim Studies of Other Religions: The Medieval Period” , in G.J. van Gelder and E. de Moor, eds., The Middle East and Europe: Encounters and Exchanges (Orientations 1; Amsterdam, 1992), pp. 10-38. C h a p t e r 11: Translation of: C.H. Becker, “Christliche Polemik und islamische Dogmenbildung” , Zeitschrift für Assy riologie 25 (Leipzig, 1911), pp. 175-95. Trans­ lation by Mark Muelhaeusler; Copyright ©2004 Ashgate Publishing Ltd. C h a p t e r 12: Translation of: Claude Cahen, “Histoire économico-sociale et islamologie: le problème préjudiciel de l’adaptation entre les autochtones et l’Islam” , Correspondance d ’Orient (Brussels, 1961), pp. 197-215; repr. in idem, Les peu­ ples musulmans (Damascus, 1977), pp. 169-88. Translation by Philip Ditchfield; Copyright ©2004 Ashgate Publishing Ltd. C h a p t e r 13: Daniel Pipes, “Mawlas: Freed Slaves and Converts in Early Islam” , Slavery and Abolition 1 (London, 1980), pp. 132-77. C h a p t e r 14: Gladys Frantz-Murphy, “Conversion in Early Islamic Egypt” , in Yusuf Raghib, ed., Documents de l ’Islam médiéval: nouvelles perspectives de re­ cherche (Cairo, 1991), pp. 11-17. C h a p t e r 15: Jean de Menasce, “Problèmes des mazdéens dans l’Iran musulman” , in Gernot Wiessener, ed., Festschrift für Wilhelm Eilers (Wiesbaden, 1967), 220-30. Translation by Philip Ditchfield; Copyright ©2004 Ashgate Publishing Ltd.

Every effort has been m ade to trace all the copyright holders, b u t if any have been inadvertently overlooked the publishers will be pleased to make the necessary arrangem ent a t the first opportunity.

PUBLISHER’S NOTE T he pagination of articles originally published in English has been m ain­ tained for this volume. In articles tran slated into English, the original pag­ ination has been indicated in the tex t in bold-face type.

GENERAL EDITOR’S PREFACE Since the days of Ignaz Goldziher (1850-1921), generally regarded as the founder of Islamic studies as a field of m odern scholarship, the form ative period in Islamic history has rem ained a prom inent them e for research. In G oldziher’s tim e it was possible for scholars to work w ith the whole of th e field and practically all of its available sources, bu t m ore recently the in­ creasing sophistication of scholarly m ethodologies, a broad diversification in research interests, and a phenom enal burgeoning of the catalogued and published source m aterial available for study have combined to generate an increasing “com partm entalisation” of research into very specific areas, each w ith its own interests, priorities, agendas, m ethodologies, and controversies. W hile this has undoubtedly led to a deepening and broadening of our un­ derstanding in all of these areas, and hence is to be welcomed, it has also tended to isolate scholarship in one subject from research in o ther areas, and even m ore so from colleagues outside of A rab-Islam ic studies, not to m ention stu d en ts and others seeking to fam iliarise them selves with a particu lar topic for the first tim e. The Form ation o f the Classical Islam ic W orld is a reference series th a t seeks to address this problem by m aking available a critical selection of the published research th a t has served to stim ulate and define the way m odern scholarship has come to understand the form ative period of Islamic history, for these purposes taken to m ean approxim ately AD 600-950. Each of the volumes in the series is edited by an expert on its subject, who has cho­ sen a num ber of studies th a t taken together serve as a cogent introduction to the s ta te of current knowledge on the topic, the issues and problem s p articular to it, and the range of scholarly opinion inform ing it. A rticles originally published in languages other th an English have been tran slated , and editors have provided critical introductions and select bibliographies for fu rth er reading. A variety of criteria, varying by topic and in accordance with the judge­ m ents of the editors, have determ ined the contents of these volumes. In some cases an article has been included because it represents the best of current scholarship, the “c u ttin g edge” work from which future research seems m ost likely to profit. O ther articles—certainly no less valuable contributions— have been taken up for the skillful way in which they synthesise the s ta te of scholarly knowledge. Yet others are older studies th a t— if in som e ways now superseded— nevertheless m erit atten tio n for their illustration of thinking or conclusions th a t have long been im p o rtan t, or for the decisive stim ulus

xii

GENERAL EDITOR’S PREFACE

they have provided to scholarly discussion. Some volumes cover them es th a t have emerged fairly recently, and here it has been necessary to include articles from outside the period covered by the series, as illustrations of paradigm s and m ethodologies th a t may prove useful as research develops. C hap ters from single au th o r m onographs have been considered only in very exceptional cases, and a certain em phasis has been encouraged on im p o rta n t studies th a t are less readily available th an others. In the present sta te of the field of early Arab-Islam ic studies, in which it is routine for heated controversy to rage over w hat scholars a generation ago would have regarded as m atters of simple fact, it is clearly essential for a series such as this to convey some sense of the richness and variety of the approaches and perspectives represented in the available literatu re. An effort has thus been m ade to gain broad international participation in edi­ torial capacities, and to secure the collaboration of colleagues representing differing points of view. T hroughout the series, however, the range of possi­ ble options for inclusion has been very large, and it is of course impossible to accom m odate all of the outstanding research th a t has served to advance a particular subject. A representative selection of such work does, however, appear in the bibliography compiled by the editor of each volume a t the end of the introduction. The interests and priorities of the editors, and indeed, of the G eneral Ed­ itor, will doubtless be evident throughout. Hopefully, however, the various volumes will be found to achieve well-rounded and representative synthe­ ses useful not as the definitive word on their subjects— if, in fact, one can speak of such a thing in the present sta te of research— bu t as introductions com prising well-considered points of dep artu re for m ore detailed inquiry. A series pursued on this scale is only feasible with the good will and cooperation of colleagues in m any areas of expertise. T he G eneral E ditor would like to express his gratitu d e to the volume editors for th e investm ent of their tim e and talents in an age when work of this kind is grossly underval­ ued, to the tran slato rs who have taken such care with the articles entrusted to them , and to Dr John Smedley and his staff a t A shgate for their su pport, assistance and guidance throughout.

Lawrence I. C onrad

INTRODUCTION Muslims and Others R o b e r t H o y la n d

In t h e c o u r s e of the third and fourth centuries AD the M iddle E ast w it­ nessed the loose-knit em pires of the Rom ans (in the west) and the P a rth ia n s (in the east) give way to the integrated em pires of the B yzantines and the Sasanians. W hereas the ruling elites of the form er pair were largely indiffer­ ent to the beliefs of the masses, those of their successors shared their creed w ith the m ajority of their subjects and sought not only to control political power w ithin their lands, but also to prom ote religious uniform ity. M ore­ over, w arfare between them assum ed an increasingly religious character and religious difference frequently becam e equated with political dissidence, the result being persecutions. This drive tow ards greater integration and confor­ mity, which intensified during the sixth century as a result of an escalation of the conflict between the two superpow ers, provoked those sectarian groups jealous of their own independence to establish a certain distance betw een them selves and im perial culture. G radually, and especially in pluralist Iraq, they transform ed them selves into com m unal organizations w ith their own schools, law courts, places of worship, religious hierarchy, and so on. They were effectively socio-legal corporations ordered along religious lines. It was w ithin this environm ent th a t Islam grew up and, n aturally enough, it continued and even extended these trends (C hapter 1 ). It divided up the world prim arily along religious lines, seeing only believers (ahl al-islãm ) and infidels (ahl al-kufr). T he la tte r were generally left— indeed expected— to m anage their own affairs and to conduct them selves according to their own laws and beliefs. The only m ajor dem and m ade of them was th a t they pay a special ta x (jizy a ) to dem onstrate their twin sham e of having been conquered and having rejected the tru e religion and its prophet M uham m ad. This laissez-faire a ttitu d e was noted of the M uslim conquerors by a northM esopotam ian resident w riting in the 690s: T heir robber bands went annually to d istan t p a rts and to the islands, bringing back captives from all the peoples under the heavens. O f each person they required only trib u te , allowing them to rem ain in w hatever faith they w ished___ T here was no

xiv

INTRODUCTION distinction between pagan and C hristian, the faithful was not known from a Jew .1

For their p a rt the early M uslims preferred to distance them selves from the conquered population, living in separate garrison tow ns and eschewing the custom s and practices of others. B ut the possession of w ealth and power by the M uslims m eant th a t the conquered peoples would inevitably seek them out, w hether to win their support in internal conflicts, or to earn a share in their privileges and riches, or simply to seek a living in their employ. So the garrison tow ns becam e cosm opolitan cities in which M uslims and nonM uslims interacted in a variety of different ways. Some exam ples of this will be given here and the select bibliography a t the end will suggest fu rth er directions for study.

Independence of Non-Muslims from Muslims In retu rn for paying their taxes the non-M uslims received a guarantee of protection (dhim m a) with regard to their lives and property and the right to practise their faith w ithout hindrance (C hapter 2 ). Since certain Q u r’anic verses (e.g. 22.17, 98.1) distinguish between the Jews and C hristians as possessors of a recognized scripture (ahl al-kitãb), and polytheists (ahl alshirk), some M uslims argued th a t only the form er qualified for protection w hereas the la tte r should be fought to the death. B ut as the M uslims pushed fu rth er east, vanquishing such peoples as the Z oroastrians and th e Hindus, any initial objections were soon brushed aside and the category of people qualifying for protected sta tu s (ahl al-dhim m a) expanded to com prise p retty much all non-M uslims. T hus M uham m ad ibn al-Qasim (d. 96/715), the first M uslim general to conquer an Indian town and to face the problem of w hat to do with its population, ruled th a t their holy places were “akin to the churches of the C hristians and the synagogues of the Jew s and the firetem ples of the Z oroastrians ” .2 And this judgem ent form ed the theoretical basis for subsequent Muslim tolerance of Hindus and their worship. The only m ajor exception to this principle involved non-M uslim A rabs, who, w hether the pagan A rabs of M uham m ad’s A rabia or the C hristian A rabs ^ o h n bar Penkaye, K tã b ã d-rïsh m e llë , ed. and trans. A. Mingana in his Sources syriaques (Leipzig, 1907), 147/175, 151/179. Compare the words of the first caliph Abü Bakr (r. 11-13/632-34) to his generals on how to treat the conquered peoples: “Let them bring tribute as determined between you, and let them be left in their faith and their land” ( Chronicon ad 1234 >trans. Andrew Palmer in The Seventh C en tury in W est-S yrian Chronicles , Liverpool 1993, 145). 2Al-Balâdhurï, Futüh al-buldãn, ed. M.J. de Goeje (Leiden, 1866), 439.

INTRODUCTION

xv

of the Fertile Crescent, were som etim es the targ e t of M uslim m issionary efforts .3 To qualify for protection and tolerance the various non-M uslim com m u­ nities were expected to be in possession of the legal w herew ithal to exercise autonom y (C hapters 3 and 4). This would seem to have been an early con­ cern, since the Q u r’an already dem ands th a t “the adherents of the Gospel judge by th a t which God has revealed therein” (5.47). And in one early C hristian-M uslim debate tex t an A rab general issues to his C hristian inter­ locutor the ultim atum : “Show me th a t your laws are w ritten in the Gospel and th a t you conduct yourselves by them , or else subm it to the law of the M uslim s ” .4 This was problem atic for the C hristians, since th e Gospel is m uch less explicit ab o u t m ost areas of hum an social intercourse th a n the Torah and the Q u r’an and is little interested in legal m atte rs. U nder pres­ sure from Islam eastern C hristian clerics set about rectifying this. One of the earliest such au th o rs whose works have survived is an eighth-century Iranian archbishop, Isho‘bokht of Fars, who w rote a six-volume tra c t called sim ply Com position on the Laws by its Syriac tra n sla to r. It is not m erely a collection of canons, b u t an a tte m p t to system atize and codify the C hristian law, tre a tin g jurisprudence (Volume 1 ), m arriage (2-3), inheritance (4), and co n tract (5-6). Interest in such subjects is not found in pre-Islam ic C hris­ tian legal w ritings. T here are some hints of related activity by au th o rs of the late seventh and early eighth centuries ,5 but it is the work of Isho4bokht which first em barks upon a proper tre a tm e n t of the m atte r, and so he is som ething of a pioneer. W h a t spurred him to this innovation is clear from his introduction: W hereas the Jews in every place have one law, as also the error of the M agians, and likewise also those who now rule over us (i.e. the M uslim s), am ong the C hristians the laws which are d eter­ mined in the land of the Rom ans are distinct from those in the land of the Persians, and they in tu rn are distinct from those in the land of the A ram aeans, and different from Ahwaz, and dif­ ferent in M ayshan, and likewise also in other p la c e s... Because of t h is ... [I am w riting this book ].6 3For examples see Robert Hoyland, Seeing Islam as Others Saw It (Princeton, 1997), 352-54. 4F. Nau, “Un colloque du patriarche Jean avec l’émir des Agaréens” , JA Ser. XI, 5 (1915), 252/262. 5See P. Crone, “Islam, Judeo-Christianity and Byzantine Iconoclasm”, JSAI 2 (1980), 71-73 (arguing that certain legal texts were composed in response to Islam). 6 Isho‘bokht, Maktbânütâ d - ‘al dïnë, in E. Sachau, ed. and trans., Syrische Rechtsbücher

xvi

INTRODUCTION

B ut Isho‘bokht did not reach this conclusion simply by observation, rath e r it was pointed out to him, as we see from the heading of one of th e chapters of his first book: “Concerning w hat is said by the Jews and the M uslims (hanpë) th a t the C hristians have no laws and it is not possible th a t their lives be conducted w ithout laws ” .7 The usual Jewish accusation had been th a t the C hristians did not hold to the law of Moses and not th a t they had no laws, so the question m ust have originated with the M uslims ( “those who now rule over us” ). And indeed, it is only in the churches of M uslimruled lands th a t we find such developed C hristian law codes. These went far beyond the usual chapters on doctrine, scripture, sacram ents and hier­ archical adm inistration, including sections on m arriage and divorce, dowries and settlem ents, inheritances, degrees of consanguinity, debts and loans, selling and buying, contracts and partnerships, pledges and oaths, and so o n .8 Besides this basic requirem ent there were a num ber of rules for social conduct by which the non-M uslims were expected to abide. These m ost fam ously appear listed in th e so-called “P act of ‘U m ar” , which p u rp o rts to be a lette r sent a t the tim e of the Muslim conquests from the C hristians of Syria to the caliph ‘U m ar I requesting protection and prom ising observance of certain obligations. The docum ent has provoked much discussion both as regards its authenticity and its significance. Earlier scholars tended to regard it as a late invention and as an indication of the discrim ination and isolation endured by non-M uslims of later tim es. M ore recently it has been argued th a t the list does reflect the conditions of the earliest period of th e conquests and th a t its contents were intended for the benefit of the M uslims ra th e r th an for th e detrim ent of the non-M uslims (C hapter 5 ) .9 Faced w ith a massive m ajority population of non-M uslims, the conquerors in stitu ted m easures to erect boundaries between them selves and the conquered peoples so as to prevent their assim ilation after the fashion of the G erm anic conquerors of Rome and so m any C entral Asian conquerors of C hina (C h ap ter 6 ). For example, one item on the list concerns the belt known as the zunnãr. Since C hristians wore such a thing before Islam, it is evident th a t the initial aim (Berlin, 1907-14), III, 8 -1 0 (1.1). 7 Isho‘bokht, Maktbänütä d - ‘al dine, III, 20 (l.X IV ). 8See Richard B. Rose, “Islam and the Development of Personal Status Laws among Christian Dhimmis”, M W 72 (1982), 159-79. 9On the radically divergent answers to the question of the nature of Muslim/nonMuslim co-existence and pertinent bibliography see Mark Cohen, “Islam and the Jews: Myth, Counter-myth, History” , in Shlomo Deshen and Walter P. Zenner, eds., Jews among Muslims: Communities in the Pre-Colonial Middle East (Basingstoke, 1996), 50-63.

INTRODUCTION

xvii

of the regulation was not specifically to hum iliate C hristians, b u t ra th e r to m ake it possible to distinguish them from M uslim s .10

Recourse of Non-Muslims to Muslims Though in theory non-M uslim com m unities acted as sep arate entities w ithin th e M uslim sta te , with their own laws and leaders, in practice th e M uslim authorities frequently becam e involved in the internal affairs of th e vari­ ous com m unities, largely a t the instigation of the non-M uslim s them selves (C hapter 7). For example, in his fight against the m utiny of Persian and east A rabian C hristians in the 650s, the eastern C hristian leader Isho‘yahb III appealed to “the local governors and also to the governor of th a t tim e who was over the local governors” . His successor, George I, cam e to face charges brought against him before the M uslim authorities by a disgruntled archbishop. And in E gypt a certain T heodore of A lexandria, a leader of th e Chalcedonian C hristian comm unity, “went to D am ascus to the chief of the M uslims, nam ed YazTd ibn M u‘äwiya (r. 6 0 -6 4/680-83), and took from him a diplom a em powering him over the people of A lexandria and M ary u t and all its environs and declaring th a t the governor of E gypt had no au th o rity over him, for he had paid Yazid much m oney” . And on his retu rn T heodore “tyrannized the Coptic patriarch A gathon and troubled him ” .11 However, com m unity leaders also ran the risk of occasional unw anted intervention in their affairs, as happened to the patriarch H nanisho‘ I (r. 686-93) during the second A rab civil war. Based a t al-M adä’in in southern Iraq, H nanisho 4 found himself under the rule of rebels a t al-K üfa who were seeking to overthrow the Um ayyad dynasty (660-750). T he pro-U m ayyad governor of Iraq, ‘Ubayd Allah ibn Ziyäd, wooed John of D asen, archbishop of Nisibis, prom ising him th a t “if you will accom pany me, I will depose him (H nanisho4) and establish you as patriarch in his place ” .12 John was pre­ sum ably courted for his connections a t Nisibis, which was a hive of sedition during the civil war. C ertainly Bishr ibn M arw ãn, b ro th er of the caliph ‘A bd al-M alik (r. 65-86/685-705), winner of the civil war, seems to have m ade some agreem ent with John, for he forcibly installed him in place of H nanisho 4.13 Subsequently Jo h n ’s partisans in Nisibis, led by an aristo­ 10See Mark Cohen, “At the Origins of the Distinctive Dress Regulation for Non-Muslims in Islam”, in Lawrence I. Conrad, ed., The Byzantine and Early Islamic Near East I V (Princeton, forthcoming); Hoyland, Seeing Islam, 364. 11Hoyland, Seeing Islam, 200 - 202 : Isho‘yahb and George; “History of the Patriarchs”, ed. and trans. B. Evetts, PO 5 (1910), 5: Theodore. 12John bar Penkaye, 156/184. 13Man ibn Sulaymãn, Kitãb al-majdal, ed. and trans. H. Gismondi (Rome, 1899), 63/56, states that John bribed ‘Abd al-Malik and his brother Bishr.

xviii

INTRODUCTION

cratic Iranian C hristian physician nam ed M ardanshah, aided M uham m ad ibn M arw ãn, another brother of ‘Abd al-M alik, to recapture the city. The p a rty of H nanisho 4 was driven out and M ardãnshãh was en trusted w ith the ad m in istratio n .14 A fter the death of John in 695 the new governor of Iraq, al-H ajjäj ibn Yüsuf, suspended the election of a new patriarch, th u s insti­ gating a Sasanian-style policy of governm ent involvement in church affairs.

Dialogue of Non-Muslims with Muslims T he boundaries between confessional com m unities were patrolled by reli­ gious authorities whose task was to reinforce allegiance to their respective com m unities. This they did by urging exclusive attendance of the in stitu ­ tions of th a t comm unity, by issuing laws prohibiting interaction and inter­ m arriage with non-m em bers, by prom oting distinctive insignia and symbols, and by dissem inating propaganda against the adherents and beliefs of all other groups. M uch of this propaganda would have been for internal con­ sum ption only, as is clear from the som ew hat stale and one-sided C hristian anti-Jew ish literatu re of pre-Islamic tim es, b ut in the early Islamic period a considerable proportion of the apologetic w riting seems to derive from real debate. And this is graphically confirmed by the com m ent of th e M uslim scholar ‘A m r ibn B ahr al-Jahiz (d. 255/868) regarding C hristians: They hunt down w hat is contradictory in our traditions, our reports with a suspect line of transm ission and the am biguous verses of our scripture. Then they single out the weak-m inded am ong us and question our common people concerning these th in g s ... and they will often address them selves to the learned and the powerful am ong us, causing dissension am ong the m ighty and confusing the w eak .15 A num ber of factors favoured such debate in the early ‘A bbäsid era (ca. 750-950): the cosm opolitan n ature of Baghdad and its province, the caliphs’ patronage of scholarship, the emergence of A rabic as a lingua fra n ca , the universal deploym ent of dialectical reasoning based upon categorical defi­ nitions, and the proliferation of converts and apostates, which m eant th a t there were m any with a genuine knowledge of two religions and w ith a real will to cham pion one over the other. B ut also, quite simply, there were 14Ibid., 63-65/56-57. 15A1-Jãhiz, Al-Radd ‘alã l-nasãrã, in A-S.M. Härün, ed., R asä’il al-Jãhiz (Cairo, 196479), III, 320. This very illuminating text has been translated by J. Finkei in JA OS 47 (1927), 311-34 (partial) and by I.S. Allouche in Hespéris 26 (1939), 129-53 (full).

INTRODUCTION

xix

m a tte rs th a t needed debating. Islam prom pted questions th a t had not pre­ viously arisen, such as w hat were th e a ttrib u te s of a tru e prophet and how could one recognize an authentic scripture, and one can observe these and o th er questions being broached in an original way (C h ap ter 8 ). For in­ stance, in a treatise aim ing to dem onstrate th a t C hristianity was the tru e religion, th e theologian T heodore Abu Q u rra (d. ca. 820s) introduces a small thought-experim ent: Let us say th a t I grew up on a m ountain ignorant of the n a tu re of people, and one day, on account of some need th a t presented it­ self to me, I went down to the cities and to the society of people, and I perceived them to be of different religions... (T heodore reviews the nine principal creeds of his d a y )___Reflecting on the doctrine of each one of them , I perceived all of them to be in agreem ent on three points and in disagreem ent ab o u t them to o ---- 1 reflected again and said to myself: It befits G od, in His goodness and grace, when He perceived th a t His creation had deviated from the worship of tru th , th a t He send to them a mes­ senger and a book to make them realize th a t and to retu rn them to it from their errors. B ut there came a plurality of m essengers and b o o k s ... .It is appropriate th a t there should be am ong them only one tru e one in accordance with w hat is known of G o d ’s grace and providence for His creation, but w hat is the trick for recognizing this one ? 16 Two younger contem poraries of Theodore, the aforem entioned M uslim w riter al-Jãhiz and the Jewish thinker Däw üd ibn M arw än al-M uqam m is (d. ca. 860s) also posed them selves this question .17 The form er, in his essay “On the Proofs of Prophecy” , postulates two categories of proofs: those p e rta in ­ ing to “sensory perception” and those based upon a “cogent tra d itio n ” , both cases requiring the involvement of the intellect for purposes of verification. T he m ost im p o rta n t sensory experience in the determ ination of prophethood was a m iracle perform ed by the contender: 16Theodore Abü Qurra, Fi wujüd al-khãliq, ed. Ignace Dick (Jounieh and Rome, 1982), 200 , 211 - 12 . This tract is discussed by Sidney H. Griffith, “Faith and Reason in Christian

Ã'a/ãm”, in S.K. Samir and J.S. Nielsen, eds., Christian Arabic Apologetics during the (Abbasid Period (Leiden, 1994), 1-43. 17And it prompted the mid-ninth-century Zoroastrian author Mardänfarrokh ï Ohrmazddãdãn, who had “traversed many lands and seas” in search of the truth, to com­ pose his “Doubt-Dispelling Exposition” ( Shkand-gumämg wizär), on which see Hoyland, Seeing Islam, 511-12.

XX

INTRODUCTION T he signs of messengers, peace be upon them , and their m iracles are more w orthy of atten tio n and renown and of being considered a compelling argum ent for hearts and m inds th an their preaching and their laws. Indeed, we know th a t Moses, peace be upon him, was unknown and of no repute except on account of his wonders and miracles, and the sam e is tru e of Jesus, peace be upon him. W ere it not for th a t, they would have been ju st like th e rest whose death and birth pass unnoticed .18

For al-M uqam m is, too, the execution of miracles was a prerequisite for belief in a prophet, and he fu rth er stipulated th a t the content of the prophecy m ust be in accord with logic and common sense, and th a t “the trad itio n ab o u t him should not come from one direction, but rath e r from several q u a rte rs ... nor from only one nation, nor in only one language” (C hapter 9 ) . 19 These three works illustrate very well two salient characteristics of the polemic conducted by C hristians, Jews and M uslims in the eighth to ten th centuries. Firstly, the com batants of each p arty entered the inter-confessional arena w ith the sam e intellectual arm ory: scriptures, au th en ticated trad itio n s and dialectical reasoning based upon categorical definitions. T he la tte r was the m ost im p o rtan t, as it enabled the debate to cross sectarian lines, and it is noticeable th a t the three tra c ts m entioned above each open w ith an excur­ sus on the natu re of knowledge and tru th and on the procedure for deriving them . Secondly, as m entioned above, the debate was clearly a real one. The question of how to recognize a tru e prophet, given such careful reflection in the above texts, was scarcely considered by pre-Islamic C hristian and Jewish authorities and was clearly provoked by Muslim claims a b o u t M u h am m ad ’s prophetic credentials .20 M oreover, we very often find in dispute tex ts th a t argum ents put forw ard by one p arty are taken up and refuted by another. This does not m ean, however, th a t we have records of actual discussions, only th a t the authors of the texts, though purveyors of literary fictions, had tested their m etal in the field. T he roots of these controversies between the M uslims and their subject peoples went back to the late seventh and early eighth centuries when Islam first began to present itself as “the religion of tru th ” , so challenging other 18A1-Jahiz, Ft hujaj al-nubüwa, in Hârün, R asä’il al-Jãhiz, III, 259. 19Al-Muqammis, ‘Ishrün maqãla, ed. and trans. Sarah Stroumsa (Leiden, 1989), 265.

20For discussion see Sarah Stroumsa, “The Signs of Prophecy: the Emergence and Development of a Theme in Arabic Theological Literature”, HTHR 78 (1985), 101 14.

INTRODUCTION

xxi

fa ith s .21 B ut the debate only gathered m om entum once A rabic, established as th e adm inistrative language of the em pire by late U m ayyad tim es, had become accepted as the international m edium of scholarship. W hereas only eight au th o rs are known to have polemicized in Syriac against Islam from th e seventh to the th irteen th century in M uslim-ruled lands, and even fewer in Greek, as m any did so in A rabic in the first ‘A bbãsid century (750-850) alone .22 T he emergence of A rabic as a lingua franca and th e patro n ag e of scholarship by the early ‘Abbãsid rulers sponsored a kind of Islamic “en­ lightenm ent” , fuelled by the transm ission of Greek learning into A rabic, and m ade Iraq of the ninth and ten th centuries a centre of lively alterca­ tions am ongst Jews, C hristians, M uslims, Z oroastrians, M anichaeans and pagan philosophers over the natu re of tru th and knowledge (C h ap ter 10 ). T he contest itself was conducted in an alm ost gentlem anly fashion: It is hoped th a t you will tre a t us fairly in the discussion and th a t you will negotiate with us as brothers who share in the goods they inherit from their father. All of them share in them ; nothing belongs to one to the exclusion of the other. So we and you should be on a par in the discussion .23 B ut though the dialectical style of the disp u tan ts becam e ever m ore honed and their argum ents ever more refined, the key objections were never over­ come. Jew s and M uslims could not forgive C hristianity its dilution of G o d ’s unity and ascription to Him of a son; M uslims and C hristians rem ained opposed to Judaism on the subjects of abrogation of the law and falsifica­ tion of the scriptures; and Jews and C hristians persisted in their rejection of M u h am m ad ’s prophethood and his claim to have brought a revelation 21 This expression (dm al-haqq) first appears on the gold coins of ‘Abd al-Malik dated 77/696. That such pretensions spurred non-Muslims to write is shown by the ninth-century Christian author Abu Rã‘ita, R asã ’il, ed. and trans. Georg Graf (Louvain, 1951; CSCO 130-31 Scr. arabici 14-15), I, 1 : “You have asked me to write for you a treatise in which I should make clear to you what is obscure to you of the doctrines of peoples and their claims about the correctness of what they hold to, especially the doctrine of the Muslims and their description of the excellence of their religion, its nobleness and its superiority over other religions.” 22See Sidney H. Griffith, Arabic Christianity in the Monasteries of Ninth-Century Pales­ tine (Aldershot, 1992), especially articles 1-3. 23Abü Rä‘ita, R asä’il, I, 3-4. This at least is the approach of polemicists writing in Syriac and Arabic; writings in Greek, whether from within or outside the Muslim empire, tend to adopt a more hostile tone. See also Sarah Stroumsa, “Ibn al-Räwandl’s S ü ’ adab al-mujãdala; the Role of Bad Manners in Medieval Disputations” , in Hava Lazarus-Yafeh, ed., The Majlis: Interreligious Encounters in Medieval Islam (Wiesbaden, 1999), 66-83.

xxii

INTRODUCTION

from G od, arguing th a t he was not announced in the scriptures and had worked no miracles. Yet in the process each was subtly transform ed and draw n a few steps tow ards its opponents’ position (C hapter 11 ). For ex­ ample, C hristianity came to place greater stress on m onotheism ; th u s th e religious encyclopaedia known as the Book o f the Tower begins its exposi­ tion of C hristianity [Bab 2.1) with the words: “T he acme of faith is the oneness of G od” (dhurwat al-ïmân tawhïd A llah). And later M uslims cam e to p u t forw ard a “Jesus-like p o rtra it of M uham m ad’s prophecy ” .24

Conversion of Non-Muslims to Islam W h a t m otivated religious authorities constantly to strive for the fidelity of their flock was the ever-present specter of apostasy. This was also a worry for Muslim leaders, especially in the first one or two centuries when M uslim num bers were small, but in the long run it was the non-M uslim com m unities th a t witnessed a steady loss of their m em bers. There are indications th a t the M uslim A rab conquerors initially tended to think along genealogical lines: in order to be an A rab and so to share in the immense privileges conferred on A rabs since the conquests, one m ust be born as one. B ut the num erousness of the non-A rabs who wished to enjoy the sam e benefits as their new m asters and th e Q u r’a n ’s stress on the universality of M uham m ad’s message (e.g. 4.79, 7.158, 34.28) m eant th a t it was not easy to refuse into their ranks anyone who lacked the right descent .25 And in general it was accepted th a t one could become a m em ber of the conquest society by adopting the A rab faith. This fact constitutes one of the m ost im p o rtan t preconditions for th e emergence of Islamic civilization, since the skills and sciences of the non-M uslim s contributed enorm ously to its developm ent (C hapter 12 ). T h a t the conquered peoples could join the ruling elite does not explain how, why and when they availed themselves of this option, and these are difficult questions to answer. Some idea of the procedure for converting to Islam, in the early ‘Abbäsid period a t least, is given by a late eighth-century chronicler who was a monk in northern M esopotam ia. W hile he was a t Edessa, he heard about a deacon of th a t district who, despite th e entreaties of “all the notables and priests” , m ade up his mind to apostatize and sought out “a certain m an from am ong the A rabs th ere” , asking th a t “he become 24See Bo Holmberg, “A Reconsideration of the Kitãb al-Majdal” , PO r 18 (1993), 25573: on the Book of the Tower; Stroumsa, “The Signs of Prophecy” , 114: portrait of Muhammad. 25The Qur’an emphasizes that it was revealed in Arabic ( 12 .2 , 13.37, 16.103, 26.195, 39.28, 41.3, 42.7, 43.3, 46.12), but otherwise does not mention the Arabs except for the fickle bedouin (aVãò). Appeal is most commonly made simply to “those who believe” .

