The Quran in its Historical Context - Reynolds et al

THE QUR’AN IN ITS HISTORICAL CONTEXT The Qur’an in Its Historical Context is a remarkable work of primary source scholarship on the Qur’an. While most studies address the Qur’an through the retrospective lens of later Islamic commentaries, the present work presents a contextual perspective. The reader is challenged herein to consider, first, the great debates over the meaning of the Qur’an and, second, the new research that claims to present a definitive solution to those debates. In Part 1, the authors consider, and advance, theories for a new understanding of the Qur’an’s interpretation. The question of Christoph Luxenberg’s Syro-Aramaic reading is debated, as is the importance of newly discovered early Arabic inscriptions. In Part 2, the authors place the Qur’an within the Late Antique religious milieu, demonstrating its conversation with Jewish and Christian literature. In Part 3, the authors consider the Islamic tradition of Qur’an interpretation, and ask how scientific research relates to religious tradition. Collectively the essays herein present a new approach to the study of the Qur’an. This approach will allow scholars to shed new light on the Qur’anic passages that have been shrouded in mystery and debate. It will also illuminate the Qur’an’s relationship to Judaism and Christianity, thereby demonstrating the Qur’an’s place in a shared Jewish–Christian–Islamic tradition. As this collection of distinguished authors represents a distinct sub-field within Qur’anic Studies, students and specialists will welcome this volume in order to get to know the state-of-the-art methods within this specific sphere of scholarship. Gabriel Said Reynolds is Assistant Professor of Islamic Studies and Theology at the University of Notre Dame, where he specializes in Qur’anic Studies. ROUTLEDGE STUDIES IN THE QUR’AN Series Editor: Andrew Rippin University of Victoria, Canada In its examination of critical issues in the scholarly study of the Qur’an and its commentaries, this series targets the disciplines of archaeology, history, textual history, anthropology, theology and literary criticism. The contemporary relevance of the Qur’an in the Muslim world, its role in politics and in legal debates are also dealt with, as are debates surrounding Qur’anic studies in the Muslim world. LITERARY STRUCTURES OF RELIGIOUS MEANING IN THE QUR’AN Edited by Issa J. Boullata LOGIC, RHETORIC AND LEGAL REASONING IN THE QUR’AN God’s arguments Rosalind Ward Gwynne THE DEVELOPMENT OF EXEGESIS IN EARLY ISLAM The authenticity of Muslim literature from the Formative Period Herbert Berg TEXTUAL RELATIONS IN THE QUR’AN Relevance, coherence and structure Salwa M. El-Awa BIBLICAL PROPHETS IN THE QUR’AN AND MUSLIM LITERATURE Robert Tottoli SÁFI COMMENTARIES ON THE QUR’AN IN CLASSICAL ISLAM Kristin Zahra Sands MOSES IN THE QUR’AN AND ISLAMIC EXEGESIS Brannon M. Wheeler THE QUR’AN IN ITS HISTORICAL CONTEXT Gabriel Said Reynolds THE QUR’AN IN ITS HISTORICAL CONTEXT Edited by Gabriel Said Reynolds First published 2008 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2007. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Routledge is an imprint of the Taylor & Francis Group, an informa business © 2008 Gabriel Said Reynolds, selection and editorial matter; the contributors, their own chapters All rights reserved. No part of this book may be reprinted or reproduced or utilized 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. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-93960-3 Master e-book ISBN ISBN10: 0–415–42899–8 (hbk) ISBN10: 0–203–93960–3 (ebk) ISBN13: 978–0–415–42899–6 (hbk) ISBN13: 978–0–203–93960–4 (ebk) CONTENTS List of images Notes on contributors Foreword by Daniel A. Madigan List of abbreviations Map: locations cited in the present volume Introduction: Qur’anic studies and its controversies vii viii xi xiv xv 1 GABRIEL SAID REYNOLDS PART 1 Linguistic and historical evidence 27 1 The Qur’an in recent scholarship: challenges and desiderata 29 FRED M. DONNER 2 Epigraphy and the linguistic background to the Qur’an 51 ROBERT HOYLAND 3 Recent research on the construction of the Qur’an 70 GERHARD BÖWERING 4 Reconsidering the authorship of the Qur’an: is the Qur’an partly the fruit of a progressive and collective work? 88 CLAUDE GILLIOT 5 Christian lore and the Arabic Qur’an: the “Companions of the Cave” in Surat al-Kahf and in Syriac Christian tradition SIDNEY GRIFFITH v 109 CONTENTS PART 2 The religious context of the late antique Near East 6 The theological Christian influence on the Qur’an: a reflection 139 141 SAMIR KHALIL SAMIR 7 Mary in the Qur’an: a reexamination of her presentation 163 SULEIMAN A. MOURAD 8 The Alexander Legend in the Qur’an 18:83–102 175 KEVIN VAN BLADEL 9 Beyond single words: Ma’ida – Shaytan – jibt and taghut. Mechanisms of transmission into the Ethiopic (GY‘Yz) Bible and the Qur’anic text 204 MANFRED KROPP 10 Nascent Islam in the seventh century Syriac sources 217 ABDUL-MASSIH SAADI PART 3 Critical study of the Qur’an and the Muslim exegetical tradition 223 11 Notes on medieval and modern emendations of the Qur’an 225 DEVIN J. STEWART 12 Syriac in the Qur’an: classical Muslim theories 249 ANDREW RIPPIN Bibliography Index of Biblical verses Index of Qur’anic verses Index of people, places and subjects vi 262 283 285 289 IMAGES 1 2 3 4 5 6 7 8 9 10 11 Epitaph of Imru’ al-Qays, Nemara (S. Syria), 328 CE Dedication to Obodas, ‘Ayn Abada/En Avdat (S. Palestine), ca. second century CE Funerary text, Hegra (NW Saudi Arabia), 267 CE Building text, Zebed (N. Syria), ca. 512 CE Graffito, Jabal Usays (S. Syria), 529 CE Building text, Harran (S. Syria), 569 CE Funerary text, Nebo (Jordan), mid-sixth century CE Graffiti, Umm Judhayidh (NW Arabia), ca. fourth to fifth century CE Graffiti, Umm Judhayidh (NW Arabia), ca. fourth to fifth century CE Graffito, Qa‘ al-Mu‘tadil (NW Arabia), AH 24 Poetry quotation, Mecca (Arabia), AH 98 vii 53 54 54 55 55 56 56 61 62 62 64 CONTRIBUTORS Kevin van Bladel is Assistant Professor of Classics at the University of Southern California. His publications include “Heavenly cords and prophetic authority in the Qur’an and its Late Antique context,” Bulletin of the School of Oriental and African Studies 69 (2007), 223–47 and “The Iranian characteristics and forged Greek attributions in the Arabic Sirr al-asrar (Secret of Secrets),” Mélanges de l’Université Saint-Joseph 57 (2004), 151–72. Gerhard Böwering is Professor of Islamic Studies at Yale University (USA). His publications include: Mystical Vision of Existence in Classical Islam (Berlin: de Gruyter, 1980) and Sulami’s Minor Qur’an Commentary (Beirut: Dar al-Mashriq, 1995; 1997). Fred M. Donner is Professor of Near Eastern History in the Oriental Institute and the Department of Near Eastern Languages & Civilizations at the University of Chicago (USA). His publications include: Narratives of Islamic Origins: The Beginnings of Islamic Historical Writing (Princeton: Darwin Press, 1997) and “From believers to Muslims: Communal self-identity in the early Islamic community,” Al-Abhath 50–1 (2002–3), 9–53. Claude Gilliot is Professor of Arabic and Islamic studies at the University of Provence, Aix-en-Provence (France). His publications include: Exégèse, langue, et théologie en Islam. (Paris: Vrin, 1990) and “Le Coran, fruit d’un travail collectif?” in D. De Smet, G. de Callatay and J.M.F. Van Reeth (eds), Al-Kitab. La sacralité du texte dans le monde de l’Islam, Acta Orientalia Belgica, 2004, 185–223. Sidney Griffith is Ordinary Professor in the Department of Semitic and Egyptian Languages and Literatures at the Catholic University of America (USA). His publications include two collections of his studies: Arabic Christianity in the Monasteries of Ninth-Century Palestine (Aldershot: Ashgate/Variorum, 1992) and The Beginnings of Christian Theology in Arabic (Burlington, VT: Ashgate/ Variorum, 2002). Robert Hoyland is Professor of Arabic and Middle East Studies at the University of Saint Andrews (Scotland, UK). His publications include: Seeing Islam as viii CONTRIBUTORS Others Saw it (Princeton: Darwin, 1997) and Arabia and the Arabs (London: Routledge, 2001). Manfred Kropp is Professor of Semitic and Islamic Studies at Johannes Gutenberg-University, Mainz (Germany). His publications include: Die Geschichte der “reinen Araber” vom Stamme Qahtan. Aus dem Kitab Namwat at-tarab fi ta’riw æahiliyyat al-‘Arab des Ibn Sa‘id al-Ma∫ribi (2nd edition: Frankfurt: Lang, 1982) and Der siegreiche Feldzug des Königs ‘Amda-Seyon gegen die Muslime in Adal. CSCO 538–9 (Louvain: Peeters, 1988). Daniel A. Madigan is Director of the Institute for the Study of Religions and Cultures at the Gregorian University, Rome. His publications include The Qur’an’s Self-Image: Writing and Authority in Islam’s Scripture (Princeton: Princeton University Press, 2001), and “The Limits of Self-referentiality in the Qur’an,” in Stefan Wild (ed.), Self-referentiality in the Qur’ân; Diskurse der Arabistik 11 (Wiesdbaden: Otto Harrassowitz, 2006), 59–70. Suleiman A. Mourad is Assistant Professor of Religion at Smith College (USA). His publications include: Early Islam between Myth and History: al-Hasan al-Basri (d. 110/728) and the Formation of His Legacy in Classical Islamic Scholarship (Leiden: Brill, 2005) and “From Hellenism to Christianity and Islam: The origin of the palm tree story concerning Mary and Jesus in the Gospel of Pseudo-Matthew and the Qur’an,” Oriens Christianus 86 (2002), 206–16. Gabriel Said Reynolds is Assistant Professor of Islamic Studies and Theology at the University of Notre Dame (USA). His publications include: A Muslim Theologian in the Sectarian Milieu: ‘Abd al-Jabbar and the Critique of Christian Origins (Leiden: Brill, 2004) and “A reflection on two Qur’anic words (Iblis and Judi) with attention to the theories of A. Mingana,” JAOS 124 (2004) 4, 1–15. Andrew Rippin is Professor of Islamic History and Dean of the Faculty of Humanities, University of Victoria (Canada). He is editor of The Blackwell Companion to the Qur’an (Malden, MA: Blackwell, 2006). His articles on the Qur’an and tafsir have been gathered in The Qur’an and Its Interpretative Tradition (Aldershot: Ashgate/Variorum, 2001). Abdul-Massih Saadi is Assistant Professional Specialist at the University of Notre Dame (USA). His publications include: “The letter of John of Sedreh: A new perspective on nascent Islam,” Karmo 1, 1998, 1, 18–31 (Arabic and Syriac); 1, 1999, 2, 46–64 (English). Samir Khalil Samir is Director of the Centre de Documentation et de Recherches Arabes Chrétiennes (CEDRAC) at Université St. Joseph (Beirut, Lebanon) and Professor at the Pontificio Istituto Orientale (Rome, Italy). His publications include: One Hundred Questions on Islam (San Francisco: Ignatius Press, 2007; also available in Italian, Spanish, French and Polish) and Rôle culturel des chrétiens dans le monde arabe, Cahiers de l’Orient chrétien 1 (Beirut: CEDRAC, 2005). ix CONTRIBUTORS Devin J. Stewart is Associate Professor of Arabic and Islamic Studies at Emory University (USA). His publications include: Islamic Legal Orthodoxy: Twelver Shiite Responses to the Sunni Legal System (Salt Lake City: Utah University Press, 1998) and “Muhammad b. Jarir al-Tabari’s al-Bayan ‘an Usul al-Ahkam and the genre of Usul al-Fiqh in ninth century Baghdad,” in J.E. Montgomery (ed.), Abbasid Studies (Leuven: Peeters, 2004), 321–49. x FOREWORD Three decades ago, when a number of revisionist approaches to the Qur’an and early Islamic history were proposed within the space of a few years, the questions at issue and the mutually incompatible answers proposed seemed of interest almost exclusively to scholars. Yet the publication of Christoph Luxenberg’s Die syro-aramäische Lesart des Koran excited a great deal of popular and journalistic attention, without for the most part finding a serious hearing among scholars. The present collection and the rich and varied conference from which it sprang, while generally unconvinced by Luxenberg’s work, takes it as a jumping-off point to try and re-focus the complex, even chaotic, field of contemporary Qur’an scholarship. There is little doubt that the pseudonymous work has been useful in sparking renewed attention to the relationship between the Qur’an and the lingua franca of much of the Middle East in the years when Islam was emerging. However, because of the hermeneutical and theological presuppositions underlying Luxenberg’s work, and the lack of a clear understanding of how languages function and develop, one wonders whether his approach is not muddying the waters rather more than clarifying them. Luxenberg proposes to “unlock” for us the original meaning of the Qur’an – until now, he claims, held prisoner by the failure of the Islamic tradition (and other scholars who have effectively followed it) to appreciate the role of Syro-Aramaic in the formation of the Qur’an and its language. Behind that aim lies an assumption that the real meaning of the Qur’an is to be found in the text itself, or rather behind the text in the mind of the original author. His work shows little appreciation of the notion that the meaning of a text is not simply found in the mens auctoris but rather in the mens lectoris or, better, in the complex relationship between the text and its readers in their contexts. He seems to espouse the reconstructionist hermeneutic of a Schleiermacher or a Dilthey, leaving unexamined the important critique made by Gadamer and the vast amount of reflection on hermeneutical issues that has taken place in the last century. This same hermeneutical naiveté drives the popular interest in Luxenberg’s reconstructions of what he sees as the original text. It is perhaps instructive to note that the vast majority of journalistic attention was devoted to Luxenberg’s xi FOREWORD proposed re-reading of Q 55:72, substituting white grapes for the fair virgins who, according to the traditional reading of the word hur, await the faithful in Paradise. This is significant not simply because it represents the return of a classical topos of anti-Muslim polemics. It reflects a perception that the modern phenomenon of Muslim suicide-bombing has its origins in a particular reading of the Qur’an – a tendentious one at that – and that the way to deal with the current crisis in relations with Muslims is simply to discredit such a reading. In the popular mind the Qur’an is a ticking bomb that needs somehow to be defused. Yet there seems to go along with this very little awareness of the many other factors, political, economic, and cultural, that fuel the rage that sometimes expresses itself in violence. Unfortunately, the resolution of the current dramatic situation will not be philological. It no less crucial for the scholarly community than for the public at large to recognize that the reconstruction of Islamic origins or of earlier layers of the Qur’an, even to the extent that these are possible, will neither answer all our questions about Islam as it is lived, nor provide us with a solution to the conflicts that rive our world. There is a widespread tendency even within religions to see them as communities born in all their natural integrity, only to gradually degrade and decline as they develop over time. Every reform is seen, therefore, as a return to the innate purity and energy of the original religious impulse. And the temptation for those of us who study the fascinating phenomenon of early Islam is to imagine that discovering and reconnecting with the truth of the initial moment can somehow transform the present. However, the initial singularity (to borrow a term from cosmology) that would eventually give rise to the new and distinct religious community was, like other moments of religious innovation, not a point of stability and calm perfection, but rather a moment of instability which has constantly sought new forms of equilibrium. There is no returning to the beginning; we can only deal with things as they have become. The “real” meaning of the Qur’an is to be found in what it actually says to actual people – in a privileged way to the community of believers for whom it is more than an ancient text. For several of the contributors to this volume the avocation of Qur’anic studies goes beyond academic interest and the pursuit of scholarly excellence. It is a way of engaging seriously with other believers on what to them is scripture, and more than scripture. Because of the way sacred texts function in believing communities, significant “new readings” of the Qur’an will more likely emerge from an engagement with the “world in front of the text” than from the archaeology of the “world behind the text.” Qur’anic studies might well benefit from the continuing discussion in Biblical studies over the just equilibrium between, on the one hand, a historical-critical analysis that excavates and continually fragments the text and, on the other, a reading that takes the text seriously as a canon. There is space and a need for both, but in Qur’anic studies as in Biblical studies the former has tended to dominate. The popular, and even some of the scholarly, response to Luxenberg’s work seems to reflect two conflicting hopes – the more negative is a desire to see the xii FOREWORD foundations of the Qur’an discredited, and along with them Muslim faith. The more positive is the hope for a new reading of the Qur’an that would form the basis of a constructive relationship between those who read it in faith and those who may not regard it as revelation. New readings are generated not simply by analysis, that is, by breaking down the text. Rather they result from catalysis, that is, by establishing new links and relations among the elements of the text itself and with the context in which it is read. In this process of catalyzing new readings the philologists and historians, even if not believers, also have their role – not because they possess some privileged insight into the “real meaning” of the text, but because they can offer to the believing reader an eye for the rich complexity of the textual elements and a knowledge of the history of their reception and interpretation in the many contexts that together make up the Muslim tradition. One senses that some Qur’an studies – quite explicitly in the cases of Lüling and Luxenberg – are competing for possession of the text. The claim that the underlying structure and numerous elements of the text are originally Christian seems to reveal a desire to dispossess the Muslim community of its foundation and greatest treasure. Yet even if the case were proven, what would that indicate? Again comparisons with Biblical studies might be instructive. The observation that several elements of the Hebrew Bible are a reworking of material from other religious traditions of the ancient Near East does not mean that the Bible has no unique and coherent vision of its own. The observation that the Infancy and Passion narratives in the Synoptics seem to be taking their cue from elements in the prophetic literature and the Psalms rather than simply from an historical memory of events does not thereby discredit the gospels. New and fruitful readings of the Qur’an will not result from competitive and conflictual analyses. They will be catalyzed by a collaboration among all those who, whether believers or not, take the text seriously not simply as an historical artifact but as a canon of scripture for a contemporary community. Daniel A. Madigan SJ Pontifical Gregorian University Rome, Italy xiii ABBREVIATIONS BSOAS CSCO EI2 Bulletin of the School of Oriental and African Studies. Corpus Scriptorum Christianorum Orientalium. The Encyclopaedia of Islam, Second Edition, Leiden: Brill, 1954–Present. EQ The Encyclopaedia of the Qur’an, ed. J. McAuliffe, Leiden: Brill, 2001–6. FV A. Jeffery, The Foreign Vocabulary of the Qur’an, Baroda: Oriental Institute, 1938; Leiden: Brill, 2007. GAL C. Brockelmann, Geschichte der arabischen Litteratur, Leiden: Brill, 1937–49. GAS F. Sezgin, Geschichte des arabischen Schrifttums, Leiden: Brill, 1967–2000. GCAL G. Graf, Geschichte der christlichen arabischen Literatur, Rome: Biblioteca Apostolica Vaticana, 1947. GdQ1, 2, 3 1, first edn: T. Nöldeke, Geschichte des Qorans, Göttingen: Verlag der Dieterichschen Buchhandlung, 1860; 2, second edn, Nöldeke’s revised work being titled therein Über den Ursprung des Qorans and including F. Schwally, Die Sammlung des Qorans, ed. and rev. F. Schwally, Leipzig: T. Weicher, 1909, 1919; 3, third ed., including Schwally and G. Bergsträsser and O. Pretzl, Die Geschichte des Koran-texts, Leipzig: T. Weicher, 1938. Reprint: 3 vols. in 1, Hildesheim: G. Olms, 1970. JAOS Journal of the American Oriental Society. JNES Journal of Near Eastern Studies. JRAS Journal of the Royal Asiatic Society. JSAI Jerusalem Studies in Arabic and Islam. JSS Journal of Semitic Studies. MIDEO Mélanges de l’Institut dominicain d’études orientales du Caire. OC Oriens Christianus (Serial). SI Studia Islamica. ZDMG Zeitschrift der deutschen morgenländischen Gesellschaft. xiv Halo ANATOLIA Caspian Sea ras R iver Constantinople w r ta Pa Bab al-Lan Darband Ku rR ive r Black Sea Tigr An tio c h Edessa Mabbug Harran o pp bed Euph e l e Z A rat es afa Riv Rus er Palmyra is R iver Mediterranean Sea a Nile River dri an x Ale Baghdad Damascus ’ Bostra (Busra) rbala Jerusalem Ka fa Madaba al-Ku jaf a a N Petra al-Hir ra Cairo Elusa as l-B a Ayla Dumat al-Jandal IRAN Pe r sia n Hegra (Mada’in Salih) lf EGYPT Medina Mecca ea dS Re Qaryat al-Faw Najran AB YS SIN IA Map Locations cited in the present volume. Gu INTRODUCTION Qur’anic studies and its controversies1 Gabriel Said Reynolds As I was planning the April 2005 University of Notre Dame conference, “Towards a New Reading of the Qur’an?” on which this book is based, a number of my colleagues told me that they were suspicious of the very idea of a “new reading” of the Qur’an. The reading of the Qur’an, that is the accuracy and meaning of the words (and not the interpretation thereof), seems sufficiently clear. No current edition of the Qur’an, for example, contains an apparatus criticus, a list of words that are different, missing or added in certain manuscripts. Meanwhile, the various editions of the Qur’an printed today (with only extraordinary exceptions) are identical, word for word, letter for letter. The Arabic text of the Qur’an, then, seems to be uniform and problem-free. Meanwhile, most English translations of the Qur’an give the impression that the meaning of the Qur’an is transparent. Rarely does a translator list variants or confess that the sense of a passage is unclear. Why, then, should one consider a new reading of the Qur’an? The answer appears only when two different Qur’an translations are compared. For while alone each translation will seem beyond reproach, when compared the two translations will produce numerous – and often irreconcilable – contradictions. These contradictions usually reflect confusion among the classical exegetes. Take chapter 18 (al-kahf ), verse 9, an introduction to the account of “the companions of the cave” (the “Sleepers of Ephesus” in Christian tradition): “Have you considered that the companions of the cave and ‘al-raqim’ are among our marvelous signs?” The term al-raqim here seems to have great importance to this story, but the exegetes are deeply divided over its meaning. Some interpret al-raqim to mean “book,” others “inscription,” “tablet,” “rock,” “numbers” or “building.” Still others see it as a proper name, but of what? Suggestions include the name of a village, or a valley, a mountain, or even a dog. Translators are faced with a decision. They could choose one of the classical scholars’ suggestions, perhaps weighing them against one another on the basis of each scholar’s reputation or apparent erudition. They could look for a different 1 GABRIEL SAID REYNOLDS meaning for al-raqim, perhaps by researching other languages or place names, resources often neglected by or unavailable to the classical scholars. Finally, they could emend the word, proposing an alternative spelling (perhaps al-raqqam or alzaqim), hoping to arrive at a form whose meaning is better known. Whatever choice is made, translators of the Qur’an are compelled to develop their own reading. On the shape of the Qur’an The common belief that the Qur’an has a single, unambiguous reading is due in part to the bravado of translators, who rarely express doubt about their choices. Yet it is above all due to the terrific success of the standard Egyptian edition of the Qur’an, first published on July 10, 1924 (Dhu l-Hijja 7, 1342) in Cairo, an edition now widely seen as the official text of the Qur’an.2 Initially, however, the publication of this edition was a purely Egyptian affair. It was the work of a government appointed committee, led by Muhammad b. ‘Ali al-Husayni al-Haddad,3 that was meant to establish a uniform text for religious education in Egypt.4 Minor adjustments were subsequently made to this text in following editions, one published later in 1924 and another in 1936. The text released in 1936 became known as the Faruq edition in honor of the Egyptian king, Faruq (r. 1936–52).5 Yet the influence of the Cairo text soon spread well beyond Egypt. It has been adopted almost universally by both Sunni and Shi‘i Muslims, and by critical scholars as well, who have long since given up Gustav Flügel’s 1834 edition. Writing in 1938, Otto Pretzl noted with amazement that in his day for the first time a de facto canonical text had emerged.6 Yet the Egyptian project was never intended to be text-critical, at least as this term is commonly understood. The scholars who worked on that project did not seek to reconstruct the ancient form of the Qur’an, but rather to preserve one of the canonical qira’at “readings” (here meant in the specialized sense it has in Islamic tradition), that of Hafs (d. 180/796) ‘an ‘Asim (d. 127/745). But these qira’at are part of the history of the text, not its starting point, and the idea of a discrete number of different yet equally canonical qira’at did not develop before the fourth/tenth century, when great divisions over the Qur’anic text led Ibn Mujahid (d. 324/936), among others, to sponsor this regulatory concept. Ibn Mujahid argued that there are seven, equally valid qira’at. Others argued for ten, or fourteen. The gradual (yet never complete) acceptance of the argument for seven qira’at (often attributed to Ibn Mujahid’s use of a prophetic hadith that speaks of “seven letters” of the Qur’an)7 was generally accompanied with the caveat that each qira’a has two versions. Effectively, then, fourteen different versions were considered equally authentic, only one of which was Hafs ‘an ‘Asim. Even in this scenario there is no unanimity over the precise shape of the Hafs ‘an ‘Asim qira’a. Four different lines (turuq) of transmission are claimed for it, and discrepancies abound in the various texts claiming to transmit it. 2 INTRODUCTION In the early twentieth century, therefore, the shape of the Qur’an would have seemed anything but clear. In fact, the Egyptian government was motivated to begin the project that would lead to the Cairo Qur’an edition due to the variations (or “errors,” as an appendix to the Cairo edition describes them) found in the Qur’anic texts that they had been importing for state schools.8 In response, the government destroyed a large number of such texts by sinking them in the Nile River and issued its own text. The Cairo project thus followed in the spirit of the caliph ‘Uthman, and the governor al-Hajjaj b. Yusuf (d. 95/714), who are reported to have destroyed competing versions and distributed their own text of the Qur’an in the first Islamic century. When the scholars in Cairo decided to fix a standard text according to Hafs ‘an ‘Asim, they still had to decide which reports of it to trust. Their project, then, involved comprehensive research of the classical qira’at works.9 In fact, they conducted this research with great thoroughness and attention to detail, according to the observations of several western scholars.10 Gotthelf Bergsträsser, for example, noted that in only a small number of cases is their reading contradicted by earlier sources on Hafs ‘an ‘Asim.11 However, the Cairo text is often at odds with manuscript evidence.12 This is perhaps to be expected, given that the Cairo project was not about recovering a text as much as choosing a text. Indeed the very idea of canonical qira’at is based on religious doctrine, not textual criticism. In the paradigm of qira’at, discussion over the shape of the Qur’anic text must take place within the context of the community’s tradition. The Egyptian edition’s claim to validity is based not on antiquity, but rather on canonicity. The Egyptian Qur’an, then, should not be confused with a critical edition. The Egyptian scholars in no way sought to record the canonical variants to their text, let alone the non-canonical variants to be found in manuscripts. That task, however, was taken up soon thereafter by two western scholars, at once friends and colleagues: the aforementioned Bergsträsser, a German, and Arthur Jeffery, an Australian. Bergsträsser, professor at the University of Munich, was a student of August Fischer in Leipzig and already well-known for his continuation of Nöldeke/Schwally’s Geschichte des Qorans and his various contributions to Semitic linguistics. Jeffery, a Methodist minister, was then a Professor at the American University in Cairo, although he would move to Columbia University in New York in 1938. He would later publish the cardinal study on the etymology of Qur’anic terms: The Foreign Vocabulary of the Qur’an. At the time, however, Bergsträsser and Jeffery were busy studying qira’at literature and filming early Qur’anic manuscripts, planning for the eventual publication of a critical edition.13 A summary of Bergsträsser’s presentation at the sixth Deutscher Orientalistentag (Vienna 1930) presents their plan in detail: So schlägt der Vortragende vor, in Zusammenarbeit mit Prof. Jeffery einen kritischen Apparat zur ägyptischen Ausgabe zu schaffen, der alle in der Literatur (vor allem den qira’at-Werken und den Korankommentaren) und 3 GABRIEL SAID REYNOLDS in kufischen Koranhandschriften (die zu diesem Zweck in großem Umfang zu photographieren seien) erreichbaren Varianten und Lesarten, unter Weglassung bloß phonetischer Differenzen, in Transkription darbietet, getrennt nach konsonantischen Abweichungen innerhalb des ‘uthmanischen Textes, nach Vokalisationsmöglichkeiten dieses Textes und nach Abweichungen von ihm; und der weiter sie kritisch zu werten sucht, wobei diejenigen Abweichungen vom textus receptus herausgehoben werden, die besonders zu beachten oder ihm vorzuziehen sind. Das gesamte Material soll durch ein Wortregister und sachlich-grammatische Übersichten erschlossen werden. (G. Bergsträsser, “Über die Notwendigkeit und Möglichkeit einer kritischen Koranausgabe,” 8314) After discussion of the idea (by scholars no less than Fischer, Baumstark, Ritter and Kahle), the scholarly group at the Deutscher Orientalistentag approved the following joint statement: Der 6. Deutsche Orientalistentag 1930 in Wien begrüßt den von Herrn Bergsträsser vorgelegten Plan eines kritischen Apparats zum Koran und erklärt, daß seine Durchführung ein dringendes Bedürfnis der Wissenschaft vom Orient darstellt. Insbesondere erklärt der 6. Deutsche Orientalistentag, dass für diesen Zweck die Schaffung einer Sammlung von Photographien alter Koranhandschriften erforderlich ist. (G. Bergsträsser, “Über die Notwendigkeit und Möglichkeit einer kritischen Koranausgabe,” 83) Incidentally, we might reasonably speculate that this critical edition project was actually inspired by the work of the Egyptians, since both Bergsträsser and Jeffery were active in Egypt during the time that the Cairo text was being produced. This would be a noteworthy case of Muslim and western scholars working in cooperation, methodological differences notwithstanding. Yet the critical edition project was ultimately to be abandoned due to an extraordinary, indeed an almost unbelievable, series of unfortunate events. In August 1933 Bergsträsser died during a hiking excursion with a companion in the Alps.15 His companion, Friedrich Thiersch, recorded an account of Bergsträsser’s death in a 16-page typed statement prepared for the police immediately after the event.16 Thiersch, after admitting, “Mein Kopf ist noch völlig durcheinander,” reports that during the excursion Bergsträsser fell off a sharp incline, injuring his head.17 Thiersch continues, in an emotional tone, that since he was unable to carry the professor back alone he remained with him on the mountain, through a cold night, during which Bergsträsser expired. But Thiersch’s version of Bergsträsser’s death has not gone unchallenged. An Egyptian scholar named Muhammad Hamdi al-Bakri, in the preface to his book on Bergsträsser’s philological work, concludes that the professor was murdered 4 INTRODUCTION due to his anti-Nazi stance. Bakri describes, in dramatic fashion, how one of the German professor’s own students attacked him during that hiking excursion, tackling him and throwing him off a cliff to his death.18 Bakri, unfortunately, provides no evidence for his version of Bergsträsser’s death. Meanwhile, Bakri’s story conflicts with the claims of Günter Lüling that the Munich school of Islamic Studies was complicit with the National Socialists, although Lüling too does not produce evidence for his assertions.19 Bergsträsser’s death, it seems, will remain a mystery until new evidence is uncovered. After his death the Qur’an project was continued by the aforementioned Otto Pretzl, Bergsträsser’s colleague and successor in Munich.20 Within the next several years Pretzl brought the project close to completion, as is evident from comments that Jeffery makes in his Materials for the History of the Text of the Qur’an, published in 1937: For many years the present writer has been collecting materials for a critical text of the Qur’an, and in 1926 agreed with the late Prof. Bergsträsser to collaborate in the much bigger task of preparing an Archive of materials from which it might be some day possible to write the history of the development of the Qur’anic text. It is hoped that it will be possible to publish shortly, as one step in that plan, a text of the Qur’an with apparatus criticus giving the writer’s collection of textual variants gathered from the Commentaries, Lexica, Qira’at books and such sources. Meanwhile Dr. Pretzl, Bergsträsser’s successor at Munich, has begun to organize the Archive for the Korankomission set up by the Bavarian Academy at Bergsträsser’s initiation, and has already assembled a goodly collection of photographs of early Kufic Codices and early unpublished qira’at works. (A. Jeffery, Materials for the History of the Text of the Qur’an, vii21) Yet just as the Qur’an project neared completion, the clouds of war appeared on the horizon in Europe. When World War II broke out, Pretzl was called to military service and the project was suspended. He was never to return. On October 28, 1941 Pretzl died in a plane crash. Pretzl’s position at Munich remained empty until after the war, when it was filled by Anton Spitaler, who himself had served for years as an interpreter in the German army. Spitaler, however, who died only in 2003, shared none of his predecessor’s vision for the Qur’an project. In fact, there are some reasons to believe that he put an end to it. In a lecture given in Jerusalem in 1946 Jeffery reported: Pretzl was killed outside Sebastopol during this late War, and the whole of the Archive at Munich was destroyed by bomb action and by fire, so that the whole of that gigantic task has to be started over again from the 5 GABRIEL SAID REYNOLDS beginning. It is thus extremely doubtful if our generation will see the completion of a really critical edition of the Qur’an. (A. Jeffery, “The textual history of the Qur’an,” 103)22 With this Jeffery gave up any remaining hopes for the project. When he died in 1959 the project died with him.23 Presumably, it can only have been Spitaler who informed Jeffery of the archive’s destruction. Indeed, over two decades later Spitaler himself commented that the destruction of the archive in the war meant the end of the project for a critical edition of the Qur’an.24 And yet it is now well known that the archive survived the war, that Spitaler took possession of it and that he passed it on to his former student, now a Professor in Berlin, decades later.25 Spitaler’s conduct in this affair recently led Günter Lüling, another of his former students – and later a personal nemesis – to malign him. Lüling suggests that Spitaler did not himself have the ability to do any productive work with the films, and was too competitive to allow others access to them.26 This is certainly an unfair assertion; it is presumably a polemical response to Spitaler’s well known opposition to Lüling. It seems more likely that Spitaler simply did not believe in the feasibility of the Bergsträsser/Pretzl/Jeffery project. Spitaler alludes in one place to his belief that even Pretzl began to doubt the possibility of producing a critical edition of the Qur’an.27 Indeed, in a presentation delivered to the twentieth International Congress of Orientalists in Brussels (1938), Pretzl notes the degree to which readings of the Qur’an had been suppressed over time, “sodass heute fast nur mehr die Lesung das Hafs ‘an ‘Asim bekannt und gebraucht ist.” 28 He continues: “Aus diesen Tatsachen ergibt sich die Schwierigkeit, im nicht zu sagen Unmöglichkeit, an einen Urtext heranzukommen.”29 However, Pretzl’s pessimism was above all in regard to the value of the Islamic transmission of variants (since there is reason to suspect that most variants are the product of exegesis). In the same presentation Pretzl enthusiastically supports Jeffery’s work towards an apparatus criticus of the Qur’an. He also notes the hard work already done towards the project (“der sich okzidentalische Gelehrte seit längerer Zeit mit Eifer und Selbstentsagung widmen.”),30 even as he adds a remarkable note about a promised contribution of Eastern, presumably Muslim, scholars to it: “Es wäre zu wünschen, dass auch orientalische Gelehrte ihr Versprechen der Mitarbeit an diesen grossen wissenschaftlichen Aufgaben einlösen würden.”31 In the end we can do no more than speculate about the true intentions of Pretzl or Spitaler, and speculation is no ground for judgment. Yet this much is clear: that Jeffery – a Qur’an scholar without equal – was unable to follow this project through was not only a disappointment for him. It was a great loss for all students of the Qur’an, a loss that perhaps can never be fully redeemed. In any case, as a result of the untimely deaths of Bergsträsser and Pretzl, and of the peculiar actions of Spitaler, the only coordinated effort to produce a critical 6 INTRODUCTION edition of the Qur’an was abandoned. Thus there still exists no critical edition, an absence that Jeffery, writing in the 1930s, called “extraordinary.”32 It is certainly all the more so today. Recently, however, the question of critical research on the text of the Qur’an has once again been attracting attention. On one hand, a reference work entitled Mu‘jam al-qira’at al-qur’aniyya was published in 1983.33 The Mu‘jam provides virtually exhaustive lists of variants reported in Islamic literature to the Cairo edition of the Qur’an. These include canonical qira’at (i.e. the “seven,” the “three after the seven,” and the “four after the ten”) and non-canonical variants (shawadhdh) reported to have existed in masahif that existed before the mushaf of ‘Uthman. In this way it provides a noteworthy addition to Jeffery’s Materials for the History of the Text of the Qur’an. Nevertheless, there is reason to believe that the variants in the qira’at and masahif literature are more the fruit of exegesis than a record of primary traditions. If this is the case, then the Mu‘jam is perhaps most important for its demonstration of the degree to which Muslim scholars themselves speculated over the shape of the Qur’anic text. On the other hand, a new source of primary materials for research into the Qur’anic text was uncovered in 1972, when workers discovered the now famous cache of ancient Qur’an manuscripts in the Great Mosque of San‘a’, Yemen. The Yemen find caused great excitement in the academic community, and provoked some sensationalist reporting in the popular media.34 Yet the Qur’an fragments themselves have still not been published,35 even now that they have been restored and photographed. This, it seems, is due to the very fact that they caused so much excitement in the West. This attention was unwanted. The Yemeni authorities came under great pressure to intervene, as the impression grew in the Islamic world that they had allowed anti-Islamic Orientalists to handle, and manipulate, Muslim religious property. In any case, it is still unclear how the Yemeni find will change our understanding of the early development of the Qur’anic text. Indeed, the very scholars who worked on the project have different perspectives, in part due to their respective disciplines. H.-C. Graf von Bothmer, an art historian, argues that the San‘a’ fragments prove that a complete standard text of the Qur’an, including the Fatiha and the final two charm suras, existed in the first Islamic century,36 all the while acknowledging that the only dated fragment is from AH 357 (AD 968). G. Puin, an Islamicist, agrees that the fragments represent the earliest extant manuscripts of the Qur’an. However, he believes that they will show that the medieval Muslim scholars who established the scriptio plena of the Qur’an misread the scriptio defectiva in numerous places. He notes, for example, that in many of the fragments the long vowel “a” is represented with the mater lectionis “ya’;” thus ilah is often written ilayh.37 This seems to confirm what philologists have long suggested regarding proper names such as Ibrahim and Shaytan (for which one would expect, based on Semitic cognates, Abraham and Shatan), that the medieval Muslim masoretes read ya’ as a ya’, even when it was originally recorded as a mater lectionis for a.38 Puin suggests, therefore, that the San‘a’ 7 GABRIEL SAID REYNOLDS manuscripts in the future can serve as a basis for emendations of the standard Qur’anic text.39 This, of course, will be impossible as long as the manuscripts remain a mystery. Meanwhile, Profs Sergio Noja Noseda and François Déroche have overseen a project to publish facsimiles of a number of early Qur’an manuscripts. Thus far they have published Ms. arabe 328 from the Bibliothèque nationale de France,40 and the first half of Ms. Or. 2165 of the British Library.41 I understand that they are planning editions of other important early manuscripts, including the famous Samarqand codex. Noseda and Déroche present these manuscripts in magnified format, together with a modified rasm of the Cairo edition on the facing page. Their editions are thus valuable (although expensive!) tools for research on the text of the Qur’an. In a word, as the scholars represented in this volume met at Notre Dame in April 2005, it was evident that momentum was building for the renewal of textual research on the Qur’an. On the origin of the Qur’an Yet the conference was also convened in the midst of increasing discord in Qur’anic Studies. For in recent years a number of controversial works – including those by Günter Lüling, John Wansbrough, Yehuda Nevo and Christoph Luxenberg – have appeared to challenge the traditional theories on the origin of the Qur’an. In different ways these authors propose that the master narrative of Qur’anic origins, that which was developed in medieval Islamic literature, is historically unreliable. Contemporary scholars, they argue, should therefore be free to observe the Qur’an without gazing through the lens of that literature. Thereby they might arrive at surprisingly novel interpretations of Qur’anic passages. These controversial works have led to a sort of sub-culture in the field. The authors are generally grouped together as Revisionists. And yet their theories, besides their basic precept, actually have very little in common. This sub-culture, therefore, has not developed a methodology, much less a school, that poses an organized challenge to the current paradigm. Despite this, or perhaps because of it, they have collectively contributed to a sense of strife, even anxiety, in the field of Qur’anic Studies. For this reason Angelika Neuwirth has recently asked whether the field is now in “ein hoffnungsloses Chaos.”42 It might be added, however, that the idea of speculating over the intended meaning of the Qur’an has a precedent in Islamic tradition itself. In early exegetical works Muslim scholars carry out speculative, and often unresolved, conversations on the meaning of numerous Qur’anic passages. They are usually forthright about the extent of their disagreement, often concluding their analysis with the simple admission: “The exegetes disagree on the meaning.” Elsewhere they use the marvelous Arabic elative term “asahh” or “more correct,” to introduce their own view while not entirely dismissing that of others. Or they resign themselves with the refrain: “God knows best.”43 8 INTRODUCTION Of course, these Muslim scholars did not question the traditional narrative of the Qur’an’s proclamation or its divine origin. Yet they did see the importance of achieving a clearer, more accurate understanding of the text. In fact, this enterprise seems to be called for by the text itself, which asks, “Will they not contemplate the Qur’an?” (4:82). The 2005 Notre Dame conference was conceived as a response to this challenge, and in this way was fully in the spirit of traditional Islamic exegesis. Meanwhile, the standard academic textbooks on the Qur’an continue to follow the master narrative, introducing the history and meaning of the Qur’an according to, first, the classic Sunni Muslim sources, and, second, the classic Western sources, at the foundation of which stands the Geschichte des Qorans of Nöldeke, Schwally, Bergsträsser and Pretzl.44 If these Western sources do not share the religious perspective of the Muslim sources, they nevertheless operate within the same basic paradigm, namely, that both the history and meaning of the Qur’an are to be understood in light of the Prophet Muhammad’s biography. Thus many classic western authors simply substitute Muhammad for God when introducing the author of the Qur’an. Nöldeke himself once commented that the Qur’an “is the work of Muhammad, and as such is fitted to afford a clue to the spiritual development of that most successful of all prophets and religious personalities.”45 To give only one example, Qur’an 96, verses 1–5 (which begin, “Read in the name of your Lord who created”) is generally explained through the famous narrative of Muhammad’s call to prophecy on Mt Hira’. If traditional Muslim scholars believe that God (or the angel Gabriel) truly spoke these verses to Muhammad that day, most traditional western scholars believe that Muhammad, at least, was really on that mountain and really proclaimed these verses believing that they came from God. Neither group of scholars, however, conclude that the story of Muhammad’s call might have been written as a way of explaining Q 96:1–5, which would mean that explaining these verses through the story of Muhammad’s call would be problematic, to say the least. Revisionist scholars, on the other hand, seem to agree on one basic precept: the link between the Qur’an and the biography of Muhammad is illusory. The story of the Qur’an’s origins must be seen in a much broader perspective. Only once the Qur’an is freed from the narrow category of that biography can it be fully appreciated. While this simple idea might not appear prima facie to be radical, it nevertheless undermines both the Islamic and western traditions of Qur’anic Studies. It is no surprise, then, that Revisionism has remained a sort of sub-culture within the field. Yet this is perhaps not the only reason for their marginalization. For if Lüling, Wansbrough, Nevo and Luxenberg all challenge the master narrative of Qur’anic Studies, the new narratives that they write have little in common. Their theories form a cacophony, not a symphony, all of which contributes to the sense that Revisionists have provided no respectable alternative narrative.46 9 GABRIEL SAID REYNOLDS Of these theories perhaps the least known, but by no means the least provocative, is that of Günter Lüling, originally expressed in his 1974 publication Über den Ur-Qur’an: Ansätze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur’an.47 The title indicates Lüling’s primary thesis, that the Qur’an preserves within it earlier Christian writings. Yet Lüling also believes his theory has serious consequences for the future of Islam, for which reason he gave the English translation of his book a substantially modified title: A Challenge to Islam for Reformation. According to Lüling, the Qur’an contains four different textual levels. The first level, the original text, is a strophic hymnal, composed by the Christian community of Mecca. This community consisted both of Trinitarian and non-Trinitarian believers. The former had adopted the Nicene doctrines of the Byzantine empire, whereas the latter maintained the ancient (and true) teaching of Christ himself, that he was an angel of the divine council incarnate, sent to undermine the hierarchical, monotheistic Judaism of his day. The second level consists of passages from that hymnal which were edited and Islamized in Muhammad’s time. The third level contains those passages originally composed in Muhammad’s time, and which have an exclusively Islamic meaning. The fourth and most recent level are those passages altered by post-Muhammadan Muslims during the editing of the Arabic rasm, the process by which the Qur’anic text went from scriptio defectiva to scriptio plena. For Lüling, then, the first task of the Qur’an scholar is archaeology. He seeks to dig through the various levels of the Qur’an in order to uncover the original Christian hymnal lying underneath. This can be seen through his examination of the aforementioned chapter 96 (surat al-‘Alaq) of the Qur’an. Opposing the traditional view, Lüling argues that a bit of textual archaeology will show that chapter 96 was originally a Christian hymn. The speaker is not God but rather the believing Christian. The intended audience is