INTRODUCTION

xxiii

a M uslim (nhaggar) by his hands” . T he A rab atte m p te d to dissuade him, fearing for him should he repent the next day, b u t the deacon insisted on his sincerity. So the A rab asked w hether he denied C hrist, baptism , the cross, the eucharist and “all th a t the C hristians profess” . A fter th e deacon had ab ju red each item in tu rn , the A rab instructed him to confess belief in M uham m ad as the messenger of God, “the book which cam e down from heaven upon him ” and Jesus as the W ord and Spirit of G od, th a t “he was a prophet and not G od” , then to remove his belt and pray to the south. As th e deacon did so, a w hite dove emerged from his m outh; perceiving the loss of his soul, he bewailed his folly. F urther on the chronicler records th a t whole groups of people would apostatize “w ithout any kind of com pulsion to do so” and “they would w rite their nam es in the register ” .26 Turning to the reasons for conversion to Islam, one should first stress th a t compulsion was very rarely am ong them , as noted by the aforem en­ tioned chronicler and as is suggested by the com paratively sm all num ber of m artyrdom accounts composed in Muslim tim es (com pared to Rom an and Sasanian tim es). Only residents of frontier areas and C hristian A rabs would seem to have ever faced pressure to convert .27 T he decision to leave the religion in which one had been brought up was never one to be taken lightly, since it m eant breaking ties with one’s family, friends and neighbors. W hen Dioscorus of A lexandria becam e a M uslim, his sister w rote inform ing him th a t she could have no furth er association with him, and a Jewish wom an who becam e a C hristian was declared by her husband to be as good as de a d .28 So why did people convert? Spokesmen of the non-M uslim com m u­ nities im pressed upon their flock th a t the only advantages Islam had to offer were social and economic advancem ent, fam e and gain .29 T hus a u th o rs of m artyrdom accounts never failed to present their heroes as being tem p ted w ith offers of high office and wealth by some M uslim notable if they would only convert. 26 Chronicon anonymum pseudo-Dionysianum, II, ed. J.B. Chabot (Paris, 1933; CSCO 104 Scr. syri 53), 389-92, 385. 27Muslim nervousness about their borders, especially with Byzantium, meant that nonMuslims living there, even if not asked to convert, would face hostility and suspicion in time of conflict. See J.M. Fiey, Chrétiens syriaques sous les Abbassides (Louvain, 1980), 48-50. 28 “Le synaxaire arabe jacobite (redaction copte)”, ed. and trans. R. Basset, PO 16 (1922), 6 th Barmahat: Dioscorus; Gaonic Responsa ( Teshüvôt g e ’ônë), ed. Joel Muller (Berlin, 1881), no. 87, 20 b: Jewish woman. 29The argument was of course an old one; for example, the conversion of a priest to Zoroastrianism in the days of Khusrau II was deemed the result of “his fierce attachment to the world and his desire for its pleasures”. See “Chronicle of Siirt”, ed. and trans. A. Scher, PO 13 (1919), 467.

xxiv

INTRODUCTION

Though this has been widely accepted by scholars and contains some tru th , it does require qualification. As regards admission to high office, one’s religious persuasion was generally no bar, a t least not in the first two or three centuries of Islam when the adm inistrative and m edical pro­ fessions were dom inated by non-M uslim s .30 And as long as conversion still entailed becoming a client (mawlã ) of an A rab and so placing oneself on an equal footing with all other clients, those endowed w ith s ta tu s in their for­ m er com m unity were unwilling to risk finding them selves working alongside their form er servants and subjects. It was, therefore, m ostly those of low social s ta tu s (especially peasants), or those who had lost their form er s ta ­ tu s (especially prisoners-of-war), who converted to Islam in U m ayyad tim es (C h ap ter 13). As for taxation, in the first century of Islam it was not specif­ ically M uslims who enjoyed fiscal benefits, but rath e r those who fought “on behalf of A llah ” .31 Taxes were som ething th a t the conquered paid for the upkeep (rizq) of the conquerors in return for protection. Only w ith ‘U m ar II (r. 99-102/717-20), and only with any consistency under the ‘A bbãsids, was there an a tte m p t to base the ta x system on a distinction between M uslim and non-M uslim ,32 and even then the practice was often much messier th an the theory: Nominally they (the tax collectors) were to levy one ten th , yet even when those A rabs had sold all they possessed, it was not 30Theophanes, Chronographia, ed. C. de Boor (Leipzig, 1886), I, 430-31, says that in 758 the Muslims tried to “expel the Christians from government chanceries, but were once again obliged to entrust the same duties to them because they were unable to write numbers”. In the late tenth century al-Muqaddasí, Ahsan al-taqasim fx m a ‘rifat al-aqâlïm, ed. M.J. de Goeje (Leiden, 1877), 183, observed that in Syria and Egypt the physicians and scribes were mostly Christians. And there were still enough Christians in positions of power in Mamlük Egypt for Muslims to write tracts decrying this fact; see R. Gottheil, “An Answer to the Dhimmls”, JAOS 41 (1921), 383-457; M. Perlmann, “Notes on AntiChristian Propaganda in the Mamlük Empire”, BSOAS 10 (1940-42), 843-61. 31 Thus the Mardaites of Lebanon were to be exempted from poll tax as long as they fought for the Muslims (al-Balädhurl, Futüh al-buldãn, 159). 32See J.B. Simonsen, Studies in the Genesis and Early Development of the Caliphal Taxation System (Copenhagen, 1988), 140-50. The fact that decrees exempting converts to Islam from poll tax were commonly issued— e.g. by ‘Umar II (numerous Syriac texts: see Hoyland, Seeing Islam, 654), Hafs ibn al-Walïd ( “History of the Patriarchs”, 116) and al-Saffah (ibid., 189)—indicates that it was not a consistent policy. Rather it was done when there was a need to mobilise support (‘Umar after the disastrous Muslim defeat at Constantinople, Hafs to usurp the governorship of Egypt, al-Saffãh in the wake of the ‘Abbãsid revolution), for it was well known that “it will be enough for you to have a herald announce that taxes will be removed from whoever becomes a Muslim and 50,000 praying men will come to you” . See al-Tabari, Ta’rîkh al-rusul wa-l-mulük, ed. M.J. de Goeje et al. (Leiden, 1879-1901), II, 1024.

INTRODUCTION

XXV

sufficient to pay w hat was dem anded of them . They were e n tre a t­ ing them to take according to the law laid down by M uham m ad, their guide and legislator, and by the first kings, and to take from each one of them w hat they had: if he had w heat take w heat, if ca ttle so cattle. B ut they did not accept th a t, saying to them : “Go and sell your possessions however you like and give us our gold .” 33 Furtherm ore, the desire to avoid tax atio n did not generally lead directly to conversion, but rath e r to flight from the land. In E gypt this would m ost often m ean seeking refuge in a m onastery or some rural re tre a t, b u t in Iraq and M esopotam ia it usually m eant escape to a city where increased co n tact w ith M uslims rendered the chances of conversion high .34 Probably m ore significant in the long run for conversion was the erosion of the prestige and influence of the leaders and in stitutions of the nonM uslim com m unities. This followed on from the late U m ayyad decision to prom ote individual responsibility for paym ent of taxes over lum p sum pay­ m ents collected by m ediating bodies, and gradually to replace local officials by governm ent-appointed Muslim agents (C hapter 14). T he M uslim A ra b s’ m ilitary successes m ust have also indirectly prom oted conversion to Islam. One of their effects was the transfer of peoples from their hom elands to the overwhelmingly M uslim environm ent of the garrison cities where conversion was correspondingly m ore likely. M oreover, they challenged o th er religions’ claims to enjoy G o d ’s exclusive favour. “It is a sign th a t God loves us and is pleased w ith our faith ” , one A rab general declares to a C hristian m onk of the m onastery of Beth Hale in Iraq, “namely, th a t He gives us dom inion over all religions and all peoples ” .35 And this point, coupled w ith th e recep­ tivity tow ards interfaith discussion of the early ‘A bbãsid rulers, m ust have provoked much soul-searching and certainly produced a num ber of converts am ong the literate elite. T he ra te of conversion would have varied substantially from com m unity to com m unity. The Jews, who had long been used to living as a m inority under foreign rule, probably fared best. T he C hristians had suffered a pe33Chronicon anonymum pseudo-Dionysianum, II, 299; this concerns land-tax but illus­ trates that the theory was not always applied. Cf. ibid., II, 341: “They imposed them (extraordinary taxes) on the Muslims as well as on the Christians, for their motivation was not concern for the tradition [of Islam], but to sate their avid desire for money.” 34 If fugitives wished to avoid being rounded up and returned to their villages (cf. alTabari, T a’rïkh, II, 1122 , 1435), then they needed to find a Muslim patron. This plus greater interaction with Muslims made conversion likely. 35Ms. Diyarbakir 95, fols. l b - 2 a; for this text see Hoyland, Seeing Islam, 465-72.

xxvi

INTRODUCTION

riod of persecution under the Rom ans and could draw upon this history and the model of the early m artyrs for stren g th and com fort. T he Zoroastrians, however, had always enjoyed sta te patronage and were unable to cope w ith being political and religious underdogs, so m ore quickly succum bed to Islam ization (C hapter 15). The speed of the la tte r process was also linked to the potential for interaction and interm arriage with M uslims. In E gypt, where the Muslim presence was m eager for the first two centuries of Islam, conversion was very slow; but in K hurasan and Iraq, which bore th e b ru n t of early M uslim settlem ent, opportunities for social intercourse were num erous and conversion more frequent. Finally, there is the question of when apostasy first occurred and when it becam e widespread. We hear of defections to the A rabs already a t th e tim e of the conquests, though our sources frequently m ake no distinction between conversion and collaboration.36 One seventh-century Syrian monk declares to his Jewish disputant: “We C hristians, though enslaved for m any years and worn down by troubles, shall not deny God. And if some C hristians have denied Him, they are not as m any as did so am ong you in Babylon.” 37 And the Coptic au th o r John of Nikiu lam ents th a t in the course of the M us­ lim invasion “m any of the E g y p tia n s... denied the holy orthodox faith and life-giving baptism , and em braced the religion of the M uslim s” .38 By the late seventh century apostasy had become an im p o rtan t issue in C hristian w ritings, dem anding the atten tio n of church authorities.39 And in the sec­ ond half of the eighth century we begin to hear of incidents of large-scale apostasy to Islam. In the days of the patriarch Michael (r. 743-67) “those who denied C hrist num bered 24,000” .40 A couple of decades later a sim ilar outbreak occurred in M esopotam ia, according to th e aforem entioned local 36E.g. the Persian cavalry corps called the Asãwira are said to have converted in 638, but fight in the 680s under a certain Mäh AfrTdhün; those transferred from al-Basra to Antioch by Mu‘äwiya appear as non-Muslims, and their leader in the time of Hishãm, Hassän ibn Mãhawayh, is obviously a first-generation convert (references given by P. Crone, Slaves on Horses, Cambridge 1980, 237-38 n. 362). It is perhaps because they were exempted from poll tax that later Muslims assumed they had converted. But it is also true that for some— e.g. for the 15,000 Muslim soldiers of Egypt who “believed in Christ and were baptised” during the first Arab civil war (Sebeos, History, trans. Robert Thomson, Liverpool 2000, 154)—collaboration and conversion went hand in hand. 37“Dialogue against the Jews”, PG (ed. J.P. Migne) 89 (Paris, 1865), 1236A -B . 38 John of Nikiu, Chronicle, trans. R.H. Charles (London and Oxford, 1916), 201, cf. 182. 39For examples see Hoyland, Seeing Islam, 100-101, 162-63, 267. 40 “Le synaxaire arabe jacobite”, 16th Barmahat. This was probably due chiefly to the decree of Hafs ibn al-WalTd, who was trying to gain support for his recapture of the governorship of Egypt, that all who converted would be exempted from the poll tax.

INTRODUCTION

xxvii

chronicler: “form ing groups of twenty, th irty and a hundred m en, two and three hundred, w ithout any kind of compulsion to this, they w ent down to H arran to the governors and becam e M uslims (m haggrïn)”.41 Only a very few years later the caliph al-M ahdl decreed th a t all converts who subse­ quently returned to their form er faith were to be pu t to d e a th ,42 confirm ing th a t it was indeed the second half of the eighth century th a t saw apostasy reach significant proportions. T hough the prospect of the death penalty m ust have p u t a brake on the num ber of those forsaking Islam, legal w ritings illustrate th a t all religious com m unities of M uslim lands were afflicted by the phenom enon of apostasy and had to m ake provision for it in their legislation. A fundam ental question was w hether renegades should be ad m itted back into the fold, to which the answ er was generally affirm ative as long as the offender did some form of penance .43 Those who refused to recant were either ostracized or, in th e case of those defecting from Islam, p u t to d e a th .44 A nother key issue was inheri­ tance, the essential point here being to prevent the hem orrhage of p roperty o u t of the com m unity. Accordingly, legislators of the various confessions ruled th a t ap o states may not inherit from their form er co-religionists ,45 and th a t their e sta te was forfeit and to be divided am ong their heirs .46 T hen 41 Chronicon anonymum pseudo-Dionysianum, II, 385. 42Elias of Damascus, “Passion” in F. Combefis, Christi martyrum lecta trias (Paris, 1666), 181. Not only out of regret did people reconvert; the caliph ‘A ll ibn A bl Talib (r. 656-60) put it to one Christian who had become a Muslim, then apostatized, that he was merely doing so to qualify for an inheritance or marriage, then would return to Islam. See ‘Abd al-Razzãq al-San‘ãnT, Musannaf, ed. H.-R. al-A‘zamT (Beirut, 1970-72), VI, 104. 43The earliest Christian witness is Jacob of Edessa (d. 708; see Hoyland, Seeing Islam, 163). On the Zoroastrian side see Manushchihr (9th c.), Dãdistãn ï dënïg, ed. T .D. Anklesaria (Bombay, 1913), no. 40, and Emëd I Ashawaliishtân ( 10 th c.), R ivây at, ed. B.T. Anklesaria (Bombay, 1962), no. 26. The earliest extant Muslim authority is probably Mãlik ibn Anas (d. 795), M uwatta’, ed. M.F. ‘Abd al-Bãql (Cairo, 1951), II, 737, who adduces a report in which ‘Umar I reprimands Abü Müsä al-Ash‘arT for not offering an apostate three days to repent before beheading him. 44E I 2, s.v. “Murtadd” (by W. Heffening, 1993). Manushchihr, Dãdistãn t dënïg, no. 40, opines that apostasy by an adult is “worthy of death”, but presumably this was not applied. 45For instance, the ruling that a son who is of a different religion to his father may not inherit from him is found in Simeon of Rewardashir, Canons no. 18, in Sachau, Syrische Rechtsbücher, III, 249; Gaonic Responsa no. 11 , 4b; Emëd ï Ashawahishtãn, Rivãyat, no. 4. 46The wealth of an apostate from Islam, were he executed or in Byzantine territory, went to his Muslim heirs (e.g. ‘Abd al-Razzãq, Musannaf, VI, 104-105; X, 338-39). Gaonic Responsa no. 87, 20 b, records an argument over whether the estate belonging to a woman who converted to Christianity should go to the husband or to the family. Emëd ï Ashawahishtãn, Rivãyat, no. 4, states that if the apostate possesses property,

xxviii

INTRODUCTION

there was the question of w hat to do in the eventuality of one’s spouse apostatizing, an action which was widely held to con stitu te grounds for di­ vorce .47 Beyond this there were num erous special cases to consider, such as the apostasy of the enslaved, the inebriated and the insane, w hat to do a b o u t crim inal offences com m itted by or upon an ap o state, and so o n .48 Shortly after the m a tte r of apostasy begins to feature in our sources, ac­ counts of the trials endured by individual C hristians a t the hands of M uslims become increasingly com m on .49 The purpose of these works was twofold: to provide role models and heroes to further the fight against apostasy, and to serve as anti-M uslim propaganda. Thus, w hether rep en tan t a p o states or com m itted Muslim converts to C hristianity, m any of the m arty rs are por­ trayed as having purposefully sought out their fate and as having been de­ liberately provocative tow ards the Muslim authorities. And in the interview betw een the m arty r and the prosecuting Muslim th a t com m only features in these accounts, the form er disdains the m aterial advantages of w ealth and s ta tu s th a t are the only inducem ents of Islam, and he is shown to choose the gifts of C hristianity, which consist in virtue, tru th and redem ption. T h a t this choice is right is confirmed by the miracles th a t frequently a tte n d the m a rty r’s death. The first examples of such w ritings are little m ore th an rep o rts of how the m arty r died. Gradually, however, they become m ore “whoever from among of the people of the Good Religion seizes it first, is entitled to it” , though accepting that “in our era this is difficult to practise”. 47Isho‘bokht, Maktbãnutã d - ‘al dfne, III, 56 ( 2 .XI); Isho‘ bar Nun, Canons, no. 114, in Sachau, Syrische Rechtsbücher, II, 168; al-Shâfi‘ï, Kitãb al-umm (Bulaq, 1903-1908), VI, 149-51. For Muslim lawyers there could be no divorce until the woman’s waiting period ( ‘idda, three menstrual cycles) had expired, up till which time either partner could still repent and then remain married. In Zoroastrianism there was also the question of the social status of a married woman whose brother had converted to Islam (Emëd ï Ashawahishtãn, Rivãyat, nos. 1- 2 ). 48 These and many other questions are posed and answered in the A si of al-Shaybãnl (d. 189/805), an annotated translation of which is given by M. Khadduri, The Islamic Law of Nations (Baltimore, 1966), 195-229. On the non-Muslim side some other rulings are: someone who betrays a convert to Islam who has returned to Christianity is excom­ municated (Isho‘ bar Nun, Canons, no. 124, II, 172); a priest or deacon who apostatises then returns to Christianity must do a long atonement (ibid. no. 117, II, 170); incitement to apostasy is a sin (Manushchihr, Dãdistãn ï dënïg, no. 40); dissuading someone from apostasy is a meritorious act (ibid., no. 41); one may make an invocation for the soul of an apostate who returned to Zoroastrianism (Emëd I Ashawahishtãn, Rivãyat, no. 26). 49 Muslim sources do mention Jewish and Zoroastrian converts to Islam who subse­ quently renege and are martyred, but no accounts of them seem to have been composed in their own tradition and in Muslim sources they tend only to be adduced in support of a legal point with few details ever being given. For further information and references see Hoyland, Seeing Islam, 336-86.

INTRODUCTION

xxix

developed, borrow ing them es and ideas from earlier m artyrdom literatu re, incorporating m iracles and other signs of divine approval, establishing plots and conventions and so on until, by the end of the eighth century, they have become a fully-fledged literary genre and achieved great popularity.

Conclusion D espite the efforts of the religious authorities to lim it contact betw een their flock and other confessional groups, this was not so easy in practice. In­ deed, one of the m ost significant aspects of the early Islamic period was, as reported in the above-quoted sta te m e n t by a north-M esopotam ian resident, th a t “there was no distinction between pagan and C hristian, the believer was not known from a Jew ” . This initial indifference of the M uslim s to divisions am ong the peoples whom they conquered, when com pounded w ith the flight and enslavem ent of an appreciable proportion of the population and w ith the elim ination of internal borders across a huge area extending from northw est Africa to India, m eant th a t there was considerable hum an interaction across social, ethnic and religious lines. Furtherm ore, the M us­ lims did not restrict the access of non-M uslims to any public places or to any professional occupations, so a t work and a t play they were likely to come into contact with each other. And the Q u r’an itself was quite liberal as regards inter-confessional fraternization. For exam ple, it rules th a t: G ood things are lawful to you, and the food of th e scriptural peoples is allowed to you, as is your food to them . And you are perm itted to m arry virtuous women of the believers and those of the scriptural peoples once you have provided them w ith their dues, living chastely with them w ithout fornication and w ithout taking concubines (5.5). It is easy to find exam ples of such interaction in our sources. A t debates it would often be the case th a t “a num erous crowd is present: Jew s, Hel­ lenes (probably m eaning the “m any Saracens” who are later said to be in atten d an ce), Sam aritans, heretics and C hristians, for th e place is public and in full view” . C hristian priests were perm itted to “give th e blessing of the saints to M uslims and pagans” , to “teach the children of M uslims, H arranians and Jew s” , and to give com m union “to a C hristian wom an who of her own free will m arries a M uslim ” . The num erous non-M uslim secretaries in th e M uslim bureaucracy would often become very close to their m asters; th u s A thanasius bar G um ãyê, secretary to 4Abd al-4AzTz ibn M arw ãn, gover­ nor of E gypt (r. 65-85/685-704), was “not only his scribe, bu t the m anager

XXX

INTRODUCTION

of all his affairs” . And from the repeated rulings of religious authorities con­ dem ning such behavior, we can assum e th a t it was common for all parties to frequent public b ath s and taverns together, to a tte n d one a n o th e r’s festivals, and to m arry, dine, and do business deals across confessional lines .50 T he onslaughts of C hristendom (especially the C rusades), S h iism (especially the Ism ãcTlT Assassins), and the Mongols ushered in a tim e of greater concern for conform ity and less tolerance of diversity, but one should not assum e this to be a constant feature of Islamic history. And certainly for th e period of chief relevance to this book, the first three centuries of Islam (ca. 7 th 10 th centuries), inter-confessional exchange was, to the chagrin of religious leaders, all too common.

50 For references and further discussion see Hoyland, Seeing Islam, 11-12 and n. 4, 162, 604, 150, 147-49. For interaction in chess playing and in taverns see Alan Jones, “A Bridge between Two Communities” , and Philip Kennedy, “Abü Nuwäs, Samuel and Levi”, in Studies in Muslim-Jewish Relations (ed. Ronald L. Nettler) 1 (1993), 31-35, and 2 (1995), 109-25 respectively.

B IB L IO G R A PH Y

T he litera tu re on the subject of “M uslims and O thers” is too vast to survey here, especially given the current tendency to use early Islamic au th o rities and events to inform contem porary debates ab o u t th e tolerance/intolerance of Islam, the s ta tu s of non-M uslims in a m odern Islamic sta te , and the possibility of M uslim /non-M uslim dialogue. We shall list only a few of th e m ore essential a n d /o r recent studies, and leave the reader to follow the recom m endations of these authors for fu rth er reading and to consult the Index Islam icus (under “Dhim m i” , “non-M uslim ” , “Islam and oth er Religions” , etc.). In t r o d u c t o r y W

o r k s/ S urveys

Bat Ye’or. The Dhimmi: Jews and Christians under Islam. London and Toronto, 1985: a negative presentation (see the article of Cohen cited in n. 9 above). Bosworth, C.E. “The Concept of Dhimma in Early Islam” , in Benjamin Braude and Bernard Lewis, eds., Christians and Jews in the Ottoman Empire (New York and London, 1982), 37- 51. Busse, Heribert. Islam, Judaism, and Christianity: Theological and Historical Affiliations. Princeton, 1988 (German original 1988). Chevallier, D., and Miquel, A., eds. Les arabes, du message à l ’histoire (Paris, 1995), Chap. 6 (“Les musulmans et les autres”). Ducellier, Alain. Le miroir de l ’Islam: musulmans et chrétiens d ’Orient au Moyen Age, Vlle-Xie siècles. Paris, 1971; rev. and expanded as Chrétiens d ’Orient et Islam au Moyen Age. Paris, 1996. Encyclopaedia of Islam. Leiden, 1966- 2001. s.v. “Ahl al-Kitab” (by G. Vajda, 1960), “Dhimma” and “Djizya” (by C. Cahen, 1965), “Mawlã” (by P. Crone, 1991). Hoyland, Robert. Seeing Islam as Others Saw It: a Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam. Princeton, 1997. Humphreys, R. Stephen. Islamic History: a Framework for Inquiry (Princeton, 1991), Chap. 11 ( “Non-Muslim Participants in Islamic Society”). Hussain, Shawkat. “Status of Non-Muslims in the Islamic State” , Hamdard Islamicus 16 (1993), 67- 79. Irmscher, Johannes, ed. Rapports entre Juifs, Chrétiens et Musulmans: eine Sammlung Forschungsbeiträgen. Amsterdam, 1995.

xxxii

BIBLIOGRAPHY

Kallfelz, Wolfgang. Nichtmuslimische Untertanen im Islam: Grundlage, Ideologie und Praxis der Politik frühislamischer Herrscher gegenüber ihren nichtmus­ limischen Untertanen mit besonderem Blick auf die Dynastie der ‘Abbäsiden. Wiesbaden, 1995. Lewis, Bernard. The Jews of Islam (Princeton, 1984), Chap. 1 (“Islam and Other Religions”); an expanded version of his “L’Islam et les non-musulmans” , Annales 35 (1980), 784-800. Noth, Albrecht. “Möglichkeiten und Grenzen islamischer Toleranz” , Saeculum 29 (1978), 190-204. Paret, Rudi. “Toleranz und Intoleranz im Islam” , Saeculum 21 (1970), 344-65. Rubin, Uri, and Wasserstein, David J., eds. Dhimmis and Others: Jews and Christians and the World of Classical Islam. Tel-Aviv, 1997. Stillman, Norman. The Jews of Arab Lands: a History and Source Book. Philadel­ phia, 1979. Tritton, A.S. The Caliphs and their Non-Muslim Subjects. London, 1930.

S p e c ia l is t T

o p ic s

Adang, Camilla. Muslim Writers on Judaism and the Hebrew Bible. Leiden, 1996. Bulliet, Richard. Conversion to Islam in the Medieval Period: an Essay in Quantitative History. Cambridge, Ma., 1979. Caspar, Robert, et al. “Bibliographie du dialogue islamo-chretien” , Islamochristiana 1 (1975) onwards: a useful survey of the relevant primary and secondary literature. Cohen, Mark. “What was the Pact o f ‘Umar? A Literary-Historical Study” , JSA I 23 (1999), 100-31. Dennett, Daniel. Conversion and the Poll Tax in Early Islam. Cambridge, Ma., 1950. Fattal, Antoine. “La nature juridique du statut des Dhimmis” , Annales de la faculté de droit de VUniversité Saint Joseph de Beyrouth 1956, 139-54. ________ Le statut légal des non-musulmans en pays d ,islam. Beirut, 1958. Friedmann, Yohanan. “Classification of Unbelievers in Sunni Muslim Law and Tradition” , JSAI 22 (1998), 163-95. ________ “Interfaith Marriages in Sunni Muslim Law and Tradition” , JSA I 24 (2000 ), forthcoming.

BIBLIOGRAPHY

xxxiii

Gero, Stephen. “Early Contacts between Byzantium and the Arab Empire” , in M.A. Bakhit, ed., Proceedings of the International Bilad al-Sham Conference 4.2 (Amman, 1987), I, 125- 32. Gervers, Michael, and Bikhazi, Ramzi. Conversion and Continuity: Indigenous Christian Communities in Islamic Lands. Toronto, 1990. Gil, Moshe. “Dhimmi Donations and Foundations for Jerusalem (638- 1099)” , JESHO 27 (1984), 156- 74. Goitein, S.D. “Evidence on the Muslim Poll Tax from Non-Muslim Sources” , JESHO 6 (1963), 278- 95. Guessous, Azzeddine. “Le rescrit fiscal de ‘Umar b. ‘Abd al-‘Aziz: une nouvelle appreciation” , Der Islam 73 (1996), 113- 37. Khoury, Adel Theodore. Les théologiens byzantins et L ’Islam: textes et auteurs, VlIIe-XIIIe s. Louvain and Paris, 1969. _________“Die rechtliche Stellung religiöser Minderheiten im Vorderen Orient” , Zeitschrift für vergleichende Rechtswissenschaft 84 (1985), 105- 17. Khoury, Paul. Matériaux pour servir à l ’étude de la controverse théologique islamo-chrétienne de langue arabe du VIII au X II siècle. 5 vols. WürzburgAltenberge, 1989- 99. Lapidus, Ira. “The Conversion of Egypt to Islam” , IOS 2 (1972), 248- 62. Lasker, Daniel J., and Stroumsa, Sarah. Jerusalem, 1996.

The Polemic of Nestor the Priest.

Lev, Y. “Persecutions and Conversions to Islam in Eleventh-Century Egypt” , Asian and African Studies 22 (1988), 73- 91. Morony, Michael. “The Effects of the Muslim Conquest on the Persian Population of Iraq” , Iran 14 (1976), 41- 55. _________“Conquerors and Conquered: Iran” , in G.H.A. Juynboll, ed., Studies on the First Century of Islamic Society (Carbondale and Edwardsville, 1982), 73- 87. Noth, Albrecht. “Die literarisch überlieferten Verträge der Eroberungszeit als his­ torische Quellen für die Behandlung der unterworfenen Nicht-Muslims durch ihre neuen muslimischen Oberherren” , in Tilman Nagel et a/., eds., Studien zum Minderheitenproblem im Islam (Bonn, 1973), 282- 314. _________“Minderheiten als Vertragspartner in Disput mit dem islamischen Gesetz: die ‘Nachkommen der Juden von Khaibar’ und die Gizya” , in id. and Hans Robert Roemer, eds., Studien zur Geschichte und Kultur des Vorderen Orents. Festschrift für Berthold Spuler (Leiden, 1981), 289- 309. Rahmatallah, M. The Treatment of the Dhimmïs in Umayyad and (Abbãsid Peri­ ods. Baghdad, 1963.

xxxiv

BIBLIOGRAPHY

Rissanen, Seppo. Theological Encounter of Oriental Christians with Islam during Early ‘Abbãsid Rule. Abo, 1993. Rubin, Uri. “Qur’an and Tafslr: the Case of ‘an yadin”, Der Islam 70 (1993), 133-44. Spuler, Bertold. “Der Verlauf der Islamisierung Persiens” , Der Islam 29 (1950), 63-76. Turki, Abdelmagid. “Situation du ‘tributaire’ qui insulte l’Islam au regard de la doctrine et de la jurisprudence” , SI 30 (1969), 39-72. Waardenburg, Jacques. Muslim Perceptions of Other Religions: a Historical Sur­ vey. Oxford, 1999. Ward, Seth. “A Fragment from an Unknown Work by al-Tabari on the Tradition ‘Expel the Jews and Christians from the Arabian Peninsula5” , BSOAS 53 (1990), 407-20. Weigert, Gideon. “A Note on the Muhtasib and Ahl al-Dhimma” , Der Islam 75 (1998), 331-37. Young, W.G. Patriarch, Shah and Caliph: a Study of the Relationships of the Church with the Sasanid Empire and the Early Caliphate up to 820 AD. Rawalpindi, 1974.

1

RELIGIOUS COMMUNITIES IN LATE SASANIAN AND EARLY MUSLIM IRAQ« Michael G. Morony

The emergence o f a segmented society composed o f separate religious communities is usually associated with Islam and marks a major change in forms of social organization which was part o f the general trans­ formation taking place in the Middle East from the fourth to the ninth centuries. This change produced a society composed o f religious communities which amounted to social corporations with their own legal institutions which gave sanction to matters o f personal status such as marriage, divorce, and inheritance. The existence o f such communi­ ties was fundamental to the formation o f Islamic society and serves as the single most important distinction between it and Hellenistic society. The appearance o f such communities was a matter o f both organization and identity, and from a personal point o f view this meant the replace­ ment of other means o f identification based on language, occupation, or geographical location by a primary identity based on membership in a religious community. It should be said at the outset, however, that such transfers o f identity were never as total or precise as they would seem in theory and that ethnic identities managed to survive, sometimes because they coincided with religious communities which gave them institu­ tional form. Apart from possible Hellenistic antecedents (eg. the politeuma), conditions in Iraq in the late Sasanian period were especially favorable for this transformation. As a result, important precedents were estab­ lished there which are crucial for an understanding o f the nature o f the corporate society emerging in the early Muslim period. i) The basis of this article is a paper, “Religious Communities in Umayyad ‘Iraq” delivered at the sixth annual meeting of the Middle East Studies Association of North America at the State University of New York, Binghamton, in November of 1972.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

2 114

M. G. MORONY

There are several criteria which point to the existence of religious communities in the late Sasanian period. The first is the spread of a primarily religious personal identity and religiously sanctioned way of life among Jews and Christians. The religious leaders themselves drew the ordinary faithful into operating communities of law reinforced by forms of religious education that emphasized doctrinally laden liturgy, weekly congregational worship, and concepts of communal property. From an institutional point of view, the Jews were organized into the most fully developed community in Sasanian Iraq complete with a system o f religious law, urban institutions, schools, and synagogues. The religious scholars who produced the Talmud were able to apply the religious law to the life o f the community and to enforce the dietary laws because rabbis trained in the schools were employed by the exilarchs as judges in the rabbinical courts (beth din), as market inspectors and as local administrators who collected the poll tax. Talmudic Judaism was also spread by the disciples o f the rabbis while the semi­ annual assemblies o f the schools served as important and effective means of popularizing the rabbis’ concept of a life ordered by religious law among the lay community. This desire to apply Talmudic law was directly responsible for the formation o f separate urban Jewish commun­ ities where a disciplined liturgical and legal life could be realized most effectively with the result that the change from a rural agricultural society to an urban society was already underway among the Jews of Iraq in the late Sasanian period1). By the early seventh century the Christians in Iraq had been divided into separate Nestorian and Monophysite communities. The Nestorian identity was formalized by the way the Christology and exegesis o f Theodore of Mopsuestia was embedded in the liturgy and teaching of the village and monastic schools and imparted to those trained for 1) The primary source for the Jews of Sasanian Iraq is the Babylonian Talmud. The most recent discussion based on all sources is that of J. Neusner, A History of the Jews in Babylonia, (Leiden, 1970), especially volume 5, pp. xviii-xix, 23, 55, 148-50, 214-15, 245, 265-8. See also S. W. Baron, Social and Religious History of the Jew (New York, 1958).

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

3 11 5

leadership as well as to laymen. Communal discipline was stiffened by the spread of a reformed coenobitic monasticism subjected to rules and by the foundation of new monasteries from the 58o’s to the 620’s. In the same period the practical problems o f church government led to the extention o f canon law to cover marriage, property, and inheritance, first for the clergy (in 544 at the synod of Mar Abã I), and then for the laity (at the synod of Ishö'yahbh I in 585). This coincided with the emer­ gence o f the concept o f church property as community property, with references to communal assemblies composed o f the clergy and faithful at the levels of the village parish and diocese, and, by 576, with the designation of the priest as religious judge1). For their part, Monophysites had begun to establish village schools and by the end o f the third decade of the seventh century had come to form a community o f their own in Iraq as part o f the Jacobite Church o f Syria2). Magians might also be described as forming an imperfectly developed community in late Sasanian Iraq. They were united among themselves and isolated from the members o f other religions around them by a concept o f themselves as a community (dën). They had their own hierarchic, ecclesiastical organization parallel to the state, a distinctive cult, a body o f scripture, and a religious law administered by mõbadhs and hirbadhs that sanctioned certain social customs such as the exposure o f the dead, the prayer o f silence, and endogamy. But Magianism was essentially a ruling-class religion, and Magian Persians formed a kind o f religious caste in Sasanian Iraq. The creation o f a larger group identity was inhibited by the nature o f the cult itself which only required the participation of the priesthood and by the way religious education tended to be confined to the upper classes. While state support may have helped to discourage the development o f completely autonomous

1) J. B. Chabot, Synodicon Orientale, (Paris, 1902), pp. 82-4, 102-3, 123-4, 158, 181-2, 335-7, 359-61, 418,441. 2) F. Nau, “Histoires d’Ahoudemmeh et de Marouta, métropolitains Jacobites de Tagrit et de 1’Orient”, Patrologia Orientalîs, V. 3 (1909), pp. 7-96. For bothNestorians and Monophysites see J. Labourt, Le Christianisme dans l ’empire perse sous la dynastie sassanide, (Paris, 1904).

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

4 I l6

M. G. MORONY

institutions, the way the Magian cult and education were maintained privately within priestly families is o f supreme importance1). Secondly, the formation of such closely knit communities increasingly isolated the members of one group from those of another. The bounda­ ries created between religious groups by separate bodies of law are indicative of the rising barriers to interfaith communication at the end o f the Sasanian period. The defensiveness associated with this development is well symbolized by the reference in the Nestorian synod o f Joseph in 554 to the canons as “high walls, impregnable fortresses, protecting their guardians against all danger” 2). For Jews this meant that the witness of apostates and sectarians (minnim) was not admitted in the rabbinic court, and that the rabbis opposed the sale o f wine to gentiles. Among Christians the social barriers created by restrictions on marriage in canon law were specifically intended to distinguish Nestorians from the members o f other religions. The establishment o f forbidden degrees of kinship for marriage by the synod o f Mar Abã I in 544 was aimed at the Magians while at the same time the prohibition o f the marriage o f a man to his brother’s wife was directed at the Jew s3). N ot only were Nestorian priests forbidden to marry unbelievers in 554, but in 585 the marriage o f any Nestorian with a Monophysite was forbidden 4). The final separation between Nestorians and Monophysites in Iraq occured in the second decade of the seventh century when mixed congregations and monasteries were purged by both sides. Thereafter villages tended to be all Nestorian or all Jacobite. About the same time the gnosticized pagans o f lower Iraq who were becoming Mandaeans were raising barriers between themselves and the J ew s6). Thirdly, the structure of existing religious communities tended to become more rigid because o f the way they received official or semi1) A. Christensen, U lran sous les Sassanides, (Copenhagen, 1944); R. de Ménasce, “L’Eglise mazdéenne dans l’empire sassanide” , Cahiers d’histoire mondiale, V. 2 (195 îT»pp- 554-565. 2) Chabot, pp. 9 7 , 3 55• 3) Ibid., pp. 82-84, 335-37. 4) Ibid.,lpip. 102, 158, 360,418. 5) E. Yamauchi, Mandaic Incantation Texts, (New Haven, 1967), pp. 33-34, 45.

5

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

117

official recognition by the Sasanian state. The importance o f Iraq for defense and revenue combined with the fact that the province was in­ habited largely by non-Persians and non-Magians led the later Sasanians to experiment with means to ensure loyalty by granting military pro­ tection and a degree of religious toleration in return for the payment o f taxes. Tradition reflects this state o f affairs in the story that Hurmizd IV (579-59°) compared the state to a throne and told the hirbadhs that just as the throne needed all four legs to stand, the Sasanian state needed to protect its non-Magian subjects (incidentally described as the two rear legs of the throne), especially in Iraq, in order to survivex). In practice the religious leaders were made responsible for the behavior o f the members o f their own communities. The exilarch was recognized as the head o f the Jewish community which he represented to the state. He was usually allowed to govern the Jews through his own system o f admini­ stration. The importance o f good behavior as well as a suggestion o f the causes o f government intervention is provided by Rabha’s admonition to the people of Mahoza, “I pray you, see that there be concord among you, in order that you shall have peace from the government” 2). In the case o f the Nestorians there were two important consequences o f such toleration which established precedents for Christians as members o f a subject religion. In the first place, the ecclesiastical organization o f the Nestorian Church corresponded to the structure o f the Sasanian state and to the Magian hierarchy o f mõbadhs. This meant not only parallels between ecclesiastical and civil provinces, but the identification of the extent o f the church with the borders o f the Sasanian empire and the application o f Sasanian hierarchic concepts to the church. Under these influences the hierarchy o f the church was regularized, there was a new insistence on the observance o f the cursus honorum, and the Nestorians adopted Sasanian-style hierarchic titles 1) T& 2in,Ta’rïkhar-Rusuln>al-Mulük,(L£iàen., 1879), V. 1, pp. 990-1; T. Nöldeke, Geschichte der Perser und Araber %ur Zeit der Sasaniden, (Leiden, 1879), p. 268; Chris­ tensen, p. 442; A. Scher, “Histoire nestorienne” , Patrologia Orientalis, V. 7 (1950) p. 195. 2) M. Rodkinson, The Babylonian Talmud, (Boston, 1918), V. 13, p. 24.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

6 X1 8

M. G. MORONY

such as “Father o f fathers” , “ Shepherd o f shepherds”, or “Chief of shepherds” for the catholicos patterned after the Sasanian King o f kings and mõbadhãn-mõbadhx). Correspondence was also expressed in terms of a theory of double sanction, ecclesiastical and secular, for church government. The state was expected to enforce the decisions of Nestorian synods and the Nestorians were expected to express their loyalty to the Sasanians by praying for the monarch and by using excommunication against rebellious Christian subjects2). In the second place, toleration brought with it the requirement for royal permission for the building o f churches and monasteries, for Christian burial, for the promulgation o f monastic rules, and for the election of the catholicos. In the latter case the desire of the government for stability and loyalty through the nomination o f the catholicos by the king conflicted with the desire of the clergy for ecclesiastical autonomy and led to irregular elevations and the involvement o f aristocratic Persian laymen at court in church politics. This situation was responsible for the emergence of two factions which each sought to control the church: the clergy and monks who favored canonical elections, and the lay aristocrats who favored a situation in which they could exercise patronage as astrologers and physicians. The way this conflict was institutionalized at Nasibin towards the end of the sixth century made that city an important epicenter o f the conflict. At about the same time that reformed coenobitic monasticism with strict rules was being introduced to the monastic communities on Mt. Izla above the city, the Nestorian school in Nasibin with its medical branch fell under the in­ fluence of Henäna of Adiabene who was trying to work out a theologi­ cal compromise with the Monophysites. The conflict, which was only aggravated by the involvement o f the Monophysites and by the war between the Persians and Byzantines, was at its height in the first three decades of the seventh century, and it was largely because o f it that from 1) Chabot,pp. 103, n o , 112,196, }6i, 368, 370,456,527,534. 2) Chabot, pp. 131, 391; I. Guidi, “ Chronica Minora I” , Corpus Scriptorum Cbrtstianorum Orientalium, (Louvain, 195 5), V. 1, pp. 15-16, V. 2, p. 15 ; A. Scher, “Histoire nestorienne” , Patrologia Orientalis, V. 13 (1919), pp. 4 4 1 - 2 , 491.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

7 II9

609 to 628 Khusrau Parviz refused his permission for the election o f a catholicos. Parallel to these developments the Sasanian tribute tax which pro­ ductive non-combatants were required to pay in return for military pro­ tection was beginning to be applied as a poll tax to non-Magians in Iraq. The exemption of the royal family, the high nobility, soldiers, hirbadhs, secretaries, and others in royal service from the payment o f the poll tax by Khusrau Anüshirvän meant that in Iraq it was paid mostly by the non-Persian, non-Magian part o f the population. The exemption o f the hirbadhs raised the question o f the exemption o f those engaged in religious professions. The rabbis who collected the poll tax from the Jews were in a position to exempt themselves, but the Christian clergy and monks were still trying to escape the tax at the end o f the Sasanian period1). The effects o f the Muslim conquest on the religious communities in Iraq were contradictory. In the first place there was no official recog­ nition of either Jews or Christians as religious communities by the Muslim state at the time o f the conquest in the same sense that they had constituted communities in the late Sasanian period. Nor was there any official distinction made among non-Muslims. Christians, as such, were not treated as protected people. There were raids on both Monophysite and Nestorian monasteries ; monks were killed, taken prisoner or driven to take refuge elsewhere. Large numbers o f Monophysite Arabs were killed, enslaved, or converted to Islam. All non-Arabs and non-Muslims in Iraq were treated as subjects liable to the tribute tax, and in mixed cities such as Hira and Mada’in (Ctesiphon and Seleucia) tribute was imposed on the entire population without regard for religious differ­ ences. This picture, drawn from a combination o f Arabic and Syriac sources, is corroborated in a rather stylized way by the Nestorian writer, Bar Penkãyê, towards the end of the seventh century who says that the Arabs at the time o f the conquest were only interested in raiding, 1) Neusner, V. 5, pp. 172, 272; Rodkinson, V. 12, pp. 181, 223; Tabari, V. 1, p. 962.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

8 120

M. G. MORONY

taking captives, and tribute and were willing to allow their new subjects to believe anything they wished. He deplored the demoralizing conse­ quences of undiscriminating toleration in the reign of Mu'awiya (660-680) when “there was no difference between pagan and Christian; the faithful was not distinct from a Jew” 1). One result o f the conquest was to break the relationship which had existed between the Nestorians and their rulers under the Sasanians. The organizational correspondence between church and state ceased to exist under Muslim rule because the Nestorian church provinces perpetuated Sasanian conditions which did not necessarily correspond to the Muslim administrative system. By 648 the seat o f the catholicos had returned to Mada’in although that city was no longer the capital of the state or even of the province. Sasanian-style hierarchic titles were also dropped after the conquest. The catholicos George I (661-680) called himself only “catholicos, patriarch o f the East.” Henanishõc I (686-693) referred to himself only as “catholicos, patriarch by the grace o f God” 2). Under these circumstances the late Sasanian pattern of relationships between the state and the religious communities began to be restored by the initiative of the religious leaders themselves who assumed they should deal with their new rulers in the same way they had with the Sasanians. Although the account is legendary, it is worth noticing that the question o f the exilarchate was brought to the attention of the Muslim authorities by Bostanai (d. ca. 660) himself whose position had been usurped by his guardian during his minority. Only then did the Muslim government confirm his power to appoint judges and the heads o f the Talmudic schools at Sürä, Nahardea, and Pumbaditha. In the same way, Nestorian leaders managed to have themselves exempted from the poll tax at the local level before such exemptions were recog­ nized in theory. In one case, in the caliphate o f ‘Umar I (634-644), priests and deacons at Kaskar had been made liable to pay the poll tax and a certain Rabban Theodore, who was a teacher at Kaskar, petitioned 1) A. Mingana, Sources Syriaques, (Leipzig, 1908), pp. 175, 179. 2) Chabot, pp. 215, 221, 245, 480, 490, 515; C. Sachau, Syrische Rechtsbücher, (Berlin, 1907), Y. 2, pp. 12-13.

9

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

121

the ‘ãmil, N u‘mãn b. Muqarrin, and secured an exemption for them in w ritingL). In another instance, at mid-century, Sabhrisho’ the metro­ politan bishop of Bëth Garmê (Bãjarmã) who had cured the two demonpossessed daughters of the governor o f the district, requested and received an exemption from the poll tax in writing for monks, priests, and students (anyone who wore wool, whether tonsured or n o t2). By the time of Mu'äwiya the Nestorian Arabs (‘Ibãd) o f Hira were em­ ployed as tax collectors by the Muslims, and in 676 Nestorian canon law prohibited a Christian tax collector from collecting the poll tax from a bishop3). On the other hand, the lack o f interest on the part of the Muslim government in the internal affairs of the religious communities in Iraq encouraged the continued operation and development o f autonomous systems of religious law. The only exception was in the case o f the Magians. Because of their official status under the Sasanians, fire temples were treated as state property and confiscated at the time o f the conquest. There are no indications o f the survival o f a public Magian cult, organization or hierarchy in early Muslim Iraq. A typical example of the fate of Magian institutions in Muslim Iraq is provided by the description, in the eighth century, o f a ruined fire temple by the side of a road in Marghä infested by devils in the form o f black ravens 4). But, apart from the Muslim attack on the official aspects of Magianism, there was no interference in private matters. Individual Magians were allowed to pay tribute along with everyone else, and Magian religious and social traditions were preserved in the private practice of upper-class Magian 1) Scher, V. 13, pp. 598-9. Nu'man b. Muqarrin had been appointed ‘ãmil of Kaskar by Sacd after the conquest. He remained there until he was sent to Nihãvand by ‘Umar in 642. Abü Yüsuf, Kitab al-Kharãj, tr. E. Fagnan, Livre de IHmpöt fancier, (Paris, 1921), V. i, p. 50; Ibn Sa'd, Kitãb at-Tabaqät al-Kabtr, (Leiden, 1909), V. 6, p. n ;T ab arl, V. i,p . 2596. 2) Scher, V. 13, pp. 632-3. 3) Chabot, pp. 225-6, 490. The question of whether or not monks capable of working should be subject to the poll tax continued to be discussed in Islamic law (Abü Yüsuf, p. 188). 4) E. A. Wallis Budge, The Book of Governors: The Historia Monastica of Thomas Bishop of Marghâ, (London, 1893), V. 2. p. 599.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

10 122

M. G. MORONY

families which survived into the eighth century at Hira and Basra and other towns of the Sawãd and in isolated villages in northern Iraq. At such places Magians continued to practice the family cult, endogamy, the exposure of the dead, and provided religious education for their children1). Otherwise, the situation favored the continued communal develop­ ment o f both Jews and Christians. The religious and social regulations in the Qur’an did not apply to non-Muslims and the Qur’an itself (Süra 5143-48 ; 62:5) regards the Torah and the Gospel as the sources of divine law for Jews and Christians to be used by them for judgment. The survival o f communal institutions from the Sasanian period such as the religious schools and court systems, synagogues, churches and monasteries, and the Jewish Rabbinate and the Christian ecclesiastical structures led to an increase in religious authority and to the division between religious and secular powers in the communities themselves by the Umayyad period. For the Jews this meant the emergence o f the split between the rabbis and the exilarch and the rise of the Gaonim as the real religious author­ ities in the community. The background to this development lies in the completion o f the Talmud by the ear) y sixth century followed by the interregnum in the office o f exilarch in the early seventh century when the heads of the schools effectively acquired religious authority for themselves. After the restoration of the exilarchate in the person of Bostanai exilarchs were never able to fully reassert themselves because o f the absence of state support for their religious authority in early Muslim Iraq. The raising of the legal issue which provided the pretext for the break coincided with the first period of Umayyad rule in Iraq when the sons of Bostanai by his Jewish wife, Hisdai and Hanina bar Adai, who succeeded him as exilarchs down to 689 sought to disinherit and claim 1) The only indication of Muslim interference in private Magian affairs is the case in 643 when the caliph ‘Umar instructed Bajalab. ‘Abda al-Anbãrl, who was secretary for Jaz5b. M u ’ä w iy a at Basra to separate every Magian from his wife and children and to forbid the recitation of the liturgy (zamzama). The order was quickly rescinded. Abü Yüsuf, p. 199; V. Büchner, “Madjus”, Encyclopaedia of Islam (1), V. 3, p. 99; Ibn Sa‘d, Y. 7(1), p. 94.

11

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

1 23

as their own slaves the children of Bostanai by a Persian princess. The rabbis insisted on the rights of Bostanai’s children by his Persian wife, but the real issue was over who had the authority to interpret and apply religious law in matters o f family life and inheritance. While the sons o f Bostanai deposed the heads of the schools and appointed their own partisans, their ultimate inability to impose their will with regard to their half-brothers provides the best evidence that the recognition by the Muslim government o f the exilarch’s right to appoint judges and the heads of the schools did not extend to authority over the interpretation of the la w x). It was in this period, and largely a consequence o f this conflict, that the rabbis took advantage of the lack o f intervention by the Umayyad government to institutionalize their independent religious authority for themselves in the gaonate by reviving the principle of continuing revelation to justify new gaonic decrees which abrogated and transcended Talmudic laws. The most instructive example is provided by the joint decree by Mar R. Huna at Süra and Mar R. Rabba at Pumbaditha about 670 that a wife could sue for divorce without losing her property rights. This is not only significant o f the freedom of the rabbis to adjust the law on their own authority, but since the purpose of the decree was to make it unnecessary for Jewish women to go to Muslim courts to obtain divorces, it is also an indication of the way this kind o f religious authority favored the creation o f barriers between the communities2). Ibn Da’üd set the year 689 as the beginning of the gaonic period, and that year does seem to have been a turning point in the emergence of the gaon since it marks the end of the first persecution o f the rabbis by the sons of Bostanai and the elevation o f Hanina o f Nahar Pekod as head of 1) A. Goode, “The Exilarchate in the Eastern Caliphate” , The Jewish Quarterly Review, V. 31, (1940), pp. 155-56. 2) H. Graetz, History of the Jews, (Philadelphia, 1941), V. 3, p. 92. This legal issue may have been related to the point of Muslim law that allowed a divorced woman full maintenance during the legal period of separation (cidda) which was based on a variant reading of the Q ur’an transmitted by ‘Abd Allah b. Masüd at Küfa. N. Coulson, A History of Islamic Law, (Edinburgh, 1964), p. 31; Ibn Sa'd, V. 6, p. 4.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

12 124

M. G. MORONY

the school o f Pumbaditha (689-697). There was a second persecution from 7x9 to 730 when Natronia b. Nehemiah (Mar Yanka) was head o f the school at Pumbaditha and drove the rabbis opposed to the house o f Bostanai to take refuge at Süra. This resulted in the permanent elevation of the gaon o f Süra over his colleague at Pumbaditha for the rest of the gaonic period. From 730 on the division between religious and secular powers in the community and the dual rule o f exilarch and gaon may be regarded as established. Judges for the rabbinic courts were appointed jointly by the exilarch and gaon while each gaon had the right o f judicial review over the judges under his jurisdiction. The office of exilarch had become mainly ceremonial, his revenues were limited to districts and towns designated by the state, and he lost his claim to religious authority and his jurisdiction over criminal cases. By the eighth century as well, dynastic succession was beginning to be under­ mined by the concept of election. The Gaonim held the deciding voice in the choice o f a successor from among the family o f the exilarch and the investiture ceremonies institutionalized the subordination o f the exilarch to the Gaonim in religious matters. In a similar way the emergence o f the maphrian as the head o f the Jacobite Church in Iraq and the East has more to do with ecclesiastical independence than with state recognition and is comparable to the rise of the gaonate among the Jews. The Monophysites in the East had only established formal ties with the Jacobite Church in the West in 629 after the end o f the war between the Persians and Byzantines. At that time Märüthä (629-645) had been elected grand metropolitan o f the East with his seat at Takrit. It was Märüthä who secured terms for the citadel of Takrit at the time of the Muslim conquest and who created the position that was later called maphrian. This was only a term of honor at first, like gaon, and, significantly, only began to be used as an official title under Barishõ’ (669-683) in the reign o f Mu'äwiya. The Nestorians also continued to strengthen the bonds o f their community by the development and application o f canon law. As early as the 640’s the marriage laws were reiterated by Isho'yahbh o f Adiabene, but the important developments only resume in the reign o f

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

13 1 25

Mu'äwiya. The synod o f George I (661-680) extended church sanction to betrothal by requiring the contract to be made legal by a benediction pronounced in the presence o f the cross, and Christian women were forbidden to marry pagans x). By that time the Muslim government had adopted the practice o f appointing judges for the Nestorians and the synod required litigation between Christians to be judged in the church by judges chosen by the bishop with the consent o f the community. N o one was to take it upon himself to act as judge unless he was obliged to do so by the government. From other considerations Ziyäd was probably responsible for this situation. This practice was eventually recognized in the Iraqi legal tradition and, according to Abü Hanlfa, the appointment by the government of a non-Muslim as a judge among the members of his own religion was valid2). The best description o f the organization and legal life o f the Nestorian community is provided by the correspondence o f the catholicos Henan!shõc I (686-693) during the second civil war. The system o f legal administration was based on church districts at the local level o f the village parish where the division between religious and civil authority occured among the Nestorians. There were two officials in each district, both subject to the catholicos: the priest who served as judge, and a civil ruler (shõltanã ‘almanayã) who probably enforced the priest’s decisions and may have represented the Nestorians before the local Muslim authorities3). Legal decisions were appealed to the catholicos by unsatisfied litigants, and the legal responses o f Henanishõc served as precedents on which general statements o f law were based. Several interesting and important points emerge from this correspondence. Written documents were regarded as the most acceptable evidence once 1) Chabot, pp. 223-4, 487-8; R. Duval, “Iso'yahb Patriarchae III Liber Epistularum” , CSCO, V. 11, pp. 15 3-4, V. 12, p. 114. 2) Chabot, pp. 219-20, 484-5; E. Tyan, Histoire de l ’organisation judiciaire enpays d’Islam, (Leiden, i960), p. 90. 3) Sachau, V. 2, pp. 26-7. The Arabie sources describe the Christians of Najrãn who were resettled near Küfa as being under the authority of their bishop, civil ruler (cäqib) and notables. Abü Yüsuf, p. i n ; Balãdhuri Kitãb Futüh al-Buldãn, (Leiden, 1866), p. 66.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

14 X2 6

M. G. MORONY

they had been sealed by the ecclesiastical inspector. Poll taxes were imposed as a lump sum on an entire village, and it was up to the villagers themselves to compute and collect them according to differences in wealth. The tax of deceased people was also incumbent on the village if they had not been sealed before they died1). Taxation practices were used as precedents to derive principles for inheritance by analogy as in the case in which H^anishõ' decided that since the land and poll taxes incumbent on the estate were paid by each heir in proportion to their share, any debts on an estate should also be divided among the heirs in proportion to their share o f the inheritance. The Church even claimed the right to inherit the estates o f those who died intestate “in order to take responsibility for taxes levied on them” 2). This post-conquest tendency towards nearly independent religious communities with a minimum o f interference or authorization by the Muslim government was offset by the way conflicts carried over from the Sasanian period were combined with the insistence of religious leaders and others on approaching their new rulers on the same terms and for the same purposes as they had the Sasanians. This is particu­ larly clear in the case o f the Nestorians where the church leaders and aristocratic laymen invited state intervention by involving the Muslim government in their internal quarrels. Muslims were drawn into the hostilities between Nestorians and Monophysites in Iraq by the Nesto­ rian catholicos îshô'yahbh III (647-658) who is said to have bribed the Muslim government in order to prevent the Jacobites from building a church of their own in Mawsil. The continuation o f the conflict between the Nestorian factions for the control of the Church was given added significance after the Muslim conquest by the Persian aristocrats o f upper Iraq who managed to pre­ serve their local influence and their lands by putting their property under monastic ownership while they continued to control it by acquiring positions of leadership in the monasteries or in the Church3). 1) Sachau, V. 2,pp. 14-15,16-17,42-43,184. 2) Ibid., pp. 18-19, 36-37. 3) J. Chabot, “ Le livre de la chasteté” , Mélanges d'archéologie et d’histoire, V. 16