not Muhammad but his fellow Christians. God is not calling Muhammad to recite the Qur’anic revelations that Gabriel will give to him. The believer is exhorting his community to pray: “Recite in the name of your Lord who created . . . .” Meanwhile, Lüling re-reads the mysterious term zabaniya of verse 18, traditionally understood as the proper name for an angel of punishment in hell, as rabbaniyya. He argues that it is cognate to Aramaic rabbouni (“lord”), the title by which Jesus is called in Mark 10:51 and John 20:16. In Lüling’s reading both terms, Qur’anic rabbaniyya and Biblical rabbouni, are references to the great angels of the divine council, among whom is Jesus.48 Lüling proposes a similar reinterpretation of the Qur’anic references to mushrikun “those who associate (something with God),” a term traditionally seen as a reference to the Meccan pagans against whom Muhammad preached. In fact, Lüling concludes, this term is an echo of the internal debate that raged between Christians in pre-Islamic Mecca. The faithful followers of Christ, who held an angel-Christology, labeled their Trinitarian opponents with this term, accusing them of associating Christ and the Holy Spirit with God. In Lüling’s reading this 10 INTRODUCTION term would be meaningless if applied to pagans, who did not even pretend to profess monotheism.49 This much seems reasonable, but ultimately readers may be discouraged by the many ways in which Lüling’s new reading for the Qur’an is woven inextricably into the fabric of his larger religious vision. Lüling’s religious vision, meanwhile, encompasses not only Islam, but also Judaism, Christianity, tribal cults, and the history of the Semitic and Prussian (!) peoples.50 Meanwhile, Lüling’s own career has been marked by conflict and polemic. Denied habilitation at Erlangen, Lüling appealed the case in the academy and ultimately in civil courts. He has since made it clear that he believes the German academy systematically excluded him.51 At the head of the campaign against him was none other than the aforementioned Anton Spitaler. According to Lüling, Spitaler and others were threatened by his theological method. Lüling believes that Islamic Studies should be reunited with its original partner in the German academy, Christian theology, or at least theology in a broader sense.52 It might be pointed out, however, that Lüling’s method is not only theological. He also uses philology, to reconstruct the Ur-Qur’an (the original document from which the current Qur’anic text derives, as he sees it), and history, to reconstruct the lost Christian community of Mecca.53 This is notably different from the approach of John Wansbrough, who argues that there was no Ur-Qur’an, and that the very idea of Mecca as the birthplace of Islam is a myth.54 Wansbrough’s thesis begins with a basic problem of the non-Qur’anic sources. The earliest Arabic Islamic literary sources for Islamic origins are the biographical, exegetical, jurisprudential and grammatical texts written under the ‘Abbasids. Most of these texts, it is true, contend to be transmitting material from early generations, often from the Prophet Muhammad himself. This contention, however, has long been challenged by western scholars, most famously by Ignaz Goldziher in the late nineteenth century.55 The antiquity of the Qur’an, on the other hand, had not been challenged before. The Qur’an, at least, was thought to be a faithful record of Muhammad’s preaching in the Hijaz, western Arabia, in the early seventh century. Wansbrough thought otherwise. He argued that the final form of the Qur’an should not be dated prior to the period when the Qur’an became a source for biography, exegesis, jurisprudence and grammar: “Logically, it seems to me quite impossible that canonization should have preceded, not succeeded, recognition of the authority of scripture within the Muslim community.”56 This canonization happened in the second/eight or third/ninth century, not the early seventh century, and in ‘Abbasid Iraq, not in the Hijaz. The origin of the Qur’an is to be found together with the origin of the other Arabic Islamic literary sources. It is for this argument that Wansbrough became known. Therefore, when more recent manuscript studies (not least of which is the Yemeni find) appear to suggest an early date for the codification of the Qur’an, some scholars have concluded that Wansbrough’s “thesis” has been disproved. It should be noted, however, that Wansbrough’s ideas far transcend this particular argument, although 11 GABRIEL SAID REYNOLDS his writing style, at once profoundly learned and terribly obtuse, presents a serious obstacle to the communication of his thought. Nevertheless, it is worth the effort, for Wansbrough’s description of the literary nature of the Qur’an presents a meaningful challenge to the standard hermeneutic used in reading it. In Wansbrough’s reading, the very literary nature of the Qur’an, above all its formulaic nature, reveals its origins: “Exhibiting a comparatively limited lexical range, those formulae serve to confirm the impression of a composition made up of originally unrelated pericopes.”57 According to Wansbrough, these pericopes were consolidated into a scripture in response to a sectarian environment, an environment where Christians and Jews, not pagans, challenged the young Arab religion. In other words, it was in ‘Abbasid Iraq that Muslims set about establishing their own scripture, along with their own salvation history.58 Neither the former, nor the latter, reflects the actual experiences (wie es eigentlich gewesen war) of a prophet in seventh century Hijaz, but rather something closer to nostalgia.59 Wansbrough’s view means, among other things, that the traditional idea of the chronology of the Qur’an, or, more generally, the idea of Meccan and Medinan suras, is misguided.60 The importance of this contention cannot be underestimated, for this idea has been a formative influence in both Islamic and western scholarship. Nöldeke spent much of his efforts in the first volume of the Geschichte trying to establish a precise chronological order of suras. The 1924 Cairo Qur’an edition, following the traditional Islamic practice, includes the marker “Meccan” or “Medinan” at the head of each sura. More recently, Angelika Neuwirth followed this division in her Studien zur Komposition der mekkanischen Suren.61 Wansbrough’s skepticism of the historicity of the ‘Uthmanic recension of the Qur’an is also worth taking seriously. By Wansbrough’s reading, the reports of Qur’anic variants in Islamic literature are not recollections of ancient texts that differed from the ‘Uthmanic text.62 Rather, they are the outcome of exegesis. He argues that most of the variants mentioned in this literature are so minor – quite unlike variants to Biblical texts – that they appear to be the product of grammatical or syntactical speculation or emendation.63 Wansbrough is not alone in this argument.64 Finally, Wansbrough challenged the traditional notion of classical Arabic. Semiticists long believed classical Arabic, in large part due to its robust system of inflections and broad sound system, to be an ancient language, probably the closest language to Ur-Semitic. Indeed, it was this belief, in part, that led to the rise of Arabic Studies in Europe, for it was thought that Arabic preserved ancient Semitic characteristics that could explain certain riddles in Biblical Hebrew and Aramaic. Accordingly, most scholars in the nineteenth century, informed by the model of Latin and Romance languages, held that modern Arabic colloquials were neither ancient nor independent (as French, Spanish and so on are not independent of Latin), but something like mischievous children of pure classical Arabic. The Qur’an, on the other hand, was generally thought to preserve this classical language. Nöldeke agreed with this idea (although he found the Qur’an contain 12 INTRODUCTION numerous grammatical errors) but it was challenged by Karl Vollers, who held that the Qur’an was originally written in a non-inflected tribal dialect, and only later adjusted to match classical Arabic.65 Wansbrough agrees with neither Nöldeke nor Vollers.66 He holds against Nöldeke that classical Arabic was a late construction, developed by the grammarians of second/eight century contemporaneously with the codification of the Qur’an; he therefore criticizes Nöldeke’s presentation of the grammatical errors of the Qur’an, since in his view there was not yet any standard from which to deviate. Classical Arabic was the new language, the colloquials were ancient. On the other hand, he holds against Vollers that there existed no classical Arabic (or poetic koiné) according to which an original colloquial Qur’an could have been adjusted. Instead, the Qur’an was written together with the development of that new language. As aforementioned, Wansbrough’s argument for the late codification of the Qur’an has not been widely accepted, but neither has it been universally rejected. Among its defenders is Yehuda Nevo, a controversial Israeli archaeologist who spent much of his career studying early Arabic inscriptions in the Negev desert. Unlike Wansbrough, who relied on literary analysis of the Qur’an and exegetical literature, Nevo based his theories on the religious terminology in those inscriptions. In his “Towards a Pre-History of Islam,”67 Nevo argues that the Arabic inscriptions found during his Negev expeditions, primarily at a site named Sede Boqer, reveal a progressive religious development – from an indeterminate monotheism to formal Islamic doctrine – during the first two Islamic centuries. The theories of Nevo, who died in 1992, were edited by Judith Koren and recently published in their Crossroads to Islam.68 Here, however, the scope of argument is still broader. Like Wansbrough, Nevo and Koren deeply mistrust the Islamic sources, which they believe were the product of a community creating, not recording, a history. Yet unlike Wansbrough, Nevo and Koren believe that the actual succession of events that led to the Qur’an and Islam might be reconstructed. This is the task of Crossroads to Islam. The key to its execution is material evidence: “We argue that postcontemporary sources cannot, per se, be accepted at face value, but must be checked against contemporary evidence. This evidence may include written accounts . . . . But even better are material remains from the period in question. A rock inscription presents no problems of transmission history.”69 In Part 1 the authors argue that the Islamic conquests were no conquests at all. The Byzantine emperors intentionally implemented a policy (begun even by their Roman predecessor Diocletian, r. 284–305) of gradual withdrawal from the eastern Provinces, placing Arab tribes in their absence to act initially as clients ( foederati) and ultimately as independent, but subservient, rulers. This policy extended to religious affairs, as the Byzantines encouraged heterodox (especially monophysite but also East Syrian/Nestorian) Christianity in those provinces, to ensure that they would not leave the faithful behind. Heraclius (r. 610–41), meanwhile, imposed the doctrine of Monotheletism so that when it was revoked after the planned withdrawal (at the Sixth Ecumenical Council, 680), even the 13 GABRIEL SAID REYNOLDS heretofore orthodox Christians in the abandoned Provinces would now likewise be heretics. In Part 2 Nevo and Koren turn to the rise of the Umayyad Empire. Mu‘awiya, by their reading, stands as the first historical ruler of the Arab Empire. But quite unlike his portrait in the Islamic sources, Mu‘awiya was no more than the warlord, or strongman, who emerged triumphant among the Arab foederati in Syria. Muhammad and the first four caliphs, on the other hand, are characters of myth. As proof for this contention the authors turn to the early Christian sources, pointing out that neither Muhammad (once the reference to a prophet in the Doctrina Jacobi is explained away), nor the rightly guided caliphs nor any of the famous battles of the Islamic futuh works appear in them. They also call as witnesses the early Arab coins, noting that the earliest coins (i.e. pre-‘Abd al-Malik) still have Byzantine iconography, reflecting the Arabs’ continued clientage to Byzantium. Only with ‘Abd al-Malik (r. 65/685–86/705) does that imagery disappear, and the name of Muhammad appear. It is Muhammad and the Qur’an that Nevo and Koren finally address directly in Part 3. They contend that until the time of ‘Abd al-Malik (and, not coincidentally, the Sixth Ecumenical Council), Arab religion was of two kinds. The general populace, and especially the Bedouins, continued to practice paganism; Sede Boqer was itself a pagan devotional site even through the time of Hisham (r. 105/724–125/743, although this contention has been widely rejected by other archaeologists). The elite, meanwhile, held to a sort of primitive monotheism, perhaps a sort of Abrahamism or Judaeo-Christianity. Yet with their increased separation from Byzantium by the time of ‘Abd al-Malik the Arabs needed their own national history, and their own prophet, for which reason the name Muhammad begins to appear on coins and on the Dome of the Rock. The Qur’an, however, was not fully codified in the Umayyad period.70 The ‘Abbasids were moved to create a new scripture in part due to the Arab religion’s increasing incompatibility with Christianity, but above all due to the development of the legal code. In place of the Byzantine based law code of the Umayyads, the ‘Abbasids developed the system of shari‘a now known which, while preserving certain elements of precedent, was theoretically rooted in the idea of exegesis and scholarly consensus (ijma‘), not unlike Rabbinic Jewish jurisprudence. Exegesis, however, requires a formalized scripture. Thus Muslim scholars codified certain homiletic sayings, or “prophetical logia” (as Wansbrough puts it), that had been circulating, and canonized them as the Word of God. Here Nevo and Koren are evidently following Wansbrough, who commented, “The employment of scriptural shawahid in halakhic controversy required a fixed and unambiguous text of revelation . . . . The result was the Quranic canon.”71 This process, they conclude, must have been completed by the time Ibn Hisham (d. 218/833) composed his adaptation of Ibn Ishaq’s (d. 150/767) sira, in light of the preponderance of Qur’an-based (historicized) narratives therein.72 The implications of Wansbrough’s and Nevo’s theories on the historical development of the Qur’an are thus fundamentally opposed to that of Lüling. 14 INTRODUCTION Whereas Lüling argues that the genesis of the Qur’an occurred much earlier than is usually thought, with a heretofore unknown “Ur-Qur’an,” Wansbrough and Nevo argue that this genesis occurred much later, with the collection of prophetical logia under the ‘Abbasids. With a scholar known under the pseudonym Christoph Luxenberg, the scene is again set back in time. In his Die syro-aramäische Lesart des Koran; Ein Beitrag zur Entschlüsselung der Koransprache,73 Luxenberg, like Lüling, turns to the text of the Qur’an and finds indications of earlier Christian writings. Unlike Lüling’s, however, Luxenberg’s reading of the text is not done in the shadow of a grand religious vision. His reading is also more controlled. If Lüling aims for a reformation of Islam, Luxenberg aims for an Entschlüsselung, a “decipherment,” of the Qur’anic language.74 If Luxenberg’s method is at all affected by a religious vision, it is simply the conviction that Syriac Christianity was important to the development of the Qur’an. This is hardly unprecedented, as Alphonse Mingana,75 Arthur Jeffery76 and Tor Andrae77 were all of the same opinion. More recently Sidney Griffith (one of the contributors to the present volume), building on the work of Georg Graf, has contributed new evidence to this question, by arguing convincingly that the Bible had not been fully translated into Arabic before the rise of Islam.78 This would mean that many Arabic speaking Christians relied on Aramaic/Syriac scripture at the time of Islam’s origins. One might expect, therefore, that the first “Arabic” scripture, the Qur’an, would in fact be in an “aramäisch-arabische Mischsprache,” as Luxenberg puts it.79 Perhaps what the Qur’an means when it describes itself as ‘arabiyyun mubin (16:103, 26:195) “clear [or, better, ‘demonstrative’] Arabic” is not that it is in pure, classical Arabic, but that it is in the language of its people. Perhaps this phrase reflects the text’s pride at being the first effort at an Arabic scripture. Luxenberg, meanwhile, never seeks to work out the historical plausibility of his syro-aramäische Lesart, but instead limits himself to emending the received text of the Qur’an. While Luxenberg’s opponents often criticize this point, it seems to me quite justifiable, even prudent, in light of our uncertain historical knowledge of Islamic origins. Of course, most of Luxenberg’s opponents have more confidence in the historical reliability of the master narrative of Islamic origins than he does. In any case, Luxenberg argues that his emendations produce a fundamentally improved text: If the above philologically underpinned analysis has demonstrated that on the basis of both philological and objective criteria the Koran text has been misread and misinterpreted to a degree hitherto considered unimaginable, then the inevitable consequence is the need for a fundamentally new reading of the Koran. The findings of the present study have created the prerequisites for such a reading. (Luxenberg, The Syro-Aramaic Reading of the Koran, 332–3) 15 GABRIEL SAID REYNOLDS Not everyone agreed with Luxenberg’s claim. In fact, his work provoked an unusually polemical reaction from some. François de Blois concludes that Luxenberg’s new reading of the Qur’an, “is a reading that is potentially attractive only in its novelty, or shall I say its perversity, not in that it sheds any light on the meaning of the book or on the history of Islam.”80 De Blois, a highly competent philologist, has weighty criticisms to a few chosen examples of Luxenberg’s method. Yet the reader of his review is still left dazed by his palpable hostility to this book. Indeed, other competent philologists find Die Syro-Aramäische Lesart to be a scholarly book with some proposals of great consequence for our understanding of the Qur’an.81 Part of this hostility against Luxenberg may be due to the amount of attention that his work has received in the popular press. It has been featured in leading newspapers in North America and Europe, twice in the New York Times (once on the front page).82 Moreover, the attraction of the popular press to Luxenberg’s book is closely related to his argument that the Qur’anic term hur should be seen not as a reference to the virgins of paradise, but rather, in light of Aramaic hur (“white”), to grapes, the fruit par excellence of the heavenly garden;83 one article in a German magazine was published with the rhyming title “Weintrauben statt Jungfrauen” (“Grapes instead of Virgins”).84 By focusing on this argument, moreover, some journalists have found a medium by which to dramatize Islamic teachings of jihad and the sexual rewards of paradise, themes long the focus of anti-Islamic polemic. One journalist asks whether “martyrs arriving in paradise may regard a bunch of grapes as a letdown.”85 Scholars such as de Blois, on the other hand, have attempted to draw attention to weaknesses of Luxenberg’s method, two of which are salient. First, in Die syro-aramäische Lesart, Luxenberg consults very few sources. The only Muslim exegete whom he cites consistently is Abu Ja‘far al-Tabari (d. 310/923). Meanwhile, although in the Introduction Luxenberg surveys earlier critical studies of the foreign vocabulary of the Qur’an – including those by Geiger, Fraenkel, Nöldeke, Horovitz, Mingana and Jeffery – he seldom integrates that work into his analysis. Second, in explaining the manner in which Qur’anic vocabulary emerged from Syriac, Luxenberg turns from orthography to phonology and back again, according to the exigencies of any particular argument. This not only seems too convenient, it suggests that Luxenberg does not have a coherent vision of how Syriac exercised such a profound influence on the Qur’an. One might add a third criticism: Luxenberg’s use of Syriac is largely based on modern dictionaries, especially Payne-Smith’s Thesaurus Syriacus (1879–1901), Manna’s Vocabulaire Chaldéen-Arabe (1900) and Brockelmann’s Lexicon Syriacum (1928). While some definitions therein are based on early Syriac texts (and thus conceivably could explain the original meaning of Qur’anic words), others are based on later Syriac (and could even be influenced by Arabic); only Payne-Smith consistently provides citations that could act as safeguards against this trap. On the other hand, Luxenberg’s method is not entirely fruitless. A number of his re-readings are intellectually compelling. For example, in the Cairo text 16 INTRODUCTION Qur’an 37:103, which relates to Abraham’s sacrifice of his son, reads wa-tallahu li-l-jabin, something like “he laid him at (to? by?) the forehead.” Luxenberg, skeptical of this awkward reading, interprets talla, in light of Syriac tla, as “bind” and re-reads jabin (a hapax legomenon) as Syriac habbin, “firewood” (which can be done without altering the Arabic rasm). A much more satisfying phrase emerges: “He tied him to the firewood.”86 More recently, Luxenberg has proposed a re-reading of sura 97 (al-Qadr).87 Most scholars continue to follow the traditional opinion that the first verse of this sura (inna anzalnahu fi laylati l-qadr) refers to the descent of the entire Qur’an on one night (even if this opinion contradicts the doctrine of asbab al-nuzul, according to which the Qur’an was revealed piece by piece, not all at once). Luxenberg agrees that this verse is referring to the descent of the Word of God, but argues that Christ, not the Qur’an, is intended. In fact, the vocabulary of this sura is strongly reminiscent of the Biblical description of the birth of Christ. The terms “night,” “angels,” “spirit” and “peace” all appear in this five-verse sura. Meanwhile, Luxenberg argues that “shahr” of verse three is better read sahr, corresponding to Syriac shahra, or “night vigil.”88 The verse comes to mean, idiomatically, “Christmas night is better than a thousand night vigils.” Needless to say, this interpretation of surat al-Qadr is profoundly different from the traditional reading. Such re-readings of the Qur’anic text will certainly fail to convince scholars who believe that the Qur’an must be read through the lens of Islamic biography and exegesis. Luxenberg’s method might appear cavalier, as he ignores centuries of scholarly tradition. Indeed, many of Luxenberg’s critics are quick to insinuate that his work is vulgar.89 To scholars who see the Qur’an within its Late Antique context, on the other hand, Luxenberg’s method is, at least, worthy of consideration. His work should not be cast aside on ideological grounds. Instead, each of his proposed re-readings should be considered individually. Importance of the present book It is precisely the relationship between the Qur’an and Late Antiquity that was at the heart of the 2005 Qur’an conference at the University of Notre Dame. In recent decades most scholars of Islamic Studies have neglected this relationship, or ignored it altogether. The authors of most works in the field continue to observe the Qur’an from what may be called a retrospective viewpoint. They look back at the Qur’an from the perspective of medieval Islamic scholars. To put it in another manner, Qur’an studies and tafsir studies have gotten all mixed up. This trend, it seems, is increasing. It is salient in recent introductions to the Qur’an,90 and is evident from the amazing fact that most graduate students in Qur’anic Studies are taught to study the Qur’an with Arabic alone; no longer are they expected to study classical or other Semitic languages. The work of the Revisionist scholars surveyed earlier can be seen as a reaction to that trend. It is perhaps an over-reaction. If these scholars have demonstrated just 17 GABRIEL SAID REYNOLDS how easily the master narrative of the Qur’an can be cast aside, they nevertheless have not succeeded in writing a coherent counter narrative. Indeed, as aforementioned, on some points the models of the Revisionists are fully incompatible. The profound role that Luxenberg sees for Syriac appears nowhere in the historical models of Lüling (although it is noteworthy that both scholars independently perceive a Christian subtext in the Qur’an). Meanwhile, the historical model of Lüling is fully at odds with that of Nevo. That of Lüling is based primarily on pre-seventh century Meccan Christianity, that of Nevo is centered on paganism and indeterminate monotheism of seventh/eight century Palestine and Syria. Finally, the work of Wansbrough implies that the very idea of reconstructing the history of Islamic origins is misguided.91 The work of Revisionist scholars, it might be added, has led to increased polemic in the field of Qur’anic Studies. Those who entertain the views of Revisionists are accused (ironically) of being unrepentant Orientalists (in the worst sense of the word). Those who reject their views are accused of being dogmatic and gullible. Part of the controversy surrounding the work of Revisionists is due to the fact that, excepting Wansbrough, these scholars have largely worked independently of the academic community. They appear to be mavericks. Yet one might counter that the dominance of the master narrative of Islamic origins is in part responsible for thier isolation. Meanwhile, the confusion they have generated might ultimately benifit Qur’anic Studies, if it leads future scholars to address the profound questions yet to be resolved in the field. In any case, with each passing year the need for a new movement in Qur’anic Studies, a movement at once innovative and cooperative, becomes more evident. It was in response to this need that we organized the 2005 Qur’an conference at Notre Dame. On one hand, the conference called for a critical evaluation of the issues raised by Revisionist scholars, an evaluation according to the theories and arguments of those scholars, not their supposed agendas. On the other hand, it encouraged scholars to develop studies on the Qur’an from a contextual – not a retrospective – viewpoint; in other words, to approach the Qur’an in light of the Late Antique context in which it arose. This goal of the conference, then, was to develop a coordinated and promising movement to confront the “hoffnungsloses Chaos” of Qur’anic Studies.92 The present book is the fruit of the Notre Dame conference. The articles in Part 1 all confront, in unique ways, the great problem that has been posed by the recent controversies in Qur’anic Studies: the integration of philology and history. If Luxenberg purposefully avoids an historical consideration of his philological theories, Profs. Donner, Hoyland, Böwering, Gilliot and Griffith take up that task here, and more. Prof. Donner provides an original evaluation of the challenges facing critical Qur’an Studies, as he looks back through the history of the field and forward into its future. Prof. Hoyland introduces what is perhaps the most original and promising material evidence for the Qur’an’s origins, early Arabic inscriptions, and analyzes its relationship to current scholarly theories. Prof. Böwering assesses the ultimate worth not only of the Luxenberg and 18 INTRODUCTION Lüling theories, but also of the idiosyncratic theories of Joseph Azzi, a provocative Lebanese scholar little known outside of the Middle East. The question of the historical plausibility of Luxenberg’s theory is again taken up by Prof. Gilliot, who illuminates evidence within the Arabic Islamic tradition that suggests the Qur’an was not the work of one author. Prof. Griffith, meanwhile, focuses on the Syriac Christian tradition, particularly on the Syriac antecedents to the Cave narrative of Q 18. Thereby he argues that while the Qur’an’s relationship with its linguistic and religious context is intimate, its meaning cannot be reduced to a Syriac subtext. In Part 2 the authors focus on the Qur’an’s relationship with that context, above all with Christianity. In this, too, the present book is exceptional. As mentioned before, the great majority of publications, particularly in recent decades, contain a retrospective viewpoint. Here, however, scholars use their knowledge of regional history, religion and language to provide a compelling contextual perspective on Islamic origins. Prof. Samir describes how Christian theology – and not simply narrative or nomenclature – is reflected in the Qur’an in unexpected ways. The Qur’an’s relationship to Christian religious tradition is again addressed by Prof. Mourad, who shows how canonical and non-canonical traditions on Mary are present in the Qur’an. Prof. Van Bladel turns to a second narrative in Q 18, that of Dhu l-Qarnayn. He demonstrates in detail how the Qur’an alludes to the story of Alexander the Great as told in an early seventh century Syriac Christian text, while describing the particular historical circumstances in which this text may have been received by early Muslims. Prof. Kropp, meanwhile, elucidates how other Qur’anic pericopes reflect particular religious traditions of Ethiopic Christianity. Finally, Prof. Saadi considers how the earliest Syriac Christian reports of Islam might shed light on our understanding of the genesis of the religion and its scripture. The final two articles, which make up Part 3 in the present book, ask the reader to consider the importance of critical study of the Qur’an. Prof. Stewart turns to medieval Muslim scholarship to show that speculation over emendation of the Qur’anic text has an important precedent within the tradition. He argues that difficulties in the text calls the reader to emendation, while proposing that the test of validity for such emendation can be found in the Mu’tazili phrase, sukun al-nafs, “ease of mind.” Prof. Rippin, meanwhile, looks back to the Islamic tradition of investigating the foreign origin of Qur’anic terms (the so-called kalimat dakhila), showing that the work of critical scholars today is not per se novel. Yet he also suggests that contemporary attempts to uncover the original context of the Qur’an are no less speculative than their medieval precedents, as “faith in the historical record rivals faith in the divine.” The present book, therefore, is designed to be a robust yet sober challenge to the field of Qur’anic Studies. It is a book that honors the Qur’an by appreciating its complexity. Hopefully it will encourage others to take up this same task. Thus while the great Qur’an project of Bergsträsser, Jeffery and Pretzl came to a premature end, perhaps with this book a new project is just beginning. 19 GABRIEL SAID REYNOLDS Notes 1 I am obliged to Professors Manfred Kropp, Andrew Rippin, Christopher Melchert and Patricia Crone for their insightful and judicious comments on this article. 2 This was the not the first printed edition of the Qur’an, which was instead that commissioned by Muhammad ‘Ali in Egypt in 1833, in the face of great opposition of religious scholars who argued that the mechanical processes of printing were inappropriate for the Word of God. On this see M. Albin, “Printing of the Qur’an,” EQ, 4, 269–70. 3 For the names of the other principal scholars, a precise description of the two original editions of the mushaf and details regarding the work of the scholarly committee, see G. Bergsträsser, “Koranlesung in Kairo,” Der Islam 20, 1932, 3–4. 4 Haddad was known for his response to Taha Husayn’s Fi l-shi‘r al-jahili, in which he argues that the contemporary reading of the Qur’an is precisely that of the Prophet Muhammad, and cannot be attributed to the various Arab tribes. See Haddad’s al-Suyuf al-sahiqa, Cairo: Matba‘at al-Ma‘ahid, 1344, and the biographical tribute to him by Bergsträsser: “Koranlesung in Kairo,” 13–23. 5 See Albin, “Printing of the Qur’an,” EQ, 4, 272. 6 “Es ist eine Merkwürdigkeit dieses einzigartigen Buches, daß sich auch diese noch belassene Freiheit weiter eingeschränkt hat, so daß heute fast nur mehr die Lesung des Hafs ‘an ‘Asim bekannt und gebraucht ist.” O. Pretzl, “Aufgaben und Ziele der Koranforschung,” Actes du XXe Congrès international des orientalistes, Bruxelles 5–10 septembre 1938, Louvain: Bureaux du muséon, 1940, 328–9. Reprinted in R. Paret (ed.), Der Koran, Wege der Forschung 326, Darmstadt: Wissenschaftliche Buchgesellschaft, 1975, 411–12. 7 In fact it is unclear whether Ibn Mujahid ever applied this hadith to his theory. See Ch. Melchert, “Ibn Mujahid and the establishment of seven Qur’anic readings,” SI 91, 2000, 5–22. 8 Bergsträsser, “Koranlesung in Kairo,” 4. The third edition is dated 1347 (1928/9). 9 For a list of the principal works, see Bergsträsser, “Koranlesung in Kairo,” 5–10. 10 Bergsträsser was deeply impressed by the Cairo edition, commenting that there “ragt eine alt-islamische Wissenschaft lebenskräftig und leistungsfähig in unsere Tage herein; er ist ein Dokument für den überraschend hohen gegenwärtigen Stand der ägyptischen Koranlesungswissenschaft.” “Koranlesung in Kairo,” 10. 11 Bergsträsser notes that his colleague (and successor) Otto Pretzl provided him with a list of variants between the orthography of the Cairo text and that recommended by Abu ‘Amr al-Dani (d. 444/1053) in his Muqni‘. “Koranlesung in Kairo,” 11, n. 1. 12 Bergsträsser, “Koranlesung in Kairo,” 32. 13 See G. Bergsträsser, “Plan eines Apparatus Criticus zum Qoran,” Sitzungsberichte der Bayerischen Akademie der Wissenschaften, January 1930, vol. 7, 3ff. Reprinted in Der Koran, Wege der Forschung vol. 326, 389–97. Cf. “Über die Notwendigkeit und Möglichkeit einer kritischen Koranausgabe,” ZDMG 84, 1930, 82–3. 14 ZDMG 84 (1930). 15 According to the obituary notice in Der Islam, Bergsträsser died near Watzmann, in the mountains just north of the Austrian border, along the Berchtesgaden national park. 16 Thiersch’s statement is unpublished. It is marked with a stamp from the Frankfurt police and is filed in the Orientalisches Seminar der J-W. Goethe-Universität Frankfurt am Main. The statement opens with the header “München, 19 August 1933” and ends with the footer “München den 22 August 1933.” 17 According to Thiersch, only hours before his own accident Bergsträsser was discussing mountain accidents with him and a third young companion named Strobl, while they took a break over coffee and Butterbrot. 20 INTRODUCTION 18 See Muhammad Hamdi al-Bakri, Usul naqd al-nusus wa-nashr al-kutub, muhadarat al-mustashriq al-almani Birjstrasir, Cairo: Matba‘at Dar al-Kutub, 1969. Bergsträsser was a guest professor at the University of Cairo during the academic year 1931/2. The Egyptian/German Egyptologist Makram Rizq quotes Bakri’s description of the incident in his brief fascicle entitled “Freundschaftsmord unter Orientalisten im August 1932” (both Bakri and Rizq place Bergstrasser’s death in 1932). Rizq agrees with Bakri’s view of Bergsträsser’s death. He comments that “die damaligen ‘Neo-Nazis’ für ihn einen heimtückischen Mordplan entworfen haben. Diesem freundlich-kannibalischen Plan nach, durfte die Liquidierung von Bergsträsser nicht nach Mord riechen, sondern nach Zufall-Tod, nach Unfall-Zufall. Dafür fand man auch gleich einen ZufallOrientalisten unter den Studenten von Bergsträsser selbst.” As with Bakri, Rizq does not name the student he accuses of Bergsträsser’s murder. As far as I can tell, this fascicle was self-published and distributed personally by Dr Rizq. 19 G. Lüling comments, “Awful as it is to say it, German Arabistics and Islamics, together generally labeled ‘Semitic Philology’, were peculiarly vulnerable to the all-permeating personnel policy of the Third Reich.” “Scholarly criticism of the Koran,” The Journal of Higher Criticism 3, Spring 1996, 1, 79. Lüling continues by describing his research into the Nazi files of German Semiticists, which – according to his account – he was allowed to view on the condition that he would omit the names of those associated with that regime. He comments: “More or less everyone had been involved,” (p. 80), and gives the names of only two scholars who were not: Walther Braune and Hans Heinrich Schaeder, both of Berlin. Meanwhile, Lüling describes Anton Spitaler of Munich as the “epigon” (p. 95) of Nöldeke, a scholar whom Lüling accuses of anti-semitism (p. 78). 20 See O. Pretzl, “Die Fortführung des Apparatus Criticus zum Koran,” Sitzungsberichte der Bayerischen Akademie der Wissenschaften, 1934, vol. 2. 21 Leiden: Brill, 1937. 22 A. Jeffery, “The textual history of the Qur’an,” Journal of the Middle East Society (Jerusalem) 1, 1947, 35–49; reprint: The Qur’an as Scripture, New York: Moore, 1952, 89–103; cf. Idem, “The present status of Qur’anic studies,” Middle East Institute: Report of Current Research, Spring 1957, 2. 23 Jeffery provides a detailed assessment of work on the Qur’an through 1957 – including the publications related to the Bergsträsser project – in his “The present state of Qur’anic studies.” 24 “Das Koranarchiv der Bayerischen Akademie der Wissenschaften besaß Fotokopien der beiden Hss., sie wurden mit dem größten Teil der Bestände des Archivs durch Kriegseinwirkung vernichtet. Dadurch wurde die Weiterführung des geplanten Apparatus Criticus zum Koran schon von den Unterlagen her unmöglich gemacht.” Spitaler, “Die Nichtkanonischen Koranlesarten und ihre Bedeutung für die arabische Sprachwissenschaft,” in a note added in 1972 to the reprint of this brief statement, Der Koran, Wege der Forschung 326, 413, n. 2. The original statement is printed in Actes du XXe Congrès international des orientalistes, bruxelles 5–10 septembre 1938, Louvain: Bureaux du muséon, 1940, 314–15. Cf. A. Neuwirth, Grundriß der arabischen Philologie, in W. Fischer and H. Gätje (eds), Wiesbaden: Reichert, 1982–92, 2, 112. 25 I understand that a project on the text of the Qur’an is currently being planned at the Freie-Universität Berlin. In November 2005 a seminar was held there during which a model format for the presentation of a Qur’anic sura with variants was proposed. 26 Lüling believes that Spitaler kept the microfilms secret with the intention of writing a definitive work on the text of the Qur’an, “what in the end he was obviously incapable of achieving.” See A Challenge to Islam for Reformation, Delhi: Molital Banarsidass, 2003, xxi, n. 8. 27 “Was in der gedruckten Fassung kaum oder gar nicht zum Ausdruck kommt, deutete er in mündlichen Unterhaltungen gelegentlich vage an, nämlich die Zweifel, die ihm mit 21 GABRIEL SAID REYNOLDS 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 der Zeit durch den Zuwachs und die fortschreitende Kenntnis des Materials immer häufiger und stärker kamen, ob die Koranhandschriften wirklich, wie Bergsträsser auf Grund der spärlichen ihm zur Verfügung stehenden Unterlagen geglaubt hatte, entscheidend dazu beitragen könnten, eine authentische älteste Gestalt des Korantextes herzustellen.” A. Spitaler, “Otto Pretzl,” ZDMG 96, 1942, 163–4; cf. The similar statement of August Fischer, “Grammatisch schwierige Schwur- und Beschwörungsformeln,” Der Islam 28, 1948, (1–105) 5, n.4. Pretzl, “Aufgaben und Ziele der Koranforschung,” 328. Ibid., 329. Ibid., 329. Ibid., 329. A. Jeffery, Materials for the History of the Qur’an. The Old Codices, Leiden: Brill, 1937, 4. Writing in 1910, I. Goldziher draws attention to Rudolf Geyer’s review of K. Voller’s Volkssprache und Schriftsprache im alten Arabien, in which Geyer pleads for “a real critical edition of the text.” See I. Goldziher, Introduction to Islamic Theology and Law (Vorlesungen über den Islam), trans. A. and R. Hamori, Princeton, NJ: Princeton University Press, 1981, 29, n. 37 and R. Geyer, “Review of K. Vollers, Volkssprache und Schriftsprache im alten Arabien,” Göttingische Gelehrte Anzeigen 171, 1909, 51. Mu‘jam al-qira’at al-qur’aniyya, ‘Abd al-‘Al Mukarram and Ahmad ‘Umar (eds), Kuwait: Dhat al-Salasil, 1403/1983. T. Lester, “What is the Koran?” The Atlantic Monthly 283, January 1999, 1, 43–56. The most complete description of the fragments yet published is in H.-C. Graf von Bothmer, K.-H. Ohlig and G.-R. Puin, “Neue Wege der Koranforschung,” Magazin Forschung 1, 1999, 33–46. See also G.-R. Puin, “Observations on early Qur’an manuscripts in San‘a’,” in S. Wild (ed.) The Qur’an as Text, Leiden: Brill, 1996, 107–11. More information on the San‘a’ manuscripts, including photos, can be found at www.unesco.org/webworld/mdm/visite/sanaa/en/present1.html. “Neue Wege der Koranforschung,” 46. Ibid., 39. Other evidence, however, suggests that ya’ is not a mater lectionis in these two names. Pre-Islamic Arabic inscriptions of Jabal Usays (usually dated to the sixth century AD), which appear not to represent a at all, contain the ya’ in Abraham’s name, suggesting that Arabs pronounced the last syllable thereof “him” in the pre-Islamic period. Meanwhile, M. Kropp argues, through an Ethiopic etymology, that Arabs pronounced the first syllable of shaytan as a diphthong from an early time. See his “Der äthiopische Satan ⫽ Maitan und seine koranischen Ausläufer mit einer Bemerkung über verbales Steinigen,” OC 89, 2005, 93–102. “Wenn aber innerkoranische Parallelen für eine Formulierung fehlen, wenn der Sinn dunkel ist, wenn die traditionellen Exegeten für eine Stelle drei, zehn oder zwanzig Interpretationen anbieten – kurz: wenn die Philologie unserer Zeit gefordert ist, dann ist es legitim, bei der Interpretation des Korantextes zurückzugehen bis auf das Schriftgerüst der frühesten Überlieferung: Die Korane von Sanaa!” “Neue Wege der Koranforschung,” 39. Sources de la transmission manuscrite du texte coranique, I. Les manuscrits de style hiæazi, le manuscrit arabe 328, in F. Déroche and S.N. Noseda (eds), Lesa, Italy: Fondazione F. Noja Noseda, 1998. Sources de la transmission manuscrite du texte coranique, I. Les manuscrits de style hiæazi, Le manuscrit Or. 2165 (f. 1 à 61), in F. Déroche and S.N. Noseda (eds), Lesa, Italy: Fondazione F. Noja Noseda, 2001. A. Neuwirth, “Zur Archäologie einer Heiligen Schrift: Überlegungen zum Koran vor seiner Kompilation,” in C. Burgmer (ed.), Streit um den Koran, Berlin: Schiler, 2004, 82. 22 INTRODUCTION 43 On this see N. Calder, “Tafsir from Tabari to Ibn Kathir: Problems in the description of a genre, illustrated with reference to the story of Abraham,” in G.R. Hawting and A.A. Shareef (eds), Approaches to the Qur’an, London: Routledge, 1993, 101–40. 44 GdQ1, 2, 3. An Arabic translation has recently appeared: Tiyyudur Nuldaka, Ta’rikh al-Qur’an, trans. G. Tamer, Beirut: Konrad-Adenauer Stiftung, 2004. 45 T. Nöldeke, “The Koran,” Encyclopaedia Britannica, 9th edition, London: Britannica 1891, 16, 597. Nöldeke’s article has recently been reprinted in Ibn Warraq (ed.), The Origins of the Koran, Amherst, NY: Prometheus, 1998, 36–63. 46 The following presentation of the work of Revisionist scholars might be compared with the sensationalist article of M. Bright: “The Great Koran Con Trick,” New Statesman 130, 12/10/01. Note also, in the following edition (131, 12/17/01), the angry responses of several scholars whom Bright quotes in his article. 47 Über den Ur-Qur’an: Ansätze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur’an, Erlangen: Lüling, 1974 (2nd edition, Erlangen: Lüling, 1993); Recently translated and expanded as A Challenge to Islam for Reformation; Die Wiederentdeckung des Propheten Muhammad: eine Kritik am “christlichen” Abendland, Erlangen: Lüling, 1981. 48 See A Challenge to Islam for Reformation, 73ff. It is worth noting that Luxenberg, while agreeing that the traditional Islamic historicization of surat al-‘Alaq is unreliable, arrives at a sharply contrasting interpretation of this sura, one example of the cacophony heard from Revisionist scholars. Luxenberg argues that sura 96 can be seen, with the insight of Syro-Aramaic, as a unified text written originally in the Christian context. It contains, not unlike the New Testament epistles, the exhortations of a Christian pastor (n.b. v. 19, “prostrate and approach,” which Luxenberg argues is an invitation to the Eucharist). See Die syro-aramäische Lesart des Koran, 2nd edition, Berlin: Schiler, 2002, 311–33 (1st edition, Berlin: Das arabische Buch, 2000; references later are to the second edition); English translation: The Syro-Aramaic Reading of the Koran: A Contribution to the Decoding of the Koran, Berlin: Schiler, 2007. 49 On this argument cf. G.R. Hawting, The Idea of Idolatry and the Emergence of Islam, Cambridge: Cambridge University Press, 1999. Hawting, however, argues that the religious idioms of the Qur’an more likely developed in Mesopotamia, not in the Hijaz. 50 Lüling argues that the spirit of liberal theology in Prussia is “predestined to lead to worldwide reconciliation” (A Challenge to Islam for Reformation, xxvii) and that Prussians are “part of the Old European bipartite tribal society which was originally Semite” (p. lii). 51 See the preface to A Challenge to Islam for Reformation and G. Lüling, “Preconditions for the Scholarly Criticism of the Koran and Islam, with Some Autobiographical Remarks,” Journal of Higher Criticism 3, Spring 1996, 1, 73–109. 52 Lüling himself laments “the dominance of Trinitarian Christian confessions in our theological faculties,” A Challenge to Islam for Reformation, 73. 53 For the latter point see his Die Wiederentdeckung des Propheten Muhammad and idem, “A New Paradigm for the Rise of Islam and Its Consequences for a New Paradigm of the History of Israel,” Journal of Higher Criticism 7, Spring 2000, 1, 23–53. 54 See Qur’anic Studies: Sources and Methods of Scriptural Interpretation, Oxford: Oxford University Press, 1977 (reprint: Amherst, NY: Prometheus, 2004); The Sectarian Milieu: Content and Composition of Islamic Salvation History, Oxford: Oxford University Press, 1978. 55 See, for example, I. Goldziher, “Neue Materialen zur Litteratur des Überlieferungswesens bei den Muhammedanern,” ZDMG 50, 1896, 465–506; idem, “Über die Entwicklung des hadith,” Muhammedanische Studien, Halle: Niemeyer, 1888–90, 2, 1–274. 56 Wansbrough, Qur’anic Studies, 202. 57 Ibid., 12. 23 GABRIEL SAID REYNOLDS 58 Elsewhere, Wansbrough speaks of the aspirations of Muslim scholars (here referring primarily to sira and tafsir) to develop not an eschatology, but a “protology,” a myth of original religious purity. Thus Islamic tradition looked to develop the beginning, not the end. See Wansbrough, The Sectarian Milieu, 147. 59 Wansbrough, Qur’anic Studies, 50. 60 Ibid., 2, 21. 61 Berlin: De Gruyter, 1981. 62 Wansbrough, Qur’anic Studies, 225. 63 Ibid., 44. 64 Cf. Spitaler, “Die Nichtkanonischen Koranlesarten und ihre Bedeutung für die arabische Sprachwissenschaft,” 315. Reprinted in Der Koran, Wege der Forschung, 413–14; A. Fischer, 5. 65 See K. Vollers, Volksprache und Schriftsprache im alten Arabien, Strassburg: Trübner, 1906; T. Nöldeke, Neue Beiträge zur semitischen Sprachwissenschaft, Strassburg: Trübner, 1910, 1–5. 66 Wansbrough, Qur’anic Studies, 87–117. 67 Y. Nevo, “Towards a pre-history of Islam,” JSAI 17, 1994, 108–41. 68 Y. Nevo and J. Koren, Crossroads to Islam: The Origins of the Arab Religion and the Arab State, Amherst, NY: Prometheus, 2003. 69 Nevo and Koren, Crossroads to Islam, 8. 70 “The evidence in general suggests that the Qur’an was canonized only under the ‘Abbasids. Thus the rock inscriptions initially preserve only non-Qur’anic locutions, including some with (to use Nabia Abbot’s phrase) ‘Qur’anic flavor.’ The tombstone inscriptions, which start in the AH 170s, likewise contain only a few Qur’anic verses, at least down to AH 200.” Ibid., 342–3. 71 Wansbrough, Qur’anic Studies, 208. 72 Nevo and Koren, Crossroads to Islam, 347–8. 73 English translation: The Syro-Aramaic Reading of the Koran: A Contribution to the Decoding of the Koran, Berlin: Schiler, 2007. 74 Luxenberg claims to limit himself to examination of those passages that puzzle the Muslim interpreters: “Nur dort, wo der Kontext offensichtlich unklar ist, wo die arabischen Korankommentatoren quasi am Ende ihres Arabisch sind . . . .” Die syro-aramäische Lesart des Koran, 24. This claim, however, is often belied by sections where Luxenberg addresses Qur’anic passages that most Muslim interpreters find quite evident, not only in regard to their meaning but also to the occasion of their revelation, such as ch. 96 (surat al-‘Alaq) (pp. 311–33). 75 A. Mingana, “Syriac influence on the style of the Kur’an,” Bulletin of the John Rylands Library 11, January 1928, 1, 77–98. Mingana argues (p. 80) that 30 percent of foreign vocabulary in the Qur’an comes from Ethiopic, Hebrew, Persian and Graeco-Roman languages combined, while 70 percent comes from Syriac/Aramaic alone. 76 In FV, Jeffery consistently returns to Syriac, or Christian Palestinian Aramaic, in tracing the etymology of Qur’anic terms; cf. D.S. Margoliouth, “Some additions to Jeffery’s Foreign Vocabulary,” JRAS, 1939, 53–61. 77 T. Andrae, Les origines de l’islam et le christianisme, trans. J. Roche, Paris: AdrienMaisonneuve, 1955; originally published in German as “Der Ursprung des Islams und das Christentum,” Kyrkshistorisk årsskrift 23, 1923, 149–206; 24, 1924, 213–25; 25, 1925, 45–112. 78 GCAL, 1:144; S. Griffith, “The Gospel in Arabic: An inquiry into its appearance in the first Abbasid century,” OC 69, 1985, 126–67. 79 Luxenberg, Die syro-aramäische Lesart des Koran, 334. 80 F. de Blois, Journal of Qur’anic Studies 5, 2003, 1, 97. 81 Compare, for example, the reviews of R.R. Phenix, Jr and C. Horn, Hugoye 6, January 2003, 1; R. Brague, “Le Coran: sortir du cercle?” Critique, April 2003, 232–51; 24 INTRODUCTION 82 83 84 85 86 87 88 89 90 91 92 J.J.G. Jansen, Bibliotheca Orientalis 60, May-August 2003, 478–80; C. Gilliot, “Méthodes et débats: Langue et Coran: une lecture syro-araméenne du coran,” Arabica 50, 2003, 3, 381–93; M.F.J. Baasten, Aramaic Studies 2, 2004, 2, 268–72 A limited selection thereof: M. Naggar, Neue Zürcher Zeitung, Feb 3, 2001; “Virgins? What Virgins?” The Guardian, Jan 12, 2002; A. Stille, “Scholars Scrutinize the Koran’s Origins,” New York Times, Mar 2, 2002; R.