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

15 127

Consequently, instead of taking advantage o f the change o f regime to secure real ecclesiastical autonomy following the Muslim conquest, the nature of the conflict between the Nestorian factions had the effect of recreating the conditions of the late Sasanian period. Significantly, the pattern emerges first at Nasibin immediately after the conquest. The metropolitan Cyriacus had been involved in the triumph o f the monas­ tic-clerical faction in the elevation o f Ishö'yahbh II (628-643) as catholicos in 628. About 640 his enemies at Nasibin brought charges against his followers to the Muslim amir who allowed them to plunder his cell and the residence of the metropolitan1). The aristocratic party seems to have recaptured the office of catholicos in the person o f Ishö'yahbh III who appointed two other Persian aristocrats, both named George, as metropolitan bishops of Nasibin and Perãth dhe Maishãn (al-Furãt). The elevation o f George the monk to the office o f catholicos as George I about 66 x coincided with the beginning o f Umayyad rule in Iraq and was the signal for the ecclesiastical revolt o f his bishops. George of Perãth brought accusations against the catholicos before the Muslim govern­ ment and George I was imprisoned by an Arab governor who tried to extort money from him a). It is not surprising, then, that the synod held by George I in 676 rejected the concept o f church-state correspondence and declared that ecclesiastical procedures such as ordination were not to conform to secular practices but to canon law 3). That this was only a futile attempt to reverse the prevailing state of affairs is reflected by the judgment of (1896), pp. 271-2; Elias of Nasibin, “Opus Chronologicum” , CSCO, V. 62, pp. 56-57, V. 63, p. 31 ; Thomas of Marghã, V. 1, pp. 82,124, V. 2, pp. 150-151, 170-180, 637. 1) Guidi, V. I , p. 31, V. 2, p. 26. 2) Thomas of Marghã. V. 2, pp. 181-2, 187-8; Chabot, Synodicon, p. 480. There are two possibilities as to who this governor was. George of Peräth was in the jurisdiction of ‘Abd Allah b. ‘Amir b. Kuraiz who was governor of Basra from 661 to 665. But the catholicos at Mada’in was under Küfan jurisdiction and Mughira b. Shu ’ba was in his second term as governor of Küfa from 662 to 669. Other considerations would seem to support the identification of Mughira as the Muslim governor who had dea­ lings with the Nestorians. 3) Chabot, Synodicon, pp. 220,485.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

16 128

M. G. MORONY

Bar Penkãyê that the lack of persecution in the reign of Mu'äwiya had proven disastrous for the Nestorians. Bishops were involved in public affairs, anti-canonic quarrels, civil tribunals, and contested the rule among themselves, while the other church leaders “only understood that they were men and that they governed men” 1). The involvement o f the Nestorians in the closing phase o f the second civil war in Iraq brought the first real intervention o f the Muslim state into their internal affairs and effectively completed the process by which the Nestorians themselves forced the Muslim government to apply Sasanian methods towards them. Although the catholicos Henan!shõc at Mada’in was under rebel rule, first under Mukhtär at Küfa from 685 to 687 and then under Mus'ab b. az-Zubair from 687 to 690, he seems to have been neutral. His concern for the well-being of his flock took the form of ad hoc arrangements and appointments in lower Iraq and independent legislation for the community. It was the metropolitan o f Nasibin, John of Dãsen, who took advantage o f the situation to acquire the office o f catholicos for himself. The Umayyads had defeated Mus'ab in 690 and from 690 to 693 Bishr b. Marwãn, a brother o f the caliph ‘Abd al-Malik, was governor of Küfa. Nasibin remained in rebel hands and the bribe which John o f Dãsen is said to have made to Bishr to secure the deposition of Henanishõ‘ and have himself made catholicos was probably contingent on John delivering Nasibin to the Umayyads. At any rate, in 693, for the first time under Muslim rule, Bishr deposed Henanishõ‘ and elevated John of Dãsen. John’s partisans in Nasibin led by an aristocratic Christian Persian physician named Mardanshãh then helped ‘Abd al-Malik’s other brother, the governor of the Jazira, Muhammad b. Marwãn, take the city. This alliance only lasted until the death o f John early in 695. Bishr had already died and had been replaced by Haj jäj who refused to allow the election o f a new catholicos, again marking the first time under Muslim rule that permission was necessary. The hardening o f the government’s position under Hajjäj coincided with the fall of Mardanshãh and his family at Nasibin. 1) Mingana, p. 178.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

17 129

H^anishõ* continued to exercise a kind o f shadow catholicate in semiretirement at the monastery o f Mar Yünân near Mawsil until the death o f Hajjäj when they were able to elect Selibhazekhä (714-728)1). The point to be made here is that this kind o f intervention was a consequence of the second civil war and was part o f the Umayyad restoration in Iraq which introduced other aspects o f Sasanian statecraft in addition to the requirement o f government permission for the election of a catholicos and the elevation or deposition o f the catholicos for political purposes. In general non-Muslims involved Muslims in their internal quarrels by the bribery of government officials and denunciation. According to both Sherira and Ibn Da’üd the exilarchate was acquired by bribing the authorities2). Under such circumstances it seems only natural that the revival o f Sasanian administrative practices and theories by Umayyad governors in Iraq should have included the methods by which the Sasanians had dealt with the members o f subject religions. This seems to have been especially true o f Umayyad governors such as Ziyäd and Hajjäj whose authority over all o f Iraq and its dependencies in the East put them in charge o f territory that nearly corresponded to that o f the Sasanian empire. Ziyäd held the combined governorship o f Basra, Küfa and the East from 669 to 672 and it seems too much to suppose that the cluster o f events in the life o f the religious communities that occured around 670 were not related to Ziyäd’s co-ordination o f the four Sasanian bureaus (military, finance, correspondence, and docu­ ments) for the purposes o f Muslim administration, to his formation o f a Sasanian-style guard alongside the shurta in Basra, or to the way he was served by a chamberlain (häjib) when he held audience. By the same token, Hajjaj’s treatment o f the Nestorians should be put in the context o f the revival o f the Sasanian court for the redress o f grievances (mazälim), the formal review o f troops in the Sasanian manner and the justification o f the authoritarian rule o f the Umayyad restoration by 1) F. Baethgen Fragmente syrischer und arabischer Historiker, (Leipzig, 1884), pp. 32-38; Elias of Nasibin, V. 62,pp. 55-6,149,153,155, V. 63,pp. 31,72-4; Mingana, p. 184; Sachau, V. 2,pp. vi-xiv, 30-33. 2) Neusner, V. 5,p. 136.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

18

130

M. G. MORONY

an appeal to fatalism that served the same political purposes as Zurvanism did under the last Sasanians. This development was paralleled by the reapplication o f the principle o f correspondence between church and state administrative districts. This appears first in the Nestorian church provinces along the upper Tigris with the formation o f the metropolitanate of Mawsil or Äthör by combining Adiabene (Hazza), Bëth Garmë and the districts north o f the greater Zab river to correspond to the Muslim province of Mawsil. This process seems to have begun in the reign o f Mu'awiya when, about 670, Bëth Garmë was subordinated to Adiabene1). It was completed by the early eighth century when Selibhazekhã consecrated a metropolitan bishop for Hazza and Mawsil who had his seat at the monastery o f Rabban Qusra in M awsil2). While the set o f relationships which had existed between the Nestorians and the last Sasanians was reconstructed vis-à-vis the Muslim government in this way during the Umayyad period it should be pointed out that the royal nomination of the catholicos and the conflict between nomination and election was only reintroduced under the ‘Abbãsids in 775. Nor is there any indication that the Muslim government ever intervened actively in the affairs o f the Jewish community until after the advent o f the ‘Abbãsid dynasty and then it was to support the Gaonim in the context o f the Karaite schism. This picture o f the nature o f religious communities in early M u slim Iraq would be incomplete without some consideration of Muslims as a ruling religious community. Although the social realities of M uslim Arabs in this period have more to do with tribalism, Islam contained within itself the concept of a community that replaced the bonds o f kinship with a bond o f faith. The working out o f this concept among Muslims in Iraq where they forged the bonds o f a religious community after the fashion o f the people around them lies at the heart of the emergence o f an Islamic society3). 1) Chabot, “chasteté” , pp. 67-68, 280-281; Duval, V. 11, p. 208, V. 12, p. 151. 2) Sachau, V. 2, p. xiv; Scher, V. 7, p. 201 3) See Cl. Cahen, “Histoire économico-sociale et islamologie. Le problème pré-

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

19 131

Muslims brought the foundations o f their own religious law, a form o f congregational worship, and the mosque with them to Iraq where the development o f their own vocations of religious leadership, education, and legal administration combined their own background with local patterns of organization. At first the Muslim community was virtually identical with the army and its dependents settled in the garrison towns where communal solidarity was expressed in congregational worship, and the leadership of the prayer by the governor as imãm made him the official head o f the Muslim community in his province. But the natural religious leaders of the community were a group o f first and second generation Muslims, the Companions (ashãb, sahãba) and Followers (tâbi un), whose authority was based on their ability to remember and interpret the Qur’an and the practices o f Muhammad. At first, as long as ‘Umar followed a policy of Islamic priority in his appointments, there was no difference between religious and secular leaders. Abü Müsä al-Ash‘ari was both a Companion and governor o f Basra from 2 2 /6 4 3 to 2 9 /6 4 9 and he produced his own system of vocalization (mushaf) for the Qur’an. But the reassertion of aristocratic tribal leadership in the caliphate o f ‘Uthmãn reduced these religious leaders to an unofficial or at best semi-official interpretation and development of Islamic requirements for the benefit o f converts. Their own behavior set examples for converts to follow as in the way the Follower (tãbi‘i) Ibn Abi Laylã declined the opportunity to dry his hands after performing the ritual ablution (wudü’) at Küfa. A degree of official recognition was extended to them by Ziyäd who paid stipends o f between three hundred and five hundred dirhams per year to five hundred shaykhs at Basra when he became governor there in 4 5 / 66$. The way early M u slim religious spokesmen set religious usages in an authoritative manner by their own example and interpretation was sanctioned by the assertion o f the infallibility of the community and was too close to the way the rabbis and their disciples developed and spread Talmudic law by their own prac­ tices and teaching to have been coincidence. It seems to have been a comjudiciel de l’adaptation entre les autochtones et l’Islam” , Correspondance d’orient, V. 5 (1961), pp. 1-19.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

20 132

M. G. MORONY

bination of the preservation of tribal sunna by pre-Islamic tribal shaykhs with the kind of authority exercised by the rabbis over the Oral Law. The emergence o f the qâdï as a judge for Muslims was part o f the breakdown or delegation o f the powers of the amir which began as early as 642 in Iraq and was completed by 658. The position o f the qâdï had been established by the reign of Mu'äwiya, and beginning with Ziyãd, Umayyad governors regularly appointed their own qãdis. At the same time, the office o f qâdï tended to conform to the general pattern o f religious justice in late Sasanian Iraq and amounted to a combination o f the pre-Islamic tribal arbiter (hakam) with the administration of religious law by rabbis, priests, or mõbadhs. The fact that the only conscious comparison made by Muslims was to the mõbadh was probably due to the fact that both Magianism and Islam were rulingclass religions, but it is also indicative o f the growing recognition o f the primarily religious jurisdiction o f the qãçlí by the early eighth century. By that time it had also become customary for judgment to be rendered in the mosque after the example o f the Magians, Jews and Christians. A kind o f popular religious education also took place in the mosque beginning in the caliphate o f ‘All (656-660). As preachers, storytellers, and teachers o f the Qur’an, the religious leaders popularized often tendentious points o f theology and law embedded in local systems o f vocalizing the Qur'an (masahif), the text of which was used for litur­ gical purposes in the same way that Christians used liturgy to create sectarian identities. The nature o f early Muslim religious leadership had several impor­ tant consequences. In the first place, because their status depended on their knowledge o f scripture, these spokesmen favored the literal interpretation of the Qur’an and its application to the public life o f the community. The issues o f power and authority raised by this attitude led to open conflict with the state as early as the end o f ‘Uthmän’s caliphatex). The roots o f the “pious opposition” of the Umayyad period as well as the literalism and militancy o f the Khawärij lie in this conflict 1) M. Hinds, “The Murder of the Caliph ‘Uthmän” , InternationalJournal of Middle East Studies, V. 3 (1972), pp. 450-469.

21

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

IJ3

which created the division between religious and secular leadership in the Muslim community and broke down the original identity o f religion and politics in Islam. The awareness o f such distinctions seems to be indicated by the fact that in 657 the Khawãrij at Harüra’ found it appropriate to choose separate amirs for the war and for prayer1). Consequently, the division between religious and secular authority emerges as a characteristic o f the Umayyad period among Muslims and non-Muslims alike. Much like the Jews and Christians in Iraq, Muslim religious law and ritual developed without direct state support. Secondly, certain ethical biases inherent in the literal application o f the Qur’an contributed to a more developed concept o f a Muslim com­ munity based on faith instead of kinship and pressures for its realization. The Khawãrij in particular sought to redress real social and economic grievances by insisting on fiscal justice and the complete equality o f all Muslims. As a result they attracted non-Arab Muslims (mawãli) even before the Shi‘a did, and one hears o f the first band o f Persian mawãli to become Khawãrij at Küfa when Mughira b. Shu'ba was governor for Mu‘ãwiya (662-669). These ideas were eventually worked out by the moderate Khawäri and the Murji'a at Basra in the late seventh and early eighth centuries and were effectively applied by the 'Abbãsids. Thirdly, the Khawãrij were the first to really apply the Qur’änic requirements for the protection o f non-Muslims with a revealed scrip­ ture (ahl al-kitãb). Their scrupulous regard for the rights o f nonMuslims while at the same time they considered Muslims who failed to agree with them as sinners or apostates outside the protection o f Islam is well illustrated by the famous account o f the rising o f the Nãji Khawãrij in 38/658. They came across two natives o f lower Iraq near Niffãr and when one of them, a Muslim dihqãn called Zadhänfarrükh, said he regarded ‘All as the rightful caliph they killed him as an unbeliever (kãfir). But when the companion o f the unfortunate dihqãn told them that he was a non-Muslim they let him go because, as they said, they had 1) Tabari, V. 1, p. 3349. It is worth noticing that the responsibility for prayer and war were also separated at Küfa by Hajjäj in 86/705. Ibid., V. 2, p. 1182.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

22 134

M. G. MORONY

“no way against him” 1). It was the literalist interpretation of the Qur’an by the pious and the Khawärij that was responsible for adding the requirement o f the possession of a revealed scripture in order to be able to purchase toleration to the other aspects of the poll tax derived from the Sasanians. By the early eighth century this had produced the Muslim distinction between the land and poll taxes. As a result this tendency must be added to the revival o f Sasanian policies towards the members of subject religions in the Umayyad period because it rein­ forced the emergence of dhimmi status in Islam. The impression that is given here o f a society composed of selfcontained, isolated religious communities should be qualified in one important respect. The bonds of a religious communal identity were effectively limited by their dependence on those who followed religious professions, and on the communication o f the outlook and forms o f a religious identity to the other members of the community by those leaders whose authority and status depended on its acceptance. The rabbis, möbadhs, Christian monks and clergy, and to a certain extent even the pious opposition all represented vested interests and through­ out the entire period the distinctions built up by religious leaders were broken down by the ordinary faithful in their everyday relations with the members of other religions, especially through commerce and participation in each other’s festivals. These distinctions were also minimized by the way conversions continued to introduce outside influences among both Muslims and Nestorians. The fact that Jewish women were going to Muslim courts to obtain divorces in the reign o f Mu'awiya, and the reaction of the rabbis, has already been noticed. Bar Penkâyë was equally critical o f the contacts of Nestorians with nonNestorians. Among the “impurities” which crept into the church in the reign of Mu'awiya he lists “commerce with the infidels, union with the perverse, relations with the heretics, and friendship with the Jews” 2). A specific example of what he was objecting to is provided by the synod of George I in 676 which complained that Christians, after taking the 1) Tabari, V. I , p. 3423. 2) Mingana, p. 180.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY RELIGIOUS COMMUNITIES

23 135

sacrament, would go straight from church to Jewish taverns to drink wine. This was especially deplored because there was no lack o f taverns run by Christians where they might satisfy their desire for wine accord­ ing to their own custom 1). Consequently, the existence of practical limitations to communal barriers are as important to an understanding o f early Muslim society as the existence of the formal barriers themselves. There was really never any time when intercommunal contact was entirely cut off, and the effectiveness o f Talmudic restrictions among Jews must be measured by the fact that Jews were still selling wine to Christians in the 670’s. 1) Chabot, Synodicon, pp. 225,489.

2

DHIMMAH IN QUR’AN AND HADITH Mahmoud Ayoub

Concepts often undergo a process of reification which reduces them from the status of broad and dynamic principles to that of mere names or designations of things and ideas. We shall study this process in one of the most important concepts in Muslim society. The lexical meanings of the term dhimmah as defined by some of the classical Arabic lexicographers will be examined. The study then turns to some of the general applications of this term in the Qur'an and hadith, where it is used to characterize the relationship of humanity to God and man’s responsibility in this relation­ ship. Instances in which the term dhimmah was used to characterize general human interrelations, and more particularly those based on the principles of Islamic faith and prophetic tradition (sunnah), will also be examined. Finally, the designation of Jews and Christians by the term ahl aldhimmah (people of dhimmah) will be analysed. The study shows that dhimmah as a concept was first used to designate moral and spiritual relations among communities of faith, but was later reduced to a mere name or designation of subordinate communities. We shall limit the analysis to hadith traditions which reveal the general attitude of Muslims toward ahl aldhimmah (Jews and Christians) and leave out the practical legal implica­ tions of this attitude. Both sunni and shi‘i traditions will be considered with the aim of presenting a broad and coherent picture rather than a comprehensive one. I Al-Jawhari, the author of one of the earliest Arabic lexicons , 1 defines dhimmah and dhimam, its alternate form, as sanctity (hurmah). Dhimmah also signifies surety, or assurance of safety or protection (aman). In its verbal form, dhamma, the term has a negative meaning, to decry a person or

M ahm oud Ayoub is affiliated with the School o f Graduate Studies, Center for Religious Studies, University o f Toronto. 1. Abu Nasr Isma‘il b. Hammad al-Jawhari, Taj al-lughah wa-sihah a l-a ra b iy y a h , transmitted by Sheikh Abu Muhammad Ism a‘il b. M uham m ad b. ‘A bdus alNisaburi, edited by Nasr al-Hurini (Cairo: Bulaq Press, 1365 A .H .), II: 287.

26

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Dhimmah

173

thing, to find it unworthy (madhmum). It can also mean to fall behind because of fatigue or inability, as when a camel is unable to advance because of fatigue. In another noun form, madhammah, or madhimmah, it means shame because of revoking the sanctity of a covenant or pact. The same form also signifies obligation or indebtedness from which a person must seek to free himself. Negative qualities such as miserliness or lying bring madhammah or blame to the person possessing them. Both Lisan aWArab2 and Taj al-‘Arus 3 generally agree with Jawhari, whom they freely quote. They lay special stress, however, on the covenantal aspects of the term. Thus, dhimmah means both the agreement Çahd) and the sanctity (hurmah) of the agreement. The author of Lisan al-Arab states: “The people of dhimmah are also called dhimmah because they enter into a covenant of protection with Muslims.”4 Based on a hadith of ‘Ali, dhimmah implies a binding obligation to fulfill a promise .5 Relying on other examples of hadith usage, Taj al-Arus adds: “ . . . everyone has a covenant of safety and protection with God . . . ” which man can revoke through acts of disobedience to divine injunctions .6 A final signification, which admirably denotes its dynamic and multifaceted application, is the use of the term to mean a banquet or wedding feast. Here a term generally denoting the depth and variety of human and divine relationships also denotes the well-known Arab ideal of hospitality .7 II In contrast with this broad use of the term, the Qur'an employs the word dhimmah only twice and in a very narrow and limited sense, both times in one single passage revealed shortly before the Prophet’s death and intended to purify the hajj ritual of any pre-Islamic characteristics and prohibit the non-Muslim Meccan Arabs from performing it .8 The passage begins by dissociating God and His Apostle from the mushriks (associators of Mecca). The word bara'ah (dissociation or freedom from something) means the exact opposite of dhimmah in its primary sense. Bara’ah in this context

2. Abu al-Fadl Jamal al-D in Muhammad b. Markram b. Manzur al-Ifriqi alMisri, Lisan al-‘arab (Beirut: Dar Sadir, 1375/1965), XII: 221. 3. Muhibb al-D in Abu al-Fayd al-Sayyid Muhammad Murtada al-Husayni alWasiti al-Zabidi al-Hanafi, Taj al-'arus min jaw ah ir al-qam u s, 1st ed. (Cairo: alM atba‘ah al-Khayriyyah, 1306 A .H .), VIII: 301. 4. Ibn Manzur, XII: 221. 5. Ibid., p. 221; and al-Zabidi, VIII: p. 301. 6. Ibid. 7. al-Zabidi, VIII, p. 301. 8. Qur’an, 9:1-3.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 174

Arab Studies Quarterly

signifies the end of a pact or agreement ('ahd) obligation .9 This declaration of the end of the pact was to be made at the time of the hajj.10 Two kinds of people are then distinguished: those who have kept their agreement, and those who have either revoked it outright or acted hypocritically. The term of the pact with the former must be honored for the duration of the sacred months . 11 Because the mushriks were nonbelievers, and hence because their lives were not morally regulated by worship (salat) and almsgiving (zakat), their word cannot be trusted: “ . . . were they to prevail over you, they would regard in you neither consanguinity (illan) nor dhimmah . . . ” 12 Here dhimmah signifies a covenant that should be honored, as the word illan is coupled with it for greater emphasis. Two verses later, the same charge is repeated with even greater emphasis: “They would not observe in a believer either illan or dhimmah\ those are treacherous people .” 13 In this verse the relation of dhimmah is narrowed to the personal level, and another of its opposite qualities—transgression (Vtida’)—is stressed. Throughout the passage, however, the possibility of repenting is recognized and pleaded for . 14 These verses were explained in later hadith tradition. After the last pilgrimage before the Prophet’s death, Abu Bakr and then ‘Ali were sent to Mecca to deliver the following declaration: “The dhimmah of God and His Apostle is free* (bari’atun) from every associator (mushrik). Journey in the earth for four months (i.e., the sacred months); let no mushrik perform the hajj after this year; nor let any naked man henceforth circumambulate the House (i.e., the Ka‘abah ).” 15 This hadith introduces an important element to the general scope of the term dhimmah, the dhimmah of God and His Apostle. Let us first consider the nature and implications of God’s dhimmah or covenant with man. Two general aspects of this divine-human relationship may be distin­ guished. The first manifests the divine providential succor which is granted to man through prayers of supplication (du‘-a') and thanksgiving. In this context dhimmah generally denotes divine protection which extends beyond the short span of human life on earth. It is related on the authority of Ibn al9. For the general context o f these verses see A bu ‘Isa b. M uhamm ad b. ‘Isa b. Sawrah al-Tirmidhi, Sunam al-Tirm idhi, 1st ed., edited by ‘Izzat ‘Ubayd al-D a “as (Hums: M atabi4 al-Fajr al-Hadithah, 1387/1967), VIII: p. 245. 10. Ibid. ; and Qur'an 9:1 ff. 11. See Qur'an, 9:5. These are the four sacred m onths during which was is prohibited. 12. Ibid., 9:8. 13. Ibid., 9:10. 14. See Ibid., 9:3, 5, and especially 11. 15. al-Tirmidhi, VIII: 245.

27

28

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Dhimmah

175

Asqa‘ that the Apostle of God uttered the following prayer at a funeral: “Oh God, behold this man (fulan), son of so-and-so, He is now in your dhimmah (protection) and within the rope {habt) of your domain ('fiwar). Protect him, therefore, from the trial of the grave and the torment of the fire, for you are worthy of fulfilling your covenant and worthy of all justice and truth (haqq). Forgive him and have mercy upon him, for you are all-forgiving, allmerciful.” 16 Here man is called upon not simply to ask for divine favor and protection but to be thankful when these are granted. Ibn Hanbal relates on the authority of Abu Umamah, who related on the authority o f ‘Umar, that the Prophet said: “ . . . whoever acquires a new garment should say (as it reaches his neck), ‘Praise be to God who clothed me with that which covers my nakedness and gives me comeliness in my life’ then give his old garment in alms; such a person will be in the protection (dhimmah) of God, in His domain (jiwar) and His safe-keeping (Kanaf), be he alive or dead, be he alive or dead .” 17 The second aspect of the dhimmah of God with man is one of reciprocity and mutual faithfulness. With regard to man, this means responsibility to God and for fellow human beings. It is therefore related in a number of hadiths, with varying degrees of emphasis, that any person who regularly offers the dawn prayers, or keeps prayers in general, would be in God’s protection (dhimmah). Most hadiths add the warning: “ . . . let not God demand of you anything in violation of His dhimmah and thus throw you face down into the fire of hell.” 18 As for man’s responsibility toward others, the Prophet said, on the authority of ‘Abdullah Ibn ‘Umar: “Whoever hoards food for forty nights would be totally dissociated (barV) from God the exalted and God would likewise be dissociated from him. The people of any land, moreover, among whom a man wakes up hungry, would be dissociated from God’s covenant (dhimmah).”'9 In both examples dhimmah is a covenant of protection between God and man. As man has no power over God, he must satisfy God’s terms of the

16. Abu ‘Abdullah M uhammad b. Yazid Ibn Maja, al-Sunan (Cairo: Ihya alKutub al-‘cArabiyyah, 1372/1952), VI: 480. See also the Prophet’s prayer for travel in al-Tirmidhi, IX: 134; and Ahmad Ibn Hanbal, al-M u sn ad, 1st ed. (Beirut: Dar Sadir, 1389/1969), II: 401. 17. Ibn Hanbal, 1:44. 18. Abu al-Husayn Muslim Ibn al-Hajjaj b. Muslim al-Qushayri al-Nisaburi, alSahih, edited by Sheikh M uhamm ad al-Zihani et al. (Cairo: M ustafa al-Babi alHalabi and Sons, 1377 A.H.), 1:250. For other hadiths on the same theme, see also M uhammad b. Isma‘il al-Bukhari, al-Sahih, 1st ed. (Cairo: al-M atba‘ah al-Bahiyya al-M isriyyah, 1343 A.H.), 1:54-55; al-Tirmidhi, VI:332; and Muhammad b. ‘Ali b. alHusayn b. Musa Ibn Babawaih al-Qumm i, known as al-Saduq, Man la yahduruhu al-faqih (Najaf: Dar al-Kutub al-Islamiyyah, 1376/1957), 11:367. 19. Ibn Hanbal, 11:33.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Arab Studies Quarterly

176

covenant by protecting others. Indeed, on man’s keeping of God’s dhimmah depend the order and well-being of the world. Thus we are told: “When four (things) spread in the world, four (other things) appear. When adultery spreads, earthquakes occur. When tyranny spreads among rulers, rain is withheld from heaven. When dhimmah is revoked, the people of association are allowed to prevail over the people of Islam. When almsgiving (zakat) is withheld, poverty prevails.”20 Ill God’s dhimmah, or covenant relationship with man, is given a human dimension by being identified with the dhimmah of His Apostle. In the Qur'an the Prophet spoke on God’s behalf. Thus, his dhimmah was in fact God’s dhimmah. Through the prophetic tradition (sunnah) men can interiorize the dhimmah of God and His Prophet and make it their own. It is then their responsibility to honor the covenant of God and His Apostle, but without having the prerogative of claiming to understand or implement fully this divine prophetic covenant. It rather must remain an ideal goal to be sought after, but Yiever fully attained. It is related that whenever the Prophet appointed a man to head an army in war he instructed him to fear God and deal kindly with his fellow Muslims. Then, addressing the rest of the army, the Prophet continued: “ . . . fight in the name of God and in the way of God. Fight those who disbelieve in God, but do not overindulge, commit acts of treachery, mutilate people, or kill a child.” The hadith continues: “ . . . if you besige a fortified settlement and its inhabitants request that you give them the dhimmah of God and His Prophet, do not give them either the dhimmah of God or that of His Prophet. Rather, make with them the covenant (dhimmah) of yourself, your father, and your fellows. For it is better for you to violate your own dhimmah than to violate the dhimmah of God and His Prophet .” 21 The concept of dhimmah as a principle of honor governing human relations was not introduced by Islam. Rather, Islam gave it a divine moral basis which elevated it beyond any considerations of human power and prestige. Yet even before Islam, Arab society regarded bonds of dhimmah as sacred. This ancient attitude is well illustrated by an anecdote related on the authority of ‘A’ishah, daughter of Abu Bakr and wife of the Prophet. Her father decided to leave Mecca before the hijrah under pressure from the

20. Abu Ja‘far b. Muhammad b. Ya‘qub b. Ishaq al-Kulayni al-R azi, a l-U su lm in al-kafi, edited by ‘Ali Akbar al-Ghifari (Tiheran: M aktabat al-Saduq, 1381 A.H.), 11:448. See also Saduq, 1:332. 21. al-Tirmidhi, V:338.

29

30

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Dhimmah

177

people of Quraysh, who protested that his piety was drawing their youths to the new religion. On his way out of the city he met a man of status who established a bond (‘aqd) of protection with him. It was agreed that Abu Bakr would remain in his house, provided that he practice his new faith in the privacy of his home. Since Abu Bakr was unable to restrain himself, the man requested that he be released from their pact because “ . . . I do not wish the Arabs to hear that I have failed a man with whom I have established a bond.” The man asked literally that his dhimmah be returned to him .22 The Islamic character of dhimmah may be further illustrated by two important sayings attributed, respectively, to ‘Ali and his great-grandson Muhammad al-Baqir, the fifth shi‘i Imam. After the battle of Siffin and the revolt of the Kharajites against ‘Ali, he sent them the following warning: “ . . . beware that you shed no innocent blood, waylay travellers, or wrong a dhimmah.23 If you do, we shall wage war against you; for God loves not those who transgress .” 24 Al-Baqir declared: “Any man who grants another protection (aman) in a pact (dhimmah) and then kills him shall come on the Day of Resurrection bearing the banner of transgression .”25 In both traditions, the reason for keeping the covenant is not to preserve one’s honor and prestige in society but to fulfill God’s command. Thus, dhimmah becomes an obligation primarily to God; it becomes a religious rather than a social principle. In contrast with pre-Islamic Arabic society, where dhimmah was the privilege of the rich and powerful, Islam made it a universal obligation, binding on all Muslims, male and female. In a very important hadith, related on the authority of a number of the Prophet’s companions and by both sunni and shi‘i traditionists, the Prophet declared: “ . . . the dhimmah of all Muslims is one by which even the lowliest of them can live.” 26 This statement was prompted by the fact that Umm Hani’, sister of ‘Ali, gave a pact of protection to two men which the Prophet honored; he went even further by stipulating that the dhimmah of a Muslim slave is binding on other Muslims. No doubt this attitude contributed to the designation of

22. Al-Bukhari, 11:25. 23. Referring to the people o f the book, who were som etim es called simply dh im m at A llah. See below and Ibn Manzur, X II:221. 24. Ibn Hanbal, 1:86. 25. al-Kulayni, V:31. 26. al-Tirmidhi, V:306; al-Bukhari, 1:210, and 11:127-28. See also A bu‘Abd alRahman b. Shu‘ayb al-N asa’i, al-Sunan, 1st ed. (Cairo: M ustafa al-Babi al-Halabi and Sons, 1383/1964), VII1:18—22. A l-N asa’i significantly substitutes the term m u'minum (believers) for m uslim un. See also Kulayni, 1:403-4 and 542 for a different reading and interpretation.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Arab Studies Quarterly

178

Jews and Christians as ahl al-dhimmah, people of the dhimmah of God and His Apostle, and by extension of all the Muslims. IV The Qur'an refers to Jews and Christians as ahl al-kitab. The designation of ahl al-dhimmah is nowhere employed, nor is the term dhimmah used in any way to characterize the relationship of Muslims to Jews and Christians. The Qur'an uses the term in its strictly Arabian pre-Islamic context. Its absence with regard to Jews and Christians may suggest a certain level of equality of faith between the people of the book and Muslims, who are also people of the book in the strictest sense. The term ahl al-dhimmah underwent a gradual process of development both in definition and in usage. In this process it lost most of its dynamic aspects and was finally reduced to a legal term designating peoples with whom Muslims had special relation­ ship. The attitude of Muslims toward ahl al-dhimmah likewise underwent a process of change from that of a sacred responsibility or divine charge to one of a legal relationship. Thus, a man who slapped a Jewish merchant for declaring the superiority of Moses over the rest of mankind in an oath was brought to answer before the Prophet by his opponent. The Jew protested confidently: “ . . . I have a dhimmah and covenant Çahd)\ why then did this man slap my face?” The Prophet reproached the offender for his excessive zeal, and forbade the making of any claims of superiority for any prophet of God over another .27 This early attitude and the change that was taking place is further illustrated by the following anecdote: A well-known companion saw a group of ahl al-dhimmah standing in the hot Damascus sun as punishment for not paying fully their land tax. He went to the governor and protested. “I bear witness that I heard the Apostle of God say, ‘God the exalted will torment on the Day of Resurrection those who torment men in this world.’ ” 28 It is easy for a people in authority to abuse their power and violate their sacred trust. Thus, in order to safeguard the sanctity of this divine trust, its violation was declared to be a heinous sin. It is related on the authority of the famous companion Abu Hurayrah that the Prophet declared: “Whoever kills a man who has the dhimmah of God and the dhimma of His Apostle shall not smell the fragrance of paradise, though it may be sensed from the distance of a seventy-year journey .”29 27. 28. 29. as in

Al-Bukhari, 11:154 Ibn Hanbal, 111:403. Ibn Maja, 11:896. This tradition appears in all six canonical collections as well most others.

31

32

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Dhimmah

179

This attitude continued to prevail at least during the period o f ‘Umar, the second caliph. In his final counsel to his people he charged his successor “ . . . with the dhimmah of God and the dhimmah of His Apostle (i.e., Jews and Christians), that he fulfill their covenant, that he fight behind them (i.e., in their defense) and that they be not burdened beyond their capacity .”30 It is significant to observe that Jews and Christians in this tradition are referred to as “the dhimmah of God” and not as ahl al-dhimmah. In what appears to be a later version related by Ibn Hanbal this injunction is put in more practical terms: “ . . . I charge you to do well toward the people of your dhimmah, for they are the source of sustenance for your children and the covenant of your Prophet .”31 This may be compared with a later designation of Jews and Christians in a tradition which shows both the older and more positive attitude as well as the change it underwent: Two of the companions were sitting in Qadisiyyah when a funeral passed by. They rose in reverence for the dead but were told: “ it is one of the people of the earth There was hence no need to show such deference. One of the two men replied that a funeral of a Jew once passed by the Prophet and he rose. When told that it was only the funeral of a Jew, he answered: “Is it not a soul?”32 It is clear that the attitude toward ahl al-dhimmah had to be reevaluated from time to time by the standard of the Sunnah of the Prophet and his first companions. Men had to be reminded of their original pledge under Islam to protect others and show kindness toward them. We are told that the pious caliph ‘Umar II, Ibn ‘Abd al-‘Aziz, wished to follow the example of his namesake in the treatment of people of other faiths. Thus, he ... established a bond of dhimmah with peoples of other religions . . . .”33 Here we see that the covenant (dhimmah) of Muslims with those of other faiths had to be renewed in accordance with the norms established by the Prophet and his immediate successors. In the Qur'an a distinction is made between the associators (mushrikun) of Mecca and the people of the book (ahl al-kitab). This distinction was retained in early hadith tradition as well. Thus, in a section in Bukhari dealing with marriage regulations between Muslims or new converts to Islam and non-Muslims a distinction is made between a mushrik (man or woman) and a nasrani (Christian man or woman). It is interesting to observe that the former is referred to as harbi, one with whom a state of war exists .34 30. Al-Bukhari, 1:160. 31. Ibn Hanbal, 1:51. 32. Al-Bukhari, 1:151. 33. Abu Dawud Sulaym an b. al-A sh‘ath al-Sijistani al-Azdi, al-Sunan , 1st ed., edited by ‘Izzat ‘Ubayd al-D a “cas and ‘Adil al-Sayyid (Hums: M uhammad ‘A li alSayyid, 1392/1972), 111:364. 34. Al-Bukhari, 111:170.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 180

Arab Studies Quarterly

As pre-Islamic religion soon disappeared from Arabia the term mushrikun was applied to Christians. It gradually became synonymous with the word Christian and has largely replaced the more positive designation of ahl alkitab. The shi‘i attitude toward ahl al-dhimmah is ambivalent at best. It deserves special attention for several reasons. First, shi‘i hadith collections are relatively late; hence they may reflect the general attitude which prevailed at the time. Second, the shi‘i attitude changed from a more or less positive one in traditions attributed to the early imams to an extreme attitude in later traditions. It must also be added that the shi‘i attitude was that of an oppressed and embittered community and therefore reflected a particular psychology. This ambivalence is well expressed in a curious and puzzling bit of double advice which ‘Ali is said to have given to a man whom he appointed as an officer in Qadisiyyah. In public he counselled the man as follows: “Guard well your kharaj (land tax), that you do not overlook one dirham (a coin of little value) of it.” He then asked the man to see him privately before he left. In private he told him: “ . . . that which you have heard from me before was only a ruse. Do not ever beat a Muslim, Jew, or Christian on account of one dirham of land tax, or sell a best of burden because of a dirham in arrears. This is because we are commanded to take what they voluntarily give.”35 A tradition attributed to the fourth im am, Zayn al-‘Abidin, enjoins that a Muslim consider the elders of the Muslim community as his father, their old women as his mother, their youths as his brothers, and their children as his own. It continues: “ . . . as for the dhimmah (i.e., A hl al-dhimmah) you should accept from them only that which God has accepted from them. Do not wrong them so long as they are faithful to God’s covenant.”36 This irenic and wise counsel may be contrasted with a tradition attributed to the sixth imam, al-Sadiq, in explanation of the prophetic hadith: “There is no child but that he is born with thtfitrah (i.e., in a state of pure faith); thereafter his parents turn him into a Jew, a Christian, or a Magian.” The sixth imam comments: “ . . . the Apostle of God gave (protection) and accepted the poll tax (jizvah from those people on the condition that they would not turn their children into Jews or Christians. As for the descendants of ahl al-dhimmah, today they have no dhimmah (covenant).” 37 This being the case, a Muslim should not have any dealings with a man of ahl al-dhimmah, nor should he show any amity toward him .38 Qur’anic verses allowing intermarriage

35. Ibn Babawaih, 36. Ibid., 11:381. 37. Ibid., 11:27. See an important disciple 38. See al-Kulayni,

11:13. the follow ing h a d ith , also related on the authority o f Zirarah, o f the fifth and sixth im am s. V:286; and al-Saduq, 111:145.

33

34

-

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

Dhimmah

181

between Muslims and the people of the book, or in any way encouraging social intercourse between members of the two communities, were either declared abrogated by other verses or interpreted in ways that would give the exact opposite sense from that intended by the text .39 A complicating factor in assessing the shi‘i attitude toward other communities, including other Muslims, is the principle of taqiyyah (dissimu­ lation). The great scholar al-Tusi, the shaykh of the shi‘i community, uses the principle of taqiyyah to harmonize contradictory traditions. Those hadiths which agree with the actual text of the Qur'an and general Muslim views he dismisses as being only concessions to prevailing sentiment: “the view of al-ammah (the general community ).”40 It must be added that in other ways shi-‘ism has shown an unusual catholicity toward Judaism and Christianity, dictated by the universal principle of the imamate. The mother of the twelfth Imam was herself a Byzantine Christian; through her an important spiritual link is established between Christian and Islam. It must be concluded that the harsh attitude of the shi‘i community in its formative period probably does not represent the imam’s thought on the matter, but the frustration of an oppressed minority. It may also reflect the general situation in the fourth century of the hijrah (tenth-eleventh century A.D.), during which shi‘i hadith tradition was largely codified. In modern everyday speech, the word dhimmah has become synonymous with honor, decency, and truthfulness. One who is without dhimmah is not a true human being; he is one who refuses to live within the dhimmah which God decreed for humankind. He refuses the divine covenant of servanthood. As we have seen, the human dhimmah stands beneath the divine dhimmah, which must at all costs not be violated. This is the meaning of the injunction of the Prophet to the armies of the Muslim community who were told not to presume themselves able to fulfill the dhimmah of God, and hence not to invoke it in applying the imperfect standards of human justice. That the human dhimmah must always be judged by the divine, primordial dhimmah and challenged by its demands of justice and decency is beyond doubt. In a sense, therefore, the people of dhimmah embodied this challenge, as may be inferred from the statement attributed to ‘Umar, the second Caliph. Yet as the term became reified into a technical legal concept, it lost its dimension of transcendence. Consequently, the people of dhimmah lost their challenge as

39. See al-Kulayni, V: 11 and 351, and VT.240. 40. See Abu Ja‘far M uhammad Ibn al-Hasan al-Tusi, al-Istibsar f i m a ukhtulifa fih min al-akh bar, 2nd ed., edited by al-Sayyid Hasan al-M usawi al-Khurasani (Najaf: Dar al-Kutub al-Islam iyyah, 1375/1956), I:179ff. for good examples.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 182

Arab Studies Quarterly

it became a legal and not a divine charge. The question of how well or badly the Muslims treated their Jewish and Christian subjects as ahl al-dhimmah is, to say the least, a complex one, which, in any case, must be answered from within the historical realities of all three communities.

35

3

THE LEGISLATIVE AUTONOMY OF CHRISTIANS IN THE ISLAMIC WORLD Néophyte Edelby

Introduction in the seventh century, Oriental Christian communities have stood in a special relationship to the laws and jurisdiction of the governments under which they live, and this situation has had implications for the members of those communities in their legal dealings with one another. Partly for historical reasons, and partly in response to the imperatives of dogma, the public law of Islam has regarded religious groups as separate nations, subject to the state but enjoying a measure of legislative and jurisdictional autonomy. Legislative and jurisdictional “immunity” , the term th a t is commonly used to designate the privileged position enjoyed by foreigners and protected persons in the O ttom an Empire under the regime of the Capitulations, has also been applied to the legal situation of Christians in the Islamic world. I would suggest th a t the term “autonom y” is more accurate, a t any rate with reference to the period prior to the Capitulations and the secularization of O ttom an law. Immunity involves a concept of exception or privilege; the fact th a t non-Muslims are subject to legislation and jurisdiction other than those of the state has been seen in term s of personal immunity, a privilege granted by the state within limits determined by the state itself and revocable at will. This view of the m atter is consistent with a dogma cherished by some contem porary observers, namely, th a t the state is the sole source of positive law. However, it is quite at variance with the conception of state and law held by the putative grantor of the immunity in question, Islam itself. As we shall see in the further course of this study, the legal autonom y enjoyed by the Christians was not a concession on the part of the Islamic state, sanctioned by treaties, pacts or Capitulations concluded [2 ] a t the time of the conquest. On the contrary, it was a logical outgrow th of the Islamic conceptions of the state as a theocracy and of all law as being religious in nature. Christians were subject to their own law and their own jurisdiction because, not being part of the Islamic community, they could not be subject [1 ] E v e r

s in c e t h e

A rab

c o nq uests

38

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 2

Néophyte Edelby

to either its secular or its religious laws. The distinction between secular and religious law is unknown to classical Islamic legal thought in any case. T hat autonomy is not in any way comparable to regional autonomy, the autonomy of a state th at is part of a federation, or the autonomy of a defeated state. In the Islamic world, Christians by no means constitute “a state within a state” , as has sometimes been asserted .1 After the Arab conquests, the Christians continued to live by their own laws not because the new government acknowledged th at their communities possessed some sort of residual sovereignty, but because in the view of the conquerors, Islamic law was not applicable to them either de facto or de jure. Moreover, the legislative autonomy enjoyed by the Christian communi­ ties did not, in the eyes of the Islamic state, comprise an autonomous power to “make” laws, as the etymology of the term might suggest. For Islam, law can only be deduced from its revealed sources. Except in the very limited field of “administrative policy” , Islam does not acknowledge th a t any human institution possesses the power to legislate. The expression “the legislative autonomy enjoyed by the Christians” must thus be understood to mean their power to “follow” their own laws, which the Islamic state regarded as an integral part of their religion. 1. The following pages are intended primarily to serve as a starting point for subsequent investigations of the internal history of the secular law of Christian communities in the Arab countries. In 1880, Bruns and Sachau published their Syrisch-Römisches Rechts­ buch, an edition and translation of two recensions of a fifth-century SyroRoman code of law, and thereby confronted the Western world with the problem of the secular legislation of the Christian communities in Islamic countries. Bruns himself was among the first to suggest th a t a form of au­ tonomous law, as supposedly reflected in the above-mentioned code, had survived to some extent in the eastern provinces of the Empire th a t had fallen into the hands of the A rabs .2 He was followed by many others: the Syro-Roman code, with its specifically provincial features (which attracted a good deal more attention than their actual im portance warranted), was uncritically viewed as the [3] authority on which courts in those Christian jurisdictions were presumed to have relied during the entire period of the high Middle Ages. 1For example, Karl Georg Bruns and Eduard Sachau, Syrisch-Römisches Rechtsbuch aus dem fünften Jahrhundert (Leipzig, 1880), II, 173: “Die Christen im Chalifat bildeten einen Staat im Staate”. 2Ibid.

39

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

3

The audacious hypothesis put forward by Ludwig M itteis in his 1891 study, Reichsrecht und Volksrecht in den östlichen Provinzen des römischen Kaiserreichs, sparked renewed interest among Romanists in unofficial Ro­ man law, beginning with the Syro-Roman code. This outlook was still preva­ lent in 1907, when Sachau published three new Syriac versions of the code, identified as R I, R II and R III, and also a large number of Nestorian legal treatises and collections of laws 3 dating from a period extending be­ tween the early sixth century and the late ninth century. All this m aterial was accepted as confirmation of M itteis’ views on the origins of provincial traditions and their importance in Roman law .4 These newly published works, together with the Oriental nomocanons, which had been edited some years earlier, appeared to show th a t the civil law of Oriental Christian communities had been a continuous entity extending from the Syro-Roman code, the Syriac version of which was thought to date from the late fifth century, to the great nomocanonical compilations of the thirteenth and fourteenth centuries (including the canonical collection of the Melkites, the anonymous Nomocanon of the M aronites, and those of ‘AbdTshü‘, Ibn al-‘Assãl, Bar Hebraeus, Abu 1-Barakãt and others). To date, these documents have never been subjected to any serious scholarly examination. During the First World War, the Italian Roman­ ist Evaristo Carusi used them in an attem pt to confirm a theory of his own on the interdependence of the legal systems used by the peoples of the M editerranean basin. Carusi asserted th a t the legal literature of the Ori­ ental Christian communities constituted evidence of a common Syrian legal tradition th at had survived in the lands conquered by Islam. T h at tradi­ tion, he claimed, had materially influenced not only Romano-Byzantine law (this aspect of Carusi’s theory was thus partly in agreement with M itteis’ hypothesis based on the Syro-Roman code), but, even more importantly, on the development of Islamic law. Carusi thought he had dem onstrated this from the Syro-Roman code and the Christian Arab papyri from the high Middle Ages, supplemented by the great nomocanonical collections of the thirteenth and fourteenth centuries. He had observed unmistakable analo­ gies [4] between those works and Islamic law, and as he never entertained so much as the possibility th a t elements of the latter might have found their way into Christian law, he took this as tangible evidence of the persistence 3Eduard Sachau, ed., Syrische Rechtsbücher, 3 vols. (Berlin, 1907-15). 4Mitteis, in fact, had used the discovery of the three new manuscripts of the SyroRoman code even before their publication, in an important study entitled “Über drei neue Handschriften des syrisch-römisches Rechtsbuches,” which was published in the Abhand­ lungen of the Berlin Academy for 1905.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

40 4

Néophyte Edelby

of a Syrian legal tradition and its influence on the formulation of Islamic law. The celebrated Orientalist C.A. Nallino had no difficulty in refuting this line of reasoning, dem onstrating th at, on the contrary, the analogies in ques­ tion simply reflected the fact th a t parts of Islamic law had been adopted en masse into Christian law. Carusi had argued for the existence of a common Syrian Christian law th a t had exerted a decisive influence on the devel­ opment of Islamic law. The evidence he had presented in support of this thesis, however, had been faulty in many respects. Nallino’s attack on th a t evidence led him to deny th a t there had ever been any such Syrian legal tradition or a fortiori any such influence. For our part, we believe th a t while the reality of a native Syrian legal tradition paralleling official Roman law may be open to question, there is no reason at all to doubt the existence of a system of law, both theoretical and practical, th at was used by Christian communities. Doubt arises only as to the precise nature of th a t law. Was it Roman law as it had come down from antiquity in the form of Justinian’s well-known compilation? Was it, in slightly altered form, the system of the Syro-Roman code? Was it some kind of customary law th a t is no longer extant? Was it a system of law th a t was common to all Christians living in Islamic lands, or did it vary from region to region and from community to community? As yet we have no satisfactory answers to any of these questions. We may readily imagine th a t this system of law used by the Christian population did not remain unchanged during the six centuries separating the Arab conquest from the first nomocanons in the thirteenth century. The Christian law, after all, had been fragm entary to begin with, had no written tradition on which to rely, and had no public authority to enforce it. It is thus hardly surprising th at during th at long interval it should have been influenced by the practice of the conquerors in areas unrelated to religion as such. Nor is it surprising th a t the Islamic authorities themselves should have imposed some features of their system of law on their Christian subjects, despite the fact th at this was contrary to the principle of the personality of law in the Islamic system. And indeed, once we accept these as possibilities, it becomes clear from the available documentary evidence th a t Islamic law was progressively, imperceptibly assimilated by the Christian communities, and also th at some Islamic standards were imposed on those communities for administrative reasons, both by religious leaders and by state authorities. Carusi began with the observed fact th a t the system of secular law found in the extant nomocanons is characterized by a great many features th a t are also found in Islamic law, but he was undoubtedly mistaken in concluding

41

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

5

from this th a t the latter had borrowed those features [5] from a Syrian Christian system of law antedating the compilation of the nomocanons; in most instances, the reverse was the case. It is not clear, however, th a t Nallino was therefore justified in denying the existence of a system of secular law used exclusively by Christians. The quarrel between Carusi and Nallino had at any rate one useful out­ come: it became clear th a t the starting point for any investigation of the development of Oriental Christian law must be the question of whether, and to what extent, the Oriental Christian communities enjoyed the right to be governed by a system of law other than th a t of the state. C.A. Nallino himself ultimately came to realize this. His article, “Libri giuridici bizantini in versioni arabe cristiane dei secoli X II-X IIP ’,5 was an extreme statem ent of his theory th at Oriental Christian legal collections had been exclusively literary in nature. It ended, however, on this less dogmatic note: It is our hope th a t this study, like its 1923 predecessor on Bar Hebraeus, may serve to strengthen the scholarly world’s growing awareness of the need to investigate the history of the legal liter­ ature and social life of Oriental Christian communities in greater depth, and to identify the origins and scope of their works, as an essential preliminary to the task of writing histories of law and comparative law. In the absence of the necessary preparatory work, reconstructions of th a t nature are likely to be completely erroneous. T hat was a very balanced assessment. The same point had been made as early as 1909 by Josef Partsch, in an article in the Zeitschrift der SavignyStiftung fü r Rechsgeschichte analysing the new Syriac sources th a t had been published by Sachau .6 In 1925, yet another scholar, Willi Heffening ,7 5Published in Rendiconti della R. Accademia dei Lincei, Cl. sc. mor., ser. VI, 1 (Rome, 1925), 101-65; reprinted in his Raccolta di scritti editi e inediti, edited by his daughter Maria Nallino (Rome, 1939-48), IV, 324-82. 6Josef Partsch, “Neue Rechtsquellen der Nestorianischen Kirche”, Zeitschrift der Savigny-Stiftung für Rechsgeschichte, Rom. Abt., 30 (1909), 355-98, esp. 360: “Bevor die Rechtswissenschaft diese Kirchlichen Aufzeichnungen juristichen karakters mit dem Gefühl der Sicherheit verarbeiten kann, wird sie sich die Frage stellen müssen, inwiefern hier weltliches Recht der nestorianischen Christen zur Aufzeichnung kommt. Liegen hier erzwingbare Normen vor, für deren Durchsetzung der Staat seine Machtmittel zur Verfügung stellt? Diese Frage, die bisher nicht ganz deutliche Antworten gefunden hat, kann meines Erachtens mit voller Schärfe gelöst werden.” 7Willi Heffening, Das islamische Fremdenrecht bis zu den islamisch-fränkischen Staatsverträgen (Hannover, 1925; Beiträge zum Rechts- und Wirtschaftsleben des islamis-

42

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 6

Neophyte Edelby

pointed out th at the exact scope and limits of the jurisdictional autonomy [6 ] enjoyed by the Oriental Christians would have to be determined before any deductions about the value of their legal writings could validly be drawn. The discussion in the pages below seeks to address precisely th a t ques­ tion. 2. Furthermore, a study of the legislative autonomy enjoyed by Christian communities in Islamic lands should also be a prerequisite for any research on the origins of Islamic law and, in particular, on the possible influence of Roman-Byzantine law on its development. The latter issue, of course, has been extensively discussed .8 It seems fair to say th a t the state of our knowledge at present is not adequate to enable us to reach a conclusion. To be sure, Islamic law did not spring up full-blown ex nihilo, as one of those examples of “parthenogenesis” th a t are accepted as so many dogmas by some historians of Islam. Like any human institution, Islamic law is a product of history, shaped by the environment in which it has developed and the civilizations with which it has been in contact. To say this is not in any way to deny its unquestionable originality or its distinctive nature, just as to acknowledge Oriental influences on the Roman legal system is not to deny the originality of th a t system. However, in order to arrive at an accurate assessment of the influence of Roman law on the development of Islamic law, we should have to know more than we know today about, first, the institutions of pre-Islamic Arabia, second, Sasanid law, since the first schools of fiqh grew up within th a t system ’s sphere of influence, third, Roman-Byzantine law as it was applied in the Asiatic provinces, and, fourth, rabbinical law, the influence of which has actually been regarded by some scholars as having been predominant. Specifically, Roman law cannot have exerted any influence either through official legal collections, of which no traces are to be found in our Arabic sources, or through the schools of law of Alexandria and Beirut, which had ceased to exist before the conquests. The only channel through which the presumptive influence could have been exerted was thus contact with the non-Muslim religious communities, the Christian communities in particular, which continued under Muslim domination to live in accordance with the traditions and practices they had followed during the preceding period. It is clear, then, th a t before we shall be able to decide whether those traditions chen Orients, I), 118: “Christen und Juden besassen im islamischen Staate eine Eigengerischtsbarkeit,” and in n. 2 : “Der Kompetenzbereich dieser Gerichte wäre näher zu untersuchen, vor allem ob ihre Urteile vollstreckbar waren”. 8Cf. Abdel-Rahman Hassam, “Le droit musulman et le droit romain”, Archives d ’histoire du droit oriental 4 (1949), 301-21.

43

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

7

and practices may have influenced the development of Islamic law, we shall have to determine just how much legislative autonomy those communities enjoyed. 3. Lastly, this study may shed fresh light on the issue of the “personal statu s” of non-Muslims under Islamic rule, which has been extensively de­ bated in the Arab countries in recent years. In particular, we shall find th a t the special situation of non-Muslims in th a t area is not, so far as its origins go, either an abdication of part of the sovereignty of the Islamic state by virtue of agreements concluded at the time of the conquest, or a form of privilege defined as protected by the public authorities. We shall also find th a t the distinction between personal status (with respect to which the state allowed the Christians a very large measure of legislative auton­ omy) and real status (which legal practice sought to bring into constantly closer conformity to Islamic law) predates by several centuries the wellknown Hatt-i humãyun of 18 February 1856. In addition, we shall see th a t there are no grounds for the view th a t Christians have been governed, during the entire period of Muslim domination, by Roman-Byzantine law ,9 with its blindness to the entire assimilation process th a t culminated, in 1856, in the non-Muslim communities’ autonomy being restricted to m atters of personal status.

I. Origins 1. The Thinking of the Prophet The origins of the Islamic sta te ’s recognition of the legislative autonom y of Christian communities in Islamic lands is attributable to two factors, one doctrinal and the other historical. The doctrinal factor is to be found in M uham m ad’s political and religious teaching, according to which Islam, with all its panoply of tenets, law and observances, could not be imposed upon the devotees of other tolerated re­ ligions; these, although subject to Islam, constituted separate, autonomous theocratic communities. The Arab conquests added a contingent factor to this concept of the personal nature of law, which at the outset had been exclusively religious and theoretical in nature. W ithin a period of less than ten years, the Arabs found themselves absolute m asters of a vast empire and heirs to the Greco-Roman and Sasanid civilizations, with their highly evolved legal systems, which the conquerors were shrewd enough to leave 9E.g. François Kozman, “Les chrétiens d’Égypte ont-ils adopté et suivent-ils jusqu’à nos jours la législation justinienne... in Acta Congressus Iuridici Internationalis, Romae 1924, II (Rome, 1925), 172.

44

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Néophyte Edelby

8

undisturbed. The indigenous peoples continued to follow the same laws and apply to the same courts as before. The autonomy th a t had been a theoret­ ical feature of M uham m ad’s teaching thus came unexpectedly and immedi­ ately to be put into practice under the pressure of events. The Umayyad dynasty organized the conquered lands, progressively absorbing the lower levels of administration, and thereby came to intervene more frequently in the domestic lives of the subject peoples; yet it introduced no fundam ental change to the legal position of the Christian communities, which contin­ ued to enjoy, virtually without restriction, the autonomy th a t the conquests had left them. Classical Islamic jurisprudence, which developed beginning early in the ‘Abbãsid period, left this situation nearly unaltered, apart from explicitly linking the de facto situation created by the conquests with the doctrinal principles taught by Muhammad. The Semites were unique among ancient peoples in th a t they never con­ ceived of any form of social organization other than theocracy, with the gods as the sole source of law. Islam developed in response to various forces, including this innate tendency characteristic of all the Semitic peoples, the combined example of the Byzantine Empire, the Sasanid monarchy and the Jewish diaspora, the fact th a t there was no central organization in Ara­ bia and the circumstances attendant on the advent of the new “nation of M uhamm ad” . All these factors contributed in their several ways to the shaping of Islam as an exceptionally unitary theocracy. According to the theocratic view, law cannot be regarded as a reflection of social conditions at a particular period in history or as an expression of the will of human legislators. God is the immediate originator of all law, because he is recognized as the head of Islamic society. “God ordaineth th a t which he pleaseth .” 10 Originating as it does from God, law can be known only through revelation. The Q ur’an was thus, in the time of M uhammad, the only forts cognoscendi of law, and also of dogma, morality and religious practice. Religion and law are so intimately united in Islam th a t in different con­ texts we find the same terms used to designate them both .11 The word sharï‘a, which as a classical legal term denotes Islamic law, is used in the Q ur’an in its technical sense of “clear road ” ,12 “general standard of life” ,13 encompassing the entire scope of the revelation. In M uham m ad’s thinking, religion and law are not merely closely interrelated, they are identical. 10Sürat n Sürat 12Sürat 13Sürat

al-Mä’ida (5), v. 1. al-Shürä (42), vs. 11 , 20 . al-Jäthiya (45), v. 17. al-Mä’ida (5), v. 52.