-P. Droit, “Et si les vierges célestes du Coran n’étaient que fruits blancs?” Le monde, May 5, 2003; “Challenging the Qur’an,” Newsweek, July 28, 2003; “Der Fuchs und die süßen Trauben des Paradieses,” Süddeutschen Zeitung, Feb 24, 2004, 15; L’espresso, no. 11, Mar 12, 2004; Tageszeitung, no. 7331, April 10, 2004, 21; N. Kristof, “Martyrs, Virgins and Grapes,” New York Times, Aug 4, 2004. Luxenberg, Die syro-aramäische Lesart des Koran, 254–94. Inamo 23–4, 2000, 66–72. Kristof, “Martyrs, Virgins and Grapes.” Luxenberg, Die syro-aramäische Lesart des Koran, 179–89. C. Luxenberg, “Nöel dans le Coran,” Enquétes sur l’Islam, Paris: Editions Desclée de Brouwer, 2004, 117–38; Cf. Luxenberg’s “Weihnachten im Koran,” a German version of the same argument: Streit um den Koran, 35–41. Cf. Luxenberg’s most recent contribution, an examination of the Umayyad era Qur’anic inscriptions in the Dome of the Rock: “Die frühe Islamgeschichte nach inschriftlichen und numismatischen Zeugnissen,” in K.-H. Ohlig and G.-R. Puin (eds), Die dunklen Anfänge: Neue Forschungen zur Entstehung und frühen Geschichte des Islam, Berlin: Schiler, 2005, 124–47. See Payne-Smith (ed.), Thesaurus Syriacus, Tomus I, Oxford: E Typographeo Clarendoniano, 1879, Tomus 2, 1901, 2, 4075. “His book is not a work of scholarship but of dilettantism.” De Blois, 97. Note the recent introductions to the Qur’an: N. Robinson, Discovering the Qur’an, Washington, DC: Georgetown University Press, 2004; M. Abu-Hamdiyya, The Qur’an: an Introduction, London: Routledge, 2000; M. Sells, Approaching the Qur’an, Ashland, OR: White Cloud, 1999; M. Abdel Halem, Understanding the Qur’an, London: Tauris, 1999; Cf. the approach of Michael Cook: The Koran: a Very Short Introduction, London: Oxford University Press, 2000. In his review of P. Crone and M. Cook’s Hagarism, Wansbrough comments, “My reservations here, and elsewhere in this first part of the book, turn upon what I take to be the authors’ methodological assumptions, of which the principal must be that a vocabulary of motives can be freely extrapolated from a discrete collection of literary stereotypes composed by alien and mostly hostile observers, and thereupon employed to describe, even interpret, not merely the overt behaviour but also the intellectual and spiritual development of the helpless and mostly innocent actors. Where even the sociologist fears to tread, the historian ought not with impunity be permitted to go.” BSOAS 41, 1978, 156. In this regard the Notre Dame conference is akin to that of a seminar convened on January 22–25, 2004, at the Institut für Islamwissenschaft of the Freie Universität in Berlin, under the leadership of Angelika Neuwirth. 25 Part 1 LINGUISTIC AND HISTORICAL EVIDENCE 1 THE QUR’AN IN RECENT SCHOLARSHIP Challenges and desiderata1 Fred M. Donner Introduction Qur’anic studies, as a field of academic research, appears today to be in a state of disarray. Those of us who study Islam’s origins have to admit collectively that we simply do not know some very basic things about the Qur’an – things so basic that the knowledge of them is usually taken for granted by scholars dealing with other texts. They include such questions as: How did the Qur’an originate? Where did it come from, and when did it first appear? How was it first written? In what kind of language was – is – it written? What form did it first take? Who constituted its first audience? How was it transmitted from one generation to another, especially in its early years? When, how, and by whom was it codified? Those familiar with the Qur’an and the scholarship on it will know that to ask even one of these questions immediately plunges us into realms of grave uncertainty and has the potential to spark intense debate. To put it another way, on these basic issues there is little consensus even among the well-trained scholars who work on them. I am not speaking here of the kind of routine difference of opinion or tension that exists in the study of all scriptural traditions between those who take that scripture as a source for their belief and life, and those who study it as a text on the basis of historical, literary, sociological, and theological analysis. Rather, I refer to the fact that so many scholars over the past century, despite deep learning, serious commitment to understand the Qur’an, and on the basis of sophisticated and subtle methods, nevertheless remain so lacking in consensus on these basic issues. This lack of consensus grew from seeds that were sown at the very beginning of scholarly examination of the Qur’an by Western scholars in the nineteenth century. For many years, however, the majority of Western scholars adopted a view of the Qur’an and its origins that followed in most of its details the view presented by the Islamic tradition itself. Although there were always a few Western scholars who expressed reservations about such traditional views, 29 FRED M. DONNER their attitudes constituted for more than a century nothing more than a muted, if persistent, minority position in Western scholarly circles. Given the difficulty of wringing reliable evidence about the seventh century out of our exiguous sources for Islam’s beginnings, many scholars who held skeptical views about the traditional stories of the Qur’an’s genesis seem to have preferred simply to hint at their misgivings and otherwise to pass over the question in silence; for however much one may have doubted the traditional view, the difficulty of actually assembling the evidence for a convincing alternative view was clear to everyone, and few felt encouraged to take on such a challenge. Most who did choose to challenge traditional views, moreover, focused on one particular aspect of the problem, rather than addressing comprehensively the question of the Qur’an’s genesis and development. For example, scholars such as Geiger, Andrae, Bell, and Torrey 2 looked at the question of the relationship between the Qur’an and the JudaeoChristian scriptural tradition; Vollers, Blachère, and Fück3 grappled with the question of what the language of the Qur’an originally was; and Bergsträsser, Pretzl, Jeffery, Beck, and Cook4 considered the question of the many variant readings about which the Islamic tradition tells us. It was not until the 1970s that a number of books appeared that questioned more bluntly and comprehensively the traditional view of the Qur’an’s origins and early development as a text. Although vastly different from one another, the works of Günter Lüling (1974), John Wansbrough (1977), and Patricia Crone and Michael Cook (1977) all posed fundamental challenges to the traditional vision of Islam’s origins, including the genesis of the Qur’an.5 Although each of these contributions met with considerable resistance at first, and none has provided us with a satisfactory alternative interpretation, they together forced scholars in the field of Qur’anic studies to confront the simplistic view derived ultimately from Islam’s own dogmas about its origins, and to open the door – albeit sometimes grudgingly – to the possibility of assembling a new, and perhaps radically different, understanding of how the Qur’an text came to be, by drawing on textual and other evidence more secure than that provided in the Islamic tradition’s own narratives. The field of Qur’anic studies was, then, in some ferment already from the early 1970s onward. This ferment was for the most part restricted to the limited circles of Western specialists in Qur’an studies (who were few) and did not have much impact on the broader public. The publication in 2000 of the first edition of Christoph Luxenberg’s book Die Syro-Aramäische Lesart des Koran,6 however, generated considerable controversy, even in popular circles, although the book is no more revolutionary than those just mentioned by Lüling, Wansbrough, Crone and Cook. Perhaps Luxenberg’s work made more waves among the public because it is presented in a manner that makes it relatively easy to understand what he is getting at. Lüling’s prose is also generally clear, but he frames his attempted reconstruction of the Qur’an in the context of complex discussions of theology that are difficult for those without deep theological training to follow. Crone and Cook’s Hagarism is convoluted in argument and not an easy read, even for those well-informed about early Islamic history. Wansbrough’s Qur’anic Studies, 30 THE QUR’AN IN RECENT SCHOLARSHIP in some ways the most radical of all, is so opaque in style that even native speakers of English are often unsure of what he meant to say; its implications and even its basic hypotheses have only begun to reach a wider audience through the work of numerous mufassirin, some of whom, such as our colleague Andrew Rippin, are widely respected scholars,7 some others of whom are decidedly not, such as the pseudonymous religious polemicist “Ibn Warraq” who seems to champion Wansbrough’s ideas in pursuit of his own personal religious agenda.8 In the following remarks, I will make a few observations on what I take to be five fundamental and interrelated questions about the Qur’an on which significant scholarly disagreement persists, to four of which, at least, Luxenberg’s Syro-Aramäische Lesart has contributed. Given the present state of our understanding of these issues, my comments will frequently take the form of raising further questions or pointing out inconsistencies, rather than proposing decisive conclusions. These observations will be followed by a few suggestions for things we might do to help us move scholarly Qur’anic studies to the next stage. Five questions 1 Can the Qur’an as we have it today be traced back to some kind of original version? That is, was there an “Ur-Qur’an” (to borrow the felicitous phrase first given prominence by Günter Lüling) that appeared in some well-defined and restricted historical context (i.e. a particular time and place), a “closed” text that stood as a precursor to today’s Qur’an? That there was such an “Ur-Qur’an” is, of course, the firm view of the Islamic tradition itself, which considers the Qur’an as we know it to be the exact, literal transcription of God’s word as revealed to Muhammad in the seventh century; in the tradition’s view, the Qur’an of today is, in fact, identical with the “Ur-Qur’an.” A long and distinguished series of Western Qur’an scholars has also subscribed to the view that there was an early textual archetype for the present Qur’an, although most broke with the traditional Islamic view to the extent that they considered the possibility, at least, that minor changes may have entered the text between its earliest days and its codification in the form of the Qur’an we know today. Nöldeke, Schwally, Bergsträsser, Bell, Beck, and more recently Watt, Neuwirth, Lüling, and Burton,9 despite profound differences on many points, all agreed that what we read today is derived in some way from a prototype text, an Ur-Qur’an, from the time of Muhammad (or even before, as Lüling has proposed). Luxenberg, too, shares this view. The main dissenter from this view has been Wansbrough, who with his followers (including Andrew Rippin and G. Hawting10) has argued that the Qur’an we have today does not go back to an early archetype, but rather represents the fruit of a long and slow process of crystallization spanning two centuries or more, during which the Qur’an as we know it was pieced together from disparate materials circulating in “the community.” Wansbrough explicitly rejected the possibility of 31 FRED M. DONNER knowing what, if anything, existed in the time of Muhammad (irrelevant to him in any case since he considered the Qur’an that eventually evolved to have emerged not only much later, but in Mesopotamia, not in Arabia). A more fundamental clash of views than this can hardly be imagined. Much of the evidence for the respective views, however, stems from debates over questions relating to the later transmission and/or canonization of the text, to which we shall return later.11 2 What was the nature of the original “Ur-Qur’an,” assuming it existed? One of the few points about which, as far as I know, all scholars are in agreement about the Qur’an concerns its nature: it is clearly intended as a source of religious and moral guidance for its audience. This is obvious in the Qur’an of today, and everyone also assumes that the “Ur-Qur’an,” even if it was in some ways different from today’s text, must also have shared this basic character as a morally and religiously didactic text. Beyond this most basic level, however, the question of the Qur’an’s nature – or, rather, of the nature of the “Ur-Qur’an” – breaks down into a number of subordinate questions that are not so readily resolved. First of all, there is disagreement on what the original religious message of the “Ur-Qur’an” may have been. Muslim tradition presents it resolutely as “Islam,” that is, it projects back to the time of Muhammad the fully developed creed that emerged probably in the second century AH, in all its details. Many Western scholars have also hewn to this line,12 but there has also been a persistent strand in Western scholarship presenting the “Ur-Qur’an” as preaching something other than Islam, or at least as having been profoundly influenced by other traditions. The main contenders have been, of course, Judaism and Christianity.13 Already in 1833, Abraham Geiger’s book Was hat Muhammad aus dem Judenthume ausgenommen? argued for formative Jewish influence on Muhammad and the Qur’an, and a century later C. C. Torrey’s The Jewish Foundations of Islam emphasized the Qur’an’s dependence on Jewish tradition so forcefully that one sometimes wonders that he did not claim outright that Muhammad had converted to Judaism.14 Crone and Cook, in Hagarism, speak of the first phase of the movement Muhammad began as “Judeo-Hagarism,” an alliance between Jews and “Arabs” (a use of modern nationalist terminology, incidentally, that I find very problematic).15 On the other hand, many other scholars have emphasized the importance of Christianity in shaping the Qur’an and its teachings. One of the most powerfully developed analyses was presented by Tor Andrae in his long series of articles entitled “Der Ursprung des Islam und das Christentum” (1923–5), in which he identified Christian influences of many kinds, particularly from Nestorians of Yemen, monophysites of Ethiopia, and especially from Syrian pietism. Andrae actually asks (but does not answer) the question, “Why did Muhammad not simply become a Christian, if he was so influenced by Christian teachings?”16 More recently, Günter Lüling argued in his Über den Ur-Qur’an (1974) that the 32 THE QUR’AN IN RECENT SCHOLARSHIP Qur’an originally consisted in part of Christian strophic hymns used as liturgy by the Christians of Mecca (the existence of whom had generally been denied in traditional scholarship). According to Lüling, Muhammad himself grew up as a member of one such Christian community (perhaps Ebionites or other JewishChristians?) before breaking with others who espoused trinitarian doctrines; he redacted their liturgical texts to expunge trinitarian concepts and form the Qur’an text we now know.17 Lüling’s method consisted in part of making small (and sometimes not so small) changes in the textus receptus of the Qur’an to “restore” the “Ur-Qur’anic” passages to their supposedly original Christian content. But to many this method seems capricious and guided by a desire to prove the hypothesis he is asserting. Nonetheless, his work is full of interesting insights of many kinds; it has, however, never received the kind of full and open examination it deserved. It was subjected to a conspiracy of silence in Germany for many years, and while some scholars outside of Germany took note of it, few were sufficiently comfortable with German to wrestle with Lüling’s often long-winded and complex argumentation. I have made my own contribution to the question of what the religious message of the “Ur-Qur’an” may have been.18 In my view, the Qur’an clearly refers to its target audience as mu’minun, “Believers” or “the Faithful” – not as Muslims – and seems to be preaching strict monotheism and rigorous observance of God’s revealed law as key components of being a Believer. The Qur’an also makes clear that “peoples of the book,” that is, Christians and Jews, if they are sufficiently righteous, may be counted among the Believers. That is, the Believers’ movement seems at first to have been a strictly monotheist but non-confessional pietistic movement that could, and did, include some Christians and Jews. In articulating this hypothesis, I attempted to draw implications and deductions from the text of the Qur’an itself, even if they run counter to the interpretation given to these passages by later Islamic tradition, which takes pains to disguise the early participation of Christians and Jews in the Believers’ movement out of which Islam crystallized during the Umayyad period. So my hypothesis falls somewhere between the Christian-influence and Jewish-influence schools, since in my view the influence of both religions (or maybe of Jewish Christianity, as numerous scholars have recently suggested)19 is likely to have been significant from a very early time. Luxenberg’s Die Syro-Aramäische Lesart clearly belongs to that strand of scholarship that focuses on Christian influences in the “Ur-Qur’an.” Like Lüling, (whom he nowhere mentions, although it is inconceivable that he did not know his work), Luxenberg adjusts or “corrects” the reading of numerous Qur’an passages to arrive at the “Ur-Qur’anic” version, but, as compared to Lüling, his method seems somewhat more systematic and objective: he first isolates passages that pose syntactic or other difficulties, and then considers whether reading them as transcriptions of Syriac words or phrases brings a better sense to the passage. In some cases he also resorts to the hypothesis that the earliest script in which the “Ur-Qur’an” was written employed writing conventions that were subsequently 33 FRED M. DONNER forgotten and misunderstood in later transmission. (We shall return to this question later.) The central theme of Luxenberg’s book is that the “Ur-Qur’an” contained whole passages not in Arabic, but rather in Syriac, but which later tradition either forgot or attempted to disguise by reading them as though they were Arabic. Luxenberg also argues that Syriac was the official language of the early Islamic state, and was replaced by Arabic only under ‘Abd al-Malik.20 We shall return later to the question of what language or languages the “Ur-Qur’an” may have been in. For the moment, we can simply note that, Luxenberg’s textbased hypothesis powerfully advances the claim that Christianity, particularly from Syria, may have been a key part of the background to the original religious message of the “Ur-Qur’an.” Another question that relates to the nature of the “Ur-Qur’an” – one that seems relatively tame by comparison to the preceding one – revolves around the role orality may have played in the formation of the original text. (By “orality,” I mean the possibility that the text originated as a recited oral performance, as opposed to a written composition. I am not referring to the question of the later oral transmission of the text, which will be considered later.) Neuwirth has recently noted that what she calls “the Qur’an’s intrinsic orality” is reflected in its style.21 On the basis of purely stylistic characteristics, it has been argued that the Qur’an is replete with both oral formulas and with folktales.22 These observations seem sensible enough, but a question that then needs to be resolved is how the orality of the “Ur-Qur’an” relates to those theories that emphasize the Qur’an as a written text, such as those of both Lüling and Luxenberg. Perhaps Daniel Madigan’s efforts to understand the meaning of the Qur’an’s use of the word kitab (normally, “book”) in a broader and looser manner will help pave the way here,23 although it appears that many scholars remain to be convinced by his argument.24 Or, do we wish to see the oral formulas as being characteristic of only parts of the Ur-Qur’an, while other parts reflect an origin rooted firmly in writing? And if so, might this imply that these different parts hail from different original contexts or communities? Such a hypothesis may seem seductive, but it cannot be held in the abstract: it could only be deemed plausible if some clear evidence is deduced from the text pointing convincingly to the different specific contexts or communities from which different parts of the text hailed. The question of orality might also be brought to bear in discussion of Wansbrough’s theory of the late stabilization of the Qur’an text. One of the pillars of his argument is the assertion that various passages of the Qur’an that relate the same information in very similar, but not quite identical, words and phrases reflect a single text that has evolved over a significant period of time. But, might such similar passages not just as cogently be viewed as transcripts of different oral recitations of the same story made in close succession, something like different recordings of a politician’s stump speech delivered numerous times over a few days or weeks? For, in such circumstances, one might expect to find exactly the characteristics that Wansbrough has latched onto in his analysis: several 34 THE QUR’AN IN RECENT SCHOLARSHIP passages that treat the same theme using similar, but not quite identical, verbal formulations.25 A third question relating to the nature of the “Ur-Qur’an” is whether it originated as a liturgical text. Numerous students, as diverse as Neuwirth26 and Wansbrough,27 allege the liturgical functions of various Qur’anic passages, and almost everyone who discusses the Qur’an notes that the very word Qur’an is derived from Syriac qeryana, “recitation, liturgy.” But if the “Ur-Qur’an” (or, to use a Wansbrough-friendly formulation, the eventual Qur’an and its presumed Vorlagen) arose to meet liturgical needs in the community, how can we explain the fact that the Islamic ritual prayer/salat requires strikingly little recitation of the Qur’an? Other than the fatiha, recitation of only one long or three short verses of the Qur’an is required, and selection of which verses to recite is left to the individual.28 Moreover, it is striking that Islam knows no liturgical calendar prescribing specific recitations during prayer for particular seasons of the year, such as is found in Christianity or Judaism. Certainly there is no lack of suitable episodes that one might have used as occasions for such specific liturgical readings, such as the hajj, Ramadan, laylat al-qadr, mi ‘raj, hijra, and so on. The only recitation required in ritual prayer is that of the fatiha, but many scholars (including some early Muslim ones) considered the fatiha not actually to be part of the Qur’an, but rather considered it a prayer that was added to the beginning of Qur’an codices. So the evidence seems to suggest not that the Qur’an originated as prayer liturgy, but rather that a few elements drawn from the prayer liturgy were used to embellish the Qur’an. The implication is that the Islamic prayer ritual and the Qur’an text, whatever it originally was, developed independently. 3 If an original “Ur-Qur’an” existed, what kind of language did it represent? And what was the relationship between the written text and that language? The debate over these questions goes back at least a century and began with observations about the orthography of the Qur’an. The language and orthography of the Qur’an of today seem to represent the result of a process of editing, in which the text of the “Ur-Qur’an” was increasingly supplied with markings to make it more readable. Early manuscripts of parts of the text – and therefore, one assumes, the “Ur-Qur’an” that presumably antedated our earliest surviving Qur’an fragments – are usually written in a script that provided only the bare skeleton of the text’s consonants (called the rasm). That is, the writing system was completely devoid of any markings for short vowels, provided indications for long vowels only intermittently, and lacked the diacritical marks that enable readers to distinguish from one another several different consonants that are written with the same basic letter form (such as j, deep h, and kh). Islamic tradition itself informs us, with what reliability we cannot be sure, that this early “defective” text was “improved” by ‘Abd al-Malik’s governor of Iraq, al-Hajjaj ibn Yusuf, late in the first century AH (ca. 700 CE), by adding various marks to the text to make it more unambiguous – such as diacritical dots to distinguish consonants written with the 35 FRED M. DONNER same basic letter-form, and alifs as mater lectionis for long a.29 Some specialists think, however, that the diacritics in Arabic may have developed first in Syria or Iraq before Islam, among Christians or Jews, under the influence of Aramaic or Syriac systems for marking diacritics, themselves beholden to Indian precursors.30 The defective script of the early text, and continuing uncertainty about how the script evolved, create profound difficulties for us in trying to determine just what kind of language the original text was intended to represent. Was it a purely literary vehicle existing only in written form and not intended to replicate actual speech, or did it attempt to capture the sound of an actual spoken text or language? Just a century ago, Karl Vollers proposed that the orthography of the Qur’an reflected the spoken dialect of Quraysh, which, he surmised, did not pronounce the glottal stop (hamz) except at the beginnings of words.31 Hence the word mu’min is written starting with m followed by w as marker for long u, because the glottal stop of classical Arabic mu’min was elided in Meccan dialect to produce the pronunciation mumin, which is how the word is written in the Qur’an (“mwmn”). This “Ur-Qur’anic” dialectal Qur’an, Vollers thought, was subsequently fitted out with vowels and diacritics, as well as with a mark (called hamza) to indicate the pronunciation of hamz, which was introduced into the recitation under the impact of eastern Arabian dialects, which still pronounced the hamz. It was also, Vollers thought, provided at this time with markings to indicate the case-endings or i ‘rab found in the poetic koiné or ‘arabiyya, a feature that continues to entertain and bring joy to all students of Arabic, right up until today. Others, such as F. Altheim und R. Stiehl, also detected in Qur’anic orthography evidence of dialect pronunciation; for example, they felt that Qur’anic words like salat, “prayer,” which is written with w in the rasm where one would expect long a, reflected a spoken vowel shift from a to o* that spread from Aramaic into Hijazi dialect.32 Several distinguished scholars (Fleisch, Blachère, Fück, and others) subsequently rejected Vollers’ views and argued that the Qur’an was not written in dialect at all, but rather in the poetic koiné of the bedouin, or ‘arabiyya.33 And it is well-known, after all, that the Qur’an itself announces that it is couched “in a clear Arabic tongue” (bi-lisanin ‘arabiyyin mubin).34 It seems to me, however, that this theory also presents us with difficulties. If we assume that the “Ur-Qur’an” was originally written to reflect the pronunciation of the ‘arabiyya or formal poetic idiom, why does the orthography of the rasm reflect a pronunciation that had no medial hamz, since medial hamz was a feature of the ‘arabiyya? And why does the orthography nowhere reflect the use of the i ‘rab or case endings, apparently a defining feature of the ‘arabiyya? Other inflected Semitic languages with equally defective writing systems, such as Epigraphic South Arabic, provided indications of these case endings when required, so we cannot claim that the practice was considered outlandish or even unfamiliar; on the contrary, one might expect such South Arabian usage to have been familiar in the Hijaz,35 so the absence of orthographic markers for case-endings in the Qur’an text becomes itself a strong indication that such endings were absent in the original language reflected by the texts. 36 THE QUR’AN IN RECENT SCHOLARSHIP In sum, the orthography of the Qur’an seems to reflect Meccan or Hijazi dialect, yet the text calls itself and is considered by most modern philologists to reflect some form of the ‘arabiyya or poetic formal language. It cannot be both, but how this paradox is to be resolved remains to be seen. Anton Spitaler argued that the orthography of some words in the Qur’an, notably salat with its puzzling w in the place of long a, reflect not dialect but the Aramaic (particularly Syriac) convention of writing the word – in other words, that salat represents what we might call a “loan-orthography,” if not a loan-word.36 Werner Diem has proposed that Qur’anic orthography reflects spoken dialect not directly, but rather through the intermediary of a presumed pre-Islamic Hijazi tradition of Arabic orthography.37 There are a few pre-Islamic inscriptions that seem to follow the orthographic conventions we find in some early Qur’an texts, notably some from Jabal Usays in the Syrian steppe southeast of Damascus,38 but other pre-Islamic Arabic inscriptions (such as those from Umm al-Jimal, Jabal Ramm, Namara, and the many Safaitic inscriptions) vary significantly from one another in both script and orthography, so the vigor of this presumed “tradition” of Hijazi writing, which in any case is not yet attested for the Hijaz, remains uncertain.39 Our understanding of these matters of Qur’anic orthography may be clarified by close examination of the most important find in Qur’anic studies of the twentieth century, the trove of early Qur’an fragments discovered in the 1970s hidden in the ceiling of the great mosque of San‘a’, Yemen, during the course of restoration work. A general description of the finds and some preliminary remarks on the orthography of these early Qur’an manuscripts have been published by those who have seen them, but this crucial body of material still awaits full study.40 In the meantime, Luxenberg has made two proposals that relate to the question of how the text was first written. First, he argues that the “Ur-Qur’an” sometimes used a single “tooth” as mater lectionis not (or not only) for long i, as it does today, but to indicate long a, rather than the alif that later was introduced to mark it; but knowledge of this early convention was subsequently lost, so that the “tooth” was then mistakenly pointed to represent y, t, or another letter that the “tooth” could represent. The simplest example of this to grasp is the Arabic form of the name of Abraham, Ibrahim, normally written in the rasm without any indication of the long a and with a y (or long i) in the final syllable. This pronunciation diverges from the way the name is pronounced in all other Semitic languages, which always have a long a in the final syllable. Luxenberg argues that in this case, as in many others, y was intended to serve as mater lectionis for long a, so that the name was actually written to render a pronunciation akin to that known from Hebrew and other languages, that is, “Abraham.” He adduces a number of other instances where, he claims, a “tooth” originally intended as a y to signal long a was misread by people who had lost knowledge of this convention and “corrected” the tooth into a t by placing diacritical dots above, rather than beneath it.41 Luxenberg’s idea that a “tooth” was once used as mater lectionis for long a seems to be supported by a variety of evidence that Luxenberg does not mention. In the specific case of the name Ibrahim, there is an occurrence of this orthography 37 FRED M. DONNER in Jabal Usays inscription number 107, dated to 528 CE.42 This may perhaps be taken as evidence of the incipient Hijazi or North-Arabian tradition of writing Arabic of which Diem spoke, which the Qur’an reflects at least in part. Both the San‘a’ Qur’ans and some other very early Qur’an leaves also show at least occasional variant orthographies that suggest the use of a tooth as mater lectionis for a. In all cases, however, this orthography is not consistent – that is, there is no surviving text which uses a tooth consistently to represent long a.43 Consequently, one must assume that most of these instances represent the survival of an earlier system that was subsequently replaced by the more familiar use of alif as mater lectionis for long a. In other words, these cases seem to be a few stray instances that for whatever reason escaped the attention of redactors “converting” the text’s orthography from one form to another that is more familiar today. Here, however, the orthography of Ibrahim poses something of a problem, because its orthography is indeed consistent,44 implying that by the redaction phase the name was pronounced with a long i or long e sound in the last syllable, so the “old” orthography was left intact. Perhaps we need to consider the possibility that the use of y to render what in classical Arabic is long a reflects in some way the dialect pronunciation of the region where this orthography first developed. If classical long a was subjected to what the Arabic dialectologists refer to as imala, or a tendency to pronounce it as e, it might be that the name Abraham was pronounced something like “Ebrahem” or “Ebrahaym,” and that the resulting long e or diphthong ay sound was rendered by a y. Rabin has noted that the Qur’anic orthography for final long a (the so-called alif maqsura) in the form of a final y, as in hatta, “until,” suggests that the word was pronounced something like “hattay,” that is, with a marked imala or shift of the long a sound to e or to a diptote ay.45 The orthography Ibrahim could be another reflection of this tendency in spoken Hijazi Arabic. But if this is so, it provides yet another indication that the Qur’an’s original orthography was closely shaped by dialect pronunciation.46 Luxenberg’s most daring proposal also relates to the Qur’an’s language. He argues, as noted already, that many passages in the Qur’an originally were not in Arabic at all, but rather in what he calls an Arabic-Aramaic “mixed language” (Mischsprache). He is, of course, right when he says that the Qur’an text contains words that are indubitably of Syriac origin, but this fact has been recognized for years – centuries, actually. Luxenberg’s notion of a “mixed language” is problematic, however, because he makes no effort to clarify what he means by it. Is he alluding to a kind of Syriac–Arabic pidgin, used perhaps in ritual or other religious contexts, that the Qur’an text has somehow captured in transcription, a pidgin that evidently never became creolized and so died out? If so, the presence of Syriac vocabulary in the Arabic matrix would imply that Syriac was the “superstratum language” and Arabic the “substratum,” since this is the normal pattern found in pidgins;47 but then we would wish to know what historical and sociological situation this might correspond to. Luxenberg’s recourse to an ill-defined concept of “mixed language” gives him an excuse to claim that the normal rules of neither of the known languages (in this case, Arabic and Aramaic) 38 THE QUR’AN IN RECENT SCHOLARSHIP applies in a particular passage of text that is difficult for us to understand. This, in turn, frees him to make capricious surmises about the meaning of a particular passage, even though those suggestions would otherwise be rejected as ungrammatical. As used by Luxenberg, in short, the hypothesis of a “mixed language” seems to be little more than a convenient pretext for high-handed interpretation of the text. Luxenberg might himself object that this is not what he meant by “mixed language.” He might have meant by the term only that it involved a situation in which passages of Aramaic, complete with its laws of grammar, are embedded in a matrix of Arabic, which except for those embedded passages functions according to the normal rules of Arabic grammar. Such a situation, however, does not strike me as anything unusual in normal language, and certainly not as something that would justify creating the dubious category of “mixed language.” For we know of countless cases in which passages in one language, even including its grammatical markers, are embedded in another matrix language, yet we do not consider the result to be a “mixed language.” (Usually we see this phenomenon when one culture needs to express, in its favored language, particular ideas for which another language and/or culture with which it is in contact has already developed a sophisticated and precise technical terminology.) The Persian and Arabic phrases relating to Islamic religion and court culture that are found embedded in Ottoman Turkish do not make the latter a “mixed language,” any more than the extensive Latin terminology and phraseology found even today in medical or legal English – habeas corpus, caveat emptor, in loco parentis, sine qua non, and so on – makes English a “mixed language.” More importantly, in such cases, the relevant, well-known grammatical rules should apply, whether we are dealing with an Aramaic passage embedded in the Arabic matrix, or with the Arabic matrix itself. Some of the negative reviews of Luxenberg’s book have, with perfect justice, castigated him for failing to demonstrate the grammatical cogency of some of his proposed emendations.48 On the other hand, even if we reject Luxenberg’s claim that the “Ur-Qur’an was couched in some kind of Arabic-Aramaic “mixed language,” we must still come to terms with the possibility that numerous passages in the Qur’an should, in fact, be read as Aramaic phrases embedded in an otherwise Arabic matrix, rather than as Arabic. This seems an eminently reasonable possibility, as likely as the many Latin phrases routinely embedded in modern English, and the proposal deserves full investigation, not the reflexive rejection it sometimes elicits.49 We know that Syriac and other forms of Aramaic had, by the late antique period, developed a highly sophisticated technical discourse in matters of theology and religious practice. Moreover, it seems almost beyond question that Christians and Jews who knew this language and terminology were in contact with people (in Arabia or outside of it) who used Arabic as their primary, or their only, language. The survival of at least one bilingual Arabic-Aramaic inscription from the century or so before Islam confirms this.50 It therefore seems not only possible, but indeed highly probable, that Arabic-speaking people who wished to express similar ideas 39 FRED M. DONNER about theology and religion would have looked to Aramaic for stock technical terms and phrases – indeed, they may have first become familiar with these religious concepts from speakers of Aramaic – and that, as with legal English, some of those borrowed phrases would include the relevant markers of Aramaic grammatical constructions. So we cannot, I think, sweepingly dismiss Luxenberg’s reconstructions of what we might call “unrecognized” Syriac phrases in the Qur’an text. Rather, each of his proposed reconstructions needs to be examined on its own merits. While some reviewers have harshly criticized a few of his emendations, the majority of them remain to be tested.51 Given what we know about the religious history of the Near East in the seventh century, and given the manifest uncertainty about just how the Qur’an text was first written and about its linguistic content, Luxenberg’s basic hypothesis, that the “Ur-Qur’an” contained elements of Syriac phraseology, certainly seems to me highly plausible and deserving of further study. 4 How was the presumed “Ur-Qur’an” transmitted? We have already touched on some of the questions regarding the early transmission of the “Ur-Qur’an” text and how it may have become the Qur’an we know today. There seems to be general agreement that in the passage from “Ur-Qur’an” to the Qur’an text familiar to us now, some editing was performed, but whether it was editing merely to make the text reflect unambiguously a wellestablished recitation, or whether the editing completely transformed the text, intentionally or unintentionally, is still subject to debate. A central question regarding the transmission of the Qur’an is whether this transmission was essentially written or oral. (This is a distinct subject from the question of the Qur’an’s original orality, discussed earlier.) Islamic tradition argues that the Qur’an’s transmission was firmly controlled by the practice of oral recitation; the revelations received by Muhammad were learned from him as recitations by his followers, and were sometime later written down to form the Qur’an text we have today. These traditional accounts usually revolve around the stories of the so-called ‘Uthmanic recension or vulgate of the Qur’an prepared at the orders of the third caliph, ‘Uthman ibn ‘Affan (d. 35/656). The doctrine of the “ ‘Uthmanic recension” was long accepted by a number of Western scholars as well, and with it the belief in an unbroken tradition of oral recitation.52 Doubts have been shed, however, on this traditional view. Burton, de Prémare, and many others have emphasized the suspect character of the traditional reports about the ‘Uthmanic recension.53 Moreover, there is mounting evidence that the Qur’an text, or parts of it at least, must at some stage in its history have been transmitted in purely written form, without the benefit of a controlling tradition of active recitation.54 This evidence takes the form of recognizing in the Qur’anic text misunderstood words, hypercorrected words (the “lectio facilior”), or stray marks which then became incorporated into the recitation, something that could only happen if the oral recitation were derived from the written text rather than the other way around. Luxenberg’s “Syriac” hypothesis, which posits that editors 40 THE QUR’AN IN RECENT SCHOLARSHIP of the Qur’an at some point did not realize that certain passages were actually Syriac and proceeded to read them as Arabic, only makes sense if the passages in question were transmitted only in written form – otherwise, the proper pronunciation of the Aramaic would have been retained. The same is true of the old use of y (“tooth”) as mater lectionis for long a, discussed earlier. Several studies by Bellamy clearly imply that the text was at least in part transmitted in purely written form without a controlling tradition of oral recitation.55 Recent work on the word furqan provides another case in which the written, not oral, transmission of the text is implied by the evidence.56 And yet, despite all this converging evidence for written transmission of at least parts of the text, the fact that the Qur’an was first written in a highly defective script implies that written copies were initially intended to serve only as a memory aid for those who already knew the text by heart, since reading it “cold” with any accuracy would have been virtually impossible. This suggests that a tradition of oral recitation was assumed by those who made these first written copies. The question of oral vs. written transmission of the Qur’an text, then, remains a very enigmatic issue and one with potentially profound implications for our understanding of the origins and history of the text. 5 How and when did codification and canonization of the Qur’an take place? As Neuwirth has pointed out, the two processes – codification and canonization – are not the same.57 Codification refers to the process by which a text is established in unvarying form, so that any variant reading is considered “wrong,” or at least not part of the codified text. Canonization refers to the process by which a text assumes a position of authority in a community. However, it is evident that the very act of codifying a text presupposes that the already existing text or texts or textual variants were seen by people as having a special authority in the community – otherwise, why would people take the trouble to codify it at all? As the word implies, the codification process is intended to define the limits of that authority, the exact nature of the claim of the text on its adherents. So codification and canonization, while different processes, are nonetheless intimately related. As noted previously, Islamic tradition explains the process of the Qur’an’s codification through the “ ‘Uthmanic recension” story (and other, similar stories), and it of course considers the Qur’an to hold canonical status from the moment of revelation. These views also were adopted by many Western scholars for a time, but on this point too scholarship in recent years has begun to diverge palpably from the traditional views. John Wansbrough’s Qur’anic Studies challenged the tradition’s views on the timetable of both codification and, by implication, canonization. His argument is that the Qur’an did not crystallize as a fully codified, invariant text until 200 or more years after the time of Muhammad.58 Wansbrough’s hypothesis remains very much under debate; a number of leading scholars have endorsed it or appear to assume its validity, but others have expressed reservations or espoused 41 FRED M. DONNER contrary views. John Burton’s The Collection of the Qur’an, published in the same year as Wansbrough’s Qur’anic Studies (1977), reached almost exactly the opposite conclusion, also on the basis of a meticulous analysis of Qur’anic evidence (albeit different evidence): where Wansbrough considered the Qur’an a text that only emerged slowly as a codified canon, that is, long after the supposed ‘Uthmanic recension, Burton felt that the text was already written down more or less as we now have it upon Muhammad’s death, that is, some decades before the supposed ‘Uthmanic recension. As Burton states in the closing sentence of his book, “What we have today in our hands is the mushaf of Muhammad.”59 It is significant, however, that the Qur’an text itself bears no telltale signs of later origin, such as we find in the hadith literature, from which we may conclude that while we still do not know exactly how or when the text coalesced or what it represents, it must have been codified relatively early (no later than the first half of the seventh century).60 On the issue of the Qur’an’s codification, we seem in fact to be faced with contradictory information, and find ourselves perhaps in something akin to the position of physicists, who are unable to choose definitively between the theory that understands light as a wave phenomenon and another that understands it as a particle phenomenon, because each theory explains some of the empirically observable characteristics of light. The available evidence on the Qur’an’s codification, similarly, seems to provide support for both the early codification and the late-codification hypotheses: a On the one hand, we know that there are large numbers of qira’at or variant readings of the Qur’an. These circulated in great numbers, but seem to fall within several internally coherent groupings associated with the major early Islamic centers: Medina, Kufa, Basra, Syria, and so on and were themselves codified with time into selections of “canonical” variant readings according to different revered early authorities. Thus one speaks of the seven canonical variant collections, of the three after the seven, the four after the ten, and of multiple recensions of each of these fourteen collections of variants. Many of these qira’at affect only the vowelings of the text, and might be explained as later quibbles imposed by Muslim Qur’an specialists on an essentially stable consonantal rasm, but some of the variants involve changes in the rasm as well.61 The existence of the variant readings shows that we cannot conceptualize the Qur’an as a text that crystallized into a single, immutable codified form at an early date (e.g. within one generation of Muhammad), even though each of the regional traditions appears to go back to a single regional archetype.62 We have to assume that various individuals and subcommunities used different variant texts, and continued to do so for centuries. (Indeed, the variants still exist and are used.) b On the other hand, even those who champion most ardently the idea that the Qur’an text lacked stability for a long time and was not codified at an early date must in turn acknowledge that the instability of the text was confined within very strict limits. While there are some significant variants in the qira’at literature, 42 THE QUR’AN IN RECENT SCHOLARSHIP we do not find long passages of otherwise wholly unknown text claiming to be Qur’an, or that appear to be used as Qur’an – only variations within a text that is clearly recognizable as a version of a known Qur’anic passage. Some early exemplars of the Qur’an place a few suras in an order different from the usual one, but we do not find pieces of a particular sura detached from their usual context and embedded in a totally different sura. This implies that, despite the text’s manifest instability on one level, it has an underlying stability on a deeper level that cannot be accidental. Rather, this “deep” stability suggests that, on that level and within those limits, the Qur’an did coalesce and acquire the status of an especially revered sacred text – a canon – quite early in the life of the community. This is evidently why, as noted earlier, the text shows no evidence of anachronisms clearly dating to a period later than the life of Muhammad.63 As regards canonization of the text, the fact that quotations from the Qur’an are used in the epistolography of the early eighth century, as shown by Wadad al-Qadi,64 not to mention the Qur’anic extracts in the Dome of the Rock inscriptions from the late seventh, show that the Qur’an, whatever its form, already enjoyed a special status and authority in the community by that relatively early date. It behooves us, then, at this stage not to be too dogmatic in our views on the timetable of Qur’anic codification. Conclusions and suggestions In my opening comments I observed that the field of Qur’anic studies is currently in disarray and lacks consensus, and in the preceding pages I have touched briefly on some – only some – of the basic issues over which serious disagreement persists among scholars who study the Qur’an. I do not want to imply, however, that this current disarray is necessarily a bad thing. Quite the contrary, it is far preferable to the earlier stage of “false consensus,” which really concealed a failure or refusal to address some burning questions in a critical way (perhaps for fear of antagonizing believers). The very openness of the current debate is healthy and may eventually lead us to a stage of real consensus on basic issues, which, if it comes to pass, will be a more durable consensus because it will be