45

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

9

It follows th a t there are as many legislative systems as there are reli­ gions, inasmuch as even after the revelation of the Q ur’an, there are various tolerated religions/peoples th at remain outside Islam. Mankind is divided into a number of religious groups, and these are coterminous with political groups. Every nation has had its apostle ,14 its warner ,15 who on the Day of Judgement will bear witness against its people if they have refused to believe ;16 every nation has had a scripture ,17 a law and a sanctuary .18 To every nation, God has set a term th at cannot be delayed or advanced by so much as an hour .19 Islam projected this religious view of nationality on to other religions. For a long time, the only social groups in the Orient were groups formed by religious communities. Not only does this theocratic conception of social groups give rise to a multiplicity of “nations” within a single geographic area, it also justifies the co-existence, within th a t area, of a number of different legal systems. Each “nation” , regardless of whether it dominates the others or is subject to them, never ceases to be autonomous in the sense of living by its own law. At the Last Day, every “nation” will be judged in accordance with its scripture .20 Here on earth, God has prescribed a standard and a rule of life for everyone .21 How can the Jews accept M uhammad as their arbitrator when they have the Torah, which contains the judgem ents of G od ?22 To each his religion ,23 in the full Semitic sense of the term , i.e. the full array of standards th a t regulate religious belief and practice, morality, and the life of the individual and the group. God revealed the Torah for the use of the Jews; it is light and guidance for the leaders of the Jewish community, who will decide disputes between their co-religionists in accordance with it .24 Not to judge in accordance with th a t which God has revealed to every “nation” is false belief, injustice, prevarication .25 God could have unified all 14Sürat Yünus ( 10), v. 48; Sürat al-Nahl (16), v. 38. 15Sürat Fätir (35), v. 22 . 16Sürat al-Nisä’ (4), v. 45; Sürat al-Nahl (16), v. 91; Sürat al-Qasas (28), v. 75. 17To Muhammad, the possible existence of any religion other than a revealed religion was inconceivable. 18Sürat al-Hajj ( 22 ), vs. 35, 66 ; Sürat al-Baqara (2 ), v. 143. 19Sürat al-A‘räf (7), v. 32; Sürat Yünus ( 10), v. 50; Sürat al-Hijr (15), v. 5; Sürat al-Mu’minün (23), v. 45. 20Sürat al-Jäthiya (45), v. 27. 21Sürat al-Mä’ida (5), v. 52. 22Ibid., v. 47. 23Sürt al-An‘äm (6 ), v. 109. 24Sürat al-Mä’ida (5), v. 48. 25Ibid., vs. 48, 49, 51.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

46

Néophyte Edelby

10

religions, but he wished to test the obedience of the faithful; let everyone, then, act rightly in accordance with his religion. The time will come when God, to whom all must return, will resolve religious disputes 26 and reward each person according to his deserts, regardless of his religion .27 “Let they who have received the Gospel judge according to what God hath revealed therein___Unto each of you have we given a law and a (different) rule of conduct ” .28 Furthermore, Islamic law is a privilege reserved for Muslims exclusively. God, in his mercy, has eased, for their benefit, the harsh Jewish and Chris­ tian laws .29 Such a privileged situation could hardly be granted to infidels. Al-Tabari (839-923), summarizes the whole of Islam’s thinking in this m atter in a single lapidary sentence: “Everyone has the law of the religion th a t he practises ” .30 Such a degree of tolerance would be incomprehensible in any other reli­ gious system. Out of respect for the earlier revelations, and out of deference to the will of God, who bestowed the diversity of religions in the world, M uhammad never entertained the possibility th a t one “nation” might im­ pose its law upon another, as to do so would ipso facto be to impose its religion upon the other “nation” , and th a t is something th a t God will not have. 2. The Outcome of the Conquests M uham m ad’s far-reaching tolerance had occasion to be applied in practice almost immediately when the Arabs conquered vast stretches of the Byzan­ tine and Sasanid Empires. Once the initial violence of armed conflict had passed, the conquerors allowed the indigenous populations a large measure of freedom, beginning with the freedom to practice whatever religion they preferred. It is now established beyond any shadow of a doubt th a t the first Arab conquerors committed no acts of religious intolerance. The conquerors were not really interested in anything but domestic peace and the regular payment of taxes. The entire civil and religious adminis­ tration of the conquered lands was scrupulously left intact. The conquerors remained fighters exclusively. Indeed, measures were adopted expressly to ensure th a t they did not lose their military qualities: [1 1 ] they were kept together in encampments, and they were forbidden to own land or engage in 26Ibid., v. 53. 27Sürat al-An‘ãm (6 ), v. 108. 28Sürat al-Mä’ida (5), vs. 51, 52. 29Sürat al-A‘räf (7), v. 156; Sürat al-Nahl (16), v. 91; Sürat al-Anbiyä’ (21), v. 107. 30Al-TabarT, Jämi* al-bayän *an ta ’wïl äy al-Qur’än (Cairo, AH 1323-29), VI, 179.

47

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

11

agriculture. The leaders were at particular pains to ensure th a t the business of tax-collecting was left in the hands of the former governm ent’s admin­ istrators, who could more readily be compelled to disgorge their takings if necessary. It must be adm itted th a t the system was shrewdly designed: it created, between the new rulers and the mass of their subjects, a sort of buffer caste th a t had to do the work of exploitation and suffer the resentment thereby engendered. W hat has just been said of the tax adm inistration, which is compar­ atively well known from papyrus documents, was equally true of other branches of the civil administration, a fortiori, inasmuch as the conquerors took even less interest in them. Along with freedom of religion, the new m asters allowed the subject population full administrative, legislative and jurisdictional autonomy. The autonomy enjoyed by the Christian commu­ nities, then, was not a concession freely granted, but a practical necessity resulting from Islamic theocratic concepts on the one hand, and from the social situations of victors and vanquished respectively on the other. No positive evidence, no concrete facts can be adduced to dem onstrate th a t the legislative autonomy of the Christian communities is as old as the period of the conquest. It is the very silence of our docum entary sources th a t appears to us to be the most telling evidence. In contrast, any allegation to the effect th a t the conquerors had restricted the legislative autonom y of the vanquished would have to be supported by hard evidence. By to day’s crite­ ria, to be sure, freedom of religion need not imply recognition of legislative autonomy in civil m atters. As we have just seen, however, law and religion were one in the eyes of the Arab conquerors; to recognize th a t a people had the right to practise its religion was equivalent to acknowledging th a t it was free to live by its own law in its entirety, including those aspects th a t would be regarded today as having no connection whatever with religion. It is true th at some scholars, confronted with the silence of the documen­ tary sources, have adduced the treaties th a t were supposedly concluded be­ tween some cities and the Arab generals. Under the term s of those treaties, the generals in question are held to have granted the inhabitants, inter alia, the right to retain their own laws and jurisdiction. Those who have in­ vestigated the origins of what are usually termed the “immunities” of the Christian communities have invariably made a point of attributing them to the texts of the capitulations of Jerusalem, Damascus or Alexandria, and above all to the celebrated “Covenant of ‘Umar” . [1 2 ] Pace those contem porary jurists who are unable to imagine any form of autonomy, within the structure of a state, th a t is not based on specific constitutional provisions or international conventions, this line of

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

48

Néophyte Edelby

12

reasoning must be regarded as invalid. Unfortunately, not one of those supposed treaties can be taken as definitely authentic. Both Christian and Muslim historians assiduously invented false documents of this kind. The former did not hesitate to make up sayings recommending the treatm ent they hoped to receive at the hands of Muslim governors and put them into the mouths of M uhammad or the early caliphs. Pious Muslims, with equal zeal, fabricated accounts and treaties in which the Prophet or the early caliphs were made to condemn what the writers regarded as the excessively lenient behaviour of those same governors. Every such document invariably ends with a long list of witnesses, a specific date and the Prophet’s formal seal. The authenticity of these documents, however, does not stand up to serious criticism. The Islamic tradition has preserved approximately ten letters ostensibly sent by M uhammad to various Christian groups in Arabia. There is also a long list of treaties supposedly concluded by the Arab generals who led the conquering armies, or by the caliph ‘Umar in person, with cities th at had submitted: a treaty with the city of Hims, a treaty with Damascus, treaties between ‘Umar ibn al-K hattãb and Jerusalem and Lydda, a treaty between ‘Amr ibn al-‘Äs and the Copts, the conditions imposed on the Christians of Syria by Abu ‘Ubayda ibn al-Jarrãh, and the best-known of them all, the renowned cahd ‘Umar (Covenant of ‘Umar), which is actually a declaration by the Christians of Syria themselves, acknowledging their obligations toward Islam. We have no space here to discuss all these texts individually. After a detailed analysis, Caetani concluded that: In general, these treaties are documents th a t were artificially composed at later dates, originating from various sources and of unequal value. An historian should not be unaware of them, but he should handle them very cautiously and beware of using any of them as a basis for any im portant conclusion; as foundations, they are not stable .31 As a rule, they portray the conditions of the Christian communities, not as it was at the time of the conquest, but as it had come to be by the early years of the ‘Abbãsid dynasty, i.e. in a rather unfavourable light. They are not invariably deliberate forgeries; [13] some of them are themes for 31Leone Caetani, Annali delTIslam (Milan, 1905-24), IV, 310.

49

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

13

student exercises ,32 and as such, they might be useful els an indication of the teaching of legal theory at a particular date about the way Christian communities should be treated. We do not deny for a moment th a t treaties (not necessarily written treaties) were concluded; we assert only th a t those treaties are not the ones th a t we are supposedly given in the works of the annalists. The function of these treaties was essentially to regulate the collection of taxes and Christian religious practices. It was extremely rare for them explicitly to guarantee freedom of religion inasmuch as th a t was taken abso­ lutely for granted. Understandably, they were even less concerned with the m atters of the religious and civil legislation th a t the Christian communities were required to apply or the jurisdiction to which they were subject. If we are to believe the Coptic chronicle of John of Nikiou ,33 ‘Amr ibn al-‘Äs, when negotiating the surrender of Alexandria with the Melkite patriarch Cyrus (one of several historical figures referred to as “Muqawqis” in the Arabic sources), undertook “not to intervene in the affairs of the Christians” . This is as accurate a way of putting it as any, and the only conclusion th a t we can draw with reasonable certainty from the silence of the documents. 3. The Umayyad Period The available documents from the Umayyad period also fail to provide us with direct evidence of the legislative autonom y of the Christian communi­ ties. However, the reality of th at autonomy is clearly apparent from 1 ) the fact th a t Christians were tried by special courts, the whole point of which was th a t Christians were subject to legislation other than th a t of the state, and 2 ) the legal collections th a t were composed a t th a t period by Church authorities, which would have served no purpose if the Christians had been subject to Islamic law. 1. The Umayyads were even more consistent than the original con­ querors in maintaining the system of adm inistration of the previous regime. The entire order of government, just as the Byzantines had left it, is rec­ ognizable beneath a thin veil of Arab or Arabised names. The only new institution introduced by the Umayyads was the office of qâdï. As far as we can tell, however, qãdis had no lasting influence on the lives of [14] Christian communities at this period. In the first place—improbable as the idea may 32A typical example is given in the Kitãb al-umm by al-Shãfi‘T (Cairo, AH 1321-25), IV, 118ff. The author presents a model treaty for the use of any Muslim general in dealing with a conquered people, with spaces left for the names of the contracting parties. 33 Chronique de Jean, évêque de Nikiou, Ethiopie text ed. and trans. H. Zotenberg (Paris, 1883), 575-76.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

50 14

Néophyte Edelby

seem—even for the Muslim population, the qâdï*s jurisdiction by no means replaced direct rule by the provincial governor; on the contrary, the qãdt*s authority appears to have been exercised merely by delegation from the gov­ ernor, who might withdraw it at any time. In the second place, the qãdt did not replace the chief of police, who was also appointed by the governor as a sort of lieutenant for criminal m atters, and wielded very broad judicial and coercive powers. In the third place, the creation of a public jurisdic­ tional institution did not bring about the disappearance of the old system of free arbitration, which not only survived but continued to be the regu­ lar procedure for settling disputes. Lastly, while the qãdt1s jurisdiction was unlimited ratione materiae, in practice it appears to have been restricted, ratione personarum, to Muslims alone. The title of qãdt l-muslimïn (judge of the Muslims) is itself significant.34 Moreover, al-Kindï’s history of the qãdts of Egypt leaves the impression th at during the Umayyad period, there was only one qãdt for all Egypt ;35 this is conceivable only on the assumption th a t the qâdï in question dealt exclusively with cases involving Muslims, who were as yet not very numerous and were in the habit of entrusting the task of settling their disputes to arbitrators of their own choosing. Canon 6 of the Nestorian synod summoned in 676 by the Catholicos M ar Gewargis I stipulates th at all disputes between Christians shall be judged in the church, in the presence of priests whom the bishop, with the consent of the community, shall designate for the purpose .36 The canon goes on to tell us: 1 ) th a t the bishop, with the consent of the community, sometimes designated ordinary laymen for th a t purpose; 2 ) th a t in the absence of duly constituted judges, some members of the community arrogated th a t power to themselves on their own authority; and 3) th a t the government sometimes designated Christians to serve as m agistrates among their co-religionists. It is doubtless to this category of qudãt al-nasãrã (judges of the Christians) th a t al-Mãwardl (d. 1056) is referring when he states th a t “it is the regular practice of governors to appoint infidel judges among their co-religionists ” .37 [15] Furthermore, alongside the ecclesiastical courts and Christian qãdts, local governors, who were usually Christians, continued to exercise their ju34A1-Kindl, The Governors and Judges of Egypt, ed. Rhuvon Guest (Leiden and London, 1912), 301, tells us that Qays ibn Abï l-‘Ãs was “the first qâdï who served as judge in Egypt among the Muslims”. 35Under the year AH 140, al-Kindf, Governors and Judges, 359, relates that Sãlih ibn ‘AlT appointed a qãdt for each district (jund), but the people complained, with the result that he had to revert to the practice of appointing only a single qâdï, Ghawth ibn Sulaymän, to hear all cases. 36J.-B. Chabot, ed. and trans., Synodicon Orientale (Paris, 1902), 484-85. 37A1-Mãwardí, Al-Ahkãm al-sultânïya, ed. Maximilian Enger (Bonn, 1835), 108.

51

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY The Legislative A utonom y o f Christians in the Islamic World

15

risdiction in Egypt until nearly the end of the first century AH. In Byzantine Egypt, the Pagarch had his own court alongside th a t of the defensor civ­ itatis; the relationship between the two jurisdictions is not yet clear. The Pagarch’s rulings could be appealed to the Duke’s court, which used the Pa­ garch’s court as its executive office, over and above the la tte r’s function as a court of law in its own right. Decisions rendered by any of these three ju ­ risdictions could be appealed to the Emperor. Between the Em peror’s court and th a t of the Duke there was an interm ediate court th a t had authority to hear cases involving sums of less than 500 solidi, and there were also the ecclesiastical courts, which were increasingly frequented by litigants .38 Interestingly, Arabic papyri from the Umayyad period confirm the survival of the same state of affairs. They depict the interplay of two civil jurisdic­ tions in Egypt: th a t of the Pagarch, who was the Christian local governor, and th a t of the Duke (am ir), who was the Arab governor of the province. At the same time, it appears from the papyri th a t when a complaint was brought before the Arab governor, he did not hear the case himself, but issued written instructions authorizing the plaintiff to bring his case before the Pagarch’s local authority .39 2. Further evidence of the legislative autonomy enjoyed by Christian communities is to be found in the collections of laws in use among the Christians at th a t period. These collections show th a t there was a Christian system of law th a t continued in use with the assent of the state. Al-MaqrTzI (1364-1442) informs us th a t down to the end of the first century a h , local custom and usage, in Egypt and elsewhere, were more prevalent than Islamic law .40 This is readily understandable in the case of the Muslim population, which as yet had no detailed legislation of its own; how much more plausible, then, th a t it should have been true of the Chris­ tian population as well, upon which the Arab conquerors had not imposed a new legal system. To the best of our knowledge, G raf stands alone in asserting, in his Geschichte der christlichen arabischen Literatur ,41 th a t for want of any leg­ islation of their own, the Christian communities applied the Islamic law 38Germaine Rouillard, L ’administration civile de l ’Égype byzantine (Paris, 1923), 145ff. 39This was doubtless a holdover from a practice that had been known in Egypt in Romain times. Cf. Louis Boulard, Les instructions écrites du magistrat au juge commissaire dans l ’Égypte romaine (Paris, 1906). 40A1-Maqnzï, Al-Mawâ‘iz wa-l-i‘tibâr f ï dhikr al-khitat wa-1-ãthãr ( al-Khitat), ed. Gas­ ton Wiet (Paris, 1911-27), IV, 143-44. 41 Georg Graf, Geschichte der christlichen arabischen Literatur, I (Vatican City, 1944), 557.

52

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 16

Néophyte Edelby

introduced by their conquerors. [16] According to him, it was not until the twelfth century th a t the Christians began to develop a civil law of their own, after the model of Byzantine law, parts of which had recently been trans­ lated into Arabic by the Melkites. This view, however, is untenable. The Christians did not begin by adopting Islamic law and subsequently devise a legal system of their own; on the contrary, they did exactly the reverse. We know th a t this was the case from the fact th a t our documentary sources from the Umayyad period show the Christians continuing to apply amongst themselves the same laws th at had governed their communities before the Arab conquests. This can be dem onstrated, especially in the case of Egypt, from papyri th a t constitute a sufficiently detailed record to enable us to study particular institutions from the Byzantine period and to follow them as they evolved under Arab rule. One demonstration of this kind, concerning contracts of purchase and sale, has been published by Louis Boulard .42 It is clear from his analysis th a t until the end of the eighth century (when contracts of sale began to be written in Arabic), conveyance, as practised among Coptic Christians, was still “the true Egyptian form of sale, dating from before the time of the Romans and left undisturbed by them ” ,43 which had outlived the Byzantines as it had outlived the Ptolemies. Islamic influence amounts to little more than a few external tokens such as the hijra year, confirmed by the names of presiding m agistrates, and the fact th at the names of the Arab amirs were used in oaths, instead of th a t of the Basileus .44 W ithin the former Sasanid Empire, the fact th a t Christian communities enjoyed legislative autonomy is confirmed, for this period, by the 25 decisions or responsa 45 of the Nestorian Patriarch HenãnTshõ* (patriarch 686-99). In the area of civil law alone, the patriarch deals with cases of the emancipation of slaves (resp. V), marriage (XXI, XXIII), the legal capacity of minors (XXIV), guardianship (XII), pledges (XIII), legitimate succession (IX, XI, XII, XIV, XV, XIX, XX, XXI, XXII), wills and bequests (XX, XXI), and judicial procedure (IV, VIII, IX, X, XVI, XXI). The very content and variety 42Louis Boulard, “La vente dans les actes coptes,” Etudes d ’histoire juridique offertes à P.Fr. Girard, II (Paris, 1913), 1-94. 43Ibid., 90. 44For example, eO[i.oXoyõ npòç xè ToO Oeoö xoö 7iavToxpáxopo wa l-cizz). The third is ethnic bigotry. The fourth is personal preference. The fifth is licentious laws (at-tarfyis f i s-sa rä ’f ) . The sixth is the phantoms o f sorcery (h ayãlãt a ssih r) and the lik e.” 46 In his K itã b a l-m a sa ’il wa l-agwibah cAmmär also lists six characteristics or traits (al-hisãl) which he thinks should indicate that any religion in which they are found is a human fabrication. They are: personal preference, licentious laws, the sword, gifts and presents o f the desirable things o f this world (al-m aw ãhib wa ca tã yã min raghã ’ib ad-dunyah), fanaticism and the

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

186 74

exercise o f ethnic bigotry (ham iyyah wa tacassub), and the observance o f the phantom s o f sorcery (m ucãyanah h a y ã l.. . min as-sihr).41 The variations in cA m m ãr’s lists o f these negative qualities merely serve to highlight the drift o f his thinking. H is argument is that if one finds any such m otives, grounds, or traits in any religion, then it is possible for him to allege that this religion is successful in the world because o f the attractiveness or the forcefulness o f these qualities, rather than because o f any recognizable divine endorsement o f that religion. cAmmár was not alone am ong the early Christian, Arabic speaking apologists in drawing up such a list o f characteristics which should enable a searcher to exclude the claim o f any religion to be G o d ’s religion if it possesses one or more o f these characteristics. The practice o f utilizing such an argument is com m on to m ost o f the Christian apologists o f the first Abbasid century whose works we know. Abü Quarrah wrote a small treatise which argues that the true religion is the one that avoids acceptance because o f license (ar-ruhsah), strength (al-c izz), the exercise o f ethnic bigotry (attacassub), and the satisfaction o f the vulgar mind (q u n if al-ca ql as-siiqi).4S In his E pistle on the C onfirm ation o f the Christian R eligion and the C onfirm ation o f the H o ly Trinity Abü R ä’itah argues that there are six categories (aqsãm ) o f inducement to religious conviction ( f tiq ã d ) which are unworthy reasons for professing any given creed. They are worldly desire (ar-raghbah), ambition (at-tam ac), overpowering fear (ar-rahbah al-qã hirah), license (ar-ruhsah), personal preference (al-istihsan), and collusion, or ethnic bigotry (ã t - t a w ã t u al-casabiyyah ).49 A nd finally the Nestorian scholar Hunayn ibn Ishäq, pursues this same line o f reasoning in the course o f his response to the challenge proposed to him by the Muslim scholar Ibn al-Munaggim. H e says, What is true and false o f all the doctrines is known only from the m otives (asbãb) for their acceptance initially. The motives for which falsehood is accepted are other than the m otives for which the truth is accepted. There are six motives for the acceptance o f falsehood. The first o f them is that the one accepting it is unwillingly com pelled to accept that to which he has been brought over. The second is that a man is willingly fleeing from oppression and violence when he cannot bear them . So because o f them he is converted to that from which he hopes for ease and com fort. The third is that one chooses might over meekness, nobility over lowliness, and strength over weakness. So he leaves his ow n religion and is converted to another one. The fourth is that the adherent o f a doctrine is a deceitful man who is crafty in speech. So he confuses and prevails over those to whom he is appealing. The fifth is that he takes advantage o f the ignorance and the lack o f manners o f those to whom he appeals. The sixth is that there should be the fervor o f natural kinship on the part o f those making the appeal. So one w ould not want to break o ff that tie that is between him self and one w ho agrees with him in religion.50 The similarities and the differences in these several lists o f unacceptable grounds for religious comm itm ent are readily evident. The overall argument that is em ployed by the apologists from the three opposing Christian sects is virtually identical in its m ain lines. There are some differences in the way in which each author deploys the argument, as well as in the term inology that each one uses to name the unsuitable m otives for faith. This

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY C om parative R eligion in the A p o lo g etics

187 75

fact suggests that the authors are not simply copying this argument from som e earlier source. Rather, once inspired by the same method o f procedure, each author carries on his apology in his own terms. We can no longer determine whose original idea it was to discuss the unworthy incentives to religious faith, and to argue that only Christianity avoids acceptance because o f one or another o f these m otives. N o such argument seems to have been employed by earlier Christian apologists. Perhaps in the form in which we find it here, the argument is an original contribution to apologetics on the part o f the anti-Muslim apologists o f the first Abbasid century. The closest analogue in Christian apologetic literature to this negative criterion for discerning the true religion, as it is deployed by cAmmãr and his fellow apologists o f the first Abbasid century, seems to be a portion o f the apologetic argument for Christianity developed by Roger Bacon (c. 1214 — c. 1292) in his O pus M ajus. The argument is to be found in the seventh part o f this work, entitled M oralis P h ilosoph ia, part 4 .51 In the course o f it, Bacon compares the religions o f the peoples whom he calls pagans, idolators, Tartars, Jews, Christians, and the Saracens. All o f them except the Christians, he argues, foster the pursuit o f human happiness in terms o f one or several o f five corrupt goals. These goals are, in his words, “ voluptas, divicie, honor, potência, fam a seu gloria n om in is.” 52 One can readily see the similarity o f this list o f corrupt goals o f human behavior to the lists o f unworthy motives for religious faith that we have been reviewing from the works o f the three Christian apologists who wrote in A rabic. Bacon says that in form ulating his argument, he is relying on al-Färäbi’s, and B oethius', development o f A ristotle’s ideas in the P o litic s.53 It is not unlikely that the inspiration for the developm ent o f this same m ode o f argument also came to cA m m är, A bu Qurrah, and Abü R à’itah from A ristotle. As we have seen, a distinguishing feature o f their apologetics is their reliance on a re-presentation o f ideas found in the general Christian, philosophic tradition which was largely inherited from the thinkers w ho created the late Neoplatonic synthesis o f thought. M otives f o r the A cceptan ce o f C hristianity cAmmãr argues that Christianity is the only religion that God endorses by means o f the divine signs that he produced at the hands o f the apostles, when they first preached Christianity throughout the world. H e claims that no trace can be found for an yon e’s ac­ ceptance o f Christianity on the basis o f any one or more o f the unworthy motives for the acceptance o f a religion that he has listed. N o such claim can be made in behalf o f any one o f the other religions, cA m m ãr maintains. Therefore Christianity alone is G o d ’s religion. Judaism and Islam are the only two non-Christian religions w hose bid for acceptance as the one true religion cAm m ár discusses in his K itã b al-burhãn. In his brief discussion o f Judaism, as we have seen, he argues that Judaism cannot be G od ’s religion, in the pre­ sent day absence o f public miraculous signs worked in its behalf, because it is possible to imagine that people may embrace it on account o f som e one or another o f the unworthy motives for religious faith. W hen it com es to Islam, however, we are unsure o f the full content o f cA m m är’s argument. As we m entioned above, the text we have is m issing at least one page. It is clear that cAm m ãr believed that ethnic bigotry and tribal collusion played a role

188

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

76 in the original acceptance o f Islam. H e mentions these two unworthy m otives for the acceptance o f a religion in the portion o f his argument that is available to u s.54 But his point o f emphasis in discounting Islam ’s claim to be the true religion is that the Q u r’an rejects the idea that M uham m ad’s preaching should be endorsed by miraculous signs. °Ammar quotes two passages from the Q u r’an to this effect, namely, al-Isrà’ (17):59 and al-Ancãm (6): 109. He also claims that according to an interpretation o f cAbd A lläh ibn al-cA bbás, a noteworthy com panion o f the prophet to whom many traditions are ascribed in Muslim sources,55 the rejection o f miraculous signs recorded in a/-A n cãm came on the occasion o f an oath sworn by Christians, Jews, and polytheists that if they should see such a sign worked at the hands o f Muhammad they would put their faith in him .56 The Christians and Jews are not in fact explicitly m entioned in the passage quoted from al-A ncãm . And cA m m àr’s purpose is not so much to argue that an oath from Christians occasioned the rejection o f miracles. Rather his purpose is to prove that any appeal to miracles made by M uslims in behalf o f Islam ’s claim to be the true religion is ruled out a p rio ri by the Q u r’an. As we have seen, cAmmár him self maintained that divine endorsement o f a religion must be in the form o f miraculous signs. There is som e evidence to suggest that Muslim apologists in his day were making reference in their arguments in favor o f Islam to stories o f miracles worked by M uhamm ad, as these are reported in popular traditions, as well as to the doctrine o f the f g à z al-Q ur'ân, that is, the inimitability o f Quranic literary style.57 °A m m ãr therefore intends to discount arguments o f this sort, as having no value in recommending Islam as G od ’s religion. The centerpiece o f cA m m ár’s account o f the motives for accepting Christianity in­ stead o f any other religion, in both his K ità b al-burhãn and his K itã b a l-m a sã ’il wa lagwibah, is a dem onstration that, given the nature o f Christianity and its demands, it is inconceivable that anyone could be brought to profess it for any one or several o f the un­ worthy reasons for accepting a religion. In this connection it is noteworthy that cA m m ár puts forward his arguments in terms o f conversion from one religion to another, rather than in reference to some hypothetical, uncom m itted context in which a person could choose between the several competing religious traditions that are recomm ending themselves. Conversion was, o f course, the experience o f most o f the first Christians. And conversion, we may suspect, is the tem ptation o f many Christians in cA m m ãr’s day. He is anxious to prove that conversion from one religious com m unity (al-m illah) to another is usually accomplished only on the basis o f one or several o f the unacceptable motives for faith — except in the instance o f Christianity. It is inconceivable, he m ain­ tains, that anyone could embrace Christianity for any o f these reasons. “ R ather,” he says o f the Christian religion, “ it forces the mind to other things, namely, signs and m iracles.” 58 This, o f course, is the whole point o f cA m m är’s apologetic argument. cAmmar makes his point by discussing each one o f the unworthy characteristics (hisãl) and m otives (asbãb), and how it has no place in Christianity. U nlike other religions, Christianity is not to be found only am ong one people, in any one country, or limited to a particular language group. In fact Christians are widespread am ong peoples who are traditionally hostile to one another. Therefore it is inconceivable that the religion could have been established by means o f som e sort o f tribal collusion am ong the nations. Regarding the sword, cAmmär says the follow ing.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY C om parative Religion in the A p o lo g etics

189 77

The whole world that is opposed to the Christian religion, the Jews, the Magus, the M uslims and others, agree that C hrist’s disciples did not conquer people by the sword, not did they use it. Rather, the most that the Jews can ascribe to them is sorcery and trickery, rather than the sword. W hile the Magüs and the Muslims think o f them in regard to the signs.59 The last sentence, obviously, expresses the very conclusion for which °A m m ãr is arguing. Bribery and cajolery cannot be considered m otives for the original acceptance o f Christianity. Its original preachers (ad-ducah), cA m m ãr reminds the reader, were poor men who had nothing to offer but their own poverty and the counsel to their followers to cultivate penury in this world. Ethnic bigotry was not an element in the establishment o f Christianity in the world. Otherwise, cAmmár claims, only Jews would have follow ed Jesus. W hereas the Chris­ tian community is made up o f people from every nation. Certainly personal preference (al-istihsan) did not dictate the acceptance o f Chris­ tianity in the beginning, cA m m ãr maintains. W hat he means by ‘personal preference’ in this connection is akin to the meaning o f the phrase “ religious preference” in Am erican civil parlance. It means not simply what one likes best in terms o f religious groups, but what one thinks is true, on the basis o f his own reasoning and personal conviction. The Arabic term istihsãn has a legal nuance, meaning “ to make a decision for a particular in­ terpretation o f the law as a result o f on e’s own deliberation.” 60 It is the element o f o n e’s own deliberation, his personal opinion (ar-ra*y), that cAm m är has in mind here. In his K itãb al masã*il wa l-agwibah he uses the intellectual subleties o f the ancient Greek systems o f thought as an example o f the operation o f al-istihsãn. A nd in this same con ­ nection he also makes reference to the profession o f m onotheism (at-ta w h id ) that is characteristic o f the doctrine “ that goes under the name o f a t-ta w h id .” 6X H e undoubted­ ly means Islam, in which the profession o f m onotheism is the basic com ponent o f faith. A bout the central doctrines o f Christianity, which cA m m är lists in this section o f his K itãb al-burhãn, he claims that it is inconceivable that istihsãn could have any part in them. He says, I think that the Christian religion is com pletely contrary to it. That is because its preachers lay claim to things, and transmit reports, which opinion would not contrive, and which would not arise in thought, nor occur in a brain, nor would a mind conceive them .62 He is speaking o f such doctrines as the virgin birth, the son o f God killed and buried, the resurrection and ascension o f Jesus, his second com ing, the worship o f som eone crucified, and belief in an unknown, heavenly reward. Christianity clearly grants no license for the pursuit o f pleasure (at-tarhis), cA m m är argues. A nyone can read its strictures about marriage and the control o f the desire for women, which cA m m är maintains is the strongest human desire. In witness o f this allegation he cites the example o f a contem porary M uslim ruler w ho went so far as to invade the hostile Byzantine territory in pursuit o f a w om an. This interesting detail by way o f example helps us to date cAm m ár. H e is presumably speaking o f the caliph alMuctasim’s expedition to Am orium in the year 838, which was remembered in Arabic poetry as an incident instigated by a w om an.63 cAm m ár goes on to cite Christianity’s well known penchant for asceticism.