achieved through the careful scrutiny of real evidence and all possibilities of interpreting it, not on a preconceived dogmatic vision. There can be no doubt that the most cherished dream of everyone who works with the Qur’an – whether academic specialist or believing Muslim (not, of course, exclusive categories) – would be the preparation of a truly critical edition of the text: that is, an edition that, working from the evidence provided by the earliest manuscript sources, comes as close as scientifically possible to the exact wording and vocalization of the original text – the Ur-Qur’an. But no sooner do we make this statement than, in light of what we have just discussed, we realize how problematic it is. For, the implication of many of the preceding reflections on recent scholarship on the Qur’an is that we face daunting problems of analysis 43 FRED M. DONNER and interpretation – orthographic, linguistic, and historical – that must be resolved before we could prepare such an “Urtext” edition with any confidence. To attempt such an undertaking before we have attained greater clarity on at least the more pressing of these problems, it seems to me, would be to risk making many false starts and possibly calamitous failure of the enterprise, as laboriously edited parts of the text were shown to be incorrect by new discoveries. Above all, we need as a community of scholars to reach some consensus on the most basic question of all, namely whether there ever actually was an Ur-Qur’an; for, as we have seen, some scholars challenge even this assumption. Another reason to be wary of the notion of preparing a critical edition of the Qur’an, in the present state of the field, is because such an enterprise is likely to become a kind of “private” or closed operation, on which a very small team of scholars work intensively, perhaps even in virtual secrecy, and probably with materials that they keep to themselves. This is, however, a formula for the production of a narrowly conceived product, shaped above all by the working assumptions of those who are “in” on the project; as time goes by there tends to be in such projects an increasing rigidity and insistence on “orthodoxy” among those working on it, even as more and more new idea arise outside the pale of those who are part of it. The parallel with the Dead Sea Scrolls editorial project is all too obvious, and that experience is not something to be replicated: as we all know, it was really when the Scrolls were “liberated,” that is, made finally available for examination by all scholars everywhere after roughly a half-century of being tightly controlled by a small coterie of scholars, that a burst of new scholarly energy in Scrolls studies was unleashed. I am not trying to suggest, however, that we do nothing – quite the contrary, there is much we can do to advance the study of the Qur’an, but preparation of a critical edition of the Qur’an is the wrong place to start at present. As a scholarly field, we are not yet ready to undertake it, because we do not yet have sufficient clarity on the fundamental issues that must be resolved first. What we need to do now, then, is to construct some tools, or working aids, that can help this and the next generation of scholars resolve those fundamental issues, and thus make possible the eventual preparation of a critical edition. To do this would be to make an enduring and indispensable contribution to Qur’anic scholarship, and it seems obvious that this should be the agenda for our generation of scholars. Our field of Near Eastern studies seems generally to be slow to provide us with proper tools and working aids – for example, we still after all these years do not have a good dictionary of Classical Arabic, much less a historical or etymological dictionary of Arabic (something that would, of course, also be of great assistance in Qur’anic studies). As far as the Qur’an and Qur’anic studies are concerned, the creation of two web-based databases might help us to answer basic still-unresolved questions about the Qur’an. The first of these would be a Qur’an Manuscripts Database (QMD). There are many basic questions about the Qur’an’s origins and earliest development that hinge on matters of simple orthography – which is, of course, anything but simple, 44 THE QUR’AN IN RECENT SCHOLARSHIP as we have seen. To address these orthographic issues we need, ideally, to examine all early manuscripts of the Qur’an, and to compare them with other important epigraphic and palaeographical sources for early Arabic orthography. We can hope that this will help us to understand, finally, just how the text was originally written, how it was recited, and the relationship of the “Ur-Qur’anic” orthography to the orthographic traditions for writing Arabic current just before and during the time of Muhammad. The most effective way to enable scholars to do this would be to establish QMD, containing a high-quality scanned image of every known early Qur’an leaf – anything prior to perhaps the third or fourth century AH. (It should, of course, also include other early Arabic inscriptions and documents.) The thousands of photographs of early Qur’ans collected in the 1930s by Bergsträsser, once thought to have perished during the bombing of Munich in the Second World War, are according to some rumors still in existence, perhaps in Berlin, and if so, they should be incorporated into this database until better, color scanned images of the original manuscripts can be secured. Photographs of the San‘a’ Qur’an manuscripts should also be posted on this site. This would finally permit researchers readily to examine all the evidence for orthographic questions as they arise. Note that the goal of the project is not to embark on analysis of the texts themselves, which would take many years, but simply to make available all the relevant information so that others could begin to make such analyses on their own. There are at least three important advantages to this project, which we can summarize under the rubrics “security,” “real-time availability,” and “flexibility.” a While most precious early Qur’an leaves are held in well-maintained library or museum collections, not all are so secure. In any case, their very nature as unique documents means that the irreplaceable information they contain is vulnerable to intentional or unintentional loss. The possibility of fire, flood, theft, or other mishap can never be dismissed even for well-maintained collections, as the recent sad fates of documents in the Chester Beatty Library in Dublin in the 1980s, the library of the Institute of Islamic Studies in Sarajevo in the 1990s, and the libraries and museums of Baghdad in the 2000s, make all too clear. One advantage of the project, then, is that the textual (orthographic) information on every early Qur’an leaf would be preserved in digital form and, therefore, secured from loss even if the original document were to perish. b Another advantage is that the information entered into the database would become available for scholars to use very quickly. As opposed to formal publication of printed facsimile volumes, which could easily consume years to prepare and would be expensive to print (and hard to sell!), large-size, high-resolution color images of individual Qur’an leaves could be scanned and mounted on the site relatively quickly once permission had been obtained. Moreover, the site would be cumulative; at first relatively few historic Qur’ans might be illustrated, but with time the collection would become more and more comprehensive. In any case, scholars could begin working with what was available on-line, and would 45 FRED M. DONNER not have to wait for other documents to begin – that is, its cumulative nature avoids the “all-or-nothing” quality that delays, and sometimes thwarts completely, many publication projects. c The final obvious advantage of a project such as QMD is the fact that it is merely providing raw materials for scholarly analysis and interpretation, which gives individual researchers the flexibility to interpret the materials in the manner that seems most compelling to them. Differences of opinion are sure to arise, which will be sorted out by the usual give-and-take of scholarly exchange, with the most cogent understandings winning out in the long term. It thus avoids the conceptual rigidity of many projects that proceed with a definite set of assumptions not only about the importance of the work, but also about how it should be carried out or the character of the results to be achieved. In some fields that may be appropriate, but students of the Qur’an still face too many basic uncertainties to be able to adopt hard-and-fast interpretive principles. The second working aid that merits inauguration would be a Hypertext Qur’an Project (HQP). Whereas QMD would be aimed at resolving problems of orthography, HQP would address the question of variant readings. Many scholars, as we have seen, have tabulated the variants known from the basic qira’at literature,65 but there are variants that are not included in these compilations (and doubtless more to be found in the manuscripts to be included in QMD). The HQP would permit a researcher to enter a reference to any Qur’an verse, and the database would exhibit all known readings for that passage, including those known only from early manuscript copies of the text. As a project, it has most of the merits just described for QMD. Solving the many problems that confront us in understanding the origins and nature of the Qur’an text is a challenge that is far too large for any single individual. Working collectively, however, scholars of the first decades of the twenty-first century may through these two projects – QMD and HQP – be able to create the conditions under which the scholarly community can finally resolve in a definitive manner some of the most pressing problems surrounding the Qur’an. By doing so, the ground would finally be readied for the eventual preparation of a critical edition of the Qur’an. Notes 1 I am grateful to the conference participants, and especially to Gabriel S. Reynolds, for corrections and helpful suggestions on the draft of this article. 2 A. Geiger, Was hat Mohammed aus dem Judenthume aufgenommen, Bonn: F. Baaden, 1833; T. Andrae, “Der Ursprung des Islams und das Christentum,” Kyrkshistorisk årsskrift 23, 1923, 149–206; 24, 1924, 213–92; 25, 1925, 45–112; R. Bell, The Origin of Islam in its Christian Environment, London: Macmillan, 1926; C.C. Torrey, The Jewish Foundation of Islam, New York: Jewish Institute of Religion, 1933. 3 K. Vollers, Volkssprache und Schriftsprache im alten Arabien, Strassburg: K. J. Trübner, 1906; R. Blachère, Introduction au Coran, Paris: G.P. Maisonneuve, 1947; J. Fück, 46 THE QUR’AN IN RECENT SCHOLARSHIP 4 5 6 7 8 9 10 11 12 13 14 15 16 17 ‘Arabiya. Untersuchungen zur arabischen Sprach- und Stilgeschichte, Berlin: Akademie-Verlag, 1950. GdQ3; A. Jeffery, Materials for the History of the Qur’an. The Old Codices, Leiden: Brill, 1937; E. Beck, “Der ‘uthmanische Kodex in der Koranlesung des zweiten Jahrhunderts,” Orientalia 14, 1945, 355–73; idem, “ ‘Arabiyya, Sunna und ‘amma in der Koranlesung des zweiten Jahrhunderts,” Orientalia 15, 1946, 180–224; idem, “Die Kodizesvarianten der Amsar,” Orientalia 16, 1947, 353–76; M. Cook, “The stemma of the regional codices of the Koran,” Graeco-Arabica 9,10, 2004, 89–104. See now Abd al-Latif Muhammad al-Khatib, Mu‘jam al-qira’at al-qur’aniyya, Damascus: Dar Sa ‘d al-Din, 2002. G. Lüling, Über den Ur-Koran, Erlangen: H. Lüling, 1974 – an expanded version of his Erlangen dissertation, Kritische-exegetische Untersuchungen des Qur’antextes, 1970; Translation: idem, A Challenge to Islam for Reformation, Delhi: Molital Banarsidass, 2003. J. Wansbrough, Qur’anic Studies, Oxford: Oxford University Press, 1977; P. Crone and M. Cook, Hagarism. The Making of the Islamic World, Cambridge: Cambridge University Press, 1977. Berlin: Das arabische Buch, 2000. A. Rippin, “Literary analysis of Qur’an, Tafsir, and Sira: The methodologies of John Wansbrough,” in R.C. Martin (ed.), Approaches to Islam in Religious Studies, Tucson: University of Arizona Press, 1985, 151–63. See especially his introduction to The Quest for the Historical Muhammad, edited and translated by Ibn Warraq; Amherst, N.Y.: Prometheus, 2000. The translations themselves are a service to scholars in that they make available in English a number of articles that were difficult of access before; but Ibn Warraq’s polemical introduction will simply confuse many, especially neophytes, as to what the goals of Western scholarship on the Qur’an actually are. See also his The Origin of the Koran, Amherst: Prometheus, 1998 and What the Koran Really Says, Amherst: Prometheus, 2002. Cf. Nöldeke, Schwally, Bergsträsser, GdQ; Bell, The Origin of Islam in its Christian Environment; Beck, “Der ‘uthmanische Kodex in der Koranlesung des zweiten Jahrhunderts; W.M. Watt, Muhammad’s Mecca. History in the Qur’an, Edinburgh: Edinburgh University Press, 1988, esp. 1–4; A. Neuwirth, “Qur’an and history – a disputed relationship: Some reflections on Qur’anic history and history in the Qur’an,” Journal of Qur’anic Studies 5, 2003, 1–18; eadem, Studien zur Komposition der mekkanischen Suren, Berlin: de Gruyter, 1981; Lüling, Über den Ur-Koran; J. Burton, The Collection of the Qur’an, Cambridge: Cambridge University Press,1977. See, for example, G.R. Hawting, The Idea of Idolatry and the Emergence of Islam. From Polemic to History, Cambridge: Cambridge University Press, 1999; Rippin, “Literary Analysis.” See sections 4 and 5 below, pp. 40–43. A few examples from a vast literature will have to suffice: W.M. Watt, Muhammad at Mecca, Oxford: Clarendon Press, 1953; idem, Muhammad at Medina, Oxford: Clarendon Press, 1956; M. Gaudefroy-Demombynes, Mahomet, Paris: Albin Michel, 1957; K. Armstrong, Muhammad, HarperSanFrancisco, 1992. A brief overview of this question, with references to other scholarship, is found in the first several pages of C. Rabin, “Islam and the Qumran Sect,” in his Qumran Studies, New York: Oxford University Press, 1957, 112–30. Geiger, Was hat Mohammed aus dem Judenthume aufgenommen; Torrey, The Jewish Foundation of Islam. See Fred M. Donner, “Modern Nationalism and Medieval Islamic History,” Al-‘Usur al-Wusta 13, 2001, 21–2. Andrae, “Der Ursprung des Islams und das Christentum,” 109. G. Lüling, Die Wiederentdeckung des Propheten Muhammad, Erlangen: Lüling, 1981, 73–89. 47 FRED M. DONNER 18 F.M. Donner, “From Believers to Muslims: Confessional Self-Identity in the Early Islamic Community,” Al-Abhath 50–51, 2002–3, 9–53. 19 See the remarks in the preceding paragraph on Lüling’s theories. See also F. de Blois, “Elchasai-Manes-Muhammad. Manichäismus und Islam in religionshistorischem Vergleich,” Der Islam 81, 2004, 31–48, and idem, “Nasrani and hanif: studies on the religious vocabulary of Christianity and Islam,” BSOAS 65, 2004, 1–30. De Blois’s excellent articles, however, seem to continue the conspiracy of silence against Lüling’s work; he makes the astonishing statement that “Arabists and students of Islam have until now paid no attention to this question [of the similarity between Qur’anic and Jewish-Christian concepts of prophecy] (“Elchasai . . . , 47), even though Lüling, whose work he disparaged in passing in a review published in 2003 (see note 47 later), had written substantively about this thirty years earlier. 20 “Licht ins Dunkel. Der Koran als philologischer Steinbruch. Ein Gespräch mit Christoph Luxenberg,” in Ch. Burgmer (ed.), Streit um den Koran. Die LuxenbergDebatte: Standpunkte und Hintergründe, Berlin: Schiler, 2004, 18–19. 21 Angelika Neuwirth, “Qur’an and History,” 12–13. 22 For example, A. Dundes, Fables of the Ancients? Folklore in the Qur’an, Lanham, MD: Rowman & Littlefield, 2003; but although Dundes’s main point appears to be welltaken, see the review by A. Rippin, BSOAS 68, 2005, 120–2, for the work’s severe limitations. 23 D.A. Madigan, The Qur’an’s Self-Image. Writing and Authority in Islamic Scripture, Princeton: Princeton University Press, 2002. 24 See, for example, the review by M. Fakhry, Journal of Islamic Studies 13, 2002, 331–2. 25 For example, we might reconsider in this way the passages on Shu ‘ayb and the “four gardens” theme, discussed by Wansbrough, Qur’anic Studies, 21–7. 26 Neuwirth, Studien zur Komposition, 263, 316; eadem, “Du texte de recitation au canon en passant par la liturgie,” Arabica 47, 2000, 194–229. 27 Qur’anic Studies, 27. 28 For example, Ahmad b. Muhammad al-Quduri, Matn al-Quduri fi fiqh madhhab al-imam Abi hanifa, Cairo: al-Maktaba al-Khayriyya, 1324/1906, 10, lines 5–8; G. Monnot, “Salat,” EI2 8,928a. 29 On diacritics and other markings added to the rasm, see A.T. Welch, “al-Üur’an,” EI 2. 5:404–9; N. Abbott, The Rise of the North Arabic Script and Its üur’anic Development, with a Full Description of the Üuran Manuscripts in the Oriental Institute, Chicago: The University of Chicago Press, 1939, 38–41; G.-R. Puin, “Observations on Early Qur’an Manuscripts in San ‘a’,” in S. Wild (ed.), The Qur’an as Text, Leiden: Brill, 1996, 107–11. On the emendations by al-Hajjaj (or perhaps by his predecessor, Ubaydallah b. Ziyad), see GdQ3, 103–4, and K.I.H. Semaan, “A linguistic view of the development of the Arabic writing system,” Wiener Zeitschrift für die Kunde des Morgenlandes 61, 1967 (22–40) 36–7; B. Gruendler, The Development of the Arabic Scripts, Atlanta: Scholars Press, 1993, seems to concentrate on the rasm and offers no observations on the development of the diacritics. 30 E.J. Revell, “The diacritical dots and the development of the Arabic alphabet,” JSS 20, 1975, 178–90. 31 Vollers, Volkssprache und Schriftsprache im alten Arabien. 32 F. Altheim and R. Stiehl, “Die Anfänge der arabischen Schriftsprache,” in their Die Araber in der alten Welt, Berlin: de Gruyter, 1964–9, 2, 357–69. For more on the orthography of Salat, see later. 33 H. Fleisch, Introduction à l’étude des langues sémitiques, Paris: A. Maisonneuve, 1947, 97–101; Blachère, Introduction au Coran, 156–69; Fück, ‘Arabiya. Untersuchungen zur arabischen Sprach- und Stilgeschichte, 3. See discussion in C. Rabin, “The Beginnings of Classical Arabic,” SI 4, 1955, 19–37. 34 For example, surat al-Shu ‘ara’ (26):195; cf. surat al-Nahl (16): 103. 48 THE QUR’AN IN RECENT SCHOLARSHIP 35 Particularly if we assume, as some do, that Mecca was first established before Islam as a trading outpost of one of the South Arabian kingdoms. 36 Anton Spitaler, “Die Schreibung des Typus im Koran. Ein Beitrag zur Erklärung der koranische Orthographie,” Wiener Zeitschrift für die Kunde des Morgenlandes 56, 1960, 212–26. Note, however, A. Mingana and A.S. Lewis, Leaves from Three Ancient Qurâns, possibly Pre- ‘Othmanic, with a list of their variants, Cambridge: Cambridge University Press, 1914, 18 (to Qur’an 9:72, which has the orthography al-Sylwt). 37 W. Diem, “Some glimpses at the rise and early development of the Arabic orthography,” Orientalia 45, 1976, (251–61), 255. 38 Muhammad Abu l-Faraj al-’Ushsh, “Kitabat ‘arabiyya ghayr manshura fi jabal Usays,” al-Abhath 17, 1964, 227–316. 39 See J.A. Bellamy, “Two pre-Islamic Arabic inscriptions revised: Jabal Ramm and Umm al-Jimal,” JAOS 108, 1988, 369–78; idem, “A new reading of the Namarah inscription,” JAOS 105, 1985, 31–48. A sampling of the tens of thousands of Safaitic inscriptions known can be perused in R. Dussaud and F. Macler, Mission dans les régions désertiques de la Syrie Moyenne, Paris: Imprimerie Nationale, 1903. 40 On the orthography, see G.-R. Puin, “Observations on Early Quran Manuscripts;” idem, “Die korrumpierte Tradition? Ein Gespräch mit Gerd-Rüdiger Puin,” in Burgmer (ed.), Streit um den Koran, 51–63; idem, “Über die Bedeutung der ältesten Koran-fragmente aus Sanaa (Jemen) für die Orthographie-geschichte des Korans,” in H.-C. Graf von Bothmer, K.-H. Ohlig and G.-R. Puin, “Neue Wege der Koranforschung,” Magazin Forschung 1, 1999, 33–46. I thank Joshua Lingel for calling my attention to the last-mentioned article and sending me the relevant section of it. 41 Luxenberg (pp. 62–5) proposes, for example, that ‘utull in Q. 68:13 is a misreading for ‘al. 42 See al- ‘Ushsh, “Kitabat ‘arabiyya ghayr manshura fi jabal Usays,” 302; on the date, see A. Grohmann, Arabische Paläeographie II, Graz, Wien, Köln: Hermann Böhlaus Nachf., 1971, 15–7. 43 On the San‘a’ Qur’ans, see Puin, note 36 earlier. For another example, see Mingana and Lewis, Leaves from Three Ancient Qurâns, 70 (to Q. 41:10), fa-qila la-ha (for fa-qala la-ha); see also Mingana’s introductory remarks on page xxxv. 44 Or virtually so; Werner Diem points out that the “‘Uthmanic recension” lacks y in Ibrahim in sura 2; elsewhere, with y: “Untersuchungen zur frühen Geschichte der arabischen Orthographie, I: Die Schreibung der Vokale,” Orientalia 48, 1979, 207–57, at p. 227. 45 C. Rabin, Ancient West-Arabian, London: Taylor’s Foreign Press, 1951, 40, 111. 46 Alternatively, one might, following Luxenberg, suggest that the final alif maqsura (or y) of hatta is another instance of using y as mater lectionis for long a, rather than evidence of imala. The inconsistency in Qur’anic orthography of long a – sometimes unmarked in rasm, sometimes marked by alif, sometimes marked by alif maqsura, sometimes perhaps marked as Luxenberg suggests by a “tooth” (y) – suggests that we still have a long way to go in understanding how the text came to be in the form it exists today. 47 See J.R. Rickford, “Pidgins and creoles,” in William Bright (ed.), International Encyclopedia of Linguistics, NewYork: Oxford University Press, 1992. A useful overview of the processes in Arabic is found in the first chapter of C. Holes, Modern Arabic, London: Pearson, 1995. 48 For example, the review by F. De Blois of Luxenberg’s Die syro-aramäische Lesart in the Journal of Qur’anic Studies 5, 2003, 92–7, or that by Simon Hopkins in JSAI 28, 2003, 377–80. 49 See, for example, the brief comment in the Introduction to A. Ambros and S. Prochazka, Concise Dictionary of Koranic Arabic, Wiesbaden: Reichert, 2004. 50 Actually a Greek-Syriac-Arabic trilingual, from Zebed (southeast of Aleppo, in northern Syria), dated 512 CE. There is also a Greek-Arabic bilingual from Harran in the Laja’, 49 FRED M. DONNER 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 south of Damascus, dating from 568 CE. On these and other early inscriptions see Grohmann, Arabische Paläographie II, 14–18, and Grüendler, Development, 13–14, with further references. Not all reviews of Luxenberg’s Syro-Aramäische Lesart have been as negative as those by de Blois and Hopkins mentioned in note 47. See Robert R. Phenix and Cornelia B. Horn in the on-line journal Hugoye 6:1, January, 2003: (http://syrcom.cua.edu/ Hugoye/Vol6no1/HV6N1PRPhenixHorn.html). This assumption seems to underlie even the works of Bergsträsser and Beck (see note 3 earlier). J. Burton, The Collection of the Qur’an. A.-L. de Prémare, Aux origines du Coran. Question d’hier, approches d’aujourd’hui, Paris: Téraèdre, 2004, 57–99. See also GdQ3, 1–119; W.M. Watt, Bell’s Introduction to the Qur’an, Edinburgh: Edinburgh University Press, 1970, 40–56. See S.K. Samir’s contribution to this volume for additional examples. J.A. Bellamy, “Some proposed emendations to the text of the Koran,” JAOS 113, 1993, 562–73; idem, “More proposed emendations to the text of the Koran,” JAOS 116, 1996, 196–204; idem, “Textual criticism of the Koran,” JAOS 121, 2001, 1–6; idem, “A further note on ‘Isa,” JAOS 122, 2002, 587–8. Fred M. Donner, “Qur’anic furqan,” JSS 52, 2007, 279–300. Neuwirth, “Qur’an and History,” 2, 13. Perhaps the clearest statement of Wansbrough’s theses is found in Qur’anic Studies, 46–7. J. Burton, The Collection of the Qur’an, 239–40. He deals with the ‘Uthmanic and other recension stories on pp. 115–89. This view is developed in F.M. Donner, Narratives of Islamic Origins: The Beginnings of Islamic Historical Writing, Princeton: Darwin Press, 1998, 35–61. Paret, citing A. Fischer, in EI2, “Üira’a.” A salient example is found in the text of Q. 3:19, where for the phrase “inna l-din ‘inda llahi l-islam” we find in Ibn Mas ‘ud’s reading “inna l-din ‘inda llahi l-hanifiyya” (Jeffery, Materials, 32). Cook, “The Stemma of the Regional Codices . . .” (note 3, earlier). Donner, Narratives, 40–9. W. al-Qadi, “The religious foundation of late Umayyad ideology and practice,” in Saber Religioso y Poder Político en el Islam, Madrid: Agencia Española de Cooperacíon Internacional, 1994, 231–73, esp. 269ff.; also eadem, “The impact of the Qur’an on early Arabic literature: The case of ‘Abd al-Hamid’s epistolography,” in A.A. Shareef and G.R. Hawting (eds), Approaches to the Qur’an, London: Routledge, 1993, 285–313. (see Chadler). See note 4 earlier. 50 2 EPIGRAPHY AND THE LINGUISTIC BACKGROUND TO THE QUR’AN Robert Hoyland In his book Die syro-aramäische Lesart des Koran Christoph Luxenberg states that “Syro-Aramaic,” that is, Syriac, his proposed Ursprache of the Qur’an, is the version of Aramaic used in Edessa and its environs (p. vii), but there is no discussion of how this language might have come to dominate in far away Hijaz to such an extent that it would form the basis of the sacred writings of its inhabitants, as Luxenberg claims. The nature of the Arabic language and script is also not discussed, rather it is simply asserted that “aside from a few pre-Islamic inscriptions of the fourth/sixth centuries CE from north Hijaz and Syria the Qur’an is the first work composed in the Arabic script” (p. 15) and “the Arabic language at the time of the emergence of the Qur’an possessed no standardized literary language, but consisted only of spoken dialects” (p. 52). This article will attempt to give very much needed discussion of these issues, which are crucial to any real exploration of the historical environment into which the Qur’an was born. To keep the article to a manageable length, I will focus on the epigraphic data, which also have the advantage of not being prey to the questions of authenticity that have dogged the literary witnesses to pre-Islamic Arabic (especially pre-Islamic Arabic poetry and tribal narratives).1 Aramaic. Aramaic is an umbrella name for a language that existed in many different varieties and forms. There is the so-called imperial Aramaic of the Achaemenid period (ca. 550–330 BCE), the official language of government for all the scribes of the Iranian empire. Then there are the numerous dialects that enjoyed a resurgence after Alexander the Great’s destruction of the Achaemenids, which were used by a wide range of peoples across the Middle East – such as the Palmyrenes, the Nabataeans, the Hatrans, the Jews of Palestine and Babylonia – either for writing or speaking or both. As the Middle East became Christianized in the fourth to sixth centuries CE, many of these dialects then became used to express the teachings of the new religion. Although the success of the Arab conquests and of Islam favored the spread of Arabic, Aramaic continued to be used across a broad swathe of the Middle East, as we are told by Gregory 51 ROBERT HOYLAND Abu l-Faraj (Bar Hebraeus), head of the West Syrian church in the east in the thirteenth century CE: Aramaic (al-suryaniyya): In it spoke God and Adam. It is divided into three dialects: the most pure is al-aramiyya, which is the dialect of the people of Edessa, Harran, and Outer Syria. Then there is al-falastiniyya, which is the dialect of the people of Damascus, the mountains of Lebanon, and the rest of Inner Syria.2 And the most ugly of the three is the Chaldaean dialect, al-nabatiyya, which is the dialect of the people of the mountains of Assyria and southern Iraq.3 The version of Aramaic used by the people of the Edessa region, referred to as Syro-Aramaic by Luxenberg and as Syriac by most modern scholars (and as al-aramiyya by Bar Hebraeus), was famous, since in it was penned a rich and distinctive Christian literature, but it was by no means the only version. In what had been the Nabataean kingdom – Palestine, Transjordan, southern Syria and northwest Arabia – a wide variety of Aramaic dialects (and scripts) were current, such as those referred to by modern scholars as Jewish Palestinian Aramaic, Nabataean Aramaic and Christian Palestinian Aramaic. However, high literary activity in this region in the fourth–seventh centuries CE, and particularly creative Christian literature (as opposed to translations), was principally conducted in Greek. In Aramaic we mostly only find translations from Greek, legal documents, and epitaphs and graffiti (overwhelmingly in rural areas). In north Syria and Mesopotamia, however, we can see that the local version of Aramaic there exhibited no such embarrassment about showing itself in public. Certainly there were many, as in Palestine, Transjordan, and southern Syria, who were bilingual, who had the privilege of acquiring a Greek rhetorical education, and who preferred to write in Greek. Yet Syriac language and culture were not submerged by the Greek; on the contrary, it became the preferred language of literacy among the Christians of this region from the fourth century CE or even earlier, and there is a wealth of both inscriptions and literary compilations composed in it from this time. It was still influenced by Greek language and culture, probably more and more so through the fifth to seventh centuries, and yet despite this it maintained its cultural standing. So whereas in Palestinian Aramaic we principally have translations from Greek Christian texts and simple inscriptions, in Syriac we have original works of theology, poetry, biblical commentary, astrology, and so on. And yet it remained largely tied to the region of Edessa and northern Syria. In particular, before Islam Syriac inscriptions are rarely found outside of this area, and then only in contexts where it is clear that they are written by pilgrims and émigrés, located on popular pilgrimage routes or in churches and monasteries for expatriate Syriac Christians.4 And none at all have been found in 52 THE LINGUISTIC BACKGROUND TO THE QUR’A N west Arabia, which is very odd if it was really, as Luxenberg assumes, the favored literary language of the people living there. Arabic. Though some form of Arabic is likely to have been in existence as early as the mid-first millennium BCE, Arabic (or rather Old Arabic, the name scholars give to pre-Islamic Arabic)5 seems to have been seldom written down until a century or so before the advent of Islam. On the very few occasions that it was committed to writing, the script of prestige in the locality concerned was employed. Thus at Dedan in northwest Arabia, before the Nabataeans arrived there (i.e. before the end of the first century BCE), an inscription was carved advertising “the funerary monument of ‘Abdsamin son of Zaydharim which Salma daughter of Aws built (allati banaha Salma bint Aws).”6 The language is Arabic, but the script is the local one in use in Dedan, a derivative of the south Arabian script. At Qaryat al-Faw, the capital of Kinda and other Arab tribes (now in modern southwest Arabia), a certain ‘Igl son of Haf‘am wrote the funerary text for his brother’s tomb (ca. first century CE) in Arabic using the script of the nearby Sabaean kingdom.7 And there are three known Arabic texts inscribed in Nabataean Aramaic script. The most famous, dated 328 CE and discovered at Nemara, in the basalt desert southeast of Damascus, is an epitaph for Imru’ al-Qays, the self-styled “king of all the Arabs,” celebrating his achievements (Figure 1).8 The second, found on a stone in Oboda (En Avdat/‘Ayn ‘Abada) in the Negev, concerns the offering of a certain Garmallahi son of Taymallahi to the god Obodas (Figure 2). He records the dedication in Aramaic, but then gives two lines of Arabic verse in praise of Obodas (though still in the Nabataean Aramaic script), which may have been part of a liturgy used in the worship of the god.9 The last text is a funerary inscription from Hegra (modern Mada’in Salih) in northwest Arabia, dated to 267 CE; it is composed principally in Arabic, but with some Aramaicisms, perhaps put in to make the text look more highbrow (Figure 3).10 There are also a few literary references to the existence of Arabic. For example, the probably fourth-century writer Uranius notes that the place name Motho means death “in the speech of the Arabs” (he arabon phone). His near contemporaries Epiphanius of Salamis and Jerome also make reference to Arabic, the former in Figure 1 Epitaph of Imru’ al-Qays, Nemara (S. Syria), 328 CE. 53 ROBERT HOYLAND Figure 2 Dedication to Obodas, ‘Ayn Abada/En Avdat (S. Palestine), ca. second century CE. Figure 3 Funerary text, Hegra (NW Saudi Arabia), 267 CE. connection with a virgin goddess whom the inhabitants of Petra and Elusa praise in the arabike dialektos and call her in Arabic (arabisti) Kaabou (cf. Arabic ka‘aba, “buxom maiden”).11 And the Jewish Talmud adduces a number of words said to be from the speech of the Arabs, and a few Arabicisms enter the Syriac language of this period.12 54 THE LINGUISTIC BACKGROUND TO THE QUR’A N Matters change with the sixth century CE, when we have a small clutch of Arabic texts written in what is recognisably the Arabic script.13 That from Zebed, southeast of Aleppo, dated 512 CE, is a short Arabic addition to a Greek-Aramaic bilingual text inscribed on the lintel of a martyrium dedicated to Saint Sergius (Figure 4). The one from Jabal Usays, southeast of Damascus, dated 529 CE, is a rock graffito by a certain Qayyim ibn Mughira, recording his despatch by “the king al-Harith” (almost certainly of Ghassan, Rome’s most important client tribe in the east) to guard this important watering hole and waystation on the route from Bostra to Palmyra (Figure 5). And the Harran text, south of Damascus, dated 569 CE, is a bilingual Greek-Arabic inscription recording the building of a martyrium for a certain Saint John by one Sharahil son of Talimu (in the Greek: Saraelos Talemou), evidently an important man in the local Christian community (Figure 6).14 Perhaps the most interesting of the clutch is the grave of Saola in Figure 4 Building text, Zebed (N. Syria), ca. 512 CE. Figure 5 Graffito, Jabal Usays (S. Syria), 529 CE. 55 ROBERT HOYLAND Figure 6 Building text, Harran (S. Syria), 569 CE. Figure 7 Funerary text, Nebo (Jordan), mid-sixth century CE. a church in Nebo, which bears his name carved in Greek letters and opposite this the “rest in peace” formula written in Arabic: bi-salam (Figure 7).15 The only plausible explanation is that the language of Saola’s family was Arabic; yet he was not, as the common perception would lead us to expect, a member of some wandering Arab tribe, but someone prominent enough in the local community to be interred in one of its churches. 56 THE LINGUISTIC BACKGROUND TO THE QUR’A N The sixth-century Usays, Harran and Nebo Arabic inscriptions are all from the former Nabataean sphere of influence, as are the second to fourth-century Avdat, Hegra and Nemara Arabic inscriptions. In addition to this we can throw into the equation two plausibly Arabic texts written in Hismaic script from the Madaba area (ca. first to third century),16 the Arabic legal phraseology evident in the first and second-century Nabataean papyri from the southern Dead Sea region,17 the predominantly Arabic nature of the toponyms in the sixth-century Petra papyri,18 and the aforementioned observations by Epiphanius of Salamis and Jerome about Arabic being spoken at Petra and Elusa. All of this makes it likely that Arabic was quite widely spoken throughout this region, and, very importantly, that it did have a written dimension, and not just an oral one.19 It would also seem probable, even necessary, that it came to be more frequently written down, and this explains the evolution of the Nabataean Aramaic script into what came to be called the Arabic script (see later in this article). The rise of the Arabic script The four aforementioned sixth-century Arabic texts are written in what can clearly be described as the Arabic script. Earlier texts composed in Arabic had, as noted earlier, used a variety of local or prestige scripts. So why this new development and who was responsible for it? There are two leading contenders: first, Arab tribes allied to Rome that had been developing incipient state structures and second, Christian missionaries who had been converting Arab tribes. Arab client tribes Comparison of letter forms makes it clear that the Arabic script developed from the Nabataean Aramaic script (see later). Constant writing of Arabic in the Nabataean script led to the evolution of the latter, as scribes introduced changes to make their task easier and their texts less ambiguous, and it is this evolved form of the Nabataean script that we call the Arabic script. Such a development is not likely to have occurred as a result of writing only a handful of texts, so it is very probable that there were many such inscriptions, and possibly documents as well (even a chancelry tradition?). Two of the most important pre-Islamic Arabic texts, Nemara and Jabal Usays, were composed by agents of Arab kings, and the Harran text by a phylarch (noted in the Greek part of the text).20 And it makes sense that it would be such characters who would have promoted the use of Arabic, endowed as they were with a measure of power and resources and perhaps also with a sense of Arab identity (especially if the reading “king of all the Arabs” is correct for the Nemara text).21 Since some were important allies of Rome, it would also be natural to suppose that they had at least a rudimentary administration, and therefore scribes at their disposal.22 57 ROBERT HOYLAND Interestingly, Arab kings are mentioned in a number of inscriptions of the third/fourth century CE. “Gadima/Jadhima king of Tanukh” (Gadimathou basileus thanouiton/Gdmt mlk tnwh), features in a ca. mid-third century bilingual GreekNabataean Aramaic epitaph of his tutor, etched on a stone found at Umm al-Jimal in modern north Jordan.23 “ ‘Amrw king of the Lakhmids” (‘Amrw lhm’dyn mlk’) appears in a bilingual Persian-Parthian monumental inscription among the vassals of the Sasanian emperor Narseh (293–302).24 A “king of al-Asd” (later Arabic ⫽ al-Azd) is mentioned in a south Arabian inscription recording the dispatch of a delegation from the Himyarite ruler Shammar Yuhar’ish (ca. 275–310), which also went to the “land of Tanukh.”25 Then there is also, of course, the aforementioned “Imru’ al-Qays son of ‘Amr, king of all the Arabs” of the Nemara inscription. Significantly, it is with such kings that Muslim Arab historical memory begins. Most Muslim universal chronicles have a section on pre-Islamic Arab history, and among the earliest sections will always be one on the kings of the Arabs. The first of these are said to be the kings of Hira, and the list of them usually begins with: Jadhima (al-Abrash), ‘Amr ibn ‘Adi of Lakhm, and Imru’ al-Qays ibn ‘Amr, apparently the same as those who appear in the epigraphic record, as cited earlier, though there are many wonderful folkloric elements attached to their biographies in the Muslim Arab accounts.26 The appearance of these Arab kings coincides with the appearance of Arab tribal names that are familiar to us from later Muslim sources, such as the Tanukh, Lakhm and al-Asd (al-Azd) of the aforementioned inscriptions, the tribes of Nizar and Ma‘add which are recorded in the epitaph of Imru’ al-Qays, and many other tribes noted in South Arabian inscriptions.27 Now, in all the tens of thousands of graffiti in Ancient North Arabian (ANA) dialects (Safaitic, Hismaic, “Thamudic,” and so on.) there appear almost no names of tribes that are also mentioned in later Muslim Arab texts.28 Evidently, there were major social changes and upheavals going on at this time. I have discussed this elsewhere, so I will not say much here beyond the fact that it reflects changes in dealings between the Roman empire and the peoples that it had conquered or that were on its borders and also processes of ethno-genesis and identity-forging. Like the Franks, Goths, Alamanni and other western groups, the Arabs were beginning to play a more major role in the Roman Empire and this was in turn changing their social and cultural makeup. Those most closely involved in the imperial power structure, such as the Ghassanids, were evidently becoming quite powerful at a local level, as is shown most dramatically in a number of texts that date events by the time in office of their leaders rather than of the emperor or the provincial governor.29 There are many accounts that relate, in a somewhat legendary character, how such kings spent their subsidies in imitating their imperial overlords, establishing luxurious courts and offering patronage to artists, a practice with a long history among imperial vassal states.30 Yet, allowing for some exaggeration, it is very plausible that such rulers established some sort of a political and administrative structure and that they patronized a degree of learning, and this could easily explain the rise of an Arabic script in the sixth century. 58 THE LINGUISTIC BACKGROUND TO THE QUR’A N Christian missionaries A second contender for the main force behind the development of the Arabic script is Christian missionaries. In the fourth century the languages of Coptic, Armenian and Palestinian Aramaic also began to be written for the first time (Coptic was written in an adapted Greek script and Palestinian Aramaic in a modified Estrangelo Aramaic script, comparable to the way Arabic was written in a modified Nabataean Aramaic script). This was done so that the key Christian texts could be preached to the masses, most of whom would not have known a high language such as Greek, in their own vernacular. Could this motive also explain the emergence of the Arabic script? Unfortunately, whereas we have many Bible translations and hagiographies in Coptic, Armenian and Palestinian Aramaic to illustrate this point, we have no such evidence for Arabic. Nevertheless, there are some points in favor of this argument. First, we do indeed have numerous accounts of Christian missionary work among the Arab tribes, in particular, tales of the virtuous lives and miraculous deeds of Christian clergy and holy men that won the hearts of many a pagan Arab. “How many Arabs,” exclaimed the biographer of Simeon the Stylite in Syria (389–459), “who have never known what bread is, but feed on the flesh of animals, came and saw the blessed (Simeon) and became disciples and Christians, abandoned the images of their fathers and served God . . . It was impossible to count the Arabs, their kings and nobles, who came and acknowledged Jesus . . . and erected churches beneath their tents.”31 It was especially acts of healing that were portrayed as the prime instigation for barbarian conversion. Thus the Arab leader Zocomus gratefully entered with his whole tribe into Christianity when a certain monk rendered his barren wife fertile. The Lakhmids held out as pagans until 593, when the king Nu‘man made his conversion as a result of being relieved by three Nestorian churchmen of a demon. And “when God wished in his bounty and generosity to save the pagans of ‘Ayn al-Namir and turn them from error, the son of the chief ’s sister fell ill and drew near to death,” thus giving the Nestorian monk Mar ‘Abda the chance to assert the supremacy of, and win round the chief ’s followers to, the “true faith.”32 Second, it is also true that Syriac Christian church authorities were involved to an increasing degree with an emergent Arab Christianity. For example, in the early fifth century CE Alexander, bishop of Mabbugh (northeast of modern Aleppo), built a church at Rusafa dedicated to St Sergius, to whom the Arab tribes of that region were much devoted, and both Jacob, bishop of Serug, and Severus, patriarch of Antioch, composed Syriac texts to celebrate this saint.33 This involvement also holds good for Christianity in Arabia. Thus Philoxenus, bishop of Mabbugh, consecrated the first two bishops of Najran in southwest Arabia in the early sixth century CE; the south Arabian martyr Elias had been a monk at the convent of Mar Abraham of Tella (east of Edessa) and had been ordained a priest by John, bishop of Tella; and Jacob of Serug and John the Psalter from the monastery of Aphtonia at Qenneshre (east of Aleppo) both penned works in honor 59 ROBERT HOYLAND of the Christians martyred at Najran in the 520s.34 So though there was no center of Syriac Christianity in the Hijaz, there certainly were Syriac Christian lines of communication passing through it. Third, the earliest dated Arabic text in Arabic script is that from Zebed (though it is out on a limb, as all other pre-Islamic Arabic inscriptions are from the Nabataean sphere of influence, see earlier), which certainly does belong to the Syriac Christian sphere of influence. The next two are from the Damascus area (Jabal Says and Harran), from which some Syriac literary material survives, though no Syriac inscriptions. However, it is also true that the Arab tribes allied to Rome operated in these areas (thus the Ghassanids had a camp at Jabiya to the southwest of Damascus and residences in Damascus itself 35), so the location of the pre-Islamic Arabic inscriptions could be used in support of either argument. Arab tribes and Christian missionaries Indeed, one could possibly combine the two arguments, for most of the Arab tribes allied to Rome posed as defenders and patrons of the Syriac Christian church to some degree.36 This connection is made by the Muslim tradition, which has men from the Christian Arab tribe of Tayyi’ derive the Arabic script from the suryani one and pass it on to Arab tribes at Hira and Dumat al-Jandal,37 and also by a mid-sixth century Syriac manuscript (BM syr. 14602) containing the signatures of the priests and abbots of the eparchy of Arabia (‘rby’). This latter document is very interesting in that the churches and monasteries to which these clergymen are attached are found not only in the Byzantine administrative province of Arabia with its capital at Bostra, but also portions of Phoenicia Libanensis, especially Damascene. In the words of Nöldeke: “Dies lässt sich nur so erklären, dass diese monophysitische Kirchenprovinz ‘Arabia’ so weit gerechnet wurde, wie die Macht der Ghassânischen Phylarchen ging;”38 that is, ecclesiastical “Arabia” was pretty much coterminous with the Ghassanid sphere of authority. The epigraphic evidence However, though it might be historically possible that Syriac would have formed the basis of the Arabic script, the surviving epigraphic evidence suggests otherwise. There are no Arabic inscriptions written in the Syriac script whereas there are quite a number of them written in Nabataean Aramaic script. A detailed examination of the change in letter forms from Nabataean Aramaic to Arabic has already been done in a study by Healey, in which he demonstrated how important Nabataean cursive (used for documents) was for the formation of the Arabic script. All I would like to add is a note about the discovery, since Healey wrote, of a number of graffiti from N.W. Arabia that provide clear examples of the Nabataean Aramaic script in transition.39 In particular, they answer the objection of the only proponent for the Syriac 60 THE LINGUISTIC BACKGROUND TO THE QUR’A N script as the source of the Arabic script rather than the Nabataean, namely that the latter is suspended from an upper line rather than resting, as with Arabic, on a lower line,40 for these transitional texts often manifest a clear horizontal lower line upon which the letters rest and by which they are connected (see Figures 8 and 9).41 This change in alignment, that is, the maintaining of a horizontal lower line, is responsible for some features of the Arabic script. For example, the classical Nabataean s/sh, an upright stroke with two crossbars (numerous times on figures 1–3; e.g. end of the second word on l.1 of Figure 1), gives, when aligned horizontally, the classical Arabic s/sh (see Figure 8, final word of l.1 and of l.2: the names Sa‘dw and Is). The classical Nabataean ‘ayn, a stroke at 45 degrees to the vertical with a crossbar or hook at right angles to it (cf. sixth and eighth words of l.1 of Figure 1), becomes lowered almost to the horizontal making it pretty much a standard classical Arabic ‘ayn (cf. second word of l.1 of Figure 8 and first word of l.2 of Figure 9: the names Sa‘dw and ‘Ubaydw). And the classical Nabataean final ya, which is quite upright (cf. first word of l.1 of Figure 1: ty), becomes lowered so that its tail starts to curve under the horizontal line of the letters (cf. first word of Figure 9, bly, and note its similarity to the third word of the early Islamic Arabic inscription shown in Figure 10, mwly42). Finally, medial het, which had been slightly erratic in form, begins to become aligned in the centre Figure 8 Graffiti, Umm Judhayidh (NW Arabia), ca. fourth to fifth century CE. 61 ROBERT HOYLAND Figure 9 Graffiti, Umm Judhayidh (NW Arabia), ca. fourth to fifth century CE. Figure 10 Graffito, Qa‘ al-Mu‘tadil (NW Arabia), AH 24. of the lower line and to take the form of the square figure-of-eight-like het (the so-called ‘ayn al-hirr) of early Islamic Arabic inscriptions (most pre-Islamic examples are as yet unpublished, but cf. the last word of l.1 of Figure 9 with the second word of Figure 10: the names Fahmw and Zuhayr). 62 THE LINGUISTIC BACKGROUND TO THE QUR’A N The date of these graffiti is somewhat vague. We have selected them because they exhibit letter forms that are between classical Nabataean (first century BC–first century CE) and the earliest Arabic (sixth and seventh centuries CE), but this still leaves a time span of some four centuries. Approximately twenty are dated, and these fall in the range 204–455 CE (assuming that the year numbers they cite are to be taken as referring to the era of Arabia, which began in 105 CE with the Roman annexation of the Nabataean kingdom). The last dated is Figure 10, which says it was written in the year 350, which in the era of Arabia would correspond to 455 CE.43 An additional date correspondence appears then to be given, namely “when ‘Amrw the king flourished (departed?).”44 If this reading is correct, then it could be an allusion to ‘Amr ibn Hujr al-Kindi, who was known to Muslim authors as an illustrious leader of the then politically important tribe of Kinda, “a man of sound judgment and sagacity.”45 We know that ‘Amr’s son al-Harith was wooed by agents of the Byzantine emperor Anastasius (491–518), so the mid-fifth century is a plausible floruit for ‘Amr. Now, even if the Syriac script didn’t inspire the Arabic script, one might still argue that Syriac literature inspired Arab sacred literature. Some degree of influence is of course likely given how important Syriac Christians were in spreading the Christian message to Arab tribes, but Luxenberg’s idea that Arabic writing was too rudimentary to support literature and so the Qur’an must be wholly derived from a Syriac Christian milieu seems excessive. It is noteworthy that the Qur’an itself is self-conscious with respect to the language in which it is written, stressing that it is “an ‘arabi recitation” (12:2) and “an ‘arabi decree” (13:37), composed in the “‘arabi tongue” (26:195, 46:12, 16:103), which has been made easy for Muhammad (19:97, 44:58) and is the language of his people (14:4). However, this may not impress the skeptic, and it is of course legitimate to question what might be meant by these terms.46 More concrete is the witness of the epigraphic record to the use of Arabic before Islam, which tells us that Arabic was of greater importance in many respects than Luxenberg allows for: a b c Arabic was evidently, as shown earlier, widely spoken in the Middle East by the seventh century CE, particularly in the region of the former Nabataean kingdom. Arabic was becoming widely written; even though not much survives, the very fact of the evolution of the Arabic script from the Nabataean script presumes frequent writing of Arabic in the Nabataean script (see earlier). Arabic had already long been used for sacred expression, as is attested by the Oboda inscription (see earlier and n. 9), and possibly also the two plausibly Arabic inscriptions from the Madaba area, both prayers to the god Sa‘b (see n. 16 and earlier). We also have Epiphanius of Salamis’ testimony as to the praises to a virgin deity sung in Arabic by the inhabitants of Petra and Elusa (see earlier). 63 ROBERT HOYLAND d Arabic had also long been used for literary expression. Note the parallelism in the Qaryat al-Faw text: “He has placed it (a tomb) under the protection of Kahl, Lah and ‘Athtar al-Shariq from anyone strong or weak, from anyone who would sell it or pawn it” and the poetic ending: “for all time without any derogation, as long as the sky produces rain or the earth herbage” (see n. 7 and earlier). And one can clearly discern deliberate grammatical and semantic Figure 11 Poetry quotation, Mecca (Arabia), AH 98. 64 THE LINGUISTIC BACKGROUND TO THE QUR’A N e arrangements in the Oboda text: “He (Obodat) acts (expecting) neither reward nor predilection. Though death has often sought us out, He afforded it no occasion. Though I have often encountered wounding, He has not let it be my destruction.”47 Furthermore there is the piece of fakhr (rhetorical boasting) at the end of the Nemara inscription: “No king matched his achievements up to the time he died” (see n. 8 and earlier). Finally one should take note of a piece of wisdom verse inscribed on a rock at Mecca in AH 98/717 CE (Figure 11). It is well known to Muslim authors, who usually attribute it to a bishop of Najran in southwest Arabia named Quss ibn Sa‘ida, or else to one of the pre-Islamic kings of Yemen.48 The attribution could be wrong, but given its early date the verse is plausibly preIslamic and this gives a little push to the idea that the mass of material that we have ostensibly going back to pre-Islamic times, in particular a vast wealth of poetry, does genuinely belong to that period. In conclusion, I would like to challenge the widely held view that Arabic was scarcely used before Islam except for orally transmitted poetry. The hackneyed image of the iceberg that is 90 percent hidden beneath water is worth adducing in this content, for the small number of currently known pre-Islamic Arabic texts are indeed but the visible tips of a now invisible, though nonetheless substantial tradition of writing and speaking Arabic. And this tradition needs to be more fully taken into account if we are to make better sense of the historical and linguistic context in which the Qur’an was revealed. Notes 1 For which see Arabic Literature to the End of the Umayyad Period, A.F.L. Beeston T.M. Johnstone, R.B. Serjeant and G.B. Smith (eds), Cambridge: Cambridge University Press, 1983, 27–113 (poetry), 114–31 (prose). For further discussion of the language situation see R. Contini, “Il Hawran preislamico ipotesi di storia linguistica,” Felix Ravenna 1/2, 1987, 25–79; R.G. Hoyland, “Language and identity: The twin histories of Arabic and Aramaic,” Scripta Classica Israelica 23, 2004, 183–99; F.Millar, “Il ruolo delle lingue semitiche nel vicino oriente tardo-romano,” Mediterraneo Antico 1, 1998, 71–94. 2 For some examples of this dialect see P. Mouterde, “Inscriptions en Syriaque dialectal à Kamed (Beq‘a),” MUSJ 22, 1939, 73–106; J. Barclay, “Melkite Orthodox SyroByzantine manuscripts in Syriac and Palestinian Aramaic,” Liber Annuus 21, 1991, 205–19 (some as late as the thirteenth century). 3 Bar Hebraeus, Ta’rikh mukhtasar al-duwal, Antun Salihani (ed.), Beirut: al-Matba‘at al-Kathulikiyya li-l-Aba’ al-Yasu‘iyyin, 1890, 18. J. Healey’s article, “Nabataean to Arabic,” is in Manuscripts of the Middle East 5 (1990–91). 4 See most recently F. Briquel Chatonnet M. Debié and A. Desreumaux (eds), Les inscriptions syriaques, Paris: Geuthner, 2004, 45–53. 5 A distinction does need to be made in that there are some small divergences from classical Arabic in these pre-Islamic Arabic texts, and more might be apparent if we had more/lengthier pre-Islamic Arabic texts to compare them fully with classical Arabic texts. See further W. Müller, “Das Altarabische und das klassische Arabisch,” in W. Fischer (ed.), Grundriss der arabischen Philologie I , Wiesbaden: Reichert, 1982, 65 ROBERT HOYLAND 6 7 8 9 10 11 12 13 14 15 16 17 17–36; C. Robin, “Les plus anciens monuments de la langue arabe,” in idem (ed.), L’Arabie antique de Karib’il à Mahomet, Aix-en-Provence: Edisud, 1991, 113–25; idem, “Les inscriptions de l’Arabie Antique et les études arabes,” Arabica 48, 2001, 545–56; and most recently M.C.A. Macdonald, “Old Arabic,” in K. Versteegh (ed.), Encylopedia of Arabic Language and Linguistics, Leiden: Brill, 2005. P. Jaussen and P. Savignac, Mission archéologique en Arabie, Paris: Leroux, 1909–22, Lihyanite no. 384. A.F.L. Beeston, “Nemara and Faw,” BSOAS 42, 1979, 1–6. Note that the south Arabian script was used for writing Arabic even into early Islamic times (Christian Robin, “Les graffites arabes islamiques écrits en caractères sudarabiques de Umm Laylà,” Semitica 26, 1976, 188–93). Beeston, “Nemara and Faw.” More recently see I. Shahid, “Byzantium and the Arabs during the reign of Constantine: the Nemara inscription,” Byzantinische Forschungen 26, 2000, 73–124; J. Retsö, The Arabs in Antiquity, London: Routledge, 2003, 467–76. Most recently see M. Kropp, “Iatromagie und der Beginn der arabischen Schriftsprache: die nabatäisch-arabische Inschrift von ‘Ayn ‘Abada,” Mélanges de l’Université Saint Joseph 55, 1997–98, 91–117. J. Healey and G. Rex Smith, “Jaussen-Savignac 17: the earliest dated Arabic document,” Atlal 12, 1989, 77–84. For these references see Retsö, The Arabs in Antiquity, 591, who also discusses the variants for the name of the goddess in Epiphanius. S. Krauss, “Talmudische Nachrichten über Arabien,” ZDMG 70, 1916, 321–53; J.B. Segal, “Arabs in Syriac literature before the rise of Islam,” JSAI 4, 1984, 104 (on yahle, gayse, and hablatha from Arabic ahl, jaysh and ibl). A. Grohmann, Arabische Paläographie 2, Vienna: Böhlau, 1971, 14–17; B Gruendler, The development of the Arabic scripts from the Nabataean era to the first Islamic century, Atlanta, GA: Scholars, 1993, 13–14. For the Zebed text see also F. Cumont, Catalogue des sculptures et inscriptions antiques des musées royaux, Brussels: Vromant, 1913, 172–5 (no. 145), and the Jabal Usays text has recently been re-read by C. Robin and M. Gorea, “Un réexamen de l’inscription arabe préislamique du Æabal Usays,” Arabica 49, 2002, 503–10. The Greek and Arabic concur for the first part of this text: “[Ar. I] Sharahil son of Talimu [Gr. phylarch] built this martyrium.” However, the Greek then says “of the holy John in the year 463 may the writer (i.e. of this) be remembered,” whereas the Arabic is usually understood to say “in the year 463 after the destruction of Khaybar by a year.” But it is odd that the Arabic text, which appears primary (note how it begins “I . . .”), does not mention the name of the dedicatee. Christian Robin has therefore suggested (in a forthcoming article entitled “La réforme de l’écriture arabe à l’époque du califat médinois”) reading instead “for [as is the sense of b‘d in Dedanite] the holy [assume mfsd is a mistake for mqds] John [i.e. Hnyn] vale [i.e. ni‘ma],” which certainly makes much better sense, though has its own problems (esp. the sense of b‘d and the fairly clear final r at the end of Hnyn/Khybr). E.A. Knauf, “Bemerkungen zur frühen Geschichte der arabischen Orthographie I: eine übersehene frühhocharabische Inschrift vom Nordostrand des Toten Meeres,” Orientalia 53, 1984, 456–8. D.G. Graf and M.J. Zwettler, “The North Arabian ‘Thamudic E’ Inscriptions from Uraynibah West,” Bulletin of the American School of Oriental Research 335, 2004, 53–89. Both are prayers to the god Sa‘b. B.A. Levine, “The various workings of the Aramaic legal tradition: Jews and Nabataeans in the Nahal Hever Archive” and A. Yardeni, “Notes on two unpublished Nabataean deeds from Nahal Hever – P. Yadin 2 and 3,” in L.H. Schiffmann, E. Tov and J.C. Vanderkam (eds), The Dead Sea Scrolls Fifty Years after their Discovery, Jerusalem: Israel Exploration Society, 2000, 836–51, 862–74. 66 THE LINGUISTIC BACKGROUND TO THE QUR’A N 18 R.W. Daniel, “P. Petra Inv. 10 and its Arabic,” Atti del XXII Congresso Internazionale di Papirologia, Florence: Istituto papirologico G. Vitelli, 2001, 1, 331–41 (esp. 338–9: “in spite of the fact that the Petra papyri are written in Greek . . . the toponyms show that society at large spoke Arabic”). 19 But it was most likely a very polyglot region. Its diversity is brought out by A. Negev, Personal Names in the Nabatean Realm, Jerusalem: Hebrew University, 1991, 2 (“342 personal names occur only in North Arabia, 327 only in Sinai, 218 only in Edom/Moab, and 117 only in the Hawran”); M.C.A. Macdonald, “Languages, scripts and the uses of writing among the Nabataeans,” in G. Markoe (ed.), Petra Rediscovered: lost city of the Nabataeans, New York: Abrams, 2003, 37–56. 20 And Figure 9 would seem to be by one ‘Amr the king (see later). 21 It is not certain; M. Zwettler, “Imra’alqays, son of ‘Amr, king of . . . ?” in M. Mir (ed.), The Literary Heritage of Classical Islam, Princeton: Darwin, 1993, 3–37, argues that ‘Arab refers here to a geographical area, not a people. Generally, it refers to the area around Hatra in modern N.W. Iraq, but it is also occasionally used of Roman Arabia (e.g. J. Parisot, “Book of the Laws of Countries,” Patrologia Syriaca 1, 1894, 602, 25: “Yesterday the Romans occupied ‘Arab and abrogated all the previous laws”). 22 See my “Epigraphy and the emergence of Arab identity,” in P. Sijpesteijn et al. (eds), From al-Andalus to Khurasan: Documents from the Medieval Muslim World, Leiden: Brill, 2006. Note, for example, the Ghassanid phylarch Abu Karib’s involvement in the settlement of a dispute between two church deacons concerning the sale of a vineyard (M. Kaimio, “P. Petra inv. 83: a Settlement of Dispute,” Atti del XXII Congresso Internazionale di Papirologia, 2, 719–24). 23 E. Littmann, Semitic Inscriptions. Princeton University Archaeological Expeditions to Syria, Leiden: Brill, 1914–49, 4A.41. 24 H. Humbach and P.O. Skjaervo, The Sassanian inscription of Paikuli 3.1, Wiesbaden: Reichert, 1983, 92. 25 W. Müller, “Eine sabäische Gesandtschaft in Ktesiphon und Seleukia,” Neue Ephemeris für Semitische Epigraphik 2, 1974, 155–65. 26 See I. Shahid, Byzantium and the Arabs, Washington DC: Dumbarton Oaks, 1984– Present, 1, 349–66 (on Hisham al-Kalbi’s writings about the pre-Islamic kings”; K. Athamina, “The tribal kings in pre-Islamic Arabia,” al-Qantara 19, 1998, 19–37; Retsö, The Arabs in Antiquity, 473–85. 27 See, for example, W. Müller, “Ergebnisse neuer epigraphischer Forschungen im Jemen,” ZDMG suppl. 3, 1977, 732, on a mid-third-century inscription whose author had been sent on a mission “to the kings of the tribes (ash‘b) of Ghassan, al-Asd, Nizar, and Madhhij,” and C. Robin and I. Gajda, “L’inscription de Wadi ‘Abadan,” Raydan 6, 1994, 113–37, on Himyarite campaigns against Murad, Iyad, Ma‘add, and ‘Abd al-Qays to the northeast of Mecca “between the land of Nizar and the land of Ghassan” (in the Himyarite year 470 = 360 CE). On these tribes in the Muslim sources see the relevant entries in EI2. 28 The only exception is Tayyi’, which appears as a self-designation in S. Campetti and E. Borzatti von Löwenstern, L’altra umanità. Origini Storia e arte dei nomadi della tenda nera, Florence: Sansoni, 1983, 16, and as a designation of others in Corpus Inscriptionum Semiticarum 5.1, Paris, 1950, no. 2795; E. Littmann, Safaitic inscriptions, Leiden: Brill, 1943, 236; V. Clark, A Study of New Safaitic Inscriptions from Jordan, PhD thesis, Melbourne, 1979, 1004. It features in Syriac literary sources, as Tayyayê, from the third century CE onward. 29 In Inscriptions grecs de la Syrie, no. 2110, for example, Flavius Seos and his son, builders of a house, refer to the reign of Mundhir. Similarly, an event in John Moschus’ Pratum Spirituale is dated to “when Nu‘man (Names), the phylarch of the Saracens, was making raids” (Patrologia Graeca, J.P. Migne (ed.), 87.3, 3024 = ch. 155) and Ms. BM Syriac 585 of the monastery of Natpha near Tadmur (Palmyra) is dated to 67 ROBERT HOYLAND 30 31 32 33 34 35 36 37 38 39 40 41 42 43 when Abu Karib, a Ghassanid, was king (W. Wright, Catalogue of the Syriac manuscripts in the British Museum, London, n.p., 1871, 2:468). Most famous is Hassan ibn Thabit’s picture of the Ghassanid court popularized by R.A. Nicholson in his A Literary History of the Arabs, Cambridge: Cambridge University Press, 1969, 53: “I have seen ten singing-girls . . . and when he would drink wine he sat on a couch of myrtle and jasmine . . . surrounded by gold and silver vessels full of ambergris and musk.” H. Lietzmann and H. Hilgenfeld, Das Leben des heiligen Symeon Stylites, Leipzig: Hinrichs, 1908, 108. Sozomen, Historia Ecclesiastica, R. Hussey (ed.), Oxford: Academic Press, 1860, 6, 38 (Zocomus); Chronicle of Siirt, ch. 60, in Patrologia Orientalis 13, 1919, 468–9 (Nu‘man); ibid., ch. 97, 586–89 (Mar ‘Abda). For these and other references see E. Key Fowden, The Barbarian Plain. Saint Sergius between Rome and Iran, Berkeley, CA: University of California Press, 1999. I. Shahid, The Martyrs of Najran, Brussels: Soc. des Bollandistes, 1971, vi-vii/45–46 R. Schröter (ed./trans.), “Trostschreiben Jacobs von Sarug an die himjaritischen Christen,” ZDMG 31, 1877, 361–405. Ibn Rusta, al-A‘laq al-nafisa, M.J. de Goeje (ed.), Leiden: Brill, 1892, 7, 326; note that al-Idrisi, Opus geographicum, A. Barbaci et al. (eds.), Naples-Rome: Istituto Universitario di Napoli, 1982, 368, says that modern Damascus was built on top of Jabiya, and Yaqut al-Hamawi, Mu‘jam al-buldan, F. Wüstenfeld (ed.), Göttingen: Brockhaus, 1866–73, s.v. “Halima,” says that “al-Harith the Ghassanid was Caesar’s deputy in Damascus”! Shahid, Byzantium and the Arabs, 3, 691–995; J.S. Trimingham, Christianity among the Arabs in Pre-Islamic Times, London: Longman, 1979. Al-Baladhuri, Futuh al-buldan, Salah al-Din al-Munajjid (ed.), Cairo: Maktabat al-Nahda al-Misriyya, n.d., 579; Ibn al-Nadim, Fihrist, G. Flügel (ed.), Leipzig: Brockhaus 1871, 4–5; cf. Ibn ‘Abd Rabbihi, al-‘Iqd al-farid, ‘Abd al-Majid al-Tarhini (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1987, 4, 240. T. Nöldeke, “Zur Topographie und Geschichte des Damascenischen Gebietes und der Haurângegend,” ZDMG 29, 1875, 420. See also T.J. Lamy, “Profession de foi,” Actes du onzième congrès international des Orientalistes, Paris: Leroux, 1897, 117–32, and more recently Shahid, Byzantium and the Arabs, 3, 821–38. At the forefront of these discoveries are the Saudi scholars ‘Ali Ghabban, Khalil al-Muaikil, Mushallah al-Muraykhi and Sulayman al-Theeb. The first three of these scholars together with Laila Nehme, Michael Macdonald, Christian Robin and myself are in the process of collating and analyzing these texts. J. Starcky in “Petra et la Nabatene,” Dictionnaire de la Bible, Supplément 7, 1966, 932–4. = S. al-Theeb, Nuqush Jabal Umm Jadhayidh al-Nabatiyya, Riyad: Maktabat al-Malik Fahd al-Wataniyya, 2002, nos. 128 and 132–3. This inscription, simply saying “I am Zuhayr mawla (client) of Ibnat Shayba” (note the diacritical marks on the letters n and z), is next to another by the same person which he dates to “the time when ‘Umar died, the year 24,” that is, 644 CE, making these the earliest Arabic inscriptions (but not the earliest Arabic texts, since there are papyri from the year AH 20/640 CE). For a full study see the article of their discoverer ‘Ali Ghabban, “Aqdam naqsh islami – mu’arrakh 24 H/644–645 M,” Arabia 1, 2003, 293–339 (the image of Figure 10 is on p. 300). The text reads: (l.1) “Indeed may there be remembered Fahmw son of (l.2) ‘Ubaydw (l.3) with goodness and peace; the year is 2 x 100 (l.4) 100 20 20 10 when there flourished (l.5) ‘Amrw (l.6) the king. Its publisher, Sulayman al-Theeb (Nuqush Jabal Umm Jadhayidh al-Nabatiyya, nos. 132–3), considered it two texts, because they are 68 THE LINGUISTIC BACKGROUND TO THE QUR’A N 44 45 46 47 48 written on two different (though adjacent) faces of the rock, but this yields nonsense (thus al-Theeb begins his commentary: “We cannot divine what made the author of this commemorative text write his text in this style . . .”). Idh jlw ‘Amrw al-malik. This is my own suggestion, and it remains tentative. However, it seems evident that some action or event concerning king ‘Amrw is intended. Al-Tabari, Ta’rikh, M.J. de Goeje (ed.), Leiden: Brill, 1879–1901, 1, 880–1; Wahb b. Munabbih, Kitab al-Tijan, Hyderabad: Matba’at Majlis Da’irat al-Ma’arif al-’Uthmaniyya, 1929, 299; Theophanes, Chronicle, C. de Boor (ed.), Leipzig: Teuberi, 1883–5, 144; Photius, Bibliotheca, trans. R. Henry (ed.), Paris: Société d’édition Les Belles letters, 1959–77, no. 3 (= account of Nonnosus, a Byzantine diplomat, who tells how his grandfather, also a diplomat, was sent to al-Harith son of ‘Amr son of Hujr of Kinda, and he and his father to Qays, “a descendant of al-Harith”). There is some discussion and references to further literature in EQ, 1, s.v. “Arabic language.” The translation is almost certainly inaccurate in parts, but it conveys the literary feel that all commentators agree is present; see n. 9 earlier and A.F.L. Beeston, “Antecedents of classical Arabic verse” in Festschrift Ewald Wagner, Beirut: Steiner, 1994, 1, 234–43. Sa‘d al-Rashid, Kitabat islamiyya min Makka al-Mukarrama, Riyad: Markaz al-Malik Faysal li-l-Buhuth wa-l-Dirasat al-Islamiyya, 1995, no. 17, with discussion of the Muslim sources for the verse. Note that the verse usually begins “The sun’s revolving prevents continuation (mana‘a al-baqa’),” not, as in Figure 11, “The sun’s revolving extinguishes the new (afna al-jadid).” And Figure 11 lacks the usual concluding verse (presumably well enough known to everyone): “I/We/You know not what today will bring, while yesterday has passed with its portion of cares.” 69 3 RECENT RESEARCH ON THE CONSTRUCTION OF THE QUR’AN Gerhard Böwering The Qur’an, proclaimed in Arabic by Muhammad in the northwestern region of the Arabian peninsula during the first quarter of the seventh century, reveals a significant relationship to the tradition of the Jewish and Christian scriptures and echoes themes found in their midrashic and apocryphal writings. No single collection of biblical writings, normative, apocryphal or midrashic, however, has been identified as the major source in which the Qur’an may have been rooted.1 To the best of our present knowledge, the Bible had not been translated into Arabic by the time of Muhammad, either in its entirety or in the form of single books.2 It is generally believed that Muhammad gathered his biblical knowledge principally, if not exclusively, from oral sources.3 This oral lore was communicated to Muhammad in his mother tongue, but its original forms were in Syriac, Aramaic, Ethiopian and Hebrew materials, as evidenced by the vocabulary of foreign origin to be found in the Arabic Qur’an.4 This foreign vocabulary formed an integral part of Muhammad’s proclamation and was understood by his audience in Mecca and Medina whom he addressed in eloquent Arabic.5 The earlier paragraph represents a summation of my present position on the single most vexing question in the field of Qur’anic studies, the question of how the Qur’an was composed and finally codified. No issue has grabbed more contemporary attention, both scholarly and non-scholarly, and no issue is more likely to generate religious controversy between Muslims and nonMuslims. In recent decades several publications have proposed radical revisionist theories of Qur’anic origins. Some of these have captured more attention than others, both positive and negative, but the response has been text specific rather than comparative and comprehensive. What I will do in the pages that follow attempts to address that lack by bringing forward a number of these studies for sustained analysis. That analysis will not be comprehensible, however, unless it is placed within the context of the last century and a half of Western scholarship on the Qur’an. The insights and consensus achieved during that period form the backdrop against which these newer studies must be understood and assessed. 70 RECONSTRUCTING THE QUR’ AN The legacy of Muslim studies on the Qur’an As the first book-length production in Arabic literature, the Qur’an stands at the crossroads of the oral, and highly narrative and poetical, tradition of pre-Islamic Arabic on the one hand and the written, and increasingly scholarly, Arabic prose tradition of the subsequently evolving civilization of Islam on the other. The beginnings of this transition from the oral to the written can be pinpointed to the time of Muhammad and are clearly reflected in the rhymed prose style of the Qur’an. This rhymed prose, the mode of speech of the pre-Islamic soothsayer’s oracles, is a characteristic mark of the Qur’an, the first sizable Arabic document to depict this form in writing. It is not known, however, whether Muhammad himself was able to read and write.6 Muslims generally claim that he was illiterate, although his professional involvement in trade and commerce might argue for some form of acquaintance with written record keeping. From the earliest centuries of Islam, Muslim scholars exhibited a particular sensitivity to inconsistencies affecting a variety of legal stipulations in the Qur’an. Acknowledging the differences found in disparate verses, they developed an intra-Qur’anic theory of abrogation that replaced the legislative force of an earlier Qur’anic verse with that of a later one.7 Hundreds of examples were cited in support of this theory. For example, the infamous “Satanic verses” were understood to have been replaced by an unrelated Qur’anic passage, with two later verses explaining the role of the Satanic interference.8 Other Muslim scholars developed a criterion, called “the occasions of the revelation,” which connected a particular Qur’anic verse or group of verses with episodes from Muhammad’s career drawn from extra-Qur’anic tradition.9 Both methods focused their analysis on individual verses, rather than on Qur’anic chapters as integral units. Another group of Muslim scholars active in later medieval times based their analysis on the assumption that the individual suras formed original units of revelation and could best be divided into two sets, “Meccan” and “Medinan,” according to whether they were revealed before or after Muhammad’s emigration from Mecca to Medina, an event known as the hijra. The Itqan of Suyuti (d. 911/1505), which incorporated these scholarly decisions, as well as many others, became the epitome of the fully developed discipline of Qur’anic studies.10 It has served to convey the legacy of such studies to subsequent Muslim generations and it has been carefully analyzed by Western scholars who long used it as the principal source for their investigations into the construction of the Qur’an. The legacy of Western studies on the Qur’an From the mid-ninteenth century Western scholars began to engage in serious literary research on the Qur’an, linking the established conclusions of traditional Muslim scholarship with the philological and text-critical methods that biblical scholarship was developing in Europe. An intensive scholarly attempt was made to create a chronological order of Qur’anic chapters and passages that could be 71 GERHARD BÖWERING correlated with the situations and varying circumstances of Muhammad’s life and career. This Western chronological approach to the construction of the Qur’an reached full elaboration in the work of Theodor Nöldeke,11 a conclusion that was then challenged by Richard Bell, and brought to a balanced adjudication by Rudi Paret’s manual of commentary and concordance to the Qur’an.12 While the Western scholarly consensus adopted the traditional distinction between Meccan and Medinan suras, it subdivided the Meccan phase of Muhammad’s proclamation into three distinct periods, taking the Medinan period as the fourth. These four periods were linked to a conception of the gradual inner development of Muhammad’s prophetic consciousness and of the emergence of his political career. Western scholarship forged this linkage through its biographical research on the life of Muhammad, which was worked out simultaneously with its research on the Qur’an.13 Since chronological research on the Qur’an and biographical research on Muhammad’s career were closely dependent on one another, the possibility of a circular argument remained a constant danger because the subjective evaluation of Muhammad’s religious development had to be read back into various and disparate Qur’anic verses from which it had been culled originally. In general, the fourfold chronology of the Qur’anic proclamation was formulated on the basis of two modes of analysis: the first related Qur’anic passages source-critically to historical events known from extra-Qur’anic literature, and the second systematically analyzed the philological and stylistic nature of the Arabic text of the Qur’an passage by passage. The most radical chronological rearrangement of the suras and verses of the Qur’an was undertaken by Richard Bell14 who suggested that the composition of the Qur’an followed three main phases: a “Sign” phase, a “Qur’an” phase and a “Book” phase. The earliest phase, which is characterized by “sign passages” and exhortations to worship God, represents the major portion of Muhammad’s preaching at Mecca, of which only a fragmentary amount survives. The “Qur’an” phase included the later stages of Muhammad’s Meccan career and the first two years of his activity at Medina, a phase during which Muhammad was faced with the task of producing a collection of liturgical recitals. The “Book” phase belonged to his activity at Medina and began at the end of the second year after the hijra, from which time Muhammad set out to formulate a written scripture. In the present Qur’an, however, each of these three phases cannot be precisely separated, because sign passages came to be incorporated into the liturgical collection and earlier oral recitals were later revised to form part of the written book. In explaining his complex system of distinguishing criteria, Bell convincingly argued that the original units of revelation were short, piecemeal passages which Muhammad himself collected into suras. He further stipulated that written documents were used in the process of redaction, a process undertaken with the help of scribes during Muhammad’s career in Medina. The watershed event for the creation of the Qur’an as sacred scripture, Bell argued, was the battle of Badr that occurred two years after the hijra. In Bell’s view, the hijra itself did not constitute a great divide for the periodization of the suras. 72 RECONSTRUCTING THE QUR’ AN None of the systems of chronological sequencing of Qur’anic chapters and verses has achieved universal acceptance in contemporary scholarship. Nöldeke’s sequencing and its refinements have provided a rule of thumb for the approximate chronological order of the suras. Bell’s hypothesis has established that the final redaction of the Qur’an was a complex process involving successive revisions of earlier material, whether oral or already available in rudimentary written form. In many ways, Western scholarship on the Qur’an reconfirmed the two assertions on which the traditional Muslim views of the Qur’anic chronology were based. First, the Qur’an was revealed piecemeal and second, it was collected into book-form on the basis of both (1) written documents prepared by scribes on Muhammad’s dictation and (2) Qur’anic passages preserved in the collective memory of his circle of companions. All methods of chronological analysis, whether those of traditional Muslim scholarship or modern Western, agree that the order of the suras in Muhammad’s proclamation was different from the order found in the present-day Qur’an where, in general, the suras are arranged according to the principle of decreasing length. Furthermore, they agree that the redaction and canonical completion was a complex process, one whose study presents a minefield of historical problems from its inception until the appearance of the final vocalized text. The ‘Uthmanic codex of the Qur’an and its assessment by Wansbrough and Burton After Muhammad’s death, the Muslim community faced three major tasks with regard to establishing the Qur’an as canonical scripture: it had to collect the text from oral and written sources, establish the skeleton of the Arabic consonantal text, and finalize the fully vocalized text that came to be accepted as the canonical standard. The traditional view depicting the way in which these tasks were accomplished covers three centuries and telescopes the history of the text into a basic scenario.15 This scenario proceeded on the assumptions that Muhammad did not leave a complete written text of the Qur’an and that the Qur’an was preserved primarily in oral form in the memory of a considerable number of Muhammad’s direct listeners, although also in written form by scribes during his lifetime. A group of companions, led by a scribe, whom Muhammad himself had employed in Medina, gathered and arranged the oral and written materials of the Qur’an in a complete consonantal text some twenty years after Muhammad’s death.16 This text, the Qur’anic codex (mushaf), came to be known as the “Uthmanic” text or codex, because it was completed in the latter years of the caliphate of ‘Uthman.17 The final, fully vocalized text of the Qur’an, however, was only established in the first half of the tenth century, after different ways of reading including slight variations of vocalization, came to be tolerated and accepted as standard.18 Since the ‘Uthmanic text was written in a scriptio defectiva that was merely a consonantal skeleton (rasm) lacking the diacritical marks which could distinguish 73 GERHARD BÖWERING certain consonants from each other, it needed oral recitation as a necessary accompaniment to secure the intended pronunciation of the text. As the Qur’anic orthography developed step by step over more than two centuries and as the linkage between the consonantal skeleton and the oral recitation became more and more defined, the deficiencies of the Arabic script were gradually overcome and the written text became increasingly free of its dependency on oral pronunciation. This process culminated with the scriptio plena of the Qur’an that provided fully pointed consonants and a complete set of vowel signs. This final, stabilized text may be termed as a textus receptus, ne varietur with the proviso, however, that no single clearly identifiable textual specimen of the Qur’an was ever established unequivocally. Today, however, for all practical purposes, only one predominant version is in general use, namely the one adopted by the Egyptian standard edition of the Qur’an produced in 1924.19 Assessing this complex history of the text, two Western scholars dismissed the traditional Muslim vision of the collection of the Qur’an in the second half of the last century, but reached opposing conclusions. John Wansbrough (1977) argued that the Qur’an was not compiled until two to three hundred years after Muhammad’s death,20 while John Burton (1977) contended that Muhammad himself had already established the final edition of the consonantal text of the Qur’an.21 Such widely differing hypotheses, as well as the fact that there is no single uniform text of the Qur’an that can provide the basis of a text-critical edition built on representative manuscripts and critically evaluated variant readings, demonstrate that the chronological construction of the Qur’an’s fixation as a written text has reached an impasse. Only the future will tell whether a possible computer analysis of the mass of manuscript material can allow scholarly research to develop a more consistent picture of the Qur’an’s textual history. In the beginning of the twenty-first century, however, the publications of two scholars, Günter Lüling and Christoph Luxenberg (a pseudonym of the actual author), have tried to overcome this impasse by proposing radical solutions to the problem of reconstructing the textual history of the Qur’an. The Ur-Qur’an of Lüling Lüling initiated his studies in the latter part of the twentieth century with three German monographs that focus upon an intense search for an Ur-Qur’an.22 He reopened the scholarly debate about the sources of Muhammad’s proclamation and asked whether Muhammad began to produce religious rhymed prose only after the defining religious experience of his call to prophecy, an event that took place when he was a man of about forty years of age. Searching the text for segments that could predate this event, Lüling looked for their roots in forms of religious worship and liturgy within the Arab environment in which Muhammad grew up and reached his maturity. Despite his multiple German publications on this topic, Lüling’s work has been largely dismissed or ignored. Undeterred by this reaction, in 2003 he published a large and laboriously titled volume 74 RECONSTRUCTING THE QUR’ AN in English that presents a compilation and reworking of his earlier studies.23 Lüling’s Challenge to Islam: The Rediscovery and reliable Reconstruction of a comprehensive pre-Islamic Christian Hymnal hidden in the Koran under earliest Islamic Reinterpretations has a basic thesis and proposes a radical hypothesis. His thesis claims that an earlier Christian hymnal underlies about a third of the present-day Qur’anic text as a hidden layer. Onto this pre-Islamic Christian layer various textual strata were superimposed. They can be rediscovered and traced to a period before and during Muhammad’s lifetime and received their final redaction by the early Muslim community after Muhammad’s death. To establish the existence of this pre-Islamic layer in the Qur’an, Lüling’s work proceeds from the hypothesis that before and during Muhammad’s lifetime there existed a substantial Christian community in the city of Mecca, the birthplace and principal theater of Muhammad’s prophetic proclamation until he migrated to Medina. In support of his basic thesis Lüling begins his arguments with the well-known fact that the earliest Qur’an codices were written in a scriptio defectiva that included only the consonantal skeleton (rasm), with the diacritical points (nuqat) being added later. At an even later date, the addition of vowel signs (harakat) created the text in scriptio plena. He then proceeds to make three assertions. First, because of its defectiveness, the original ductus of the Arabic script was subjected to such extensive manipulation that the intended meaning was obscured in the process of perfecting the Qur’anic script. Examples of this would be the deviant placement of diacritical points and vowel strokes or the intentional alteration of individual characters and words. Second, the existence of strophic refrains in the present-day Qur’an indicates that there must have been a strophic structure of original hymns, which were subsequently transformed into prose and changed in their wording in the process of redaction. They were made to fit the overall Qur’anic pattern of rhymed prose, which, because of the lack of meter, is considered prose rather than poetry (although it resembles poetry because of the consistent phenomenon of end rhymes). As postulated by Lüling, while the strophic hymns are no longer apparent in the present-day Qur’an, they can be reconstructed by establishing their intrinsic relationship with Christian Ethiopian hymns derived from Coptic originals in about 500 CE. Third, critical examination can rediscover different layers in the text of the present-day Qur’an. One third of the Qur’an is comprised of the layer of the original Christian hymns, which were transformed in the Qur’an through the superimposition of Islamic interpretations. The other two thirds include the layer of purely Islamic texts that can be traced back to Muhammad and the textual layer reinterpreted by the redactors who, in postMuhammad times, fashioned the final appearance of the present-day Qur’an. The motives for the successive revisions of the Qur’anic text, from its pre-Islamic origins to its post-Muhammad reinterpretations were, according to Lüling, a mixture of dogmatic, historical and tribal motives. The tribal Arabs of Mecca took action against a Hellenistic, trinitarian Christianity that had spread in central Arabia, particularly in Mecca. They wanted to be liberated from this foreign-dominated Christianity that had taken root in their midst for about 75 GERHARD BÖWERING two centuries and had used the Ka’ba, the central sanctuary of Mecca, as its church. In Ebionite, non-trinitarian Christianity, represented by the pre-Islamic god-seekers or hanifs mentioned in Muslim tradition, they found an ally against trinitarian Christianity and its various sectarian trends. The hanifs were attracted to Muhammad’s leadership because of his rejection of Christ’s divine sonship and their own adherence to an “angel christology” which interpreted Christ as an angelic and created being. For their part, the tribal pagan Arabs found Muhammad’s summons for a return to “the religion of Abraham, Isma’il and the tribes”24 as serving their national agenda. Muhammad, however, betrayed the Ebionite cause of the hanifs when he espoused his belligerent ways of jihad to further the national agenda of the tribal Arabs. The victory of Muhammad’s Muslim Arab community over the trinitarian Christian “associaters” (mushrikun) opened the way for a reworking of the Ur-text of the Qur’an into a classical Arabic prose text with end rhymes that removed the strophic structure of the Christian hymns and thereby obliterated the memory of its actual Christian origins. Linguistically speaking this classical prose text of the present-day Qur’an is the end product of a process. This process begins with the pre-Islamic Christian Arabic of the strophic hymns that, according to Lüling, was a vernacular devoid of case endings for nouns or modal endings for verbs but enriched with a stock of Syriac and Hebrew loanwords. Onto this literate vernacular was grafted a linguistic genre which is mistakenly assumed to reflect the ecstatic mode of Muammad’s Qur’anic utterances but which actually represents a non-literate level of language mainly marked by a preference for nominal constructions. Finally, the composite that resulted from the merger of these language levels was supplemented with a large complex of prose texts in elegant classical Arabic. Historically speaking, so Lüling argues, certain statements and anecdotes found in the extra-Qur’anic Muslim tradition confirm this whole process of orthographic, linguistic, dogmatic and historical transformation of the “Ur-Qur’an” – a text that has to be rediscovered by reconstruction – into the present-day text. To situate this development of the present-day Qur’an from the Ur-Qur’an, Lüling needs a Sitz im Leben and he finds it in the geographical regions of the central and northwestern Arabian peninsula. According to Lüling, this area was thoroughly christianized by Muhammad’s lifetime, and Mecca was a significant Christian town ruled by the Quraysh, a Christianized tribe that worshipped in the Ka‘ba, a Christian church built with an orientation toward Jerusalem. Unfortunately, this assertion of a substantial and organized Christian presence in Arabia before and during the lifetime of Muhammad cannot be documented in any Muslim or Christian sources that have been discovered to date. Lüling’s minute philological analysis of a small number of select Qur’anic passages presents text portions as superimposed on earlier text strata and illustrates the reworkings of Qur’anic passages in the process of redaction. Unfortunately, however, these complicated reworkings are intuited rather than substantiated in Lüling’s study. There is reasonable plausibility to Lüling’s basic 76 RECONSTRUCTING THE QUR’ AN claim of Jewish-Christian source materials underlying certain passages of the Qur’an. This material, so Lüling justly presupposes, cannot simply have been an oral transmission. There must have been some literary form from which it ultimately derived. Whether this form was in Coptic is a more dubious point. Lüling’s assumption of a massive Christian presence in central and northwestern Arabia also cannot be substantiated.25 Yet there can be no doubt that Christianity, in some form and to some extent, had arrived in the Hijaz and the environs of Mecca by Muhammad’s lifetime. In addition, it had a firm foothold in the regions of the Arabian peninsula bordering on Iraq, Syria and Palestine, was entrenched across the Red Sea in Ethiopia, and had major seats in al-Hira in Iraq and Najran in Yemen. The Meccan merchants, and with them Muhammad, thus had ample opportunity to encounter Christians within the Arab environment. The lectionary of Luxenberg Where Lüling leaves off, Luxenberg begins his attempt to decode the language of the Qur’an with a highly ambitious approach. His findings, first published in 2000, but now available in a revised and enlarged second edition of 2004,26 created quite a stir in the media and spawned a number of sensation-seeking articles in the popular press. That initial publication, however, has given new impulse to the study of the Qur’an in more scholarly venues because of the startling conclusions that the author drew from his analysis. Taken as a whole, his conclusions dismiss the entire edifice of Muslim Qur’an commentary as irrelevant and imply that all extant copies of the Qur’an, whether ancient or modern, include major and myriad misreadings. Luxenberg’s monograph is both narrowly philological in method and broadly speculative in its conclusions. He operates in the splendid isolation of purely philological intuitions, but disregards any form of historical– critical analysis. The results of his research provide many plausible new readings of the text, but very few of his conclusions carry the resounding ring of genuine certitude. In addition, Luxenberg consistently disregards the scholarly accumulation of almost two centuries of Qur’anic textual criticism and excuses this disregard with a profession of academic purity, namely that he thereby avoids being influenced by previous studies. Based on his inquiry of obscure Qur’anic passages, Luxenberg seeks to prove that the consonantal text of the ‘Uthmanic version was misread by the early generations of Muslims who were no longer aware that the language of the Qur’an was profoundly influenced by Syriac. To recapture the substratum of Syriac in the Arabic of the present-day Qur’an, the author diminishes the role of oral tradition in the transmission of the Arabic Qur’an, changes diacritical marks and occasionally even alters the skeleton of the consonantal text, and rifles the Chaldean dictionary for Syriac roots – albeit with scarce attention to their historical usage. This results in some astonishing readings, some of which appear plausible at times, as when elegant solutions are suggested for Qur’anic phrases that experts have been unable to render with certitude. Instances would include, 77 GERHARD BÖWERING for example, the observations on the names of Abraham, the emendation of bara’a as referring to the covenant, and the analysis of the word for Torah.27 Though not impossible, some other readings are startling as when the heavenly virgins are banished from the Qur’an by means of a semantic substitution that understands the houris as grapes, a fruit of paradise.28 Unfortunately, Luxenberg cannot document a single short sura in its actual Arabic wording that can be set side by side its supposedly Syriac substratum. While he examines three suras in their entirety, their dependence on Syriac is hardly as certain as he wishes it to be. The sura generally understood to include the first revelation to Muhammad is interpreted by Luxenberg as a liturgical invitation to the Eucharist, ending with the words, wa-sjud wa-qtarib (Q 96:19), and meaning in Luxenberg’s expanded – and very free – translation, “perform your divine service and take part in the eucharist.” His analysis, however, overlooks the strength of a parallel passage (Q 53:62) and requires a weak interpretation of iqtarib, an imperative in the eight stem, which would have to be taqarrab in the fifth stem to sustain this interpretation that Luxenberg claims on the basis of a Christian poem.29 The short sura entitled al-kawthar (Q 108) is hardly the first documented proof of the existence in the Qur’an of a passage from a NT letter. In the Syriac of the Peshitta, the perseverance in prayer in the face of the devil (1 Pt 5:8–9) uses a term for the satanic adversary, which bears no etymological affinity to the Arabic for the human adversary in the Qur’an. Furthermore, Luxenberg’s switch from the Arabic for “sacrifice” to “persevere” in Q 108:2 is simply arbitrary.30 The grounds for Luxenberg’s assertion of the appearance of Christmas in the Qur’an are even more restricted, namely a personal pronoun in surat al-Qadr (Q 97:1). In this case, a Qur’anic parallel reference (Q 44:3) directly counters Luxenberg’s interpretation which replaces the “night of power” as the standard reference for this verse with an allusion to the infant Jesus in Syriac liturgy.31 Also, it cannot be overlooked that, generally speaking, Semitic roots invite creative tinkering and that one can offer alternate explanations for Luxenberg’s readings by manipulating roots in Hebrew, Ethiopic or Syriac in comparison with Arabic. Luxenberg’s radical methodology, however, takes a much bolder leap. It equates the Qur’an with the qeryana, originally meaning lectio rather than lectionarium, the technical term for readings from scripture used in the Syriac liturgy, and identifies the Qur’an as a “lectionary” that included readings from the Old and New Testaments, as well as liturgical prayers, psalms and hymns. As Luxenberg argues, the book of the Qur’an (kitab), at least initially, resembled such a lectionary, and the “mother of the book” (umm al-kitab) was the Bible, the source-book of the Syriac lectionary. Luxenberg insists that Muhammad could probably read and write, that he traveled to Aramaic-speaking areas as a merchant and that he presumably came into contact with Aramean traders in Mecca itself.32 This social mix of Arabic and Aramaic speakers in Mecca – which for Luxenberg was originally an Aramaic settlement – facilitated the emergence of the Qur’an as its first written expression. The language of this Qur’an was created by scribes who were familiar with the cultural language of the Aramaic milieu and who 78 RECONSTRUCTING THE QUR’ AN produced a mixed language, one which blended Arabic with Syriac in such a fashion that almost a third of the Qur’an’s content forms a textual layer derived from Syriac. This hybrid Qur’anic language, which Luxenberg assumes to be a reflection of the Meccan dialect of the time, also included loanwords from a variety of other sources and can be traced, in particular, in the Meccan suras of the present day Qur’an. Furthermore – and this is decisive for Luxenberg’s argument – after Muhammad’s death the true meaning of this hybrid language was forgotten in what must have been an astonishingly wide-spread loss of memory in the Muslim world. Later generations, familiar only with Arabic but no longer with Syriac, and scribes, whose ancestors had left the Hijaz to live in the conquered areas of the Fertile Crescent, were unable to understand the original mixed language and recorded the Qur’an in the classical type of Arabic in which we have it today. The gradual disappearance of a knowledge of Syriac among Muslim Arabs began in the reign of ‘Abd al-Malik (685–705) when Syriac was replaced by Arabic as the official written language for the administration of the Umayyad empire. Luxenberg further asserts that the Arabic script offers only an image of a language but cannot help us determine whether this scriptural image was also spoken. Because of the widespread loss of memory and consequent disappearance of Syriac in the Muslim world, the oral conveyance of the Qur’an was cut short. Given this interruption in the process of transmission, Muslim exegesis of the Qur’an never operated authentically because it could not base itself on the original formulation of the Holy Book. This pessimistic assessment of the adequacy of oral tradition in the transmission of the Qur’an from the time of Muhammad to that of the fixation of the written text, allows Luxenberg to claim enormous freedom in emending the Qur’anic text. With passages he deems obscure or in need of repair, he is free to move and remove diacritics or invert the sequence of letters within a word by metathesis or graft different vowels on a penstroke (mater lectionis), in order to make Arabic words fit the Syriac roots he has in mind. By assuming a complete break in the oral transmission of the