190

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

78 Finally, cAm m är argues that it is inconceivable that anyone should accept Christiani­ ty because o f the phantom s o f sorcery. The miracles and wonders o f Christianity, he maintains, are not elusive. Rather, what is promised is achieved. The preachers fulfilled the conditions they set for them selves. The conclusion o f his argument is simply stated. The facts com pel us to dissociate Christianity from the earthly motives by reason o f which other [religions] are established, in favor o f the testim ony that it was accepted and established by clear signs from G od, and propogated among all o f the different nations by reason o f its reliably true distinguishing marks.64 Therefore, according to cAmmär, we have a com pelling, intellectually respectable reason to believe that Christianity is the only true religion. III. Islam an d the Christian A p o lo g e tic T heology o f the First A b b a sid C entury Students o f Islam have long been tantalized by the m anifold similarities between various aspects o f the Islamic science o f religious discourse f i l m al-kalãm ) and Chris­ tian, largely Greek, philosophical theology. Since so much o f this philosophical theology found expression in the Syriac monastic and scholastic traditions that were at hom e in Iraq where the Muslim sciences were born after the A bbasid revolution, it is not unreasonable that one should expect to find in the Christian traditions at least the traces o f the ideas that would be elaborated in Muslim scholarship. In this context it is not sur­ prising that modern western scholars should search am ong the writings o f the Muslim mutakallim ùn for “ parallelisms which betray dependence.” 65 Some scholars have even carried this enterprise to the point o f suggesting earlier Christian or Jewish parallels for practically every significant doctrinal formulation to be found in M uslim theology, regardless o f its own Islamic context, and sometimes heedless o f the philosophical and linguistic presuppositions in terms o f which it is phrased.66 Contrariwise, there has also been no paucity o f scholars to deny the dependence o f the m utakallim ùn on Christian sources. For the fact is that while there are many tantalizing similarities to Christian doctrines and form ulations to be found in the writings o f the Muslim thinkers, the differences are equally striking. From its inception Islamic religious discourse is recognizably and uniquely Islamic, and distinctly non-Christian in its thought, form at, and style.67 Already in 1842 this fact encouraged A ugustus Schmölders to maintain that there is no relation between the Muslim m utakallim ùn and the Christian apologists.68 A nd most recently Richard M. Frank, on the basis o f his extensive study o f the works o f som e o f the most significant Muslim m utakallim ùn from the Basrian school o f the M u ctazilah, remarks: “ In order to understand the kalãm ... one needs only the native language and tradition o f Arab Islam . . . This is not to say that the kalãm contains no parallels with and no clear dependences upon the pagan and Christian tradi­ tions that preceded it, but rather that these dependences are chiefly to be sought on a deeper level. M ost o f the basic issues, though in a real sense (and for us unavoidably) ‘G reek,’ are nevertheless framed and conceived in an Islamic m ode and must be so read.” 69

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY C om parative Religion in the A p o lo g e tic s

191 79

It is difficult, if not im possible, to isolate and to identify with any specificity the “ deeper level” on which there lie the dependences o f the Muslim scholars on the works of Christian apologists. But it is not difficult to demonstrate that the M uslim mutakallimün o f the first A bbasid century were aware o f the Christian apologists who were writing and teaching in their midst. Am ple docum entation for this fact is easily provided. We have already cited reports from the Fihrist o f Ibn an-Nadïm about the controversies o f Abü Qurrah and c Ammãr al-Basri with the M uctazilites al-Murdãr and Abü al-Hudhayl respectively.70 In addition, Ibn an-Nadim m entions more general refutations o f Christians written by such ninth century M uslim scholars as Dirar b. CAmr, Abü cIsã al-W arräq, Abü H udhayl, and Hafs al-Fard.71 When cA bd al-Gabbár (d. 1025), the famed M uctazilite doctor, com es to the anti-Christian section o f his T athbit dala*il an-nubuw w ah he lists eight members o f the M u ctazilah who had written refutations o f Christians. M ost o f them are from the tenth century. But am ong ninth century scholars he lists al-G áhiz (d. 864) and Abü Gacfar al-Iskafi (d. 885).72 It is not necessary here to list all o f the M uslim , anti-Christian texts from the first Abbasid century that are available to us. We have said enough to make the point that the major Muslim scholars o f the period were aware o f the work o f their contem porary Christian apologists.73 Recent western scholars have called attention to the basically defensive apologetic tone and character o f much o f the M uslim cilm a l-kalãm ,1AThis fact reminds us that a good deal o f the “ dependence” o f M uslim writers on Christian sources may have been in the way o f negative influence. This is to say that Muslim scholars may well have been under pressure to form ulate a system o f religious thought in defense o f their m ost important doctrines, which would not be vulnerable to the attacks o f the Christian polemicists. The differences o f their approach to a given issue are readily evident in the way in which Muslims and Christians address themselves to a topic. The issue o f the discernment o f the true religion is a case in point. We have seen that our three Christian apologists take this topic in hand by reaching back into their N eoplatonic philosophical heritage, and on this basis they elaborate a far reaching system o f apologetics. From the effects o f his creative activity they claim to be able to derive a knowledge o f the existence o f God, and o f how his perfections may be described by analogy with the perfections we may discern in his creations. The identification o f the creaturely perfections, o f course, is determined by the N eoplatonic anthropology that informs this whole intellectual enterprise. Abü Qurrah claims to be able to discern the true religion on the basis o f measuring the several religious creeds against what this philosophy teaches him to be true about God and man. W hile cA m m är al-Basri says that he cannot rely on such a comparison to teach him which o f the several religions is true. N evertheless, his list o f characteristics which would disqualify the claim o f any religion in which we might be able to conceive o f their presence, even if it was once accepted because o f evidentiary, miraculous signs, betrays the values o f the same Christian, N eoplatonic anthropology that inspires Abü Qurrah and the other Christian apologists o f the p eriod .75 The intellectual presuppositions o f the Muslim apologists are o f a different order. To state it quite simply, they are not grounded in any Greek philosophical system, but in the Q ur'an. There is a system o f thought that inspires their understanding o f the Q ur'an,

192

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

80 analogous to the role o f the N eoplatonic philosophical systhesis in the Christian interpretation o f the Bible. It is what we would perhaps call the hermeneutics o f the Arabic grammatical tradition. The influence o f Arabic grammar studies on the thought patterns o f the Muslim m utakallim ùn has been examined by Richard M . Frank. A somewhat extensive quotation from his work may make the point very clear for our present purpose. Fittingly, for it joins central elements o f the ancient Arab culture with Islam — poetry and the Koran — grammar is the first science to reach maturity in Islam — before the end o f the second/eighth century — and it does so, almost com pletely apart from earlier and alien traditions, as a peculiarly Islamic science. This attention to language, most particularly to the language o f the Koran and to the grammatical and lexical structures and the characteristics o f literary Arabic, had a profound influence on the form ation and developm ent o f the kalãm , m ost especially in the principal M uctazilite tradition o f Basra and in that o f the A scarites, not simply in their term inology but also in the manner in which many fundamental problem s o f ontology and ethics — concerning, thus, G o d ’s Unity and His Justice (atta w h id wal-cadl) — were conceived, formulated and analyzed.76 And so we must turn first to the Q u r’ân to discover the Muslim scholar’s approach to the problem o f discerning the true religion. According to the teaching o f the Q u r’ân, God has sent “ His messenger with the guidance and the religion o f truth (din al-haqq), that He may uplift it above every religion, though the unbelievers be averse” (at-Taw bah 9:33, Arberry). In context the Jews and the Christians are seen as adversaries. Muhammad is, o f course, the messenger whom God has sent. And the true religion, or religion o f truth, to remain consistent in our translation o f one o f the Q u r’ân*s standard phrases (cf. al-H ugurãt 48:28, and as-Saff 61:9), is Islãm , or submission to the message. “ The true religion with G od (ad­ din cind A llã h ),” says the Q u r’än, “ is Islã m ” (À l cIm rãn 3:19, Arberry). This is the straight, or the right religion (ad-din al-qayyim ), to borrow a n o th er‘standard phrase from the Q u r’ân (cf. for example, at-Taw bah 9:36). A nd G od ’s “ conclusive argum ent” (ial-huggah al-bâlighah, cf. al-A n cãm 6:149) in behalf o f the right religion consists in the signs (ãyãt) that are the verses o f his speech in the Q u r’ân, sent down in clear Arabic language. “ Those who do not believe in G o d ’s signs, God will not guide them , and to them is due grievous punishm ent” (an-N ahl 16:103 and 104). A s we have seen above, and as cAmmär al-Basri him self records, the Q u r’an explicitly rejects miracles o f any other sort as motives o f faith. And so most M uslim scholars based their studies on a m eticulous examination o f the language o f the Q u r’ân. Eventually the doctrine o f the icg a z alQ ur'ân assumed a role in Islamic apologetics that is comparable to the role o f the theory o f evidentiary miracles in Christian apologetics.77 Given this state o f affairs it is not surprising that the philosophic cast o f m ind that characterizes the writings o f the Muslim m utakallim ùn differs so markedly in its style and its intellectual assumptions from the N eoplatonic spirit o f the Christian thinkers.78 Religiously the most striking difference seems to be the M uslim insistence that our only true knowledge o f God comes from his own revealing speech in the scriptures. W hile the Christian remains convinced that he may com e to a limited but true knowledge o f G od by

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY C om parative R eligion in the A p o lo g etics

193 81

the use o f his own reasoning capacity. This knowledge then enables him to discern the true religion, he believes, and it further enables him to accept the scriptures and the teachings o f the church, which carry his knowledge o f G od beyond what he can acquire by his own efforts. The influences that are exerted by Christians on M uslims, and vice versa, within the sphere o f these antipodal basic assum ptions, are, then, largely the negative one o f avoiding form ulations which may fall victim to the other side’s polem ics; and the positive one that encourages the creation o f a satisfying, opposing system o f thought. Support for this conclusion exists in the not infrequent remarks we find in the writings o f the M uslim m utakallim ün to the effect that the form ulations espoused by their own Muslim adversaries, are to be found wanting precisely because they play into the hands o f the Christians. So, for exam ple, as-Sahrastäni claims that the teachings o f Abü Hudhayl and Abü Hásim about the divine attributes put one in mind o f the divine hypostases o f the Christians.79 Christian thinking certainly had a strong influence am ong M uslims when one m oves away from the atmosphere o f religious apologetics, into the realm o f Greek philosophy and logic, as fostered by the philosophers (al-falãsifah) o f the stamp o f al-Kindi (d. 873). But this influence was exercised in a non-theological fashion, that bears no relationship to the sort o f apologetics espoused by Abü Qurrah, Abü RãTitah, and cAm m ãr. Muslim influence on the Christian apologetic writers can be seen in the numerous instances in their treatises where they em ploy the language o f the Q ur'an. Clearly their intention is to invest their arguments with all o f the persuasiveness that they can derive from Muslim religious sentiments, phrased in Muslim religious language. M oreover, Abü Qurrah, Abü R ä ’itah, and cAm m ãr are adept at taking advantage o f the currency o f doctrinal disputes within Islam on certain subjects to advance their own arguments in favor o f Christian doctrines. This practice is nowhere more in evidence than in their discussions o f the doctrine o f the Trinity, utilizing the vocabulary and phraseology o f the Muslim scholars’ discussions about the divine attributes. While what is described here is hopefully a fair statement o f the intellectual relationship between the M uslim and the Christian religious com m unities during the first Abbasid century, the account should not com e to a close without a brief reference to a hitherto unmentioned theological venture that failed. W e have made the point that the intellectual background which N eoplatonism furnished for the Christian apologists is contrary to the set o f assumptions which the study o f Arabic theoretical grammar provided for the M uslim m utakallim ün. Because o f this basic antipathy o f mind sets, we have implied, the two systems o f apologetics developed in different directions. But there is evidence that prior to the developm ents in the intellectual life o f the first A bbasid century which we have been discussing, there was an attempt on the part o f som e Muslims to construct an Islamic apologetics on the basis o f a set o f essentially Neoplatonic theses. The venture failed, probably as much as for any other reason, because o f the contrariety that is evident between any N eoplatonic notion about how one may come to know anything about G od, by the use o f reason, and the Islamic view that one comes to know about God principally through his revealed speech in the Q ur'an. The venture at issue is the body o f thought that is associated with the name o f Gahm ibn Safwán (d. 746). Everything that is known about his thinking com es from the writings o f intellectual opponents who are busy refuting G ahm ’s presumed ideas, as these are

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

194 82

V

espoused by his alleged followers, the G ahm iyyah. Principal among the opponents o f the V G ahm iyyah during the first Abbasid century was the famed rigorist M uslim scholar, Ahmad ibn Hanbal (d. 855).80 A study o f the system o f thought which is represented in the sources as that o f the G ahm iyyah reveals that it can be understood as essentially Neoplatonic in its inspiration and basic assum ptions.81 What makes this point particularly significant in the present context is that it suggests that there were Muslim thinkers active during the important years o f the first Abbasid century, w hom their adversaries called the G ahm iyyah, who sought to construct an apologetic theology utilizing som e fundamental philosophical conceptions that were com m on both to themselves and to the Christians. They would have been conceptually equipped to dispute with the Christians on their own terms, and to construct a system o f M uslim apologetics that rejected the Christian claims. The telling point to be made here, in our discussion o f the influences o f the Christian apologists on the thinking o f the M uslim m utakallim ün y and vice versa, is that the theses o f the G ahm iyyah were found to be objectionable, and inconsistent with basic Islamic principles, not only by H anbalites, but by Muctazilites as well. It is important to emphasize this fact because scholars w ho want to find Christian influences in the Muslim kalãm often allege that this influence is particularly evident in the works o f the M uctazilites.82 In the intellectual milieu that we have described, the work o f the Christian apologists was an important element in the scholarly movement that was to come to fruition during the second Abbasid century in the works o f thinkers both Muslim and Christian w hose names are still remembered by scholars everywhere.

N O TES 1

r

Cf. H.A.R. Gibb, “ The Fiscal Rescript of Umar II,” Arabica 2 (1955), 2-3; M.A. Shaban, The Abbasid Revolution (Cambridge: University Press, 1970), p. 168. A. Mingana, “ Timothy’s Apology for Christianity,” in Woodbrooke Studies 2 (1928), J-162. Shorter Syriac rendition in A. Van Roey, “ Une apologie syriaque attribuee à Élie de Nisibe,” Le Museon 59 (1946), 381-97. For the Arabie versions, cf. Hans Putman, L ’eglise et l ’islam sous Timoihee I (Beyrouth: Dar el-Machreq 1975): Robert Caspar, “ Les versions arabes du dialogue entre le Catholicos Timothée I et le calife al-Mahdi,” Islamochristiana 3 (1977), 107-75.

^ A. Van Roey, Nonnus de Nisibe; traité apologétique, Bibliothèque du Muséon, 21 (Louvain: Bureaux du Museon, 1948). 4

Addai Scher, Theodorus bar Koni Liber Scholiorum, CSCO, 55 and 69 (Paris, Poussielgue, 1910 and 1912), LXIX, 231-84. I. Arendzen, Theodori Abu Kurra de cultu imaginum libellus e codice arabico nunc primum editus latine versus illustratus (Bonn: Typis Caroli Diobnig, 1877); Constantin Bacha, Les oeuvres arabes de Théodore Aboucara évêque d'Haran (Beyrouth: lmp. Alfawaid, 1904); idem, Un traité des oeuvres arabes de Theodore Abou-Kurra, évêque de Haran (Tripoli de Syrie and Rome: L’evêché greccatholique, 1905); Georg Graf, Die arabischen Schriften des Theodor Abu Qurra, Bischofs von Harran (ca. 740-820) (Forschungen zur christlichen Literatur — und Dogmengeschichte, X. Band. 3/4 Heft; Paderborn: F. Schoningh, 1910); Louis Cheikho, “ Mimar li Tadurus Abi Qurrah fi Wugüd alHäliq wa d-Din al-Qawim,” al-Machriq 15 (1912), 757-74; 825-42; Georg Graf, Des Theodor Abu

195

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY C om parative Religion in the A p o lo g e tic s

S3

Kurra Traktat über den Schöpfer und die wahre Religion (Beitrage zur Geschichte der Philosophie des Mittelalters. Texte und Untersuchungen, Band XIV, Heft. 1; Munster i.W.: Aschendorf, 1913); Ig­ nace Dick, “ Deux écrits inédits de Theodore Abuqurra,” Le Museon 72 (1959), 53-67; Sidney H. Griffith, “ Some Unpublished Arabic Sayings Attributed to Theodore Abu Qurrah,” Le Museon 92 (1979), 29-35. For Abu Qurrah’s works preserved only in Greek, cf. J.P. Migne, Patrologiae Cursus Completus, Series Graeca (161 vols, in 166; Paris: Brepols, 1857-1887), XCV1I, cols. 1461-1610. For a recent general study on Abu Qurrah cf. Ignace Dick. “ Un continuateur arabe de saint Jean Damascene: Theodore Abuqurra, évêque melkite de Harran,” Proche-Orient Chrétien 12 (1962), 209-23, 319-32; 13 (1963), 114-29. 6

7

g 9

.

-

Georg Graf, Die Schriften des Jacobiten Habib ibn Hidma Abu R a ’ita (CSCO, 130 and 131; Louvam: L. Durbeck, 1951). c Michel Hayek, ed., Am mar al-Basri, Apologie et Controverses (Beyrouth: Dar el-Machreq, 1977). Cf. the editor’s French introduction to the texts, also published in Islamochristiana 2 (1976), 69-113. Georg Graf, “ Christliche Polemik gegen den Islam,” Gelbe Hefte 2 (1926), 825-42. Cf. Bacha, 1904, p. 28.

^Bayard Dodge, ed. and trans., The Fihrist o f al-Nadim (2 vols.; New York: Columbia University Press, 1970), I, 394. 11 Dodge, p. 388. I2Cf. Cheikho, 1912, and Graf, 1913 for Abü Qurrah; Graf, 1951, CXXX, 131-159; CXXXI, 159-194 for Abu Ra’itah; Hayek, 50-56, 1-90 for cAmmár. ^Hayek, pp. 135-36. Reading ad-din for alladhina, II. 19 and 20, p. 135. U Cf. Cheikho, 1912, p. 837. 15Cheikho, p. 766. ^C f. Bacha, 1905, pp. 13 and 14. 17 c Cf. A.S. Tritton, The Caliphs and their Non-Muslim Subjects, a Critical Study o f the Covenant o f Umar (London: Oxford University Press, 1930); A. Fattal, Le statut légal des non-musulmans en pays d'islam (Beyrouth: Imprimerie Catholique, 1958). 18 Tritton, p. 12. 19 Cf. Rachid Haddad, “ Hunayn ibn Ishaq apologiste chretien,” Arabica, 21 (1974), p. 297. 20

Gerhard Klinge, “ Die Bedeutung der syrischen Theologen als Vermittler der grieschischen Philosophie an den Islam,” Zeitschrift fur Kirchengeschichte 58 (1939), 375-83.

^*Cf. Raphael J. Bidawid, Les lettres du patriarche nestorien Timothee I, Studi e Testi, 187 (Città del Vaticano: Biblioteca Apostolica Vaticana, 1956), p. 35. For a general discussion of these themes cf. De Lacy O’Leary, H ow Greek Science Passed to the Arabs (London: Routledge and K. Paul, 1949); Richard Walzer, Greek Into Arabic (Cambridge, Mass.: Harvard University Press, 1962); F.E. Peters, Aristotle and the Arabs (New York: New York Univer­ sity Press, 1968).

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

196 84 23

24

Cf. A.C. Lloyd, ‘T he Later Neoplatonists,” in A.H. Armstrong, ed., The Cambridge History o f Later Greek and Early Medieval Philosophy (Cambridge: University Press, 1967), pp. 272-325; I.P. Sheldon-Williams, “The Greek Christian Platonist Tradition From the Cappodocians to Maximus and Eriugena,” The Cambridge H istory..., pp. 421-533. Cf. also R. Walzer, Galen On Jews and Christians (Oxford: University Press, 1949).

Cf. Klinge, above, and my doctoral dissertation, “The Controversial Theology of Theodore Abu Qurrah, Melkite Bishop of Harrän; A Historical, Comparative Study in Christian Arabic Literature,” The Catholic University of America; Washington, 1978.

25Hayek, p. 22. 26Hayek, p. 26. 11Ibid. 2SIbid. 29Hayek, p. 27. 30 v. Al-Gahiz, to name one Muslim polemicist, makes this charge in regard to the Christian claim that the doc­ trine of the Incarnation is in the scriptures. “The motive (sabab) for this interpretation is entirely sin, uncritical acceptance, and belief in anthropomorphism.” J. Finkel, Three Essays o f Abu cOthman cAm r ibn Bahr al-Jahiz (d. 869) (Cairo: Salafyah Press, 1926, p. 25. 31

Cf. W. Montgomery Watt, The Formative Period o f Islamic Thought (Edinburg: University Press, 1973), p. 118.

32

Hayek, p. 27.

33Ibid. 34

Hayek, p. 29.

35Cf. Constantin Bacha, Un traité des. oeuvres arabes de Theodore Abou-Kurra, évêque de Haran (Tripoli de Syrie and Rome: L’évêché grec-catholique, 1905). Cf. also Greek opusculum 21, PG, XCVIl, cols. 1547-552. 36

Hayek, pp. 130-35.

37Cf. Watt, pp. 234-35. 38

Hayek, p. 28.

39 Cf., for example, George F. Hourani, Islamic Rationalism, the Ethics o f cA bd al-Jabbãr (Oxford: Claren­ don Press, 1971), p. 70. 40

Hayek, p. 29.

A]Ibid. 42

Hayek, p. 30.

4W

197

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Comparative Religion in the Apologetics 44

85

There is at least one page missing in the manuscript from which Hayek has copied and edited the Kitab atburhan. The text breaks off in the course of cAmmar’s discussion of Judaism. It resumes in the midst of his discussion of Islam. Hayek explains (p. 31, n. 2) that a page is missing. From this inforc mation it is impossible to determine if Ammar discussed any other religion.

45 Ibid. 46

Hayek, p. 33.

4^Hayek, pp. 136-37. ^Constantin Bacha, Les oeuvres arabes de Theodore Aboucara, eveque d ’Haran (Beyrouth: Imp. Alfawaid, 1904), pp. 71-75. 49 . . Georg Graf, Die Schriften des Jacobiten Habib Ibn Hidma Abu Ra'ita, (Louvain: L. Durbeck, 1951), CXXX, 131-32. 50 Louis Cheikho, Vingt traites theologiques (Beyrouth: Imprimerie Catholique, 1920), p. 144. ^Eugenio Massa, ed., Rogeri Baconis Moralis Philosophia (Turici: In Aedibus Thesouri Mundi, 1953), pp. 187-243. 52 Massa, p. 189. 53

In his Politics Aristotle speaks of the unlimited human desire for wealth, property, power, reputation, and so forth. Cf. Immanuelis Bekker, Aristoteles Graece II, (Berlin: G. Reimerum, 1831), II 1323a. AlFârâbi’s presentation of Aristotle’s thought on this subject would have been available to Roger Bacon in a translation by Gerard of Cremona. Cf. Angel G. Palencia, Al-Fãrãbi, Catalogo de las Ciências 2nd ed. (Madrid: Instituto Miguel Asin, 1953), p. 168. As Bacon says, Boethius also lists these objects of human desire in his De Consolatione Philosophiae, book 3. But Boethius himself mentions Epicurus, and not Aristotle, in connection with the list. Cf. Boethius, The Consolation o f Philosophy, Richard Green, trans. (New York: Humanities Press, 1962), p. 44.

54

Hayek, p. 31.

55Cf. EI2, I, 40-41. 56

57

58

cHayek, pp. 31 and 32. So far I have not found this interpretation of al-An am 6:109 attributed to Ibn alcAbbás in any Muslim source. Cf., for example, the stories referred to in Tor Andrae, Die Person Muhammeds in Lehre und Glauben seiner Gemeinde (Stockholm: P.A. Norstedt, 1917). Hayek, p. 33.

59 Hayek, p. 34. 60 61

2 R. Paret, “ Istihsan and Istislah,” EI , IV, 256. Hayek, p. 136.

^Hayek, p. 36. 63

Hayek, p. 19.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

198 86 64

Hayek, p. 40.

65 The phrase comes from Morris S. Seale, Muslim Theology;. a Study o f Origins with Reference to the Church Fathers (London: Luzac, 1964), p. 74. 66Cf. many attributions of this sort in Harry Austryn Wolfson, The Philosophy o f the Kalam (Cambridge, Mass.: Harvard University Press, 1976). Cf. Josef Van Ess, “The Logical Structure of Islamic Theology,” ed., G.E. von Grunebaum, Logic in Classical Islamic Culture (Wiesbaden: Harrassowitz, 1970), pp. 21-50; idem, “The Beginnings of Islamic Theology,” ed. J. Murdoch, and E. Sylla, The Cultural Context o f Medieval Learning (Boston: D. Reidel, 1975), pp. 89-111. 68

Cf. the quotation from Schmölders’ Essai sur les écoles philosophiques chez les arabes in Wolfson, p. 59.

^Richard MacDonough Frank, Beings and Their Attributes; the Teaching o f the Basrian School o f the Muctazila in the Classical Period (Albany, N.Y.: New York University Press, 1978), p. 5. 70

Cf. the references given in n. 10 above.

?1Cf. Dodge, I, 417, 419, 448. 72

73

c

-

:

c

Ibn Ahmad Abd al-Gabbar al-Hamdhani, Tathbit dala’il an-nubuwwah (Beirut: Dar al- Arabiyyah, 1966), I, 198. Al-Gahiz’ treatise is published in J. Finkel, Three Essays o f Abu COthman cAm r Ibn Bahr al-Jahiz (Cairo: Salafyah Press, 1926). CF. the partial French trans. by I.S. Allouche in Hesperis 26 (1939), 123-155; and partial English trans. by Finkel in Journal o f the American Oriental Society 47 (1927), 311-334.

For the works of Muslim controversialists cf. Erdmann Fritsch, Islam und Christentum in Mittelalter; Beitrage zur Geschichte der muslimischen Polemik gegen das Christentum in arabischer Sprache (Breslau: Muller & Seiffert, 1930). Cf. the extensive bibliography of Robert Caspar et al., “ Bibliographie du dialogue islamo-chretien,” Islamochristiana 1 (1975), 125-181.

74 , c Cf. Louis Gardet, “Quelques reflexions sur la place du ihn al-kalam dans les ‘sciences reiigiuses’ musulmanes,” ed. George Makdisi, Arabic and Islamic Studies in Honor o f Hamilton A.R . Gibb (Cambridge: University Press, 1965), pp. 258*69; Schlomo Pines, “ A Note on an Early Meaning of the Term Mutakallim, ” Israel Oriental Studies 1 (1971), 224-240. ^The use of the term Neoplatonism here means to indicate the melange of Platonic and Aristotelian doctrines that were adopted by the Christian thinkers and woven into the fabric of their theology. On this general subject cf. the survey by R. Arnou, “ Platonisme des pères,” DTC XII, 2, cols. 2258-392. Regarding anthropology, cf. the discussion in R.A. Norris, Manhood and Christ, a Study in the Christology o f Theodore o f Mopsuestia (Oxford: University Press, 1963). Cf. also the works cited in n. 23 above. 76 77

Frank, p. 10. r

-

?

Cf. Tor Andrae, also G.E. von Grunebaum, “ 1 a[/az,” EI „ III, 1018-20, and Richard C. Martin, “The Role of the Basrah Muctazilah in Formulating the Doctrine of the Apologetic Miracle,” Journal o f Near Eastern Studies 39 (\ 980V 17^-189

78

The classic example of a discussion illustrating the conflicting styles of thought is the debate between the Muslim mutakallim as-Sirâfi, and the Christian logician Mattã ibn Yünus. Cf. Muhsin Mahdi,

199

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY Com parative Religion in the A p o lo g etics

87

“ Language and Logic in Classical Islam,” ed. G.E. von Grunebaum, Logic in Islamic Culture (Wiesbaden: Harrassowitz, 1970), pp. 51-83. 79 v ^ c As-Sahrastani, Kitab al-milal wa l-nihal (Cairo: al-Matba ah al-Adabiyyah, 1899), I, 63. Cf. also Ibn anNadim’s story about Ibn Kulläb in Dodge, I, 448-49. 80 Cf. W. Montgomery Watt, The Formative Period o f Islamic Thought (Edinburgh: University Press, 1973), pp. 143-148. ^l Cf. Richard M. Frank, “The Neoplatonism of Gahm ibn Safwân,” Le Muséon 78 (1965), 395-424. 82 On this subject cf. A. Abel, “ La polemique damascenienne et son influence sur les origines de la theologie musulmane,” in L'élaboration de l'Islam (Paris: Presses Universitaires, 1961), pp. 61-85, and the earlier (1911) study by C.H. Becker, “ Christliche Polemik und islamische Dogmenbildung,” which appears in his Islamstudien (Hildersheim: rpt. Georg Olms, 1967), l, 432-449.