Qur’an, Luxenberg can assert the primacy of a hypothetical text, one written in the rudimentary form of an Arabic ductus without diacritical dots, a text that he claims enshrines a layer of Syriac equivalent to 30 percent of the text. Unfortunately, this postulated text of the Holy Book cannot be documented by a single manuscript, either in whole or in part. The scenario of Azzi Behind Luxenberg stands a French work by Joseph Azzi published in 2001 that was preceded by a number of publications in Arabic issued under a pseudonym.33 Azzi’s book resuscitates the question of a possible link between the Qur’an and one form of Jewish-Christianity, known as the Ebionites. This link with Ebionite Christianity has pre-occupied scholarship since the times of A. von Harnack and, later, Hans-Joachim Schoeps, who both focused on possible Jewish-Christian 79 GERHARD BÖWERING origins of Islam.34 The subject also drew the scholarly attention of such scholars as Tor Andrae and Richard Bell as well as more recently, J. Spencer Trimingham.35 M.P. Roncaglia affirmed this connection explicitly in a 1971 survey article.36 On the basis of this assumed link, which identifies the Ebionites with those Christians who are called Nasara in the Qur’an, Azzi presses an audacious claim that has yet to be substantiated. His thesis is based on the figure of Waraqa b. Nawfal,37 the cousin of Muhammad’s first wife, Khadija, who is portrayed by Azzi as Muhammad’s teacher and mentor.38 According to Muslim tradition, Waraqa was literate and became a Christian in pre-Islamic times. He copied down passages from the Christian scriptures and recognized Muhammad’s prophetic mission immediately after his first Qur’anic revelation. Azzi, however, depicts him as a priest who prepared Muhammad to follow him as the leader and head of the small Ebionite community of Mecca. Muhammad betrayed this project when he migrated to Medina and there became the head of the new Islamic state. The separation of the pupil from the teacher explains the change that took place between an original Qur’an and the codex of ‘Uthman. The original Qur’an, assembled by the priest, was intended to serve as a lectionary for the ritual worship of the Christian Arab community of Mecca. The codex of ‘Uthman, on the other hand, put together as a book after Muhammad’s death, incorporates the agenda of Muhammad’s Islamic state and constitutes the basis of our present-day Qur’an. While being educated by his mentor, so Azzi proposes, Muhammad became familiar with the contents of the Gospel of the Ebionites, known as the Gospel of the Hebrews.39 This apocryphal gospel was current among the Nazareans, that is the Aramaic speaking Jewish-Christians of Syria-Palestine, and was translated into Arabic by Waraqa. The Ebionites were known for ritual washings and a Eucharistic celebration using bread and water rather than wine. They were anti-trinitarian, denied the divinity of Christ and saw Jesus as created, begotten from the semen of a human being. They rejected the christological teachings of Paul and denied any redemptive role for Jesus. For them Jesus was a prophet following in a line of prophets leading from Adam to Moses. Jesus would rule the future age while the present age is ruled by the Devil. To sum up, Azzi’s study is highly speculative and poorly documented, but it nevertheless points in the direction of the kind of historical–critical research that would have to be undertaken to support Luxenberg’s purely philological studies. Other new voices in the debate In addition to the far ranging theories of Lüling, Luxenberg and Azzi, a number of genuinely scholarly articles have appeared in the past decade that are focused on clearly defined issues surrounding the origins of the Qur’an. Following up on his study of Tabari’s exegetical achievements, C. Gilliot wrote a series of articles focusing on the possibility of the Qur’an as a collective work that drew on a great variety of informants.40 F. de Blois pursued a line of research that disentangles the 80 RECONSTRUCTING THE QUR’ AN possible links of the Qur’an with the Elchasites, Mandeans and Manicheans. As a result of his studies, the Sabians, who are cited in the Qur’an, can no longer be understood, as previously assumed, to be identical with the Mandeans,41 but in all likelihood refer to a Manichean community that existed in Mecca or Medina.42 Following the studies of Schaeder,43 de Blois also established a more precise notion of Jewish Christians for the term Nasara in the Qur’an.44 In the same article he also analyzed the frequently studied term hanif 45 and charted its development from the pre-Islamic Christian use for pagan to the Qur’anic meaning of true follower of Abraham and to the post-Qur’anic interpretations of authentic Muslim.46 J. M. F. Van Reeth addressed the possibility that the Diatessaron could be the underlying source for the Qur’anic use of the term injil (in the singular) for the gospel of Jesus.47 A. Neuwirth wrote a variety of articles on Qur’anic rhetoric and literary structure, drew attention to the audiences addressed by Muhammad and showed the complexity of the problems of possible dependence on biblical materials.48 H. Motzki examined the collection of the Qur’an in the light of data culled from the relevant Hadith materials.49 Concluding remarks Reviewing these recent studies on the Qur’an, mainly published during the last decade, it is clear that, despite the clamor in the press, no major breakthrough in constructing the Qur’an has been achieved. The ambitious projects of Lüling and Luxenberg lack decisive evidence and can reach no further than the realm of possibility and plausibility. They have made it clear, however, that an exclusively philological approach is insufficient to break new ground concerning the origins of the Qur’an. A critical historical inquiry into source accounts of the JewishChristian legacy underlying the Qur’an is a necessary complement to the philological probings. These fragmentary accounts have to be discovered, in the first place, in the traditions of those religious communities to which the Qur’an makes reference. These religious communities were disparate and heterogeneous. The Qur’an refers to them in general terms, such as “the children of Israel” (banu Isra’il, cited forty times in the Qur’an), a reference mainly to Jews but occasionally also Christians, “the possessors of (previous) revelations” (ahl al-dhikr, Q 15:43; 21:7), and “the people of the Book” (ahl al-kitab, cited 54 times in the Qur’an), including both Jews and Christians invariably.50 There are also specific inner-Qur’anic references to Jews (occurring eight times in the plural as yahud, Q 2:113 twice, 2:120; 5:18, 51, 64, 82; 9:30, three times as hud, Q 2:111, 135, 140, and once in the singular, yahudi, Q 3:67). Then there are fourteen references to “the Nazarenes” (Nasara), and one to “the people of the Gospel” (ahl al-injil, Q 5:47). Other religious groups cited are the Sabians (Q 2:62, 5:69, and 22:17), probably Manicheans, and the Magians (Q 22:17), that is, Zoroastrians. In addition, there are references to “those who profess Judaism” (alladhina hadu, mentioned ten times in the Qur’an) and “those who have associated” (mushrikun).51 Furthermore, there are all those instances in 81 GERHARD BÖWERING which the Qur’an cites the Jewish and Christian scriptures under the names of Torah (tawrat), Gospel (injil) and Psalms (zabur) and, in eight instances, also the scrolls (suhuf) or ancient scriptures attributed in two Qur’anic passages to Abraham and Moses (Q 53:36–7; 87:19).52 All of these names of individual groups, general descriptions of religious communities and references to actual scriptures or holy books point to the religious environment with which the Qur’an was in contact, often in polemical fashion. From them it derived inspiration and among them future research has to pinpoint relevant source materials if a new vision of the Qur’an’s construction is to be achieved. To this end more scholars with cross-cultural expertise need to join ranks, uniting knowledge of nonnormative Jewish and Christian traditions prior to the advent of Islam with historical and philological analysis of early Islam against the background of the Semitic languages that are cognate with Arabic. There are not only a great many relevant inner-Qur’anic references, there are also a plethora of extra-Qur’anic references that can be found in Muslim religious literature, including Qur’anic commentary, prophetical traditions and historical writings. These references point to a variety of informants, who served as possible sources for the religious lore from which the Qur’an drew inspiration. There is Waraqa b. Nawfal, to whom Azzi has again drawn attention, but also ‘Ubayd Allah b. Jahsh and ‘Uthman b. al-Huwayrith, who have received less scholarly interest.53 To these may be added all those known as hanif, plural hunafa’,54 who are highlighted in the extra-Qur’anic literature as “God-seekers” standing in contrast to “those who have associated.” Much scholarly speculation surrounds this group of hanifs, on whom a great number of scholars have rendered widely differing judgments.55 Then there is the legend of the hermit Bahira, who is said to have predicted Muhammad’s prophetic destiny.56 Another group of possible informants of Muhammad has recently been re-examined and identified by Gilliot as “informant slaves.”57 These informant slaves were men of low birth and foreign origin who could read the scriptures and were in contact with Muhammad in Mecca in particular. Furthermore, the quest for the origins of the Qur’an may also have to take into account Zayd b. ‘Amr,58 who is pictured as Muhammad’s precursor. A very significant role in the final redaction of the Qur’an was played by Zayd b. Thabit,59 who served Muammad as his principal scribe in Medina and whose central role in the collection of the Qur’an needs a fresh analysis (together with those who assisted him in this task). Finally, more scholarly emphasis will have to be given to the first forty years of Muhammad’s life, the time before “his call” (approximately 570–610 CE). There are important and still open questions about this period that have been neglected by the recent scholarship on the construction of the Qur’an. For example, prior to his call, to what degree did Muhammad assimilate many of the religious ideas that became essential elements of his Qur’anic message? Prior to his call, to what degree had Muhammad mastered the style of rhymed prose that he uses so powerfully in his earliest Qur’anic proclamations? Can Muhammad have been under a significant Christian influence in the first forty years of his life even though the earliest Qur’anic 82 RECONSTRUCTING THE QUR’ AN proclamations in rhymed prose display hardly any decidedly Christian motifs or topics? During his lifetime, Muhammad had a good number of his Qur’anic proclamations copied down by scribes, but there is no evidence that he used foreign written source materials for the composition of the Qur’an. Until the appearance of evidence to the contrary, one has to support the position that it was oral information on which the Qur’an drew directly, even if behind this oral information there was a core of passages extracted from written traditions that were translated into Arabic from one or the other of its sibling languages. This core, however, has not yet come to light in a distinct form. The almost total absence in the Qur’an of direct parallels with the normative, midrashic or apocryphal biblical traditions60 makes it impossible to argue for a direct dependence on written sources. Essential sections of the Qur’anic message were received from the oral lore of a variety of religious communities who were rooted in the widely dispersed and non-normative Jewish and Christian traditions. Not a single written source, whether scriptural or liturgical, however, has been identified that would satisfy the search for an underlying Ur-Qur’an, whether postulated as a Christian hymnal or a Syro-Aramaic lectionary, that served as a written source book for the Qur’an. Notes 1 For the biblical narratives in the Qur’an, see H. Speyer, Die biblischen Erzählungen im Qoran, Gräfenhainichen: Schulze, 1931 (reprint: Hildesheim: Olms, 1961). 2 The earliest translations of the gospels into Arabic were made in the late second/eight century from Greek originals by Christian monks in Palestine, see S.H. Griffith, “The Gospel in Arabic: An inquiry into its appearance in the first Abbasid century,” OC 69, 1985, 126–67; A. Guillame, “The version of the Gospels used in Medina,” Al-Andalus 15, 1950, 289–96; J. Blau, “Über einige christlich-arabische Manuskripte aus dem 9. und 10. Jahrhundert,” Le Museon 75, 1962, 105–7. A. Baumstark, however, posited a pre-Islamic version of the gospels in Arabic, see A. Baumstark, “Das Problem eines vorislamischen christlich-kirchlichen Schrifttums in arabischer Sprache,” Islamica 4, 1931, 562–75; idem, “Eine frühislamische und eine vorislamische arabische Evangelienübersetzung aus dem Syrischen,” Atti del XIX Congresso Internazionlae degli Orientalisti (1935), Rome: G. Bardi, 1938, 682–4. A fragment of a Christian Arabic translation of the Psalms (termed zabur in the Qur’an) can also be traced to the second/eighth century, cf. B. Violet, “Ein zweisprachiges Psalmfragment aus Damascus,” Orientalische Literaturzeitung 4, 1901, 384–403, 425–41, 475–88. 3 R. Paret, Mohammed und der Koran, Stuttgart: Kohlhammer, 1957, 83. 4 FV; S. Fraenkel, Die aramäischen Fremdwörter im Arabischen, Leiden: Brill 1886; see also, A. Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam, Frankfurt: Peter Lang, 1984. 5 C. Gilliot and P. Larcher, “Language and style of the Qur’an,” EQ, 3, 109–35. 6 The question of Muhammad’s literacy is focused on the Qur’anic term, Ummi and its interpretations, see R. Paret, “Ummi,” Shorter Encyclopaedia of Islam, Leiden: Brill, 1961, 604; E. Geoffroy, “Ummi,” EI2, 10, 863–4. 7 Cf. J. Burton, The Sources of Islamic Law, Edinburgh: Edinburgh University Press, 1990; idem, Naskh, EI2, 7, 1009–12. 83 GERHARD BÖWERING 8 The Satanic verses (qissat al-gharaniq, cf. Q 53:19, 20) refer to a famous incident when Muhammad is reported to have revoked a Qur’anic revelation as a Satanic insinuation. See J. Burton, “Those are the high-flying cranes,” Journal of Semitic Studies 15, 1970, 246–65; U. Rubin, The Eye of the Beholder, Princeton, NJ: Darwin, 1995, 156–66; S. Ahmed, “Satanic Verses,” EQ, 4, 531–5 9 For the “occasions of revelation” (asbab al-nuzul), see A. Rippin, “Occasions of revelation,” EQ, 3, 569–73. 10 Al-Suyuti, al-Itqan fi ‘ulum al-Qur’an, Muhammad Abu l-Fadl Ibrahim (ed.), Cairo: al-Hay’a al-Misriyya al-’Amma li-l-Kitab, 1967; see also E. Geoffroy, “Suyuti,” EI 2, 9, 913–16. 11 GdQ. 12 R. Paret, Der Koran, Kommentar und Konkordanz, Stuttgart: Kohlhammer 1971, 1977 (with addendum). 13 For the principal traditional Arabic biography of Muhammad, see Ibn Ishaq, Sirat rasul Allah (recension of Ibn Hisham), F. Wüstenfeld (ed.), Das Leben Muhammeds nach Muhammad Ibn Ishaq, Göttingen: Dieterich, 1858–60, trans. A. Guillaume, The Life of Muhammad, Oxford: Oxford University Press, 1955; for a standard scholarly biography of Muhammad, see F. Buhl, Das Leben Mohammeds, trans. H.H. Schaeder, Leipzig: Quelle and Meyer, 1931 (3rd edn); for a recent assessment of the sources, see H. Motzki, The Biography of Muhammad, Leiden: Brill, 2000. 14 W.M. Watt, Bell’s Introduction to the Qur’an, Edinburgh: Edinburgh University Press, 1970, 1991; R. Bell, The Qur’an. Translated with a Critical Re-arrangement of the Suras, Edinburgh: Edinburgh University Press, 1937 (reprinted 1960); idem, A Commentary on the Qur’an, C.E. Bosworth and M.E.J. Richardson (eds), Manchester: University of Manchester, 1991. 15 R. Blachère, Introduction au Coran, Paris: Maisonneuve, 1947; A. Welch, Üur’an,” EI2, 5, 400–29. 16 R. Blachère, Introduction au Coran, 1–70. 17 ‘Uthman b. ‘Affan, the third caliph of Islam ruled from 23/644 to 35/656, see G. Levi della Vida and R.G. Khoury, “ ‘Uthman b. ‘Affan,” EI2, 10, 946–8. 18 R. Blachère, Introduction, 71–181. Actually, Muslim scholars of the Qur’an, who publish editions of all seven readings until this day, recognize other ways of reading (qira’at) in addition to the seven. There is no space to discuss this issue in the context of this article. 19 The standard Egyptian text is based on the version of Hafs (d. 190/805) for the reading of ‘Asim (d. 127/744). Some other versions with minor divergences, namely those of Warsh (d. 197/812) and Qalun (d. 220/835) for the reading of Nafi’ (d. 169/785) circulate in the northwestern regions of Africa, see A. Welch, “Üur’an,” EI 2 5, 409. 20 J. Wansbrough, Quranic Studies: Sources and Methods of Scriptural Interpretation, Oxford: Oxford University Press, 1977; for a further development of Wansbrough’s approach, see P. Crone and M. Cook, Hagarism: The Making of the Islamic World, Cambridge: Cambridge University Press, 1977; for a convincing refutation of the views of Wansbrough, see H. Motzki, “The Collection of the Qur’an. A reconsideration of Western views in light of recent methodological developments,” Der Islam 78, 2001, 1–34; M. Muranyi, “Neue Materialien zur tafsir-Forschung in der Moscheebibliothek von Qairawan,” in S. Wild (ed.), The Qur’an as Text, Leiden: Brill, 1996, 225–55; see also C.H.M. Versteegh, Arabic Grammar and Qur’anic Exegesis in Early Islam, Leiden: Brill, 1993. 21 J. Burton, The Collection of the Qur’an, Cambridge: Cambridge University Press, 1977. 22 G. Lüling, Über den Ur-Qur’an, Erlangen: Lüling, 1974; idem, Der christliche Kult an der vorislamischen Kaaba, Erlangen: Lüling, 1977; idem, Die Wiederentdeckung des Propheten Muhammad, Erlangen: Lüling, 1981. 84 RECONSTRUCTING THE QUR’ AN 23 A Challenge to Islam for Reformation: The Rediscovery and Reliable Reconstruction of a Comprehensive pre-Islamic Christian Hymnal Hidden in the Koran under Earliest Islamic Reinterpretations, Delhi: Motilal Banarsidass, 2003. 24 Actually, the Qur’anic phrase is “Ibrahim, Isma’il, Ishaq, Ya’qub and the tribes” (Q 2:136, 140; 3:84; 4:163). 25 H. Lammens, “Le Chrétiens à la Mecque à la veille de l’Hégire,” Bulletin de l’institut français d’archéologie orientale 14, 1918, 191–230; F. Nau, Les Arabes chrétiens de Mésopotamie et de Syrie du VIIe au VIIe siècle, Paris: Imprimerie nationale, 1933; J. Henninger, Spuren christlicher Glaubenswahrheiten im Koran, Schöneck: Administration der Neuen Zeitschrift für Missionswissenschaft, 1951. 26 C. Luxenberg, Die Syro-aramäische Lesart des Koran, 2nd edn, Berlin: Schiler, 2004. 27 Ibid., 95–103, 107–10. 28 Ibid., 255–94. Luxenberg’s translation of grapes for the houris of the Qur’an does not require emending the text, and may be one of his more helpful proposals. 29 Ibid., 310–33. 30 Ibid., 304–10. 31 C. Luxenberg, “Weihnachten im Koran,” in C. Burgmer (ed.), Streit um den Koran, Berlin 2005, 35–41; R. Bell noticed the similarities with the Eve of the Nativity in this sura (R. Bell, The Qur’an. Translated, 669). Luxenberg, however, makes “Christmas in the Qur’an” a certainty. 32 Luxenberg, “Weihnachten im Koran,” 17. 33 J. Azzi, Le Prêtre et le Prophète, trans. M.S. Garnier, Paris: Maisonneuve et Larose, 2001, as based on Abu Musa al-Hariri, Nabiy al-rahma, Beirut: Diyar ‘Aql, 1990 and idem, Qass wa-nabiy, Beirut: n.p., 1991. 34 A. von Harnack, Lehrbuch der Dogmengeschichte, Zweiter Band: Die Entwicklung des kirchlichen Dogmas, Tübingen: Mohr, 1909, 529–32; idem, Die Mission und Ausbreitung des Christentums in den ersten drei Jahrhunderten, Leipzig: Hinrichs, 1915, 152–8; H.J. Schoeps, Theologie und Geschichte des Judenchristentums, Tübingen: Mohr, 1949; see also A. Schlatter, “Die Entwicklung des jüdischen Christentums zum Islam,” Evangelisches Missionsmagazin 62, 1918, 251–64. 35 T. Andrae, Les origines de l’Islam et le Christianisme, Paris: Adrien-Maisonneuve, 1955; R. Bell, The Origin of Islam in its Christian Environment, London: MacMillan, 1926, reprinted 1968; J.S. Trimingham, Christianity among the Arabs in Pre-Islamic Times, London: Longman, 1979; see also C.D.G. Müller, Kirche und Mission unter den Arabern in vorislamischer Zeit, Tübingen: Mohr, 1967. 36 M.P. Roncaglia, “Éléments Ébionites et Elkésaites dans le Coran,” Proche Orient Chrétien 21, 1971, 101–26. 37 U. Rubin, The Eye of the Beholder, 103–12; C.F. Robinson, “Waraüa b. Nawfal,” EI 2, 11, 142–3. 38 For this point see T. Nöldeke, “Hatte Muhammad christliche Lehrer?,” ZDMG 12 (1858), 699. 39 According to Azzi, this Gospel of the Hebrews, which Waraqa is assumed to have translated from Aramaic into Arabic, is identical with the apocryphal gospel of Matthew of which only a Latin version of the eight or ninth century is extant, see G. Schneider, Apocryphe Kindheitsevangelien, Freiburg: Herder, 1995, 238. To be sure, the Gospel of the Hebrews was once current in Aramaic, see E. Hennecke, New Testament Apocrypha, vol. 1, Philadelphia: Westminster Press, 1963, 139–53. 40 C. Gilliot, Exégèse, langue et théologie en Islam. L’exégèse coranique de Tabari (m. 310/923), Paris: Vrin, 1990; idem, “Les ‘informateurs’ juifs et chrétiens de Muhammad. Reprise d’un problème traité par Aloys Sprenger et Theodor Nöldeke,” JSAI 22, 1998, 84–126; idem, “Le Coran, fruit d’un travail collectif ?,” in D. De Smet, G. de Callatay and J.M.F. Van Reeth (eds), Al-Kitab. La sacralité du texte dans le monde de l’Islam. Acta Orientalia Belgica, 2004, 185–231. 85 GERHARD BÖWERING 41 D. Chwolsohn, Die Ssabier und der Ssabismus, St Petersburg: Buchdruckerei der kaiserlichen Akademie der Wissenschaften, 1856. 42 F. de Blois, “The Sabians (Sabi’un) in pre-Islamic Arabia,” Acta Orientalia 56, 1995, 39–61; idem, “Elchasai – Manes – Muhammad. Manichäismus und Islam in religionshistorischem Vergleich,” Der Islam 81, 2004, 31–48. 43 H.H. Schaeder, “Nazarenos, Nazoraios,” in G. Kittel (ed.), Theological Dictionary of the New Testament, Grand Rapids: Eerdmans, 1968, 4, 874–9. 44 F. de Blois, “Nasrani and hanif: Studies on the religious vocabulary of Christianity and Islam,” BSOAS 65, 2002, 1–30. 45 U. Rubin, “Hanifiyya and Ka’ba – an inquiry into the Arabian pre-Islamic background of Din Ibrahim,” JSAI 13, 1990, 85–112. 46 K. Ahrens, “Christliches im Qoran. Eine Nachlese,” ZDMG 84, 1930, 15–68, 148–90. 47 J.M.F. Van Reeth, “L’Évangile du Prophète,” Acta Orientalia Belgica 3, 2004, 156–74. 48 For example, A. Neuwirth, “Qur’anic literary structure revisited,” in S. Leder (ed.), Story-telling in the Framework of Non-Fictional Arabic Literature, Wiesbaden: Harrassowitz, 1998, 388–420; “Zur Archäologie einer Heiligen Schrift,” in C. Burgmer (ed.), Streit um den Koran, 82–97. 49 H. Motzki, “The Collection of the Qur’an. A reconsideration of Western views in light of recent methodological developments,” Der Islam 78, 2001, 1–34. Recent Muslim contributions to the topic are M. Sfar, Le Coran, la Bible et l’Orient ancien, Paris: Sfar, 1998 and idem, Le Coran est-il authentique?, Paris: Sfar, 2000. Reference may also be made to S. Wild (ed.), The Qur’an as Text, the recent re-issue of articles on Qur’anic topics under the pseudonym Ibn Warraq, The Origins of the Qur’an, Amherst, NY: Prometheus, 1998 and idem, What the Koran Really Says, Amherst, NY: Prometheus, 2002, and the re-issue of articles by A. Rippin, The Qur’an: Formative Interpretation, Aldershot: Ashgate/Variorum, 1999, and idem, The Qur’an: Style and Contents, Aldershot: Ashgate/Variorum, 2001. R. Paret, Der Koran, Darmstadt: Wissenschaftliche Buchgesellschaft, 1975, had previously published another selection of significant scholarly articles on Qur’anic topics. A controversial treatment of the Qur’an is the anonymous Brother Mark, A Perfect Qur’an, posted on the internet 2000, and ongoing. 50 G. Vajda, “Ahl al-kitab,” EI2 1, 264–7; S.D. Goitein, “Banu Isra’il,” EI2 1, 1020–1. 51 For a recent study of the mushrikun, see G.R. Hawting, The Idea of Idolatry and the Emergence of Islam: From Polemic to History, Cambridge: Cambridge University Press, 1999. According to all evidence, the term mushrikun refers primarily in the Qur’an to pagan Arab polytheists, who associate partners with God, but it may also imply Christians with Trinitarian beliefs. 52 Within the larger frame of reference one would have to count all the scholarly work that has been done on the patriarchs and prophetical characters in the Qur’an as well as on the principal religious ideas that the Qur’an shares with the biblical tradition, whether normative, midrashic or apocryphal. This cannot be examined here, but H. Busse, Die theologischen Beziehungen des Islam zu Judentum und Christentum, Darmstadt: Wissenschaftliche Buchgesellschaft, 1988, may be cited as a lucid re-examination of these issues during the last two decades. 53 M. Hamidullah, “Two Christians of Pre-Islamic Mecca,” Journal of the Pakistan Historical Society 6, 1958, 97–103. 54 In the Qur’an, the term hanif occurs ten times in the singular and twice in the plural with reference to the true follower of the religion of Abraham. In extra-Qur’anic literature the term hanif refers either to true Muslims or to pre-Islamic seekers of God, see F. de Blois, “Nasrani and hanif,” 16–25. 55 W.M. Watt, “Hanif,” EI2, 3, 165–6; U. Rubin, “Hanif,” EQ, 2, 402–4; see also D.S. Margoliouth, “On the origin and import of the names muslim and hanif,” JRAS 86 RECONSTRUCTING THE QUR’ AN 56 57 58 59 60 1903, 467–93; K. Ahrens, “Christliches im Qoran. Eine Nachlese”; F. de Blois, “Nasrani and hanif,” 16–25. According to the legend Muhammad met the monk (rahib) Bahira in his hermitage (sawma’a; cf. Q 22:41) in Bostra, see Tabari, Ta’rikh al-rusul wa l-muluk, M.J. De Goeje (ed.), Leiden: Brill, 1879–1901, 1, 1123–5; A. Abel, “Bahira,” EI2, 1, 922–3. It may be noted, however, that, in Syriac, Bahira serves as the title for a reverential address of a monk rather than a proper name. C. Gilliot, “Informants,” EQ, 2, 512–18. M. Lecker, “Zayd b. ‘Amr,” EI2, 11, 474–5. M. Lecker, “Zayd b. Thabit,” EI2, 11, 476. To stress again, only a very small number of Qur’anic verses parallels small passages of the apocryphal gospels, and only one Qur’anic verse, Q 21:105, is a direct quotation from the Hebrew Bible (Old Testament), namely Psalm 37:29. The earliest known Muslim Arabic citation from the New Testament is the passage of John 15:23–16:1 which is presented in summary form by Ibn Ishaq (d. 767 CE) in Muhammad’s biography, see F. Wüstenfeld (ed.), Das Leben Muhammeds nach Muhammad Ibn Ishaq, 1, 149–50. For the small harvest of parallels between Qur’anic passages and the Syriac liturgy, see E. Graf, “Zu den christlichen Einflüssen im Koran,” Festschrift Joseph Henninger: Studia Instituti Anthropos Bonn, Al-Bahith 28, 1976, 121–44. 87 4 RECONSIDERING THE AUTHORSHIP OF THE QUR’AN Is the Qur’an partly the fruit of a progressive and collective work? Claude Gilliot The hypothesis that the Qur’an could be partly or wholly the fruit or the product of a collective work (German: Gemeindeprodukt) is seldom clearly expressed. It is, however, often in the background of the Qur’anic researches of many western scholars. To give only a few such examples we can mention the Austrian physician and Arabist Aloys Sprenger (1813–93),1 and more recently the late English Semitist and Arabist John Wansbrough (1928–2002),2 along with Patricia Crone and Michael Cook.3 That this idea is not more widely expressed is due to the fact that for the history of the Qur’an we are mainly still in the world of “Alice in Wonderland” or to be more in the local color, in the world of the “Marvels of Aladdin’s Lamp,” when compared with research in the field of Biblical studies, for instance. For this reason Andrew Rippin can write: “In teaching undergraduate students, I have often encountered individuals who come to the study of Islam with a background in the historical study of the Hebrew Bible or early Christianity, and who express surprise at the lack of critical thought that appears in introductory textbooks on Islam.”4 In fact, the more we become acquainted with the Arabic Islamic sources, the more we become convinced of the hypothesis that the declarations delivered by Muhammad (as coming from God) could be partly the product of a collective work at the different phases of their proclamation, before they were collected or amended to become a “recitation” and/or “lectionary” (qur’an). We say hypothesis, because, contrary to the religious Islamic thesis on the Qur’an, Western scholars or Orientalists have above all hypotheses on this issue and not a religious or an ideological thesis. I expressed my hypothesis of this collective work for the first time in an article in French published in 1998,5 and then more explicitly in the article of an encyclopedia.6 But the publication of the book of Christoph Luxenberg, Die 88 THE AUTHORSHIP OF THE QUR’AN syro-aramäische Lesart des Koran (The Syro-Aramaic Reading of the Qur’an),7 was a new starting-point. It led me to re-examine the traditional Muslim accounts on the history of the Qur’an and to analyze whether these sources contain elements in favor of Luxenberg’s thesis. The present paper is a kind of partial inventory or a summary of several articles and contributions, both published and unpublished, wherein the reader will find more details, sources and studies. Here I will address evidence for collective authorship of the Qur’an first in Islamic tradition and second in the Qur’an itself. Thereafter I will consider, in light of this evidence, the importance of the proposals that the Qur’an was particularly shaped by Syriac Christian tradition. The Islamic tradition and another history of the Qur’an uphill, or “the obscure beginnings”8 I proposed the distinction between “the reconstruction of the Qur’an uphill” and its “reconstruction downhill” within a panel organized by Manfred Kropp during the first World Congress of Middle Eastern Studies (WOCMES),9 in a paper delivered in French: “Une reconstruction critique du Coran est-elle possible?” (“Is a critical reconstruction of the Qur’an possible?”) In our perspective “the Qur’an uphill,” means the lectionary (qur’an) or parts of a lectionary, before the Islamic Qur’an. Our concern will be here above all with that lectionary. In the same way the expression “the Qur’an downhill”10 means the Qur’an after the collection and arrangement of the so-called ‘Uthmanic Qur’an, whose critical reconstruction was planned and begun in some way by Gotthelf Bergsträsser (1886–1933),11 Otto Pretzl (1893–1941)12 and Arthur Jeffery (1888?–1959).13 The Qur’an itself contains several, and the Islamic sources many, indications of another history of the Qur’an that is different from, and in some way opposed to, the official Islamic theological representation of the genesis and development of this qur’an. The informants of Muhammad The question of whether Muhammad relied on informants is connected to the scholarly discussion of the origin of the Qur’an.14 Many of the Qur’anic narratives must not have sounded new to the Meccan opponents of Muhammad. They used to say, gibing him: “This is naught but falsehood he has forged, and other folk have helped him to it [ . . . ]. They say: Fairy-tales of the ancients (or perhaps: scriptures of the ancients, asatir al-awwalin)15 that he has written down, so that they are recited to him at the dawn and at the evening” (Q 25:4,5). The locus classicus where the question of informants is dealt with in the Qur’anic commentaries is Q 16:103: “And We know very well that they say: Only a mortal is teaching him. The speech of him at whom they hint is barbarous; and this is speech Arabic, manifest (lisan ‘arabi mubin).”16 We shall see below that the usual translation of the end of this verse is: “speech Arabic, manifest,” which is 89 CLAUDE GILLIOT influenced by the theological representation of the Islamic exegetical tradition of the lingua sacra of the Qur’an. It is probably not the right translation, which should rather be: “speech Arabic, making [things] clear (mubin).” According to the renowned exegete Muqatil b. Sulayman (d. 150/767): There was a servant of ‘Amir b. al-Hadrami al-Qurashi. He was a Jew, not an Arab [or spoke bad Arabic, a‘jami]; he spoke Greek [more likely: Aramaic],17 and his name was Abu Fukayha Yasar. As the Qurayshis saw the Prophet speaking with him, they said: “Indeed, he is being taught by Abu Fukayha Yasar.” (Muqatil b. Sulayman, Tafsir al-Qur’an, 2, 48718) According to another version: “The apostle used often to sit at al-Marwa at the booth of a young Christian called Jabr, slave of the Banu l-Hadrami, and they used to say: ‘The one who teaches Muhammed most of what he brings is Jabr the Christian, slave of the Banu l-Hadrami.’ ”19 In some versions, the slave of the Banu l-Hadrami is not named. According to Mujahid b. Jabr (d. 104/722), he came from the Byzantine territories (rumi), he had a book (sahib kitab) and he spoke Aramaic (al-rumiyya, which could mean Greek, but for a Christian of these territories, probably Aramaic).20 According to al-Hasan al-Basri (d. 110/728), this slave of Ibn al-Hadrami had been a soothsayer (kahin) in pre-Islamic times.21 According to Talha b. ‘Amr (al-Hadrami, d. 152/769): “Khadija used to see Khayr (or Jabr?) frequently, and the Qurayshis said that a slave of the Banu l-Hadrami taught her and that she taught Muhammad, so the verse (i.e. Q 16:103) was revealed.”22 The accusations against Muhammad have been summed up by one of his greatest opponents, al-Nadr b. al-Harith.23 This Qur’an is naught but lies that Muhammad himself has forged [ . . . ]. Those who help him are ‘Addas, a slave of Huwaytib b. ‘Abd al-’Uzza, Yasar, a servant of ‘Amir b. al-Hadrami, and Jabr who was a Jew, and then became a Muslim. [ . . . ] This Qur’an is only a tale of the Ancients (hadith al-awwallin), like the tales of Rustam and Isfandiyar. These three teaching Muhammad at dawn and in the evening (cf. Q 25:4–5). (Muqatil, Tafsir al-Qur’an, 3, 226–724) The study of the reports on the informants has led us to the conclusion that we cannot exclude the possibility that whole sections of the Meccan Qur’an could contain elements originally established by, or within, a group of “God seekers,” of “deprived” or “have-nots” who possessed either Biblical, post-Biblical or other information. People like Waraqa b. Nawfal and Khadija may have also participated in that common enterprise under the direction of Muhammad or another individual.25 90 THE AUTHORSHIP OF THE QUR’AN The topos “Holy! Holy!” (Quddus, Quddus) or the “Helpers” of Muhammad (Khadija, Waraqa b. Nawfal, and so on) who “Made Him a Prophet” 26 Muhammad had auxiliaries who helped him during his first enigmatic experiences of revelation27 and who contributed to his establishment as a prophet. But for his first wife Khadija, he probably would never have become a prophet.28 In his Persian abridgment of the Annals of Tabari, Bal‘ami (d. 363/974) writes that Khadija “had read the ancient writings and knew the history of the prophets, and also the name of Gabriel.”29 When Muhammad had his first revelatory experience, Khadija took him to her cousin Waraqa b. Nawfal.30 The passage on this event is given in the Maghazi (or Maghazi-Sira) of the historian Ibn Ishaq (d. 150/767),31 in the section on Muhammad’s “invocation to mission” or “call” (al-mab‘ath), within a longer narrative on his call, translated here according to the version (transmission) of Ibn Hisham (d. 218/834), in the account of the storyteller (qass) ‘Ubayd b. ‘Umayr b. al-Laythi (d. 68/687).32 Then she rose and gathered her garments about her and set forth to her cousin Waraqa b. Nawfal b. Asad b. ‘Abd al-‘Uzza b. Quayy, who had become a Christian and read the scriptures and learned from those who follow the Torah and the Gospel. And when she related to him what the apostle of God told her he had seen and heard, Waraqa cried: ‘Holy! Holy! Verily by Him in whose hand is Waraqa’s soul, if thou has spoken to me the truth, O Khadija, there hath come unto him the greatest Namus [i.e. Greek nomos]33 (Tabari: meaning Gabriel) who came to Moses aforetime, and lo, he is the prophet of his people. Bid him be of good heart.”34 (A. Guillaume, The Life of Muhammad: A Translation of Ibn Ishaq’s Sirat Rasul Allah, 107) In other versions the topos “Holy! Holy!” does not appear, for instance in that of al-‘Utaridi/Yunus b. Bukayr which is in Abu Bakr al-Bayhaqi’s Dala’il al-nubuwwa.35 In al-Zuhri’s (d. 124/742) Maghazi,36 Waraqa says: “It is the Namus which God let come down upon Moses.”37 According to ‘Urwa b. al-Zubayr (d. probably 94/712),38 ‘A’isha’s nephew and another authority on the life and military campaigns (maghazi) of Muhammad, Waraqa says: “Verily the one who comes to him is the greatest Namus, the Namus of Jesus, the son of Mary . . . .”39 But in another version, the same ‘Urwa transmitting also from ‘A’isha has: “It is the Namus which God let come down upon Moses.”40 The same topos “Holy! Holy!” is also found in the account of the legendary meeting between Khadija and the hermit Bahira41 (or Sergius, Sarjis), and the account of one of Muhammad’s informants ‘Addas.42 In another report Khadija gives to Abu Bakr the order to go with Muhammad to Waraqa, and when Waraqa hears the account of Muhammad, he cries: “All-Perfect! All-perfect!”43 (sabbuh or subbuh).44 91 CLAUDE GILLIOT The case of Waraqa b. Nawfal and the topos “Holy! Holy!” could be an indication of traces of a Christian liturgy in the Arabic language among the Christian Arabs.45 It seems clear that we are here in the semantic context of a Christian liturgy (Sanctus! Sanctus! in Latin; cf. Isaiah 6:3). The case of Zayd b. Thabit, secretary of the revelation and “Editor” of the Qur’an Zayd b. Thabit probably knew Aramaic, Syriac or Hebrew, or elements of these languages before the arrival of Muhammad to Yathrib. He is sometimes credited with the knowledge of two languages (or rather two scripts) – Arabic and Hebrew – which he acquired before Islam. In some reports, however, Zayd’s study of the “Jewish script” (kitab Yahud) is said to have taken place after the emigration of Muhammad to Yathrib. The Messenger of God ordered me to study for him the script of the Jews, and he said to me: “I do not trust the Jews with regard to my correspondence” [i.e. correspondence with the Jews, written in their script]. Not even half a month passed until I used to write for him, and when they wrote to him, I read their letter. (Baladhuri, Futuh al-buldan, 47446) Or in another version with the following chain of transmitters: [ . . . ] ‘Abd al-Rahman b. al-Zinad/his father/Kharija b. Zayd: I was brought to the Messenger of God when he came to Medina. They said: “Messenger of God, this is a boy of the banu al-Najjar [He should have been eleven years old at that time], of what had been revealed unto you he knows (qad qara’a) seventeen (sixteen or ten, in other versions) suras.” So I recited to the Messenger of God, and he was pleased with it. (Ibn ‘Asakir, Ta’rikh madinat Dimashq, 19, 30247) This tradition is followed by the order to learn the script of the Jews as in the previous version. An initial question arises hence. Are these suras of the Qur’an? Would they not rather be passages of Jewish writings which Muhammad or others liked and which were used for the composition of the Qur’an? The word sura is not of Arabic origin, as the embarrassment of the lexicographers regarding it demonstrates, but stems rather from Aramaic or from Hebrew.48 This is not the only case where information that Muhammad acquired from others has been attributed to Muhammad himself by inverting the situation. In a famous episode the Christian Palestinian Tamim al-Dari transmits eschatological traditions to Muhammad on the Antichrist and the Beast (al-Dajjal wa-l-Jassasa). Yet in another version of the episode, it is Muhammad himself who informs Tamim al-Dari on this subject.49 92 THE AUTHORSHIP OF THE QUR’AN A second question also supports the hypothesis of an inverted situation. In the traditions cited above it is said, essentially, that Muhammad ordered Zayd to learn Hebrew, Syriac or Aramaic. Why not consider instead that Zayd, who had frequented the Jewish school of Yathrib, already knew one of these languages? In fact, Ibn Mas’ud describes him as a Jew: “Zayd b. Thabit was still a Jew with two locks of hair.”50 Elsewhere: “He was still in school (kuttab) with his lock of hair.”51 We are not the first to raise this problem. In fact, the famous Mu’tazili theologian Abu l-Qasim al-Balkhi (al-Ka‘bi, m. 319/931) takes this position in his critical work on traditions and traditionists, in a report transmitted by Sha‘bi (d. 103/721). The Qurayshis were literate but the Helpers (ansar) were illiterate. Therefore the Messenger of God ordered those [Qurayshis who were taken prisoner in the battle of Badr] who could not [pay ransom] (man kana la mala lahu) to teach writing to ten Muslim [helpers], among whom was Zayd b. Thabit. I asked those who were trained in the field of the life (sira) [of the Prophet] about this, among others Ibn Abi l-Zinad, Muhammad b. Salih (d. 252/866) and ‘Abdallah b. Ja‘far, and they objected to this strongly, saying: “How could someone teach writing to Zayd, who had learned it before the Messenger of God arrived [to Medina]? There were more literate men (kuttab) in Medina than in Mecca. In fact, when Islam arrived in Mecca, there were already a dozen there who knew how to write. When it was the turn of Medina, there were already twenty, among whom was Zayd b. Thabit, who wrote in Arabic and Hebrew, along with Sa‘d b. ‘Ubada, al-Mundhir b. ‘Amr, Rafi‘ b. Malik, etc.” (al-Ka‘bi al-Balkhi, Qabul al-akhbar, 1, 20252) Our hypothesis on this matter is that we face here a reversal of the situation. In Islamic times it became unacceptable that the secretary of the revelation could have written Hebrew or another script before Muhammad came to Yathrib, and the account was reversed.53 But there exist still other issues in favor of another history of the Qur’an which we have treated elsewhere at length, and can only be alluded to here. ● ● The problem of the scribes of the Qur’an, an issue that we have treated elsewhere under the sub-title of “A Qur’anic Variant and the Sleepy Scribe.”54 This issue has to be connected with the problems of the linguistic errors contained in the Qur’an.55 The ambiguities in the vocabulary of memorization (jam‘ and verb jama‘a), collection (again jam‘ and jama‘a), composition (ta’lif) of the Qur’an are great.56 As a consequence we do not know exactly how the so-called ‘Uthmanic Qur’an came into being, and we also do not know whether the Qur’anic text we have at our disposal today represents this ‘Uthmanic Qur’an. 93 CLAUDE GILLIOT ● The (supposedly) missing verses or suras, and those that God (or Muhammad) suppressed or abrogated,57 also constitute an issue of the “obscure beginnings” of the Qur’anic text. The problem of language of the Qur’an That the Qur’an contains words which are not originally Arabic, among them Aramaic loanwords, is recognized not only by Western scholars, but also by several ancient Muslim scholars.58 It is recognized as well that many proper names quoted by the Qur’an are not from Arabic origin but come from languages including Hebrew.59 A great part of the technical terms on the Qur’an as a book are also not of Arabic origin. The word qur’an is a loanword,60 as is mushaf (codex). One of the supposed “collectors” of the Qur’an, Salim b. ‘Ubayd (or: b. Ma‘qil), mawla of Abu Hudhayfa, is supposed to have been the first to give the name mushaf to the Qur’an as a collected book,61 a word he learned in Ethiopia.62 Finally, neither sura,63 nor aya,64 are of Arabic origin. The Islamic theological thesis By “theological thesis” I mean the position which was established definitively in Islam around the fourth/tenth century, but which had already existed from the end of the second/eight and the beginning of the third/ninth centuries, although not in such a formalized, theoretical format. It begins with the assertion that the language of the Qur’an is Arabic. This question found an answer in Islamic theology with a special way of interpreting the Qur’anic text itself, which says: “And We never sent a messenger save with the language/tongue of his folk, that he might make [the message] clear for them (li-yubayyina lahum)” (Q 14:4). The exegetes conclude from this quotation that the language of the Qur’an is that of Muhammad and his Companions, understood as the dialect of Hijaz, and more particularly of Quraysh. To that identification, they added a second one: the language of Quraysh is al-lugha al-fusha (which Arabists call “classical Arabic”). Two Qur’anic verses with the expression: “in plain/clear Arabic speech/ tongue” (bi-lisanin ‘arabiyyin mubinin; Q 26:195; cf. Q 16:103) played a decisive role in the development of the theological thesis. But this expression needs more reflection, because the traditional translation, given above, is misleading from the point of view of morphology, and consequently of semantics. Mubin is the active participle of the causative-factitive abana, which can be understood as “making [things] clear.” Such a way of understanding this expression is suggested by Q 14:4 with the causative factitive bayyana. But the adjectival opposition found in Q 16:103 between ‘ajami on the one hand, and ‘arabi and mubin on the other, was understood by the exegetes as “barbarous,” that is, non-Arabic (‘ajami) and indistinct (a’jami), in contradistinction with clear/pure Arabic.65 The shift of signification made by the Muslim exegetes and theologians supported 94 THE AUTHORSHIP OF THE QUR’AN the logo-centrism and the Islamic “imaginaire” of the medieval period.66 It continues to do so today! Thus a gulf lies between the theological thesis of Muslims on the language of the Qur’an and the approach of linguists. It already appears in the following declaration of one of the founders of the Arabists’ school Heinrich Leberecht Fleischer (d. 1888): “The question for us is not: What is the purest, the most beautiful and correct Arabic, but what is Arabic in general?”67 What constitutes the strength of the theological thesis for the believers is precisely what represents its weakness for critical scholars: It is only based on the Qur’anic text on the one hand and upon conviction on the other hand, without any verification of another nature.68 The Arabists’ hypotheses on the language of the Qur’an69 On the other hand, Western Arabists proposed competing explanations of the language of the Qur’an. Their hypotheses can be reduced to two. They have in common the presupposition of a diglossic situation in Arabia: that is, the coexistence of, on the one hand, the various dialects of the Arab tribes, and, on the other, a common language, which, among other things, was the vehicle of poetry, and for that reason, has been termed poetic koinè. The hypothesis of the majority of the Arabists is that the Qur’anic Arabic is, save for some “Hijazi” peculiarities, basically the same as Arabic or pre-Islamic poetry, hence the qualification of “poetic and Qur’anic koinè” sometimes given to that language, which is considered as the basis of classical Arabic.70 The German Qur’anic scholar and Semitist Theodor Nöldeke published in 1910 a contribution on the language of the Qur’an,71 in which he listed an impressive quantity of mistakes – that is grammatical forms, both morphological and syntactic – in the Qur’an that do not enter into the general linguistic system of Arabic. However, he continued to support the thought that the Qur’an, in spite of its “drawling, dull and prosaic”72 style, and in spite of his conviction that “Muhammad was at the very most a mediocre stylist”,73 was written in classical Arabic. He was followed in this by most of the Arabists, with some exceptions including Karl Vollers (1857–1909). In fact, the stronger hypothesis is originally that of Vollers. He concludes that the Qur’an was first delivered by Muhammad in the vernacular of Mecca, a WestArabian speech missing, among other features, the final declension (i‘rab),74 before it was re-written later into the common language of poetry.75 For him this language, though it is the basis of the literary classical language, is primarily an eastern Arabic speech, fitted with final declension and other features. Vollers’ hypothesis was taken up again by Paul Eric Kahle (1875–1964),76 but in a modified form.77 He does not maintain that the Qur’an was re-written, and he admits, without any further discussion, that the consonantal ductus, traditionally attributed to the Caliph ‘Uthman, represents the Arabic spoken in Mecca,78 but 95 CLAUDE GILLIOT for him the “readings” (qira’at, i.e. variae lectiones) of that ductus express the influence of the poetic language. Jonathan Owens has shown recently that the practice of the “major assimilation” (i.e. a consonantal assimilation between words) traditionally linked with the reader Abu ‘Amr (d. 154/770) does not imply linguistically the loss of the inflectional ending, but only the absence of short vowels, inflectional or not, at the ending. This means that: “[Voller’s] assumption that there was a koranic variant without case ending receives plausible support from the koranic reading tradition itself.”79 Generally things are changing with the progress in Arabic studies of sociolinguistics and historical linguistics. Arabists today have gone beyond the diglossic representation of Arabic and are in favor of a polyglossic conception of Arabic and of a continuum, even of an inherent variation. In doing so they take up again, in some way, the conception that the most ancient Arab grammarians, notably Sibawayh, had of Arabic. These last did not understand the speeches/dialects (lughat) as discrete varieties, but only as variants, good or bad, of one and the same language. In this context, the various “readings” (qira’at) of the Qur’an can be seen as the reflection of this linguistic variation.80 The ancient Christian or Syriac connection These observations about the nature of Qur’anic Arabic do not, however, imply that the language and content of the Qur’an were not influenced by other cultures or religions. In this paper I will focus on the hypothesis that the Qur’an was especially influenced by the Syriac or Aramaic connection or channel, an idea that has been present in the mind of the Western scholars since the very beginning of modern Qur’anic research, above all concerning Qur’anic loanwords, as seen above. Hartwig Hirschfeld (1854–1934), for example, concluded that “many Biblical words which occur in the Qorân, have