9 JEWISH POLEMICS AGAINST ISLAM AND CHRISTIANITY IN THE LIGHT OF JUDAEO-ARABIC TEXTS Sarah Stroumsa

I Jewish Polemics in the Arabic Middle Ages is a complex phenomenon. In order to understand its nature and function properly, one needs to address it, first o f all, as polemics. M oritz Steinschneider^ approach is a case in point. Steinschneider was well aware o f the existence o f a general body o f Judaeo-Arabic literature meriting separate treatment.1Yet, in his Polemische Literatur he avoided any such separate classification, treating books written by Jews against Muslims and Christians, by Christians and Muslims against Jews and also by Christians and Muslims against each other, as one body o f literature.2 O ne obvious reason for this procedure was the scarcity o f polemical books written in Arabic by Jews against Christianity or Islam.3 This was probably one o f Steinschneiders main incentives for adding to his book a vast appendix on all the Jewish literature against Islam, hoping thus to make up for what was lacking in Arabic.4 There is however another justification for Steinschneider^ unified treatment. The polemical literature forms one corpus not only as a literary genre used by all the participants in the polemics, but also in terms o f content. In the triangular marketplace where Muslims, Christians and Jews set up their doctrinal booths, the arguments brought up in the discussion— both oral and written— served as currency, quickly changing hands. The exact value o f a given coin can only be learned if one follows its course from hand to hand; the arguments may often best be understood in the larger context o f the inter-religious debate in Arabic, and not only in Arabic. Even the literary ingenuity o f authors who use the polemical genre can be best evaluated if we see it in the broader context ofJewish-Muslim-Christian debate. Judah Halevi, for example, portrays the King o f the Khazars as interested in Judaism only after he has both the Muslim and the Christian ^ e e M. Steinschneider, Die arabische Literatur derJuden (Frankfurt, 1902). 2Idem, Polemische und apologetische Literatur in arabischen Sprache (Hildesheim, 19662). 3See ibid., p. 4. AIbid., p. 244.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

202

242

SARAH STROUMSA

spokesmen rely on the Jews for the affirmation o f their claims.5 This literary setting enables Halevi to deal briefly with Christianity and Islam, before settling comfortably into his main concern, the vindication o f Judaism. This literary device, however, gains a historical perspective and credence when we realize, through non-Jewish polemical writings, that it was indeed common practice for both Muslims and Christians to turn to Jews for their witnesses against each other.6 In what follows I shall deal only with literary aspects o f the polemics in Arabic,7 and only with the Jewish side of these polemics. I shall, however, try to show that, even within these limits, for a proper understanding o f the Judaeo-Arabic polemical texts the wider inter-religious polemical literature is a necessary background. II The striking scarcity o f books in Judaeo-Arabic dedicated entirely to polemics with Islam has drawn the attention o f scholars, and has received various explanations. The reasons most commonly mentioned are the need for caution w hen writing in Arabic against Islam, and the absence o f the common scriptural ground that is necessary if such polemics are to thrive.8 Both these considerations were, no doubt, important factors in the shaping of polem ical literature. Maimonides's responsa provide us w ith explicit 5Kitab al-radd wa'l-datilfi'l-din al-dhalil, ed. D. H. Baneth-H. Ben-Shammai (Jerusalem, 1977; henceforward: Kuzari), p. 9:13-14. 6See, for instance, M. F. Nau, “Un colloque du Patriarche Jean avec l’émir des agaréens,” Journal Asiatique, II série, 5 -6 (1915), pp. 233, 251, 2 6 0 -2 6 1 ; K. Völlers, “Das Religionsgesprach von Jerusalem,” Zeitschrift für Kirchengeschichte XXIX (1908), p. 37; and see now G. B. Marcuzzo (ed. and trans.) Le dialogue d’Abraham de Tibériade avec ‘Abd alRahmãn al-Hãshimi àJérusalem vers 820 (Rome, 1986). On the limits o f the Christian reliance on Judaism see Sidney H. Griffith, “The Christian AdversusJudaeos Tradition and the ‘N ew Jew’, a Polemical Characterization o f Muslims in the Christian Apologies in Syriac and Arabic of the First Abbasid Century,” forthcoming. 7There is no denying that oral polemics between people of various religions and beliefs were a common practice, and the realia o f these discussions is reflected in our sources. See, for instance, al-Humaydï, Jadwa aUmuqtabas fi dhikr wulãt al-andalüs, ed. al-Tanjï (Cairo, 1352), pp. 101-102. Yet, wherever these disputations are used as the formal setting for a written polemical treatise, it becomes rather difficult to trust their authenticity; see, for instance, Vollers, Art. cit., p. 32; G. Vajda, “Un traité de polémique christiano-arabe contre les Juifs attribué à Abraham de Tibériade,” Bulletin de l'institut de recherche et d'histoire des textes 15(1967—1968), pp. 144—150. A more reliable documentation can sometimes be found in the occasional allusions to such encounters which are to be found in theological or philosophical writings. These allusions, made en passant, are less likely to be intentional inventions. (For examples o f such allusions see G. Vajda, “La finalité de la création de l’homme selon un théologien juif du IXe siècle,” Oriens XV (1962), p. 68; Sarah Stroumsa, Dâwüd ibn Marwân al-Muqammi$}s Twenty Chapters ( ‘Ishrün Maqãla) (Leiden, Brill, 1989), pp. 248—251; L. Cheikho, Vingt traités théologiques d'auteurs arabes chrétiens (Beirut, 1920), p. 68; Abu Hayyân al-Tawhîdï, al-imtâ‘ wa'1-mu'ãnasa, ed. Ahmad Amïn and Ahmad al-Zayn (Beirut), p. 218:12.)

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY JEWISH POLEMICS

203 243

testimonies that these considerations were indeed present in the minds o f Jewish writers. Even when writing in Hebrew, Maimonides cautiously limits his criticism o f Islam to vague allusions. The precise nature o f the Muslims’ “error and folly”, he says, “cannot be put in writing, because o f the vile apostates.”9 And Maimonides categorically forbids any attempt to explain the scriptures to Moslems, who do not accept the authenticity o f the Bible.10 Nevertheless, these considerations could perhaps be held sufficient to explain a total lack o f written polemics against Islam; but they lose some o f their force w hen we remeber that we do have a fair documentation o f outspoken polemics with Islam, incorporated in apologetical, exegetical and theological works.11 What is lacking is not polemics as such, but rather works whose single or main purpose is to attack Islam. Furthermore, as we shall see, the number o f books written by Jews in Arabic and directed against Christianity is not substantially larger than the number o f such books directed against Islam. The explanation for the lack o f separate polemical works is therefore probably one that applies to both Christianity and Islam. In all likelihood this explanation should be sought in the situation o f Judaism as a religion despised by both Christianity and Islam.12 In its confrontation with the two dominating religions, the polemical energy o f the Jews was mostly used for the purpose o f apology, that is to say, the strengthening of their own

8M. Steinschneider, Jewish Literature from the Eighth to the Eighteenth Century (New York, 19702), pp. 129-130; M. Perlmann, “The Medieval Polemics between Islam and Judaism,” in S. D. Goitein, ed., Religion in a Religious Age (Los Angeles, 1974), p. 122. 9“Mi-pnei posh‘ei ve-rish(ei Ysra’el”, see R. Moses b. Maimon, Responsa, ed. J. Blau, vol. II (Jerusalem, 1960), p. 726. On this designation o f the apostates, see S. D. Goitein, A Mediterranean Society, Vol. II (Berkeley, Los Angeles and London, 1971), p. 300. w Ibid., Vol. I (Jerusalem, 1958), pp. 284-285. u To the sources listed by Perlmann, ibid., pp. 137-138, one can add Judah Halevi’s Kuzari, pp. 6—9, 162—163; and Maimonides, Epistle to Yemen, ed. A. S. Halkin (New York, 1952). Oxford Ms. heb.d.62 fol. 8 is catalogued as “a fragment o f a sermon in Arabic on the Sabbath” [A. Neubauer and A. E. Cowley, Catalogue of the Hebrew Manuscripts in the Bodleian Library (Oxford, 1906), vol. II, p. 317, n* 2850, 3a]. Its actual subject, however, is the abrogation o f the law, and specific references to Islamicfiqh leave no doubt that this fragment was also part o f an explicit controversy with Islam. Another Geniza fragment which should be considered in this context is Cambridge T-S. N.S. 314.22: refuting the claims o f the upholders o f “the abrogation o f the law and (those who claim that) it was falsified” (naskh al-shar*wa-tabdiluhu), this fragment also mentions “their book and its tradition” (kitabuhum wa-khabaruhu), probably a reference to the Qur’an. 12The feeling that, although they shared with the Christians the official status o f dhimma, Jews were, in fact, much inferior to the Christians, was shared by both Christians and Muslims. See Qur’an 5:82; al-Jãhiz, al-Radd ‘alã al-nasãrã in Thalâth rasâ'ïl, ed. J. Finkei (Cairo, 1926), p. 17; Nau, art. cit., p. 37; Vajda, “Polémique,” p. 147. The Jews themselves had no illusions about their inferior social status, as the Kuzari's full title makes clear (see above, note 5).

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

204 244

SARAH STROUMSA

theological position. Only in a few cases did Jews writing in Arabic turn to a full scale attack on Islam, or, for that matter, on Christianity. O f these rare Judaeo-Arabic books which were, nevertheless, written against Islam, none is known to have survived. The Qaraite al-Qirqisãni (first half o f the tenth century) tells us that he wrote such a book.13 From the summary of its main points given in Qirqisãms Kitãb al-anwãr wã'1-marãqib, we learn that it dealt extensively with such matters as the Muslim allegation that the Jews had falsified the scriptures (tabdil; tahrif), Muhammad’s prophethood and the Muslim dogma of inimitability of the Qur’an, and the abrogation of the law (naskh). The abrogation o f the law is also the topic (and the title) o f a book written by Samuel b. Hofnl (d. 1013). From Moses Ibn Ezra's account we learn that this book also argued against the Muslim claim o f the miraculous nature o f the Qur’an (i‘jã z).14 And Ibn Balaam’s remarks show that Ibn Hofhl's Naskh al-Shar‘ contained digressions to exegetical discussions.15 Most o f the book is lost. The few existing fragments suggest that this book dealt with multiple aspects o f the problem o f naskh. These aspects included, obviously, polemics: against Karaites, against Christians, and probably against Islam.16 But it could hardly be seen as directed mainly against Islam. The most interesting o f the lost Judaeo-Arabic works criticising the Qur’an is usually considered to have been the one attributed to Samuel Hanagid (d. 1056), the surviving fragments o f which are to be found in a short treatise by Ibn Hazm (d. 1064), entitled al-Radd ‘alã Ibn al-Naghrila,17 The defiant style o f this Qur’an criticism seems, however, inconsistent with Samuel’s usual prudence and has caused scholars considerable unease.18 This unease has been expressed in various ways; from marvel at Sam uel’s exceptional boldness to cautious persistence in calling the book “the alleged

13Kitâb al-anwar wa’l-mamqib, Code of Karaite Law, ed. L. Nemoy (New York, 1939—1943), pp. 292:8, 301:5-6. 14 Kitãb al-muhãdara wa’l-mudhâkara, ed. A. S. Halkin (Jerusalem, 1975), pp. 36:40-38:3, and see Steinschneider, Polemische Literatur, pp. 102—103. 15See Harkavy, Zikhron ha-Rav ha-Ga’oti Shmu’cl ben Hofni u-Scfarav, p. 14, n. 20; p. 41, n. 114. 16I am indebted to Mr. David Sklare for the reference to these fragments, which he is preparing for publication. 17Al-Radd ‘alã Ibn al-Naghnla al-yahüdi wa-rasä’il ukhrã lïibn Hazm, ed. Ihsãn ‘Abbãs (Cairo, 1960). For the correct reading o f ha-Nagld’s Arabic name see S. Stem, (iLetoledõt R. Shmu’cl Hanagid,” Zion 15 (1950), p. 135, n. 2. 18See David Wasserstein, The Rise and Fall of the Party-Kings; Politics and Society in Islamic Spain, Î 002—1086 (Princeton, 1985), p. 201. 19E. Ashtor, Koröt ha-yehüdim bi-sefarad ha-muslemit (Jerusalem, 1966), vol. II, pp. 74-75.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY JEWISH POLEMICS

205 245

Qur’an Criticism of Samuel ha-Nagïd.”20 Others sought to solve the problem by attributing the book to Joseph, Samuels more reckless son.21 But in fact this book was not written by Ibn al-Naghrila, Senior or Junior, nor was it written by a Jew at all. The book quoted by Ibn Hazm is actually the lost Kitäb al-dãmigh of the notorious ninth-century heresiarch, Ibn al-Rawandi. A detailed demonstration o f this claim would carry us beyond the limits o f this paper: suffice it to say that a refutation of the Kitãb al-dãmigh, written by Abu ‘All al-Jubba’ï (d. 915), was still available in the eleventh century. Excerpts from it are quoted by both ‘Abd al-Jabbãr (d. 1025)“" and Ibn al-Jawzi (d. 1200),23 and these excerpts enable us to establish that what Ibn Hazm quotes is indeed the Kitãb al-dãmigh, and that it was Abu ‘Alls refuation o f the Kitãb al-dãmigh which served as Ibn Hazm’s immediate source. It is appropriate to ask how this book came to be attributed to Samuel Hanagid: One could, of course, argue that, following the model of the Kitãb al-dãmigh, Ibn al-Naghrila wrote his own refutation o f the Qur’an. But in view o f his above-mentioned prudence, this does not seem a very likely possibility. Moreover, the construction o f Ibn Hazm’s work suggests his awareness that the real author of the refutation of the Qur’an is not Ibn alNaghrila, against whom the epistle is directed. If this is indeed the case, we would have to conclude that Ibn al-Naghrila/Hanagid never wrote a refutation o f the Qur’an 24 For our own purposes the identification o f the real author o f this refutation o f the Qur’an is somewhat unfortunate, since it reduces by onethird our already short list of anti-Muslim books in Judaeo-Arabic. But the new identification also has its advantage, for it enables us to re-define the borders o f Jew ish polem ics w ith Islam. These polem ics focused on Muhammad’s prophethood and on the Qur’an in a general way. Ibn alRâwandï’s book, in contrast, included juxtapositions o f particular Quranic verses in the attempt to show that they contradicted one another. It also included derisive observations about minor details o f the Qur’an. Such

20Perlmann, “Medieval Polemics,” p. 109, and see also idem, “Eleventh-Century Andalusian Authors on the Jews o f Granada,” PAAJR 18 (1948—1949), pp. 280 ff. 21Ihsãn Abbãs in his introduction to al-Radd ‘alã Ibn al-Naghrila, pp. 7 ff.; Paul B. Fenton, “Jewish Attitudes to Islam: Israel heeds Ishmael,” Jerusalem Quarterly 29 (1983), p. 91; this solution was apparently adopted already by R. Dozy, see his Histoire des Musulmans d’Espagne (revised by E. Lévi-Provençal, Leiden, 1932), vol. Ill, p. 70. 22Al-Mughnifi abwãb al-tawlnd wa’l-‘adl, vol. XVI (Cairo, 1960), pp. 389:10-394:8. 23Al-Muntazam fi ta’rïkh al-mulük wa’l-umam (Hyderbad, 1357), pp. 99-105. 24For a detailed analysis of the evidence see S. Stroumsa, “From Muslim Heresy to JewishMuslim Polemics: Ibn al-Râwandî’s Kitãb al-Dãmigh,”Journal of the American Oriental Society 107 (1987), pp. 767-772.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

206 246

SARAH STROUMSA

criticism would have been written only by an insider; it was beyond the scope o f Jewish polemics against Islam, not so much because Jews would not dàre to write it, but rather, because it was not something that would naturally occur to them as a polemical possibility. Ill As to Jewish polemics against Christianity, although not much more was written, more has survived, and in particular our documentation o f the polemics from its beginnings is much better. These beginnings are in the ninth century, with the work o f Dâ’üd Ibn Marwãn al-Muqammis.25 His arguments against Christianity are to be found mainly in three o f his books. The first is al-Radd ‘alã al-nasãrã min tariq al-qiyãs, o f w hich we are fortunate to have a few Geniza fragments.^6 This is what might be called polemics in low style. It is a collection o f fifty questions and answers, aimed at exposing the absurdity o f Christianity. Some o f these questions refer to minor details o f the Gospels and o f the Christian dogma and practice, in a derisive attempt to show their inner contradictions. The exposition is in a defiant, scoffing tone, which resembles that o f Ibn al-Râwandï’s criticism o f the Qur’an. N ow this is the style we have just referred to as criticism o f an insider, and which we regarded as beyond the scope o f Jewish polemics against Islam: why then was it not beyond the scope o f Jewish polemics against Christianity? The explanation is to be found largely, if not entirely, in the person of al-Muqammis himself. He had been a Christian, and, therefore, after his return to Judaism, was the perfect person to write this sort o f virulent attack against his former religion.27 His second book against Christianity, the Kitãb al-Darä’a, adopts the same aggressive stance, but in a much more sober tone. It reviews the history o f Christianity and tries to demonstrate that Christianity was an invention o f Paul and o f Constantine, that it deviated from the path indicated by Jesus, and that it is full o f contradictions. From the quotations o f the Kitãb al-àarã*a in Qirqisânï s Anwäfö we can get a fair picture o f what the book was like, and

25On al-Muqammis’s life and work see Sarah Stroumsa (ed. and trans.), D ä’üd ibn Marwãn al-Muqammis, Twenty Chapters ( ‘Ishrün Maqãla) (Leiden: Brill, 1989). 26One fragment (T-S. 8. Ka4^ was published by H. Hirschfeld, “The Arabic Portion o f the Cairo Geniza at Cambridge, ”JQR, O.S. 15 (1903), pp. 167-181, 677-687. The Cambridge Geniza collection contains another portion o f the same manuscript (T-S N.S. 91.26), which is now being prepared for publication by Dr. P. Fenton. 27On al-Muqammis’s Christian period see Qirqisãní, Anwãr, p. 44:10-16. 2SAnwãr, p. 47:15-16. See L. N em oy, ‘A l-Q irqisânï’s Account o f Jewish Sects and Christianity,” HUCA 7 (1930), pp. 317-397. Al-Muqammis’s view o f the early history of the church, as well as some o f his polemical arguments, are ultimately derived from JewishChristian criticism o f Christianity (see S. Pines, “The Jewish Christians o f the Early Centuries o f Christianity According to a N ew Source,” PLASH II (Jerusalem, 1966), p. 47, n. 176). This, again, can certainly be called “inner criticism.”

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY JEWISH POLEMICS

207 247

we also have enough evidence to be able to conclude that it had a major influence on the development o f Jewish polemics against Christianity. More than a century ago, Leon Schlossberg has published an antiChristian treatise in Judaeo-Arabic, entitled “The Account o f the Controversy with the Bishop.”29 This publication is the (rather poorly done) reproduction o f a Paris manuscript, which Schlossberg believed to be a unique copy o f this work. But the Geniza collections contain quite a few fragments which clearly belong to this book.30 The book was apparently quite popular in the Middle Ages: parts o f it were incorporated into Ibn Kammùnah s comparative study o f the three monotheistic religions,31 and as early as the eleventh century it was translated into Hebrew.32 Schlossberg, who followed indications given in the treatise itself, believed the Qissa to be the Arabic version o f a sixth century Syriac treatise. His assessment was not accepted, and the book is generally believed to have been written in Arabic. Its date and precise origin, however, remain uncertain. The author o f the Qissa accuses the Christians o f corrupting the laws of Judaism, laws which were taught by the prophets and by which Jesus had lived. Like al-Muqammis, he endeavours to expose the disagreement and contradictions within Christianity, and in particular, what he regards as the polytheistic implications o f the Christian doctrines. Although I would hesitate to actually identify this book with al-Muqammis’s Kitãb al-Darâ’a, we can at least be certain that the Qissa contains many arguments, and perhaps whole passages, which are ultimately derived from this book. Perhaps we can adopt here the terminology o f art-historians, and regard the Qissa as the product o f “the school of the Kitãb al-Darã’a”. O f al-Muqammis s third book, the ‘Ishrün maqãla, the greater part has survived, and it enables us to see a third level o f polemics, namely, the integration o f the polemical influence into the theology o f Judaism. AlMuqammis’s theology is laced with attacks on Christianity, yet in his exposition o f his own views, he draws extensively from his Christian education. Fortunately for us, his borrowing is often rather wooden, so that it is easy to detect the Christian identity o f his source.33

29 Qissat Mujãdalat al-Usquf, ed. L. Schlossberg (Vienne, 1880), to which Schlossberg added a French translation entitled Controverse d'un Evêque (Paris, 1988). 3(lOn the various fragments, and on the way they relate to each other and to the text published by Schlossberg, see S. Stroumsa, “ Qissat Mujãdalat al-Usquf: A Case Study in Polemical Literature”, in J. Blau and S. C. Reif, eds., Geniza Research After Ninety Years: The Case ofJudaeo-Arabic (Cambridge, 1992), pp. 155-159. 31Sa‘d b. Mansür Ibn Kammüna, Tankth al-abhãth f ’l-milal al-thalãth, Examination of the Inquiries into the Three Faiths, ed. M. Perlmann (Berkeley-Los Angeles, 1967), pp. 57fF. 32On the Hebrew version, see Daniel Lasker, “Qissat Mujãdalat al-Usquf and Nestor HaKomer: The Earliest Arabic and Hebrew Jewish Anti-Christian Polemics”, Geniza Research After Ninety Years (above, note 30), pp. 112-118. 33See Ishrün Maqãla (above, note 7), pp. 27-33.

208

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY 248

SARAH STROUMSA IV

Later Jewish theologians were also influenced by the polemical encounter, but, more skillful than al-M uqammis, they took care to adapt their borrowings to the needs o f Jewish theology. I would like to give one example o f how this “polemical influence” was incorporated in the Arabic writings of various Jewish writers, but, before doing so, a few preliminary observations are in order. Each o f the three religions which participated in the debate had its traditional weak areas, particularly open to attack, which served as targets for the attack o f the two other participants. Thus, the anti-Jewish polemics o f both Christians and Muslims concentrated on a variety o f features, which they saw as supporting their accusation of tabdil (falsification o f the scriptures by the Jews)34 and on the abrogation of the law (Naskh): in the anti-Christian polemics of both Jews and Muslims, the main issue was the Unity o f God; and when Christians and Jews polemicised against Islam their topic was prophecy. In their refutation o f Islamic prophetology, Christians developed a set o f “negative attributes” o f the true religion;35 the true religion is the one that spreads without the aid o f military force, o f ties o f kin or nationality, o f the promise o f worldly gains or of the appeal to human fancy. As we have already said, theological arguments were passed from one religion to the other, and were adapted by each religion to suit its own purposes. It is usually impossible to determine where a particular argument started. But the cluster o f “negative attributes” o f the true religion represents a rare example o f a polemical theme the exact origin o f which can be established: It was conceived o f by Christians for the purpose of refuting Islam, and it was also used occasionally to discredit Judism. We may briefly follow the influence o f this set o f “negative attributes” in the works of four Jewish writers: (a) When al-Muqammis sets out to prove the validity o f Judaism, he bases his claim on a Christian list o f this kind. He quotes the “negative attributes” from an unidentified Christian source, stopping after each attribute to show that it applies to Judaism. Here as elsewhere his adaptation lacks flexibility. 34Maimonides in his responsa presents the accusation o f tabdil as specifically Muslim, and emphasizes the common scriptural ground which Judaism has with Christianity ( Teshüböt haRambam, ed. J. Blau, Jerusalem 1958, vol. I, pp. 284—285). It should be noted, however, that in the heat of the debate with Judaism, Christians did not disdain the use o f this accusation, regardless o f the implication it might have for their own religion. See, for example, Qirqisânï’s self-satisfied description of how he refuted such a Christian attack, Anwãr, p. 220:8—17; and M. Perlmann, “Ibn al-Mahrüma, a Christian opponent o f Ibn Kammüna,” in H . A. Wolfson Jubilee Volume (Jerusalem, 1965), vol. II, pp. 641—665. 35See Sidney H. Griffith, “Comparative Religion in the Apologetics o f the First Christian Arabic Theologians,” Proceedings of the Fourth Annual Patristic, Medieval and Renaissance Conference (Villanova, Pa.), pp. 63—87; S. Stroumsa, “The Signs o f Prophecy: The Emergence and Early Development o f a Theological Theme in Arabic Theological Literature,” Harvard Theological Review 78 (1985), pp. 101—114.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY JEWISH POLEMICS

209 249

Thus we find him implying, for instance, that the true religion is one not confined to a single nation or to the speakers o f a single language. For alMuqammis, this is a description which fits Judaism, since “this Torah exists in Syriac and Persian and Greek and Arabic/ He was obviously insensitive to the fact that this particular argument is one that can be far better employed in favor o f Christianity and against Judaism. (b) Later generations were more wary or, rather, better trained. Maimonides, in the Epistle to Yemen, also refers to translations o f the Bible, enumerating Syriac, Greek, Persian and Latin.37 Unlike al-Muqammis, however, Maimonides is careful to specify the exact, narrow implication o f the existence o f these translations: They do not constitute proof o f the veracity o f the Bible in general, but they do serve to disprove the Muslim accusation that the Jews eliminated the references to Muhammad from the scriptures, since they were all propagated long before the rise o f Islam. In other words, Maimonides uses an argument fashioned by Christians, but he re-tailors it to fit his own requirements. (c) Another use o f the originally Christian description o f true religion can be discerned in Ibn Ezra’s Kitãb al-Muhãdara.38 After a brief reference to the Muslim dogma o f i ‘jã z al-qur’an, Ibn Ezra tells us that the rhetorical excellence o f the Arabs was the cause o f their success in conquering so many civilizations. N ow the seductive linguistic beauty o f a scripture figured in the Christian list among the wrong reasons for the religions success.The Kitãb alMuhãdara addressed itself to a public to whom, no doubt, this argument was familiar. Thus when Ibn Ezra mentions Arabic eloquence as a cause o f the success o f Islam, he also reveals to his reader his actual thought about the religious relevance o f Arabic rhetorical excellence, including the inimitable beauty o f the Qur’an.39 (d) The last example comes from Judah Halevi. In the first part o f the Kuzari the King o f the Khazars asks the Rabbi to tell him about the spread o f Judaism. He explains his request by saying that every religion begins necessarily with individuals who fight to establish their views, or a king who forces his views on the multitude.4 The King’s question reveals Halevi’s awareness o f an originally Christian argument against Islam. The King’s question is an invitation to the Rabbi to say— as the Christians would o f their own religion— that the spread o f Judaism was not achieved by military force, or by any other o f the wrong

36 ‘Ishrün Maqãla, (above, note 25), pp. 268-269. 37Epistle to Yemen, p. 38. 38Muhãdara, p. 38. 39Ibn Ezra gives his reader another carefully camouflaged indication in the same direction. He claims that the prophets foretold the eloquence o f the Arabs, and he supports his claim by quoting those biblical verses usually interpreted by Jews as referring to Islam (Dan. 8:23, 7:20; Isa. 42:11), none o f which is very flattering (see Muhãdara, p. 38). 40Kuzari, p. 21.

MUSLIMS AND OTHERS IN EARLY ISLAMIC SOCIETY

210

250

SARAH STROUMSA

means. The Rabbi's reaction to this invitation, however, is outright rejection: Only religions invented by human beings spread gradually. The divinely sponsored religion does not “develop” in any way. Rather, it is created in an instant, as is the world, by the sheer will o f God.4 The Rabbi’s answer suggests not only that Halevi was familiar with the set of negative characteristics o f the true religion, but also that he was aware o f its Christian origin. He knew that this set o f arguments could be— and indeed was— also used to discredit Judaism. For this reason, then, Halevi, through the Rabbi’s answer, defines the whole question o f how a religibn propagates as totally irrelevant. This method o f adaptation to the changing polemical milieu is very common for all the participants in the religious debate. Jews, like others, were alert to the development o f new arguments by their opponents. When such an argument was perceived as potentially or actually directed against Judaism, it was not always considered wise to pick up the challenge; sometimes it was deemed safest to slip away. The elegant way o f doing this consisted in reshaping or slightly rewording the presentation o f Judaism, so as to render the opponent’s argument inapplicable and irrelevant. It is largely in this way that the polemical encounter with Christianity and Islam contributed to the shaping and development o f Jewish theology.4

41M . , pp. 21-22. 42That is to say, that Jews would “consciously rectify their position in order to avoid being molested again,” as did also Muslims and Christians; compare J. Van Ess, “The Beginnings o f Islamic Theology,” inj. E. Murdoch and E. D. Sylla, eds., The Cultural Content of Medieval Learning (Boston, 1975), p. 99. 43I wish to thank Shira Wolowsky and Frank Stuart for reading drafts o f this paper and for their useful suggestions.

10 MUSLIM STUDIES OF OTHER RELIGIONS: THE MEDIEVAL PERIOD* Jacques Waardenburg

Not only among the ancient Greeks with autors like Herodotus and Plutarch but also in medieval Islamic civilization there existed an interest in the reli­ gions of other civilizations and in religious history as such. A contrast with medieval Europe may be made, in that it was, with a few exceptions, only at the time of the Renaissance and of the voyages of discovery that people showed an interest in the mythology and religions of the Ancients, and the beliefs and religious practices of the newly discovered countries and re­ gions. Thus, if we accept that medieval Muslim scholars manifested their interest in alien religions, what can we say about the “study of religions” in medieval Islamic civilization? There were several difficulties and limitations with regard to such a study. First of all there were technical difficulties. There was very little knowledge of languages other than Arabic and Persian. The diffusion of foreign texts in manuscript form, that is to say texts from outside the lands of Islam, was exceedingly difficult. There was a lack of knowledge of the history of civilizations before and outside the world of Islam. The way in which non-Muslims were perceived empirically depended on the informa­ tion available and also upon the possibilities of direct contact between Mus­ lims and non-Muslims, either inside or outside Muslim territory, and Muslim curiosity about non-Muslim beliefs, practices and ways of life. There were also limitations of a different sort. They arose out of the views on life and the world which were predominant at the time. Such views were both “medieval” in a broad sense and “Islamic”, that is to say nourished by the Qur’ân and the Sunna and further elaborated intellectually. The notes o f this paper have been restricted to a minimum. For further references see Jacques Waardenburg,