evidently gone through an Aramaic channel.”81 However, in an article published in 1927, the great Christian Iraqi Semitist Alphonse Mingana went further and defended the thesis or the idea of a Syriac influence in the style of the Qur’an.82 His attempt did not hold the attention of most scholars for at least two reasons. First, he did not give many examples. Second, his idea ran counter to a quasi dogma among Islamists: that the Qur’an was written in so-called classical Arabic.83 Beginning with different concerns than Mingana, the German liberal Protestant theologian and Semitist Günter Lüling wrote an important study which has also been overlooked by Islamists and Arabists, including Germans: Über den Ur-Qur’an (On the Primitive Qur’an),84 which has recently been translated into English.85 Therein he calls for “the rediscovery and reliable reconstruction of a comprehensive pre-Islamic Christian hymnal hidden in the Koran under earliest Islamic reinterpretation.” The point of departure is not only the Qur’an, but Lüling’s own scholarly orientation, which is shaped by an “emphasis directed as 96 THE AUTHORSHIP OF THE QUR’AN self-criticism against the falsification of Christianity by its Hellenization [ . . . ], as well as against the falsification of the history of Judaism.”86 The theses of Lüling on the Qur’an are the following: (1) About one-third of the present-day Qur’anic text contains as a hidden ground-layer an originally pre-Islamic Christian text. (2) The transmitted Qur’anic text contains four different layers, given here chronologically: the oldest, the texts of a pre-Islamic Christian strophic hymnody; the texts of the new Islamic interpretation; historically parallel to the next layer, the original Islamic material, which is to be attributed to Muhammad; and finally the texts of the post-Muhammadan editors of the Qur’an. (3) The transmitted Islamic Qur’anic text is the result of several successive editorial revisions. (4) The presence of the successive layers in the Qur’anic text can be confirmed by material in Muslim tradition. The theses of Lüling, which have been largely ignored,87 deserve serious consideration. Yet here I will focus on the work of another Semitist, C. Luxenberg, who has taken up Mingana’s thesis on the Syriac influence and clearly outlined the heuristic. Beginning with those passages from the Qur’an that are unclear to western commentators, the method runs as follows: First check if there is a plausible explanation in Qur’anic classical exegesis (above all, Tabari’s commentary), possibly in some cases overlooked by western scholars. If not, check whether an ancient Arabic dictionary records a meaning unknown to the exegetical sources. If not, check if the Arabic expression has a homonymous root in Syriac with a different meaning that fits the context. In many cases, Luxenberg finds that the Syriac meaning makes more sense than the Arabic meaning of the Qur’anic term. It is to be noted that these first steps of the heuristic do not change the consonantal text of the Cairene edition of the Qur’an. If these steps do not work, then Luxenberg recommends changing one or more diacritical marks to see if this results in an Arabic expression that makes more sense. Luxenberg finds that many instances of problematic lexemes may be shown to be misreadings of one consonant for another. If this method does not produce results, then the investigator should change one or more diacritical points and then check if there is a homonymous Syriac root with a plausible meaning. If there is still no solution, he checks to see if the Arabic is a calque of a Syriac expression. Calques are of two kinds: morphological and semantic. A morphological calque is a borrowing that preserves the structure of the source word but uses the morphemes of the target language. A semantic calque assigns the borrowed meaning to a word that did not have the meaning previously, but which is otherwise synonymous with the source word.88 Is an ancient Christian or Syriac connection possible? A. Neuwirth has reproached Luxenberg for not having taken history and the canonizing process into consideration.89 Similarly, according to S. Hopkins: “[Luxenberg’s] book makes no attempt to place its findings in any plausible context.”90 This critique is not unfounded. History is not Luxenberg’s concern, 97 CLAUDE GILLIOT and this can be considered as a weakness in his study. But now the question is whether it is possible to furnish elements of this historical background. Our aim is not to prove the truth of the Syriac connection, and especially of Luxenberg’s thesis. Our position is that all the cases (ca. 75) presented in his study have to be examined by Arabists and Semitists. It has already been seen in the previous pages of this paper that the Islamic sources themselves contain traces of another history of the Qur’an before it became the text we know nowadays. Meanwhile, the embarrassment of Muslim exegetes unable to understand Qur’anic passages seems to suggest the existence of a text before the Qur’an in which Muhammad or others could have found some inspiration. This embarrassment is well-known. To illustrate it we will choose a limited case, that of the meaning of the hapax legomenon al-kawthar (sura 108),91 mostly understood by Muslim exegetes as the name of a river in Paradise given by God to Muhammad, a case we have treated at length elsewhere:92 a crux grammaticorum sive philologorum et interpretum. The Andalusian exegete Qurtubi (d. 671/1273) establishes a list of seventeen interpretations of this word.93 As for the exegete Ibn al-Naqib (d. 698/1298), he inventories twenty-six different interpretations in his huge Qur’anic commentary (of ninety nine volumes) entitled al-Tahrir wa-l-tahbir li-aqwal ‘ulama’al-tafsir fi ma‘ani kalam al-sami‘ al-basir,94 a source erroneously identified by Harris Birkeland.95 The eighth interpretation given by Qurtubi, according to Ibn Kaysan, is: “al-ithar” (preference, election?),96 a meaning not seen by Birkeland who had not consulted Qurtubi’s commentary. Ibn Kaysan here is none other than the Mu’tazilite exegete and theologian Abu Bakr al-Asamm (d. 200/816, or 201/817).97 Yet Qurtubi, or another before him, seems to have censored a significant part of Ibn Kaysan’s interpretation, because he writes, according to Tha’labi (d. 427/1035) in several manuscripts,98 that al-kawthar could be “a word coming from the first prophecy, meaning al-ithar,” or in the edited text “a word coming from the first scriptures (al-kutub al-ula), meaning al-ithar.”99 The possibility of this interpretation is also given by Maturidi (d. 333/944) in his Qur’anic commentary: “It has been said that al-kawthar is a word [Harf can mean: written representation of a consonant, or of a word or sentence;100 It corresponds to what linguists call “articulation.”]101 borrowed from the ancient scriptures.”102 Indeed Maturidi, along with most exegetes, is very embarrassed by this word, but shows insight when he rejects the interpretation of the “majority” [namely, a river of Paradise]: “God has promised more than that to his community.” Thus “a river of paradise” given to the Prophet cannot be considered an honor (tashrif ). The embarrassment of the exegetes on this word and on the whole sura confirms our opinion that this text has little or no meaning in its present state. This is the reason why Luxenberg proposed that it is a palimpsest of the Aramaic text of 1 Peter 5:8–9: “ . . . be vigilant; because your adversary, Satan, as a roaring lion, walketh about, seeking whom he may devour. Whom resist steadfast in the faith . . . .” There is no need to repeat all of Luxenberg’s argument here,103 but it 98 THE AUTHORSHIP OF THE QUR’AN serves as an important example of a hypothesis that must be considered in light of the confusion of Muslim exegetes. Historical background for Luxenberg’s hypothesis Still more recently another researcher, Jan Van Reeth,104 supports the idea of a “Syriac and Aramaic connection.” He emphasizes that the Qur’an’s use of Injil in the singular could reflect a desire, with ancient roots in early Christianity, to reestablish the unity of the message of Jesus. Faced with the criticism of Celsus (followed later by that of Porphyry, the Emperor Julian, the Manicheans, and plenty others), “Marcion (ca. 85–160) et Tatien (ca. 120–73) ont tenté de répondre en présentant un texte évangélique unique et en reconstruisant la vraie séquence des événements.”105 If Muhammad (or one of his informants or collaborators, in theory) had been familiar with the integrated Gospel, the Diatessaron, it is clear that, as far as he knew, the Diatessaron had probably been composed in Syriac.106 Having said that, it is not easy to know which Gospel text Muhammad could have been familiar with. However, there are a few rare direct references in the Qur’an to the Gospels. Thus Q 48:29: “Such is their likeness in the Torah and their likeness in the Gospel – like as sown corn that sendeth forth its shoot and strengtheneth it and riseth firm upon its stalk, delighting the sowers – that He may enrage the disbelievers with (the sight of) them. God hath promised, unto such of them as believe and do good works, forgiveness and immense reward.” This text combines two Gospel pericopes – Mark 4:26–7 and Matthew 12:23 – the same amalgam that the Diatessaron makes, seen for example in the Middle-Dutch translation thereof, done in the thirteenth century from a lost Latin translation, and in the Arabic translation thereof.107 Van Reeth applies the same treatment to the passages of the Qur’an which pertain to the infancy of Mary (Q 3:35–48), John (Q 19:3), and Jesus (Q 3:37; 19:22–6), showing again that “le Coran témoigne de la tradition du Diatessaron.”108 He does the same again with the Docetist version of the Crucifixion of Jesus (Q 4:157), but in this case he refers to Angel-Christology (cf. G. Lüling), notably that of the Elkesaites, declaring: “Plutôt qu’un simulacre que Dieu aurait façonné et substitué au Christ pour être crucifié à sa place, il s’agit originalement de la forme humaine que Dieu a instaurée pour Jésus au moment de l’incarnation et dans laquelle sa personne transcendante et angélique pouvait descendre.”109 Even if the Diatessaron does not explain all of the Qur’anic particularities on the life of Jesus, Van Reeth makes the following conclusion: “En se référant au Diatessaron comme l’avait fait Mani avant lui, le Prophète Muhammad pouvait souligner l’unicité du message évangélique. En outre, il s’inscrivait dans la longue lignée de Marcion, de Tatien et de Mani. Tous ont voulu (ré)tablir le vrai Évangile, afin d’en atteindre le sens original. Ils se croyaient autorisés à faire de travail d’harmonisation textuelle parce qu’ils s’assimilaient au Paraclet que Jésus avait annoncé.”110 99 CLAUDE GILLIOT The same Van Reeth offers his support, with refinements and connections, to the interpretation of Luxenberg of the so-called virgins of paradise (Q 44:54; 52:20), and ephebes of this same paradise (Q 52:24; 56:17; 76:19),111 mentioned in the present day text Qur’anic text (and which have fueled the imagination of so many Muslim men). The origins thereof are likewise in Syriac literature. They are obscured by a deformation in the present day Qur’anic text. Thus kurâ d’enbên in the Syriac version of the Apocalypse of Baruch becomes in the Qur’an hur al-‘ayn. Other historical backgrounds that could suggest an Aramaic path include the question of the origins of Arabic writing and the close relations between Mecca and Hira,112 along with the question of Biblical passages translated into Arabic and the use of Arabic in portions of Christian liturgy,113 subjects which we have discussed elsewhere and to which will we not now return. Conclusion: toward a critical restruction of the Qur’an It appears difficult to follow Nöldeke on a number of fundamental points. Essentially, when he declared: “It is important to note that the good linguistic sense of the Arabs kept them, almost entirely, from imitating the particular peculiarities and weaknesses found in the language of the Qur’an” (“Wichtig ist nun, daß der gesunde Sprachsinn der Araber sie fast ganz davor bewahrt hat, die eigentlichen Seltsamkeiten und Schwächen der Koransprache nachzuahmen”),114 he nevertheless held that, dialectal occurrences notwithstanding, this language was Classical Arabic (or better: ‘arabiyya). In this he was followed by the great majority of Islamicists, including Régis Blachère (1900–73) and Rudi Paret (1901–83), to mention only translators of the Qur’an. He continues to be followed in this presently, with some notable exceptions. We distance ourselves from this conviction of Nöldeke and no less from another: “Slight clerical error there may have been, but the Koran of Othmán contains none but genuine elements – though sometimes in very strange order,”115 or likewise: “Nothing inauthentic: the Qur’an contains only genuine passages” (“Keine Fälschung: der Korân enthält nur echte Stücke”). In fact, this proposition contains two hypotheses: On one hand, that the Qur’an we have is in fact the ‘Uthmanic codex; and on the other hand, that this ‘Uthmanic codex in fact contains the authentic revelations delivered by Muhammad. Nevertheless Nöldeke later (1909) admitted the possibility that there were interpolations in the Qur’an,116 after the publication of a contribution of August Fischer on this subject.117 Too little has been made of the work carried out by al-Hajjaj b. Yusuf in the epoch of the Umayyad ‘Abd al-Malik (r. 65–86/685–705). Without going so far as to consider, with Paul Casanova (d. 1926), that the ‘Uthmanic recension is a fable, or that it “n’a qu’une filiation fantaisiste,”118 one can presume that modifications were made to the text.119 Having said that, we are still in need of precise research, such as those that Fred Leemhuis conducts which have shown that at the beginning of the second/eight century qira’a (“variant”) and tafsir 100 THE AUTHORSHIP OF THE QUR’AN (“interpretation”) were not always separate and that the “standard text [of the Qur’an] was not yet universally accepted.”120 Meanwhile, the work of Luxenberg (and already that of Lüling before it) can be understood in the tradition of variae lectiones of the Qur’an when one distinguishes between three types of variation, namely: “minor variations,” that is different readings or interpretations of the same consonantal ductus; “major variations,” namely differences in the ductus, for example, in non-’Uthmanic codices, but also in the variants known as irregular (al-qira’at al-shadhdha); and finally, “extreme variations,” namely Arabic/Aramaic transliterations of the ductus.121 The historical basis that is lacking in the work of Luxenberg might be partly found in Syriac literature, and partly in the bond which, according to the Muslim sources, existed between Muhammad and those who enthroned him as a prophet (Waraqa b. Nawfal, Khadija, and others), or those who furnished him with reports or enriched those that he already possessed on the Bible and post-Biblical literature. In any case, these two critical reconstructions of the Qur’an – the reconstruction downhill (the materials collected by Bergsträsser and Pretzl, the rich literature of variae lectiones, increasingly edited) and the reconstruction uphill (with work such as that of Lüling, Luxenberg, Van Reeth, and so on) – must proceed in concert. As for us, we find in this research motivation for a critical reading of the Muslim sources which reflects a “lectionary” in constant evolution, perhaps until the Umayyad period: informers of Muhammad, the reception by Muhammad and his collaborators, abrogation, the “forgetting” of verses, or even suras, missing (or victim to forgetting) verses or suras, collections more or less complete, interpolations, partial correction of faults (lahn, pl. luhun) contained in the text (and which perdure in the present day vulgate), various linguistic emendations, and so on. A prophet is not created in a single day, and a holy book no less! 122 Notes 1 A. Sprenger, Das Leben und die Lehre des Mohammad, 2nd edn, Berlin: Nicolai, 1869. 2 J. Wansbrough, Quranic Studies: Sources and Methods of Scriptural Interpretation, Oxford: Oxford University Press, 1977. 3 P. Crone and M. Cook, Hagarism: The Making of the Islamic World, Cambridge: Cambridge University Press, 1977. 4 A. Rippin, Muslims: Their Religious Beliefs and Practices, London: Routledge 1990, 1, ix. 5 Cl. Gilliot, “Les ‘informateurs’ juifs et chrétiens de Muhammad. Reprise d’un problème traité par Aloys Sprenger et Theodor Nöldeke,” JSAI 22, 1998, 84–126. 6 Cl. Gilliot, “Informants,” EQ, 2, 512–18. 7 Ch. Luxenberg, Die syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der Koransprache, Berlin: Das arabische Buch, 2000 (2nd edn Berlin: Schiler, 2004) (translation: The Syro-Aramaic Reading of the Koran, Berlin: Schiler, 2007); see the review article of Gilliot, “Langue et Coran: une lecture syro-araméenne du Coran,” Arabica 50, 2003, 381–93; the very negative review of François de Blois, 101 CLAUDE GILLIOT 8 9 10 11 12 13 14 15 16 17 in Journal of Qur’anic Studies 5, 2003, 92–7; Ch. Burgmer, Streit um den Koran: Die Luxenberg-Debatte: Standpunkte und Hintergründe, Berlin: Schiler, 2004. The expression “the obscure beginnings” is taken up from the German title of: K.-H. Ohlig and G.-R. Puin (eds), Die Dunklen Anfänge: Neue Forschungen zur Entstehung und frühen Geschichte des Islam, Berlin: Schiler, 2005. First World Congress of Middle Eastern Studies, Mainz, September 8–13, 2002. This long contribution will be published under the title: “Une reconstruction critique du Coran ou comment en finir avec les merveilles de la lampe d’Aladin?” in M. Kropp (ed.), Results of Contemporary Research on the Qur’an: The Question of a Historico-Critical Text, Beirut, Orient-Institut, 2006, 33–137. For this issue, which is not dealt with here, see Gilliot, “Reconstruction,” § nos. 2–14. G. Bergsträsser, “Plan eines Apparatus criticus zum Koran,” Sitzungsberichte der Bayrischen Akademie der Wissenschaften, philosophisch-hist. Abt., 1930, Heft 7, reprint in R. Paret (ed.), Der Koran, Darmstadt: Wissenschaftliche Buchgesellschaft, 1975, 389–97. O. Pretzl, “Bericht über den Stand des Koranunternehmens der Bayerischen Akademie der Wissenschaften in München,” Mitgliederversammlung der DMG, Halle, 3. Januar 1936, in ZDMG 89, 1935, 20–1; idem, “Die Fortführung des Apparatus Criticus zum Koran,” Sitzungsberichte der Bayrischen Akademie der Wissenschaften, philosophisch-hist. Abt., 1934, Heft 5; idem, “Die Wissenschaft der Koranlesung,” Islamica 6, 1935–6, 1–47; 230–46; 290–331. A. Jeffery, “Progress in the study of the Qur’an text,” Moslem World 25, 1935, 4–16, reprint in Paret, Der Koran, 398–412; idem, Materials for the History of the Text of the Qur’an, Leiden: Brill, 1937, reprint New York: AMS, 1975; idem, “The textual history of the Qur’an,” Journal of the Middle East Society (Jerusalem) 1, 1947, 35–49; reprint in idem, The Qur’an as Scripture, New York: Moore, 1952, 89–103; A. Jeffery and I. Mendelsohn, “The orthography of the Samarqand Qur’an codex,” JAOS 3, 1942, 175–94. Th. Nöldeke, “Hatte Muhammad christliche Lehrer?” ZDMG 12, 1858, 699–708; A. Sprenger, “Ueber eine Handschrift des ersten Bandes des Kitàb Tabaqàt al-kabyr vom Sekretär des Wàqidy,” ZDMG 3, 1849, 450–6, p. 454 for Bahira; idem, “Mohammad’s Journey to Syria and Professor Fleischer’s Opinion Thereon” (with texts and translations from al-Tirmidhi and Ibn Ishâq), Journal of the Asiatic Society of Bengal 21, 1852, 576–92; idem, Leben, 2nd edn, 1, 178–204 (on Bahira); 2, 379–90 (“Wie hies der Lehrer?”); 390–7 (s.t. “Asâtyr alawwalyn, d.h. die Märchen der Alten”). One should add the sections on Waraqa b. Nawfal, 1, 81ff., 124–34; idem, “Mohammad’s Zusammenkunft mit dem Einsiedler Bahyrâ,” ZDMG 12, 1858, 238–49; Gilliot, “Les ‘informateurs’ juifs et chrétiens de Muhammad. Reprise d’un problème traité par Aloys Sprenger et Theodor Nöldeke,” JSAI 22, 1998, 84–126; idem, “Informants”; idem, “Zur Herkunft der Gewährsmänner des Propheten,” in K.-H. Ohlig and G.-R. Puin (eds), Die dunklen Anfänge, 148–69. Asatir is probably derived from Syriac; see S. Fraenkel, Die aramäischen Fremdwörter im Arabischen, Leiden: Brill, 1886, 250; GdQ1, 16, n. 4, with other possible origins; A. Mingana, “Syriac influence on the style of the Kuran,” Bulletin of the John Rylands Library 11, 1927, 88–9. Cf. the remarks of A.-L. de Prémare, “Les textes musulmans dans leur environnement,” in Cl. Gilliot et T. Nagel (eds), “Les usages du Coran,” Arabica 47, 2000, 391–408, see 400–3 on zubur al-awwalin (Q 26:196). The other places in the Qur’an which provide occasion for the exegetes to address this subject include the aforementioned Q 25, 4–5, as well as Q 26, 195; see Gilliot, “Informateurs,” § nos. 15–19, 23, 25. Al-rumiyya here seems to refer to the languages spoken in the Byzantines territories, not only Greek, but also Aramaic for Christians or Jews of Syria, Palestine, Iraq, and so on. 102 THE AUTHORSHIP OF THE QUR’AN 18 ‘Ali Muhammad Shahata (ed.), Cairo: Mu’assasat al-Halabi, 1980–9; Gilliot, “Informateurs,” § no. 12; idem, “Informants,” 513a. 19 Ibn Ishaq, Sirat rasul Allah, F. Wüstenfeld (ed.), Göttingen: Dieterich, 1858–60, 260; A. Guillaume, The Life of Muhammad: A Translation of Ibn Ishaq’s Sirat Rasul Allah (hereafter: Ibn Ishaq-Guillaume), Karachi: Oxford University Press, 1978 (1st edn: 1955), 180; Gilliot, “Informateurs,” § no. 13; idem, “Informants.” 20 Yahya b. Sallam (d. 200/815), Tafsir, Hind Shalabi (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1425/2004, 1, 91; al-Tabari, Tafsir, Ahmad Sa’id ‘Ali, Mustafa al-Saqqa et al. (eds), Cairo: Mustafa al-Babi al-Halabi, 1954–7, 14, 179, ad Q 16:103. 21 Yahya b. Sallam, Tafsir, 1, 91, ad Q 16:103; but p. 469 (ad Q 25:4): al-Hasan al-Basri said: “They mean a slave of Ibn al-Hadrami;” Ibn Abi l-Zamanin, Tafsir, Husayn b. ‘Ukasha and Muhammad b. Mustafa al-Kanz (ed.), Cairo: al-Faruq al-Haditha li-l-Tiba‘a wa-l-Nashr, 2002, 4, 200, ad Q 44:15; al-Tha‘labi, al-Kashf wa l-bayan ‘an tafsir al-Qur’an, Abu Muhammad Ibn ‘Ashur (ed.), Beirut: Dar Ihya’ al-Turath al-‘Arabi, 2002, 7, 123, has to be corrected: qala al-Hasan b. ‘Ubayd b. al-Khidr al-Habashi al-kahin, should be read: wa-fi qawli l-Hasani: huwa ‘abdun li-bni l-Hadramiyyi, wa-kana kahinan fi l-jahiliyyati; see Gilliot, “Zur Herkunft der Gewährsmänner,” § nos. 19–21. This error was taken up by Sprenger, Leben, 2, 389: “Hasan (Baçry) nennt den Abessynier ‘Obayd al-Chidhr, welcher ein Kahin war,” according to Baghawi, Tafsir (which is an abridgment of Tha‘labi’s Kashf ) ad Q 25: 4; cf. al-Baghawi, Tafsir, Khalid ‘Abd al-Rahman al-’Ak and Marwan Sawar (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1992, 3, 361. 22 Tha‘labi, Kashf, 6, 43. For the entire record see Hud b. Muhakkam, Tafsir, Balhajj b. Sa‘id Sharifi (ed.), Beirut: Dar Gharb al-Islami, 1990, 3, 201, ad Q 25:4, according to al-Hasan al-Basri and Muhammad b. al-Sa’ib al-Kalbi (d. 146/763); al-Suhayli, al-Ta‘rif wa-l-i‘lam fi-ma ubhima fi l-Qur’an min al-asma’ al-a‘lam, ‘Ali Muhammad al-Naqrat (ed.), Tripoli: Kulliyat al-Da‘wa al-Islamiyya wa-Lajnat al-Hifaz ‘ala al-Turath al-Islami, 1992, 173; Sir W. Muir, The Life of Mohammad from Original Sources, T.H. Weir(ed.) , Edinburgh: Grant, 1912, reprinted 1923, 66. 23 See Ch. Pellat, “al-Nadr b. al-Harith,” EI 2, 7, 874 24 Muqatil, Tafsir al-Qur’an, 3, 226–7; Guillaume, The Life of Muhammad, 135–6; Tabari, Tafsir, 18:182, ad Q. 25:5; al-Nuwayri, Nihayat al-arab fi funun al-adab, n.e., Cairo: Dar al-Kutub al-Misriyya, 1923–98, 16, 220, 271; Gilliot, “Muhammad, le Coran et les ‘contraintes de l’histoire’,” in S. Wild (ed.), The Qur’an as Text, Leiden: Brill, 1996, 3–26, esp. 23–6. 25 See Gilliot, “Informants,” 517. 26 See Gilliot, “Le Coran, fruit d’un travail collectif?” in D. De Smet, G. de Callatay and J.M.F. Van Reeth (eds), Al-Kitab. La sacralité du texte dans le monde de l’Islam. Acta Orientalia Belgica, 2004, 185–231, esp. 188–90; for more details see the account on the meeting of Khadija with the monk Sergius, which also contains the topos “Holy! Holy!” 27 On Muhammad’s first experiences of revelation, see GdQ1,74–89; Sprenger, Leben, 330–49; Muir, The Life of Mohammad from Original Sources, 39–54; R. Bell, “Muhammad’s call,” The Muslim World 24, 1934, 13–19; idem, “Muhammad’s visions,” The Muslim World 24, 1934, 145–54; R. Blachère, Le problème de Mahomet, Paris: PUF, 1952, 38–43; R. Sellheim, “Muhammeds erstes Offenbarungserlebnis,” JSAI 10, 1987, 1–16; G. Schoeler, Charakter und Authentie der muslimischen Überlieferungen über das Leben Muhammads, Berlin: de Gruyter, 1996, 59–117. 28 Sprenger, Leben, 1:355: “Without the love and trust of Khadija, Muhammad would never have been a prophet, and when death snatched her from him, Islam lost in purity and the Quran in dignity” (our translation); cf. Muir, The Life of Mohammad from Original Sources, 106, n. 1. 103 CLAUDE GILLIOT 29 Tabari (i.e. Bal’ami), Muhammad, sceau des prophètes, trans. H. Zotenberg, Paris: Sindbad, 1980 (originally: Chronique de Abou-Djafar-Muhammad-ben-Djarirben-Yezid, traduite sur la version persane d’Abou-Ali Muhammad Bel‘ami, Paris: Maisonneuve, 1867–74), 67; cf. Sprenger, Leben, 1, 151–2. 30 See Sprenger, Leben, 1, 124–34, 302–4; Muir, The Life of Mohammad from Original Sources, 49–50; U. Rubin, The Eye of the Beholder, Princeton, NJ: Darwin, 1995, 103–12: “The Khadija-Waraqa story.” 31 On Ibn Ishaq’s Maghazi, which never was a complete book in the modern sense of the word, and its versions, see J. Fück, Muhammad ibn Ishâq. Literarhistorische Untersuchungen, Dissertation, Frankfurt am Main, 1925, 34–43; J. Horovitz, The Earliest Biographies of the Prophet and Their Authors, L. Conrad (ed.), Princeton, NJ: Darwin, 2002 (original text published, in Islamic Culture, 1927–8, vols 1–2), 74–90; S.M. al-Samuk Die historischen Überlieferungen nach Ibn Ishaq. Eine synoptische Rekonstruktion, Inaugural-Dissertation, Frankfurt am Main, 1978, 134–9. 32 Ibn Hajar, Tahdhib al-tahdhib, n.e., Hyderabad: Da’irat al-Ma‘arif al-Nizamiyya, 1907–9, reprint Beirut: Dar Sadir, n.d., 7, p. 71. 33 Cf. Gospel of John 15:25–6, where nomos (Law) is in relation with the Paraclete; Sprenger, Leben, 1, 127, 158. Nomos is a word loaned from Christian Palestinian Aramaic, coming from Greek nomos (law); A. Sprenger, “Ueber den Ursprung und die Bedeutung des arabischen Wortes Nâmûs,” ZDMG 13, 1859, 690–701. In the Sira of Ibn Ishaq it has the meaning of law or confidant; A. Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam (gest. 218/834), Frankfurt am Main: P. Lang, 1984, 349–50. 34 We have taken up Ibn Ishaq-Guillaume’s translation, 107 (Wüstenfeld (ed.), 153), modifying here the transliteration. This passage is within a longer narrative with the following chain of authorities: Ibn Ishaq/Wahb b. Kaysan, a mawla of the Zubayrids/‘Ubayd b. ‘Umayr b. Qatada al-Laythi Ibn Hisham, al-Sira al-nabawiyya, Mustafa al-Saqqa et al. (eds), Cairo: Maktabat wa-Matba‘at al-Babi al-Halabi, 1955 (1st edn: 1936), 1, p. 238; al-Tabari, Annales, M. J. De Goeje (eds), Leiden: Brill, 1879–1901, 1, 1151–2; The History of al-Tabari, VI, Muhammad at Mecca, trans. W.M. Watt and M.V. McDonald, Albany, NY: State University of New York Press, 1988, 72. 35 al-Bayhaqi, Dala’il al-nubuwwa, ‘Abd al-Mu‘ti Qal‘aji (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1405/1985, 2, pp. 148–9, with the following chain of transmitters at the beginning of the long narrative in which our passage appears: [ . . . ] Ahmad b. ‘Abd al-Jabbar al-‘Utaridi (d. 272/886; GAS, 1, 146)/Yunus b. Bukayr (d. 199/715)/Ibn Ishaq/‘Abd al-Malik b. ‘Abdalllah b. al-Sufyan b. al-‘Ala’ al-Thaqafi. 36 On Ibn Shihab al-Zuhri and his Maghazi, see GAS, 1, 280–3; Horovitz, The Earliest Biographies of the Prophet and Their Authors. 50–66; Schoeler, Charakter und Authentie der muslimischen Überlieferungen über das Leben Muhammads, 59–79; Husayn ‘Atwan, Riwayat al-Shamiyyin li-l-maghazi wa-l-siyar fi l-qarnayn al-awwal wa-l-thani li-l-hijra, Beirut: Dar al-Jil, 1986, 69–150. 37 Abu l-Faraj al-Isfahani, Kitab al-Aghani, n.e., Bulaq: n.p., 1285/1868, 3, 14; Cairo: Dar al-Kutub, 1927–74, 3, 120; cf. al-Zurqani, Sharh al-Mawahib al-laduniyya, Muhammad ‘Abd al-‘Aziz Khalidi (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1996, 1, 399. 38 GAS, 1, 278–9; Horovitz, The Earliest Biographies of the Prophet and Their Authors, 15–29; G. Schoeler, Charakter und Authentie der muslimischen Überlieferungen über das Leben Muhammads, 79–86; idem, “Foundations of a new biography of Muhammad: The production and evaluation of the corpus of traditions from ‘Urwah b. al-Zubayr,” in H. Berg (ed.), Method and Theory in the Study of Islamic Origins, Leiden: Brill, 2003, 21–8; J. von Stülpnagel, ‘Urwa Ibn az-Zubair. Sein Leben und seine Bedeutung als Quelle frühislamischer Überlieferung, Dissertation, Tübingen, 1957; Muhammad Mustafa al-A‘zami, Maghazi rasul Allah. ‘Urwa b. al-Zubayr b. 104 THE AUTHORSHIP OF THE QUR’ AN 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 al-‘Awwam, (transmitted by) Abu l-Aswad, Riyadh: Maktab al-Tarbiya al-‘Arabi li-Duwal al-Khalij, 1981; Salwa Mursi al-Tahir, Bidayat al-kitaba al-ta’rikhiyya ‘inda l-‘Arab. Awwal sira fi l-islam: ‘Urwa b. al-Zubayr b. al-‘Awwam, Beirut: al-Mu’assasa al-‘Arabiyya li-l-Dirasat wa-l-Nashr, 1995. Abu l-Faraj al-Isfahani, Kitab al-Aghani, 3, 122; cf. Zurqani, Sharh al-Mawahib al-laduniyya. The long version of the this narrative is with the following chain of transmitters: [ . . . ] Ibn Shihab (al-Zuhri)/‘Urwa b. al-Zubayr/‘A’isha, in al-Bukhari, al-Jami‘ al-sahih, L. Krehl et al. (eds), Leiden: Brill, 1862–1908, book 65, 3, 380–1; Ibn Hajar, Fath al-bari bi-sharh Sahih al-Bukhari, ‘Abd al-‘Aziz b (ed.). ‘Abdallah Baz, numeration of the hadiths according to the editor of Muhammad Fu’ad ‘Abd al-Baqi, under the direction of Muhibb al-Din Khatib, Cairo: Dar al-Kitab al-Jadid, 1390/1970; reprint Beirut: Dar al-Ma‘rifa, n.d., tradition 4953, 9, p. 715, commentary, 716–21; Bayhaqi, Dala’il, ‘Abd al-Mu‘ti Qal‘aji (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 2, 135–7. Bukhari, Sahih, book 1, chapter 3, Krehl (ed.), 1, 5; Ibn Hajar, Fath, 1, 23, tradition no. 3; Ibn Sayyid al-Nas, ‘Uyun al-athar, Cairo: Maktabat al-Qudsi, 1937, 1, 85; Mursi al-Tahir, Bidayat al-kitaba al-ta’rikhiyya ‘inda l-’Arab. Awwal sira fi l-islam, 79. On this legend in the Christian polemic against Islam, see R.G. Hoyland, Seeing Islam as Others Saw It. A Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam, Princeton, NJ: Darwin, 1997, 476–9, with further literature, 476, n. 1; Gilliot, “Le Coran, fruit d’un travail collectif?” 188–9. Ibid., 189. al-Salihi, Subul al-huda, ‘Adil Ahmad ‘Abd al-Mawjud and ‘Ali Muhammad Mu‘awwad (eds), Beirut: Dar al-Kutub al-‘Ilmiyya, 1414/1993, 2, 233–4. E.W. Lane, An Arabic-English Lexicon, Cambridge: Islamic Texts Society Trust, 1984 (first edition: London, 1863–93), 1, 1291. See A. Baumstark, “Das Problem eines vorislamischen christlich-kirchlichen Schrifttums in arabischer Sprache,” Islamica 4, 1929–31, 562–75, 566–7; idem, “Arabische Übersetzung eines altsyrischen Evangelientextes und die Sure 21, 105 zitierte Psalmenübersetzung,” OC, 1934, 3. ser. 9, 31, 165–88 (Ps 37, 29 and Coran 21, 105). ‘Abd Allah and ‘Umar Anis al-Tabba (eds), Beirut: Mu’assasat al-Ma‘arif, 1987 (1377/1957); M. Lecker, “Zayd b. Thabit, ‘A Jew with Two Sidelocks,’ Judaism and Literacy in Pre-Islamic Medina (Yathrib),” JNES 56, 1997, (259–73) 267; reprint in Jews and Arabs in Pre- and Early Islamic Arabia, Aldershot: Ashgate/Variorum, 1999, III. Muhibb al-Din Abu Sa‘id ‘Umar b. Gharama al-‘Amrawi (ed.), Beirut: Dar al-Fikr, 1995–2000; cf. Gilliot, “Le Qur’an, fruit d’un travail collectif?” 195–6; idem, “Reconstruction,” § no. 22, both with other versions. See the sections following the note cue (see p. 92). Gilliot, Exégèse, langue et théologie en islam. L’exégèse coranique de Tabari, Paris: Vrin, 1990, 109, n. 1; M. Lecker, “Tamim al-Dari,” EI2, 10, 189. Ibn Shabba, Ta’rikh al-Madina al-munawwara, Fahim Muhammad Shaltut (ed.), Jadda: Ahmad, 1399/1979, 3, 1008. Ibn Hanbal, al-Musnad, Muhammad al-Zahri al-Ghamrawi (ed.), Cairo: al-Matba‘a al-Maymaniyya, 1895, 1, 405; Ahmad Muhammad Shakir (eds), Cairo: Dar al-Ma‘arif, 1995 (begun in 1949), 4, 58, no. 3846; Lecker, “Zayd b. Thabit,” JNES 56, 1997, 259–60; Gilliot, “Reconstruction,” § no. 24. Abu ‘Amr al-Husayni b. ‘Umar b. ‘Abd al-Rahim (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1421/2000. See A.-L. Prémare, Les fondations de l’islam. Entre écriture et histoire, Paris: Seuil 2002, 306–10. Gilliot, “Le Coran, fruit d’un travail collectif ?” 214–7: on Q 13:31, apud Tabari, Tafsir, Ahmad Muhammad Shakir (ed.), Cairo: n.p., 1969, 16, 452, no. 20410, and the reaction of the editor, 16, 2 (his introduction), 452, n. 3. 105 CLAUDE GILLIOT 55 Gilliot, Exégèse, langue et théologie, 196–7; idem, “Le Coran, fruit d’un travail collectif?” 213ff.; J. Burton, “Linguistics errors in the Qur’an,” JSS 33, 1988, 181–96, according to Abu ‘Ubayd. 56 For these two issues, see Gilliot, “Collecte ou mémorisation du Coran. Essai d’analyse d’un vocabulaire ambigu,” a paper presented at the Eighth Colloquium From Jahiliyya to Islam, Jerusalem, July 2000, to be published in JSAI; idem, “Les traditions sur la composition ou coordination du Coran (ta’lif al-qur’an),” in Cl. Gilliot und T. Nagel (eds), Das Prophetenhadit. Dimensionen einer islamischen Literaturgattung, Göttingen: Vandenhoeck & Ruprecht, 2005, 14–39. 57 See Gilliot, “Un verset manquant du Coran ou réputé tel,” in M.T. Urvoy (ed.), En hommage au Père Jacques Jomier, O.P., Paris: Cerf, 2002, 73–100; idem, “Le Coran, fruit d’un travail collectif?” 199–204. 58 al-Jawaliqi, al-Mu‘arrab min al-kalam al-a‘jami ‘ala huruf al-mu‘jam, Ahmad Muhammad Shakir (ed.), Cairo: Dar al-Kutub al-Misriyya, 1942; Suyuti, al-Itqan fi ‘ulum al-Qur’an, Muhammad Abu l-Fadl Ibrahim (ed.), 2nd. revised edn, Cairo: al-Hay’a al-Misriyya al-‘Amma li-l-Kitab, 1974–5 (1st edn 1967), chapter 38, 2, 125–43; W.Y. Bell, The Mutawakkili of as-Suyuti, a translation of the Arabic text, with introduction, notes, and indices, Dissertation Yale University, 1924; A. Sprenger, “Foreign words occurring in the Qôran,” Journal of the Asiatic Society of Bengal 21, 1852, 109–14. R. Dvořák, Ein Beitrag zur Frage über die Fremdwörter im Koran, Inaugural dissertation, München, 1884 (above all on tannur); FV; Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam (gest. 218/834); Cl. Gilliot, Exégèse, langue et théologie, 95–110; A. Rippin, “Foreign vocabulary,” EQ, 2, 226–37. 59 J. Horovitz, “Jewish proper names and derivatives in the Koran,” Hebrew College Annual, 2, 1925, 145–227 (reprint, Hildesheim: Olms, 1964); idem, Koranische Untersuchungen, Berlin and Leipzig: de Gruyter, 1926, 156–65. 60 GdQ1, 31–4; FV, 233–4; Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam (gest. 218/834), 286–91; Luxenberg, Die syro-aramäische Lesart des Koran, 54–60 (2nd edn, 81–7) 61 GdQ2, 11; Suyuti, al-Itqan fi ‘ulum al-Qur’an, chapter 18, 1, 205. For more details, see Gilliot, “Reconstruction,” § no. 15. 62 For the Ethiopian origin of mushaf (also: mashaf and mishaf), see Fraenkel, Die aramäischen Fremdwörter im Arabischen, 248; Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam (gest. 218/834), 228–30. 63 Fraenkel, Die aramäischen Fremdwörter im Arabischen, 237–8; FV, 180–2; Hebbo, Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam (gest. 218/834), 208–10. 64 FV, 72–3. For other words see A.-L. de Prémare, “Textes musulmans,” 400–6. 65 Wansbrough, Quranic Studies: Sources and Methods of Scriptural Interpretation, 98–9 (for the exegetes); Larcher, “Language, concept of,” EQ, 3, 109 (for the linguistic theorization of the exegetical debate); Gilliot and Larcher, “Language and style,” 114. 66 See Gilliot, Exégèse langue et théologie, chapters 3–4, 73–110. 67 H.L. Fleischer, “Ueber arabische Lexicographie und Ta’âlibi’s Fiüh al-lu∫a,” in Berichten für die Verhandlungen der k. sächs. Gesellschaft der W., Philol.-histor. Cl., 1854, 1–14, p. 5; reprint in Kleinere Schriften 3, Osnabrück 1964, 152–66, p. 156. 68 Gilliot and Larcher, “Language and style,” 122. 69 Gilliot and Larcher, “Language and style,” 109–24; Pierre Larcher, “Language, concept of,” EQ, 3, 108–9; Gilliot, “Le Coran, fruit d’un travail collectif?” 209–17. 70 Ibid. 71 Th. Nöldeke, “Zur Sprache des Korans,” in Neue Beiträge zur semitischen Sprachwissenschaft, Strassburg: Trübner, 1910, 1–30; French translation by 106 THE AUTHORSHIP OF THE QUR’AN 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 G.-H.Bousquet, Remarques critiques sur le style et la syntaxe du Coran, Paris: Maisonneuve, 1953. GdQ1, 107. GdQ1, 143, n. 2; this note was probably written by Schwally. K. Vollers, Volkssprache und Schriftsprache im alten Arabien, Strassburg: Trübner 1906, reprint, Amsterdam: APA, 1981, 169; M. Zwettler, The Oral Tradition of Classical Arabic Poetry, Columbus, OH: Ohio State University Press, 1978, 117–18, with discussion of this thesis; K. Versteegh, The Arabic Language, Edinburgh: Edinburgh University Press, 2001 (1st edn: 1997), 40–1. Vollers, Volkssprache und Schriftsprache im alten Arabien, 175–85. P.E. Kahle, The Cairo Geniza, Oxford: Oxford University Press, 1959 (1st edn: 1947), 141–9, 345–6; idem, “The Qur’an and the ‘Arabiya,” in S. Löwinger and J. Somogyi (eds), Ignace Goldziher Memorial Volume, Budapest: n.p., 1948, 163–8; idem, “The Arabic readers of the Koran,” JNES 8, 1949, 65–71. See Gilliot and Larcher, 123–4. Kahle, The Cairo Geniza, p. 142. J. Owen, “Id∫am al-Kabir and history of Arabic language,” in W. Arnold and H. Bobzin (eds), “Sprich doch Aramäisch, wir verstehen es!” 60 Beiträge zur Semitistik für Otto Jastrow zum 60. Geburtstag, Wiesbaden: Harrassowitz, 2002, (503–20) 504. Gilliot and Larcher, 124. H. Hirschfeld, New Researches into the Composition and Exegesis of the Qoran, London: Royal Asiatic Society, 1902, 32. A. Mingana, “Syriac influence on the style of the Kuran,” Bulletin of the John Rylands Library 11, 1927, 77–98; reprint in Ibn Warraq (ed.), What the Koran Really Says: Language, Text and Commentary, Amherst, NY: Prometheus, 2002, 171–92. For more on Mingana, see Gilliot, “Reconstruction critique,” § no. 40. G. Lüling, Über den Ur-Qur’an: Ansätze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur’an, Erlangen: Lüling, 1974 (2nd edn: 1993); review of M. Rodinson, in Der Islam 54, 1977, 321–5; G. Lüling, Die Wiederentdeckung des Propheten Muhammad: Eine Kritik am ‘christlichen Abendland,’ Erlangen: Lüling, 1981; review of books by Cl. Gilliot, “Deux études sur le Coran,” Arabica 30, 1983, 16–37. G. Lüling, A Challenge to Islam for Reformation: The Rediscovery and Reliable Reconstruction of a Comprehensive pre-Islamic Christian Hymnal Hidden in the Koran under Earliest Islamic Reinterpretations, Delhi: Motilal Banarsidass, 2003. Lüling, Challenge, LXIII. See Gilliot and Larcher, 129–30. Luxenberg, Die syro-aramäische Lesart des Koran, 10–15; Gilliot, “Langue et Coran,” 387–9; Gilliot and Larcher, 130. A. Neuwirth, “Qur’an and history – A disputed relationship. Some reflections on Qur’anic history and history of the Qur’an,” Journal of Qur’anic Studies, 2003, 1–18. Simon Hopkins, review of Luxenberg’s book, in JSAI 28, 2003, 380. Luxenberg, Die syro-aramäische Lesart des Koran, 269–76 (2nd edn, 304–10). See Gilliot, “L’embarras d’un exégète musulman face à un palimpseste. Maturidi et la sourate de l’Abondance (al-Kawthar, sourate 108),” in R. Arnzen and J. Thielmann (eds), Words, Texts and Concepts Cruising the Mediterranean Sea, Leuven: Peeters 2004, 33–69. al-Qurtubi, Tafsir, Ahmad ‘Abd al-Halim al-Barduni et al. (eds), Cairo: Dar al-Kitab al-’Arabi, 1952–67; reprint: Beirut: Dar Ihya’ al-Turath al-’Arabi, 1965–7, 20, 216–18; H. Birkeland, The Lord Guideth: Studies on Primitive Islam, Oslo: Nygaard, 1956, 56–99 (on the whole sura 108), here, p. 73; Gilliot, “L’embarras,” 40–3. Ibid., 44–6. Birkeland, The Lord Guideth: Studies on Primitive Islam. 107 CLAUDE GILLIOT 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 Qurtubi, Tafsir, 20, 217. GAS, 1, 614–15. Tha‘labi, Kashf, ms. Istanbul, Ahmet III, 76. Tha‘labi, Kashf, 10, 310. K. Versteegh, Arabic Grammar and Qur’anic Exegesis in Early Islam, Leiden: Brill, 1993, 157. We thank our colleague Pierre Larcher who has called our attention to the relation between harf and articulation. Al-Maturidi, Ta’wilat ahl al-sunna, Fatima Yusuf al-Khaymi (ed.), Beirut: Mu’assasat al-Risala, 1425/2004, 5, 526; idem, Ayat wa-suwar min ta’wilat al-Qur’an, A. Vanlioælu and B. Topaloælu (eds), Istanbul: Acar Matbaacilik, 2003, 73–4. Luxenberg, Die syro-aramäische Lesart des Koran, 269–76 (2nd edn, 304–11); cf. Gilliot, “Embarras,” 48–52; idem, “Reconstruction,” § 45–52. J.M.F. Van Reeth, “L’Évangile du Prophète,” in D. De Smet, G. de Callatay and J.M.F. Van Reeth (eds), al-Kitab, Acta Orientalia Belgica, 2004, 155–74; Gilliot, “Reconstruction,” § 51, quoting and following Van Reeth. Ibid., 158. Van Reeth, “L’Evangile du Prophète,” 155, n.1. Ibid., 161–2. Ibid., 163; cf. H. Räisänen, Das koranische Jesusbild: Ein Beitrag zur Theologie des Korans, Helsinki: Missiologian ja Ekumeniikan, 1971, 23–37; idem, Marcion, Muhammad and the Mahatma, London: SCM, 1997, 87–91, but without referring to the Diatessaron. Van Reeth, 166. Ibid., 174 Van Reeth, “Le vignoble du paradis et le chemin qui y mène. La thèse de C. Luxenberg et les sources du Coran,” published in Arabica 53, 4, 2006, 511–24; Luxenberg, Die syro-aramäische Lesart des Koran, 221–69 (2nd edn, 255–303). Gilliot, “Reconstruction,” § 26–33. Gilliot, “Reconstruction,” § 34–9. Th. Nöldeke, “Zur Sprache des Korans,” 22; Bousquet, Remarques critiques, 34. Th. Nöldeke, “The Koran,” in Sketches from Eastern History, London: Black, 1892, 53. GdQ1, 99. A. Fischer, “Eine Qoran-Interpolation,” in Carl Bezold (ed.) Orientalische Studien Theodor Nöldeke zum 70. Geburtstag, I, Gieszen: Töpelmann 1906, 33–55; see now T. Nagel, Medinensische Einschübe in mekkanischen Suren, Göttingen: Vandenhoeck and Ruprecht, 1995. P. Casanova, Muhammad et la fin du monde: Étude critique sur l’islam primitif, Paris: Geuthner, 1911–24, 127. Ibid., 141–2; GdQ3, 104, n. 1; A. Mingana, “The transmission of the Koran,” Muslim World 7, 1917, 223–32, 402–14, esp. 413–4; reprint in Ibn Warraq (ed.), The Origins of the Koran: Classic Essays on Islam’s Holy Book, Amherst, NY: Prometheus, 1998, 97–113, esp. 112–13. F. Leemhuis, “Ursprünge des Korans als Textus receptus,” in S. Wild and H. Schild (eds), Akten des 27. Deutschen Orientalistentages, Würzburg: Ergon, 2001, 301–8, esp. 307. Gilliot and Larcher, “Language and style,” 131–2. A. Mingana, “The transmission of the Koran,” 412/reprint, 112. 108 5 CHRISTIAN LORE AND THE ARABIC QUR’AN The “Companions of the Cave” in Surat al-Kahf and in Syriac Christian tradition Sidney Griffith Introduction It is something of a truism among scholars of Syriac to say that the more deeply one is familiar with the works of the major writers of the classical period, especially the composers of liturgically significant, homiletic texts such as those written by Ephraem the Syrian (c. 306–73), Narsai of Edessa and Nisibis (c. 399–502), or Jacob of Serugh (c. 451–521), the more one hears echoes of many of their standard themes and characteristic turns of phrase at various points in the discourse of the Arabic Qur’an. Conversely, Qur’an scholars in search of the origins of what they sometimes present as the “foreign vocabulary” of the Qur’an have not infrequently called attention to what they consider to be the high incidence of Syriac loan words and cognates in the Arabic idiom of the Islamic scripture. One difficulty which has attended the study of these matters has been that while most Qur’an scholars are well trained in Islamic languages such as Arabic, Persian and Turkish, they have seldom had more than a philological grasp of Syriac and almost no first-hand acquaintance with the classical literature of the language. Similarly, most Syriac scholars are deeply immersed in the study of the classical texts of the fourth, fifth and even the sixth centuries, but their grasp of Arabic is largely grammatical and lexical and they are often not at all familiar with Qur’anic or other early Islamic literature. Put another way, Qur’an scholars have often been unwilling to consider pre-Qur’anic, Syriac religious discourse as belonging in any way within the Qur’an’s hermeneutical circle. And Syriac scholars have seldom seen any reason to think that the Qur’an belongs within the textual or discursive framework of “Late Antique” Early Christian or Patristic thought. The result has been that when scholars have posited Syriac connections for some locutions in the Arabic Qur’an, be they grammatical, lexical or even 109 SIDNEY GRIFFITH thematic, they have done so almost in a vacuum, without much to say about their methodological and hermeneutical presuppositions. There have been few if any efforts to set forth well articulated and historically plausible sets of principles or hypotheses in the light of which the stipulated coincidences between Syriac and Qur’anic usages or modes of expression might find their most likely significance. From the linguistic and philological point of view, already in the early years of the twentieth century scholars working in the west were assuming that there was a Syriac background for a significant portion of the Qur’an’s Arabic wording.1 Alphonse Mingana, writing in 1927, estimated that 70 percent of the “foreign influences on the style and terminology” of the Qur’an could be traced to “Syriac (including Aramaic and Palestinian Syriac).”2 Noting Mingana’s estimate of the high incidence of Syriac etymologies for a significant portion of the Qur’an’s “foreign vocabulary,” Arthur Jeffery then wrote in 1938 that “one fact seems certain, namely that such Christianity as was known among the Arabs in pre-Islamic times was largely of the Syrian type, whether Jacobite or Nestorian.”3 He noted further that numerous early Islamic texts mention Muhammad’s contacts with both Syrian and Arabian Christians and this observation prompted Jeffery to conclude that these testimonies, coming from within the early Islamic community, “at least show that there was an early recognition of the fact that Muhammad was at one time in more or less close contact with Christians associated with the Syrian Church.”4 The effort to shed further light on the conundrums of the Arabic Qur’an by reference to Syriac lexicography and grammatical usage found its most persistent advocate at the end of the twentieth century in the work of the author who employs the pseudonym Christoph Luxenberg.5 Alleging that both traditional Muslim mufassirun and modern scholars alike have neglected what he calls the “Syro-Aramaic” roots of Qur’anic Arabic in favor of an overly exclusive reliance on Bedouin Arabic idiom for data to explain the text’s meanings, Luxenberg’s method is to examine selected cruces interpretum in the text from what he calls the “Syro-Aramaic” perspective. He changes the vowels and diacritical points of selected lexemes as necessary, to explore the possibility that with the Syriac dictionary in hand a more intelligible reading of hitherto obscure passages may be attained, often then found to be congruent with earlier, Aramaic grammatical forms or syntactical usages. In the ensemble, the over-all suggestion is made that when it is read from Luxenberg’s “Syro-Aramaic” perspective, the Qur’an can be thought to have once been a very different scripture from the one it has become in the hands of its Muslim and western commentators from early Abbasid times until now. Indeed Luxenberg’s enterprise seems, under the guise of a philological quest, to be a modern-day analog of the efforts of some earlier Arabic-speaking, Christian apologists of the early Islamic period to argue that before it was “corrupted” by early Muslims, and Jewish converts to Islam, the Qur’an was actually a book of virtual Christian meaning and sensibility.6 While this is certainly not Luxenberg’s avowed purpose, the net effect of his philological methods, to the degree that they would be deemed plausible in individual instances, is certainly to 110 CHRISTIAN LORE AND THE ARABIC QUR’ AN bring the Arabic Qur’an’s narratives much more within the hermeneutic range of Aramean Christianity in its Syriac expression than has hitherto been customary. There is a particular hermeneutical danger in the purely linguistic and philological approach to the search for the influence of Syriac on the Arabic Qur’an. For the ingenuity of the philologist can all too readily manipulate the linguistic materials into possible grammatical and lexical formulations solely on the basis of philological or orthographic considerations which leave out of account that degree of historical or cultural probability which is requisite for plausibility. In other words, from a responsible interpretive point of view, purely grammatical or etymological readings of words and phrases in the Arabic Qur’an on the basis of a presumed underlying Syriac must be supported by reference to a thematic context which would make them not only possible and plausible, but in all likelihood the carriers of the authentic, originally intended meaning. From a thematic point of view, as opposed to strictly philological or lexical concerns, Tor Andrae is undoubtedly the modern scholar who has so far the most systematically investigated what he considered to be Muhammad’s and the Qur’an’s indebtedness to the Syriac expressions of Christianity. In his well-known study of the Origins of Islam and Christianity,7 Andrae speaks confidently of what he takes to be the influences and borrowings from Syriac sources which one familiar with the idiom and dominant themes of the major texts of the classical period of Syriac literature may perceive in the Qur’an. His emphasis is on religious ideas and their characteristic formulae rather than on grammar or lexicography. Andrae first calls attention to the Christianity in Arabia at the time of Muhammad, mentioning in particular its flourishing in a Syriac form in places such as Najran and al-Hira, and among large tribal confederations such as the Lakhmids and the Ghassanids. Then, having dismissed the pre-Islamic, Arabic poets and the so-called Hunafa’ as sufficient sources for the Christian ideas and expressions he finds in the Qur’an, Tor Andrae devoted most of his study to what he calls “the eschatological piety (Frömmigkeit) of Muhammad.” He meant the Muslim prophet’s systematic thinking about the “last things,” final judgment and the hereafter, resurrection, final reward and punishment. According to Andrae, this piety of Muhammad’s was “a coherent, well-defined conception (Anschauung) which provided the most important expression of his religious personality.”8 According to Andrae, this eschatological conception articulated in finished formulae reflected a precise homiletic program ( prédication précise ⬍ bestimmten Verkündigung) with which he thought Muhammad must have been thoroughly familiar.9 In the sequel, following his detailed analyses and comparisons of passages in the Qur’an and in selected Syriac texts, Andrae argued that this “precise homiletic program” was that of the Syriac-speaking Christian community, which, in his view, served as the model for Muhammad’s own eschatological preaching.10 According to Andrae, “Whatever Muhammad received from Christianity he only learned it by way of oral preaching and personal contacts.”11 Presumably, although he does not say so explicitly, Tor Andrae supposed that Arabic was the language of the oral preaching and personal contacts. He does say 111 SIDNEY GRIFFITH that he thinks that the “Church of the East,” the so-called Nestorian church, was the source of the influences and borrowings from the Syrian Christians that went into the make-up of Muhammad’s “eschatological piety.”12 More specifically, Andrae proposed that the missionary preaching of the “Nestorians” came to Muhammad’s attention from Yemen, where a “Nestorian” mission had been established in the late sixth century.13 But Tor Andrae was also alive to what he called the “Monophysite” influence on Muhammad and the Qur’an. He found it in the Qur’an’s reflection of the Christology it rejected. According to Andrae, the Qur’an’s surprising idea that the Christian Trinity consists of God, Jesus and Mary,14 its polemic against the presumed Christian allegation that God is the Messiah15 and its reflection of an interest in the apocryphal narratives of Jesus’ infancy, all suggest a polemical response to Christian “Monophysite” interlocutors. He supposed that they were to be found in the Abyssinian associations of the early Islamic community.16 But later western commentators would posit as close a connection between Muhammad and the originally Syriac-speaking “Jacobites” as Tor Andrae had posited between Muhammad and the “Nestorians.” For example, John Bowman, pointing to the presence of the so-called Monophysites in Najran and among Arab confederations such as the Ghassanids, argued that the Qur’an’s prophetology and its biblical awareness are best explained on the hypothesis that Muhammad was in conversation with “Jacobite” Christians, among whom the Syriac Diatessaron circulated as the normal text of the Gospel. According to Bowman, Muhammad must have gained even his knowledge of Old Testament personae from a milieu in which the harmonized Gospel circulated because the only Old Testament personages named in the Qur’an are those whose names also appear in the Diatessaron. For Bowman, the weight of these observations strongly supports the view that the Christians with whom Muhammad would have been familiar, and who would have been in the Qur’an’s audience, would have been “Jacobites,” or “Monophysites,” as he called them.17 At one notable point, Tor Andrae singled out the works of St Ephraem the Syrian (c. 306–73), the early Syriac writer beloved by “Melkites,” “Jacobites” and “Nestorians” alike, as texts in which he could the most readily find Syriac vocabulary, turns of phrase and religious conceptions cognate with those to be found in the Arabic Qur’an.18 One of his suggestions in particular stirred up a scholarly storm. Andrae proposed that the houris of Paradise as depicted in the Qur’an (e.g. in al-Dukhan [44]:54; al-Tur [52]:20; al-Rahman [55]:72; al-Waqi‘a [56]:22) could be found prefigured in one of St Ephraem’s hymns De Paradiso (VII:18).19 Tor Andrae wrote One may recognize a veiled reference to the virgins of Paradise in Afrem’s saying: “Whoever has abstained from wine on earth, for him do the vines of Paradise yearn. Each one of them holds out to him a bunch of grapes. And if a man has lived in chastity, they (feminine) receive him into a pure bosom, because he as a 112 CHRISTIAN LORE AND THE ARABIC QUR’ AN monk did not fall into the bosom and bed of earthly love.”20 . . . Popular piety certainly interpreted this daring imagery in a crass and literal sense, and under such circumstances one cannot blame a citizen of pagan Mecca for doing the same thing. (Andrae, Mohammed: The Man and His Faith, 88) In 1948 Dom Edmund Beck, OSB, the modern editor of the critical editions of most of the Syriac works of Ephraem the Syrian, wrote a response to what he took to be Tor Andrae’s claim about St Ephraem’s meaning.21 Beck took it that Andrae was proposing that St Ephraem’s works suggested a heavenly reward for the celibate monk comparable to that provided by the houris of the Qur’an for the faithfully departed Muslim. So he went to some trouble to show that such could not have been St Ephraem’s meaning. Beck called attention to the faulty text of the Editio Romana of Ephraem’s hymn that Andrae had used, and then he set about explaining the imagery and symbolism of the passage in its context, by a somewhat complicated word-study of several key terms. In sum, Beck argued that St Ephraem’s imagery of the grape-vine, its stocks and shoots, evoked a vision of Paradise and a line of thinking which he thought definitively excluded any concept of the kind of delights provided by the houris. While Beck’s exposition of St Ephraem’s own thought is convincing, it seems that he did not in fact completely grasp Andrae’s point. Tor Andrae did not actually say that Ephraem envisioned anything like houris in Paradise. Rather, he suggested that “popular piety,” not to mention “a citizen of pagan Mecca,” might have been inspired by such lines as Ephraem wrote to conjure up the houris. It was Andrae’s major point, one should remember, that homiletic descriptions such as those by Ephraem, envisioning the blessings of Paradise in terms of a garden of delights, could reasonably be supposed somehow to lie behind the similar descriptions of Paradise in the Qur’an, especially if one would be prepared to concede that Ephraem’s descriptions could well have been reflected in the discourse of Arabic-speaking Christians who were in the audience of the Arabic Qur’an. In connection with a consideration of Tor Andrae’s suggestion about a Syriac background for the Qur’an’s depiction of Paradise, illustrated for him by reference to a passage in the works of St Ephraem the Syrian, it is interesting to note in passing that Christoph Luxenberg, in what has become the most widely quoted part of his work, has on philological and lexical grounds reinterpreted the key phrase in Q al-Dukhan (44):54 and al-Tur (52):20, zawwajnahum bi-hurin ‘inin, to mean not something on the order of, “We shall wed them to maidens with large, dark eyes,” or “We pair them with beautiful-eyed maidens,”22 but “We will make them comfortable under white, crystal(clear) (grape clusters),”23 claiming that so understood the expression is more consistent with the Qur’an’s own eschatological scenario. Whatever one might think of the verisimilitude of this interpretation, it is clear that it is certainly closer to St Ephraem’s image of the grape clusters which the Syrian writer says will welcome the chaste into their bosom than to the vision of the embraces of houris as conventionally imagined. 113 SIDNEY GRIFFITH While here is not the place to discuss the merits of Luxenberg’s reading, it may be just the place to point out that from a hermeneutical point of view, the case can be made that the difference in the readings is not a difference which philology or lexicography alone can plausibly reconcile. Rather, the difference here may well be the difference which Islam as a new narrative system itself makes in the presentation of traditional themes. One must point out that hermeneutically speaking, the meaning of the new Arabic Qur’an cannot authentically be reduced to the parameters of its presumed conceptual background in the language of one of its predecessor narratives. It is in this connection that one requires the aforementioned, historically plausible set of principles and hypotheses in the light of which Syriac scholars in particular might evaluate the coincidences they observe between classical Syriac texts on the one hand and on the other hand the seemingly borrowed usages or modes of expression which they can sometimes find in the Qur’an. The Qur’an and Syriac The historical and geographical setting in which the Arabic Qur’an first appeared, both in its oral form, as it circulated among Arabic-speaking Muslims during the lifetime of Muhammad, and as a written text following its collection and standardization in the decades following Muhammad’s death, favors the expectation that the Christian beliefs and practices it reflects, critiques and approves or rejects would in large part be those current in the Aramaic expressions of Christianity. The very script of North Arabic had Aramaic antecedents, be they Nabatean or Syriac. The Christianity finding its way among the Arabic-speaking populations beyond the frontiers of the Roman or Persian empires in the sixth and seventh centuries be it from Sinai, Palestine, Syria, Mesopotamia or southern Iraq, was Aramaic, and mostly Syriac, in its original homiletic and liturgical formulae. While there was also a significant presence of Abyssinian and even Coptic Christianity in Arabia in the sixth century, and traces of its characteristic idiom are not wanting even in the Qur’an, the case could nevertheless hardly be otherwise: the Christianity reflected in the Qur’an would for the most part betray its largely Aramaic and Syriac original formulation. The Arabic Qur’an had perforce to address the Christian beliefs, practices and narratives actually current among its Arabic-speaking audience, reflecting the modes of expression in which they were customarily voiced among them. It is for this reason that it is not surprising that scholars working from a philological or lexical perspective, such as Alphonse Mingana or Christoph Luxenberg, should find so many traces of Syriac usage behind the diction of the Arabic Qur’an,24 or that scholars who examine the text with the methodologies of the disciplines of the “History of Religions” or “Comparative Religion,” like Tor Andrae, should find in it echoes of a SyroAramaic religious discourse already present in its ambience. Given the historical and geographical circumstances of the people the Qur’an addressed, it is only to be expected. How else could its message have been intelligible to them? 114 CHRISTIAN LORE AND THE ARABIC QUR’ AN Given the high historical probability that the Christianity to which the Qur’an refers or the Christian diction which it reflects would have had its original expression in Aramaic and even Syriac, it is equally highly probable that among the Christians in the Qur’an’s audience its standard formulae and typical modes of expression, along with its narrative lore, originally Syriac, would all have normally been voiced among them in Arabic,25 with the traces and echoes of its Syriac origins still discernible in the newly minted Arabic expression. In other words, the Christians in the Arabic Qur’an’s audience spoke Arabic, but their distinctive Christian vocabulary disclosed its Syriac origins. Sometimes these Syriac origins appear as “Syriacisms” when such Christian locutions were quoted in the Arabic Qur’an. “Syriacisms” are words or phrases in the Arabic diction of the Qur’an which betray an underlying Syriac locution. That is to say, they are calques, or “loan translations” from Syriac into Arabic; they are not simply Syriac words used in place of Arabic words and phrases. The recognition of them requires a familiarity with classical Syriac phraseology, along with an eye for what might otherwise appear as awkwardness in the Arabic diction. A case in point, which the present writer has developed elsewhere,26 is the enigmatic phrase, thalithu thalathatin, usually translated as “third of three” in al-Ma’ida (5):73. Once the phrase is recognized as a “Syriacism,” reflecting the Syriac epithet for Christ, tlîthayâ, “the treble one,” the Qur’an can be seen to be rejecting the Christian belief about Christ in the very language in which Arabic-speaking Christians in its audience affirmed the belief, using an epithet which was awkward, even a‘jami, in its Arabic version, but full of typological significance for those who understood it in its full Christian sense, originally expressed in Syriac. An important hermeneutical ingredient in recognizing “Syriacisms” such as this one in the Arabic Qur’an is the concomitant recognition that the Qur’an repeats such locutions for its own rhetorical purposes, which may well be very different from the Christian purpose in coining or contriving the usually awkward Arabic words or phrases in the first place, or simply Arabicizing Syriac terms. For the Qur’an’s posture toward Christianity is primarily one of critique. Rhetorically, its critique is the more effective the more accurately it repeats actual Christian formulae in the context of the Qur’anic revelation that would highlight the perceived Christian inadequacy or, on the Qur’an’s principles, its falsity. Similarly, beyond the range of “Syriacisms” properly so called, genuinely Christian, even biblical lore, no doubt re-told in Arabic from Syriac originals for the benefit of Arabic-speaking Christians, is sometimes deployed in the Qur’an for the purpose of purchasing credibility for the Qur’an’s own teachings among the Arabic-speaking Christian members of its audience. Or more often, the Qur’an does not repeat or quote Christian scriptures or traditional narratives so much as it alludes to them, with the presumption that their stories, their dramatis personae and their exploits are already known to its audience. Indeed the Qur’an’s large presumption of its audience’s familiarity with the considerable amount of biblical and apocryphal scriptural material it contains, along with Jewish and Christian traditional lore, is itself the most convincing evidence of the circulation of 115 SIDNEY GRIFFITH these narratives in the Arabic-speaking milieu of the audience the Qur’an actually addresses. Hermeneutically speaking, one should approach the Qur’an as an integral discourse in its own right; it proclaims, judges, praises, blames from its own narrative center. It addresses an audience which is already familiar with oral versions in Arabic of earlier scriptures and folklores. The Qur’an does not borrow from, or often even quote from these earlier texts. Rather, it alludes to and evokes their stories, even sometimes their wording, for its own rhetorical purposes. The Arabic Qur’an, from a literary perspective, is something new. It uses the idiom, and sometimes the forms and structures, of earlier narratives in the composition of its own distinctive discourse. It cannot be reduced to any presumed sources. Earlier discourses appear in it not only in a new setting, but shaped, trimmed and re-formulated for an essentially new narrative. Syriac in the Arabic Qur’an is no longer Syriac; it may be a “Syriacism” in Qur’anic Arabic, or a narrative originally told in Syriac, which in an oral Arabic version has become a point of reference for the Qur’an’s own discourse. Allusions to, even quotations from, or structural similarities with earlier Syriac narratives do not control the Qur’an’s discourse. Rather, the Qur’an, framing the new hermeneutical horizon, casts any originally Syriac elements in its Arabic diction into the framework of meaning constructed by its own diction. In this context, the study of the Syriac prolegomena for any given passage in the Qur’an’s narrative is the study of a Syriac text which presumably, in all historical likelihood, lies behind the oral, Arabic version of a recital to which the Qur’an alludes or refers, with an assumption of its familiarity to the Qur’an’s own audience. One may think of it as a heightened form of oral “intertextuality” which envelops two closely related language communities, close to one another in time, space and scriptural consciousness. In important ways, historically speaking, due to commerce, transhumance and other cultural phenomena, the populations of the two language communities, the Arameans and the Arabs, had already been intermingled for centuries. It was only a matter of time and the accidents of history, or the providence of God, before an Arabic master-narrative arose to subsume the terms and themes of earlier Aramean narratives into a distinctively new pattern of meaning in a new text, which its community would consider divinely inspired.” The Companions of the Cave The legend of the “Companions of the Cave” or the “Sleepers of the Cave” is a good example of Christian lore current in the Syriac homiletic and liturgical tradition prior to the rise of Islam, to which the Qur’an alludes and portions of which the Qur’an includes in its own narrative for its own purposes, addressing its own Arabic-speaking audience who are presumed to be familiar with the details of the legend. For the present purpose it seems best first of all to recall the setting of the references to the legend of the “Companions of the Cave” in 116 CHRISTIAN LORE AND THE ARABIC QUR’ AN the Qur’an, then to discuss the forms in which it circulated among Christians in Syriac in pre-Islamic times, and presumably also in Christian Arabic folklore, and then to examine the recollection of the details of the legend as it survives in the earliest extant Syriac homiletic text which includes it. Against this background, one might then be in a position to make some useful observations about what there is to be learned by reading the Qur’anic passage in surat al-Kahf (9–26) against the background of the legend’s currency in the earlier, Syriac homiletic and liturgical tradition. The Companions of the Cave in the Qur’an In the Islamic scholarly tradition, as well as in western Qur’anic scholarship, there has been an abundant commentary on the narrative of the “Companions of the Cave.”27 Almost always, the commentators have looked back to the text of the Qur’an from a post Qur’anic standpoint, looking for the meaning of the narrative within the context of a developing or developed Islamic point of view; some westerners have looked for the origins of the legend in earlier Christian material, taking a more or less critical view of the Qur’an’s fidelity to this material or highlighting the coincidences.28 Be this as it may, in the present inquiry, the point of view is reversed. One approaches the narrative of the “Companions of the Cave” in the Arabic Qur’an from a pre-textual perspective, with a deep and primary familiarity not with the post Qur’anic Islamic commentary tradition but with the traditional literature of the Syriac-speaking Christians, among whose texts modern scholars have found the earliest, still extant accounts of the legend of the “Companions of the Cave.” The approach assumes, for reasons broadly stated earlier, that these texts furnish the form of the legend nearest to that in which it would in all likelihood have been familiar to the Arabic-speaking Christians in the Qur’an’s audience as Muhammad gave voice to the discourse sent down to him from God. The Qur’an’s evocation of the legend of the “Companions of the Cave” comes close to the beginning of surat al-Kahf (18:9–26), where Allah addresses Muhammad with the following question: “Do you reckon that the companions of the cave (ashab al-kahf ) and of the inscription29 are wondrously among Our signs?” (18:9). It is pertinent to notice at the outset that Allah poses this question in the context of recalling for Muhammad the purpose of the revelation of the Scripture (al-kitab ⫽ the Qur’an) to His servant in the first place, which among other things was “so that it might give a warning to those who say, ‘God has gotten a child’ ” (18:4). Similarly, at the end of the pericope of the “Companions of the Cave” Allah tells Muhammad that the youths had no protector other than God, “and He takes no one as an associate in His governance” (18:26). The next verse exhorts Muhammad to recite from the Lord’s scripture, “whose words no one will change,” and it ends with the assurance that “You will not find any recourse apart from Him” (18:27). Clearly it is in the context of commending trust in the revelation of the Qur’an that Allah reminds Muhammad of the story of “the youths who sought shelter in 117 SIDNEY GRIFFITH the cave” (18:10); He suggests that analogously with the verses of the Qur’an, this story is itself to be considered among “Our signs.” Alternatively, perhaps the message is also that just as this story is among “Our signs,” so are the verses of the Qur’an. In any event, it is notable that a purpose of the scripture is to warn “those who say God has gotten a child” (18:4) when in the next breath the text offers an extended evocation of a martyrdom and miracle story which had hitherto circulated only among Christians. And, at the end of the passage, the text intimates that an important meaning of the story is that God “takes no one as an associate in His governance” (18:26) nor is there “any recourse apart from Him” (18:27). It is hard to avoid the thought that not only is the Qur’an here using a familiar Christian narrative to enhance the understanding of the sense of the expression “God’s signs,” but that it is also proposing in the sequel that the true meaning of the Christian story corrects what the Qur’an considers to be one of the major errors of the Christian understanding. Namely, the doctrine that God has a son and that he is Jesus, the Messiah. This reading of the setting in context of the Qur’an’s evocation of the Christian legend of the “Sleepers” is not without its difficulties. To begin with, it is the received wisdom among both Muslim commentators and western scholars that the sura in which the legend is evoked is, with the exception of just a few verses, “Meccan.”30 Furthermore, it is generally assumed that Muhammad’s adversaries in Mecca were given to the polytheist view that angels and lesser goddesses could be taken to be Allah’s children. Accordingly, on this view the warning against “those who say God has gotten a child” (18:4) was not in the first place addressed to Christians but to Meccan polytheists.31 It is also the generally received opinion that Muhammad encountered opposition from Jews and Christians only in the Medinan period of his prophetic career, whereas in Mecca the attitude was more positive.32 However this may be, the fact remains that it is in this largely Meccan sura that the Qur’an evokes the memory of the Christian legend of the “Sleepers” with the clear expectation that it is already known to Muhammad and presumably also to other members of the Qur’an’s audience, a number of whom may well have been Arabic-speaking Christians. How else would one explain the currency of such a detailed reminiscence of a Christian legend, together with so many other elements of Christian scripture, doctrine and ecclesiastical lore that are to be found broadcast throughout the Qur’an?33 Not only are they present in the Qur’an, but often the text evokes them in such a way that there is evidently a presumption that the audience too is thoroughly familiar with them. At the outset it is clear that in the Qur’an’s evocation of the legend of the “Sleepers” Allah comments on the story and in the process poses rhetorical questions to Muhammad about its proper interpretation: “Do you reckon that the companions of the cave and the inscription are wondrously among Our signs?” (18:9). As the catechesis of Muhammad on this matter unfolds in the subsequent sixteen verses, one may distinguish two phases in the discourse: 18:10–20 and 21–6. In the first phase (vv. 10–20), directly addressing Muhammad in the second person, Allah recalls central elements of the legend and by interweaving stock phrases 118 CHRISTIAN LORE AND THE ARABIC QUR’ AN from other parts of the Qur’an He discloses the proper, Qur’anic understanding of the story’s central message. In the second phase (vv. 21–6), Allah addresses several points about which controversy had arisen in connection with the details of the story of the “Companions of the Cave” and issues Muhammad instructions in the imperative mood about how they are to be dealt with. The first phase of the Qur’an’s evocation of the legend of the “Companions of the Cave” itself involves two narrative stages: Allah’s statement of the central elements of the story (vv. 10–12) and Allah’s interpretive rehearsal (qass) of the authentic narrative (naba’) of the companions’ experience (vv. 13–20).34 The central elements as Allah states them are several: the youths sought refuge in the cave (v. 10); they prayed for their Lord’s mercy and right guidance (v. 10); Allah shut their ears for a number of years (v. 11); Allah roused them to know which of two parties had rightly computed the extent of their stay (v. 12). Allah’s interpretive rehearsal of the authentic narrative of the companions’ experience presents the story of the youths according to the familiar Qur’anic pattern of righteous individuals who believed in their Lord (v. 13), who called on none other as God (v. 14), whose people had taken on other gods without any show of power (v. 15). So too the youths, Allah says, after their refusal to worship any other than Allah, exhorted one another in the second person to take refuge in the cave where their Lord’s mercy would unfold for them and He would provide a way out of their predicament (v. 16, cf. similar wording in v. 10). Allah discloses to Muhammad the miracle of the sun’s movements, shifting a bit to the south of the cave’s opening at its rising and a bit to the north at its setting, while the youths were within. The text specifies that this too was one of “God’s signs” (v. 17, cf. also v. 9). Allah tells Muhammad, “You would think they were awake, but they were asleep (ruqud); We would turn them over,” while their dog stretched out his paws on the threshold. Allah assures Muhammad that had he come upon this scene he would have fled in dismay (v. 18). Allah says He roused the youths precisely so that they would ask questions among themselves; to the question about how long they had been there some said a day or so, some said the Lord knows best. Allah then recalls for Muhammad some of the further dialogue among the youths, “Send one of you into the city with money to search out and bring back to you the cleanest food, but let him act very courteously and not tell anyone about you (v. 19). If they were to get the better of you they would stone you or make you go back to their religion (milla) and then you would never thrive” (v. 20). In the second phase of the Qur’an’s evocation of the legend of the “Companions of the Cave,” Allah addresses points of controversy which had arisen about some details of the story and instructs Muhammand about how to deal with them. In the first place, God himself takes responsibility for the fact that the legend was well known among members of the Qur’an’s audience, including the information that somewhere there was a shrine or martyrion in their memory and that opinion was divided about how many companions there were. Allah says about the companions, “We alerted [people] to them so that they would know that 119 SIDNEY GRIFFITH God’s promise is true; there is no doubt about the hour, albeit that [people] dispute among themselves about their (i.e. the companions’) experience” (v. 21). The Qur’an then recalls that some people wanted to put up a building over them; those who prevailed opted for a place of worship (masjid) (v. 21). Regarding the differing estimates about the number of companions, Allah tells Muhammad, “Say, ‘My Lord knows best,’ ‘Do not engage in dispute about them,’ ‘Do not ask anyone’s considered opinion about them’ ” (v. 22). In the next two verses Allah reminds Muhammad not ever to say about something on his own recognizance, “I will do that tomorrow” (v. 23). “Remember” and “Say, ‘Perhaps my Lord will guide me’ ” (v. 24). Then, depending on how one understands the Arabic text, either Allah makes the declarative statement about the “Companions of the Cave”: “They stayed in their cave three hundred years; nine are added,” or the Qur’an is reporting that this is an estimate commonly given (v. 25). At the end, Allah tells Muhammad, “Say, God knows best how long they stayed. They had no protector apart from Him and He takes no one as an associate in His governance” (v. 26). The Companions of the Cave in pre-Islamic Syriac texts Although their story circulated in all the languages of early and medieval Christianity,35 modern scholars are not sure of the original language of the legend of the “Seven Sleepers of Ephesus,” as the “Companions of the Cave” are generally known among western Christians. The currently prominent opinion is that a record of their miraculous survival after more than three hundred years of entombment was first composed in Greek by Bishop Stephen of Ephesus between the years 448 and 450, albeit that the earliest extant texts are in Syriac and date from the sixth century. The thought is that the original “memoranda” (hypomnemata), later translated into Syriac, were first composed by Bishop Stephen in testimony to the veracity of the doctrine of the resurrection of the body, which had been denied by Bishop Theodore of Aegeae and his followers late in the fourth century.36 Nevertheless, the alternative opinion of a Syriac original, strongly seconded by Theodor Nöldeke in 1886, and bolstered by the remark of St Gregory of Tours (d. 594) that he owed his account of the “Seven Sleepers” to a Latin translation from a Syriac original, still survives. The thought among those who support this opinion is that the legend arose in the Syriac-speaking churches in connection with the “Origenist” controversies of the sixth century, in which differing opinions about the doctrine of the resurrection of the body were an issue.37 Whichever may have been the original language of the legend, the earliest extant texts in which it actually survives are in Syriac.38 The earliest Syriac texts which feature the story of the “Youths (tlayê) of Ephesus,” as the “Companions of the Cave” or “Seven Sleepers” are always called in Syriac, are two recensions of a liturgical homily (mêmrâ) attributed to Jacob of Serugh (c. 451–521),39 who spent most of his life as a monk composing homilies on biblical and other liturgical themes. Indeed the hundreds of homilies attributed to him are in the ensemble 120 CHRISTIAN LORE AND THE ARABIC QUR’ AN a virtual compendium of Aramean Christian thought and tradition, written in the classical Syriac idiom of Edessa and its environs in the crucial decades spanning the turn of the sixth century.40 At that time in the Syriac-speaking world the two parties who would in due course become two separate ecclesial communities were in the process of disengaging from one another over issues of biblical interpretation and Christology.41 Jacob of Serugh’s mêmrê circulated widely among those who would later be called “Jacobites,” the “Syrian Orthodox Church,” whose faith would be championed among the Arab Ghassanids and eventually the Christians of Najran.42 Toward the end of his life, Jacob himself wrote a letter of consolation at a time of persecution addressed to his brother Christians and confessors among the Himyarites of southern Arabia.43 The other confessional community among the Syriac-speakers, the so-called Nestorians,44 the “Church of the East,” whose faith had spread among the Arab Lakhmids and along the coast of southern Arabia, cherished the mêmrê of Narsai of Edessa and Nisibis (399–503), a rival of Jacob of Serugh at the time of the break-up of the School of Edessa in the course of the controversies precipitated by the decisions of the Council of Chalcedon (451).45 So far no trace of the legend of the “Youths of Ephesus” has been found among the mêmrê attributed to Narsai, or in any other “Nestorian” Syriac text which dates from the pre-Islamic period. In addition to the two recensions of the mêmrâ of Jacob of Serugh, two further Syriac accounts of the legend of the “Youths of Ephesus” are included in two “Jacobite” histories from pre-Islamic times. The first of them is in the Syriac epitome of the originally Greek Ecclesiastical History of Zacharias of Mytilene (c. 465–536), often called Zacharias Rhetor or Scholastikos. The surviving Syriac epitome of the History, which originally provided an account of the religious controversies of the years 450–91 from a “Jacobite” perspective, was included in a much longer Syriac chronicle compiled by an anonymous monk at Amida in the year 569.46 The second account is in a portion of the Ecclesiastical History of John of Ephesus (c. 507–86)47 which has been preserved in the later Chronicle of Dionysius of Tell Mahre (d. 845), which in turn was preserved in the Chronicle of Michael the Syrian (1126–99) and the Ecclesiastical Chronicle of Gregory Abu l-Faraj, commonly known as Bar Hebraeus (1226–86).48 It is notable that in Syriac in pre-Islamic times the legend of the “Youths of Ephesus,” as far as we can ascertain, circulated solely in “Jacobite” communities. This fact may well be significant for the study of the origins of the legend in the context of the theological controversies of the fifth century, but that inquiry is beyond the range of concerns in the present study.49 In connection with the inquiry into the Syriac background of the evocation of the legend of the “Companions of the Cave” in the Arabic Qur’an, the recognition of the fact that it circulated only among the “Jacobites” prompts one to draw the conclusion that in the Qur’an’s milieu, the narrative circulated first among Arabic-speaking, “Jacobite” Christians in the Ghassanid confederation in the Syro-Jordanian steppe land as well as in the environs of Najran in southern Arabia. From these centers it would have circulated among Arabic-speaking Christians throughout Arabia. 121 SIDNEY GRIFFITH Furthermore, given the likely, oral form of the legend’s circulation in this milieu it is reasonable to suppose that the liturgy and its wider ecclesial ambience was the primary setting in which the legend circulated among Arabic-speaking, “Jacobite” Christians. This being the case, it is also reasonable to suppose that the recollection of the details of the legend in the liturgically inspired mêmrâ of Jacob of Serugh is, textually speaking, the most likely, still extant, single narrative ancestor in Syriac in the background of the Arabic Qur’an’s evocation of the legend.50 Nevertheless, the fact remains that no single Syriac text includes every aspect of the legend as it circulated orally or as the Qur’an evokes it. A situation of intertextuality obtains, according to which one expects to find details of the narrative scattered among the several textual witnesses, with no one of them containing all the narrative features in a single record. The “The Youths of Ephesus” in the mêmrâ attributed to Jacob of Serugh The two recensions of the mêmrâ attributed to Jacob of Serugh on “The Youths of Ephesus” are transmitted in several different Syriac manuscripts, of different provenance and different dating.51 The first recension, and seemingly the earlier of the two, is known in two manuscripts, one of which, Vatican Syriac MS 115, paleographically dated to the seventh or eighth century, is among the oldest, extant collections of Jacob of Serugh’s mêmrê.52 The other manuscript containing the first recension, of uncertain date, is Codex Syriacus Nitriensis 13.53 The second recension is so far known in only a single manuscript of an uncertain later date, Vatican Syriac MS 217.54 Many of the 763 mêmrê which both Jacob of Edessa (c. 640–708) and Bar Hebraeus say Jacob of Serugh composed55 have been transmitted in several recensions. Because of their constant use in the homiletic and catechetical enterprises of the “Jacobite” communities many of these mêmrê effectively became the property of the churchmen who used them and they seem often to have altered them the better to suit their immediate purposes. Most of the alterations, as in the second recension of the mêmrâ here under discussion, are not of major significance; they are for the most part cosmetic, improving the grammar or diction, occasionally correcting a narrative detail. The story recalled in both recensions of the mêmrâ on the “Youths of Ephesus” remains a faithful recollection of the legend as we also have it in the somewhat later sixth century historical texts of Zacharias of Mitylene and John of Ephesus. At the beginning of his mêmrâ on the “Youths of Ephesus,” Jacob of Serugh evokes the liturgical setting of the composition; first he addresses himself in prayer to the Son of God and then he says, “About the children, the sons of the princes of Ephesus, I have a homily (mêmrâ) to declaim before the hearers. Pay attention to me, laborers; sing praise, sons of the bridal chamber.”56 It was probably the feast day of the youths, who were considered to be martyrs; Jacob clearly expects his congregation, perhaps monks, to be familiar with the story. 122 CHRISTIAN LORE AND THE ARABIC QUR’ AN Nevertheless, he will recall the narrative outline for them, to refresh their memories, as if to summon the very presence of the youthful saints and martyrs into the minds of the congregation. Jacob first of all reminds his audience (the “hearers”) that the trials of the youths began in the time of Emperor Decius (r. 249–51), who came to Ephesus and decreed a festival for the gods, Zeus, Apollo, and Artemis,57 requiring that everyone should place incense on their altars. When several handsome noble youths refused, they were summoned to the emperor’s presence, who demanded to know why they would not comply with his order. The text says The son of the prefect and his seven companions (habrawhy) spoke up, “We are not going to bow down to dumb idols, the work of human hands. Ours is the Lord (maryâ) of the heavens; He will help us. To Him we will bow down, and to Him offer the purity of our heart. You have as king, Zeus and Apollo, along with Artemis. We have as king, the Father and the Son, along with the Holy Spirit.” (Guidi, Testi Orientali Inediti, 1, 19, ## 31–9) The emperor deferred dealing with the youths until he would return from a journey to other parts of his domain. In the meantime, having escaped custody, the youths determined to take refuge in a “cave of stone”58 on a nearby mountain top. After they established themselves there and having prayed that the Good Shepherd would protect his sheep, Jacob says The Lord saw the faith of the beloved lambs, and He came to give a good wage for their recompense. He took their spirits and brought them up to heaven, and He left a watcher (‘îrâ) to be the guardian of their limbs. (Guidi, Testi Orientali Inediti, 1, 19–20, ## 58–62) When the emperor returned and learned where the youths had hidden, he planned to kill them there and he gave orders for the cave’s entrance to be walled up. Jacob says that at this moment Two sophists, sons of princes, were present there, and they thought that the Lord was going to raise them up. They made tablets of lead59 and placed them beside them; they wrote the names of the sons of light on them, and the reason why the youngsters went to hide in the cave, and in what time period they had fled from Decius the king. (Guidi, Testi Orientali Inediti, 1, 20, ## 69–73) The time came when “the Lord willed to awaken the sons of light.”60 Long after their entombment, a wealthy man came to build a sheepfold on the mountain top 123 SIDNEY GRIFFITH and he took the well carved stones walling up the entrance to the cave to use in the new structure. When the stones were removed, “the light came in and awakened the sons of light.”61 From this point Jacob somewhat elaborately recalls how the youths took counsel with one another and sent one of their number, Yamlîkâ by name,62 to go into the city to learn the news and to bring back food. Jacob lingers over the adventures of Yamlîkâ in an Ephesus now only barely familiar to him, with signs of the cross prominently displayed. When he attempts to pay for food with coins the youths had taken into the cave with them people are convinced that he had found a buried treasure and they are determined to force him to lead them to it. Eventually, Yamlîkâ is brought to the church, where a sophist recognizes the coin as coming from the reign of the emperor Decius, as the youth himself had claimed. The sophist, who says, “I see you are about twelve years old,” goes on to say The one of whom you speak was a very long time ago. According to the numbering and the reckoning of the Greeks, he was the king three hundred and seventy-two years ago.63 (Guidi, Testi Orientali Inediti, 1, 22, ## 152 and 154–5) Eventually, Yamlîkâ leads the people of Ephesus to his companions in their cave. Jacob says the bishop sends a message to Emperor Theodosius II (408–50), who came to see the great prodigy for himself. Jacob also says that the emperor “took up the tablet of lead and he began to read why it was the youngsters had entered and hidden in the cave.”64 Theodosius tried to persuade the youths to return to city of Ephesus, where “he would build a temple (hayklâ) over their bodies.”65 But they refused, saying they would remain where they were, and they told the emperor For your sake . . . our Lord, the Messiah, awakened us, so you could see and affirm that there is truly a resurrection. (Guidi, Testi Orientali Inediti, 1, 23, # 184) The Qur’an’s relationship to the Syriac narrative When one reads the evocation of the story of the “Companions of the Cave” in the Arabic Qur’an against the background of the accounts of the legend which are available in the pre-Islamic, Syriac sources, and in particular in the mêmrâ of Jacob of Serugh on the “Youths of Ephesus,” a number of interesting coincidences present themselves. But in the very first place it should be mentioned that viewed from the perspective of the Syriac tradition, it seems clear that the Qur’an does not present its own, full recension of the story. Rather, for its own rhetorical purposes and within the context of its own concerns, the Qur’an evokes the memory of the story, which it presumes is common knowledge among 124 CHRISTIAN LORE AND THE ARABIC QUR’ AN its audience or at the very least that the legend was known in some detail to Muhammad, with whom Allah actually speaks about it. Allah recalls the gist of the story and asks questions about its proper interpretation. And a striking feature about the text is that when the Arabic Qur’an’s evocation is read against the background of the Syriac tradition of the “Youths of Ephesus” as it is preserved in the several Syriac textual sources, a remarkable coincidence of word, phrase or narrative detail is sometimes observable. One may best observe this phenomenon by following the Qur’an’s text verse by verse, according to the outline presented earlier. The narrative setting (18:9) The Qur’an provides the immediate setting for its evocation of the legend of the “Companions of the Cave” with Allah’s question to Muhammad, “Do you reckon that the companions (ashab) of the cave and of the inscription (al-raqim) are wondrously among our signs?” (18:9). Reading from the perspective of the Syriac texts, two terms are of immediate interest in this verse. The first of them is the term “companions” to refer to the youths whose story is in the offing. It is a term which frequently appears in the Christian texts in its Syriac equivalent (habrê) to refer to the youths hidden in the cave, and for which the Qur’an’s Arabic term (ashab) may be considered an apt translation. Of more interest however is the often disputed Arabic term al-raqim, which is traditionally, very often translated “inscription,” but about which there has been much speculation regarding its actual meaning both among Muslim commentators and western scholars of the Qur’an.66 The difficulty commentators have had with al-raqim springs both from the rarity of the word in Arabic lexicography, its grammatical form in this verse, and from the perceived awkwardness of its possible meaning in the present context. Consequently, scholars have offered a broad array of alternative meanings for the troublesome word.67 Recently, James A. Bellamy, noting the possibilities for misreading the consonants in the Qur’an’s orthography, has argued that in its present form the text is corrupt and he has proposed reading al-ruqud (“sleeping,” “sleepers”) for al-raqim, a term which actually occurs in 18:18 in reference to the companions. Bellamy then takes the restored phrase to say, “The sleeping companions of the cave.”68 Alternatively, Christoph Luxenberg has argued that on the basis of their failure to recognize a common, underlying, Syro-Aramaic orthography,69 the transmitters of the Arabic Qur’an’s text changed a misunderstood, original al-ruqad (“sleep,” “slumber”) into the puzzling al-raqim, which has yielded the well known array of suggestions regarding its possible significance in the works of the later Muslim commentators. Luxenberg then takes the restored text to be saying, “Die Leute der Höhle und des Schlafes.”70 Reading the verse in question from the perspective of the several, pre-Islamic Syriac accounts of the “Youths of Ephesus,” with the traditional association in Arabic of the root consonants r-q-m with “writing,” and the understanding that 125 SIDNEY GRIFFITH al-raqim could just possibly mean “inscription” or “tablet,”71 one recalls the importance in the narrative of the “lead tablet(s)” which record the names of the youths and give an account of their entombment in the cave. There are two important moments in the narrative in which the “tablet(s)” are mentioned: the moment of the entombment and the moment when the Christian emperor arrives at the cave/tomb to verify the miracle of the resurrection of the youths.72 According to Jacob of Serugh’s mêmrâ, at the moment of the entombment Two sophists, sons of princes, were present there, and they thought that the Lord was going to resurrect them. They made tablets of lead and they set them beside them; they wrote on them the names of the sons of light, and the reason why the youths went into the cave to hide, and at what era they had fled from Decius the king.73 (Guidi, Testi Orientali Inediti, 1, 20, ## 68–73) When Emperor Theodosius II (408–50) arrived at the cave to verify the miracle of the resurrection of the youths, according to Jacob of Serugh He took up the tablet of lead and he began to read why the children had entered into the cave to hide. (Guidi, Testi Orientali Inediti, 1, 22, ## 176–7) With this background in mind, Allah’s question to Muhammad, “Do you reckon that the companions of the cave and of the inscription are wondrously among our signs?” (18:9) makes complete narrative sense. In the Syriac texts, the “Companions” really are portrayed as “Companions of the Cave” and of the “inscription,” they belong in both of them. Furthermore, even if one were to read the passage in a way that separates the youths and the “inscription” (i.e. “the Companions of the Cave and the inscription”), in the Qur’anic understanding, the “writing” (raqm // kitab) on an inscription would just as readily be among God’s “signs” (ayat) as would be the miracle of the youths’ resurrection. What is more, it seems that this reading of the verse yields more consistent intelligibility, on the hypothesis that the Syriac narrative of the legend is in the background, than does either of the suggested textual emendations or any understanding of the meaning of the term al-raqim other than one which bespeaks “writing.” There remains only the perceived awkwardness of the grammatical form of the word in Arabic. In this connection one wonders why an awareness of the Syro-Aramaic background of the Arabic diction in such a context should not suggest that the form could be understood to be a “Syriacism.” That is to say, the likely scenario would be that the form of the Syriac passive participle ( f‘îl),74 used as a substantive adjective ( fa‘îl),75 has been imported into Arabic diction to produce the 126 CHRISTIAN LORE AND THE ARABIC QUR’ AN anomalous al-raqim,76 presumably originally by an Arabic-speaking Christian with a Syriac-speaking background, who was concerned with translating the legend of the “Youths of Ephesus” into Arabic. The Qur’an simply “quoted” this usage, presumably current among Arabic-speaking Christians, along with its evocation of the rest of the legend. In this interpretation, one might plausibly claim that a philological possibility gains probability from a consideration of the historical and cultural background of the narrative. The first narrative phase (18:10–20) Stage one (18:10–12) The details which the Qur’an recalls in these verses have parallels in the Syriac tradition. Here we may note them most readily in the mêmrâ of Jacob of Serugh. The youths (al-fitya // tlayê) took shelter (awa // batw) in the cave (v. 10);77 they prayed for their Lord’s mercy and right guidance (v. 10);78 Allah shut their ears for a number of years (v. 11);79 and finally Allah roused them (v. 12).80 The Qur’an goes on to say that God roused them “to know which of the two parties rightly calculated how long they had stayed” (v. 12). As we shall see later, there is a concern in the Syriac texts to calculate the length of the youths’ stay in the cave, but there is no mention of “two parties” being involved in the reckoning. Stage two (18:13–20) Allah retells (qass) the narrative (naba’) of the “youths who believed in their Lord” (v. 13) recalling familiar Qur’anic themes in familiar Qur’anic language. In these verses the Qur’an makes the legend of the “Youths of Ephesus” its own; one might see in them the “Islamicization” of the current Christian narrative. The historical, geographical and overtly Christian frame of reference, so much a part of the Syriac tradition, is left behind in favor of highlighting the Islamic themes of the refusal of the youths to adopt the pagan practices of their people (vv. 13–15), the miraculous signs of God’s providence in their behalf (vv. 16–17), God’s personal care for the seemingly sleeping youths (v. 18), God’s raising the youths and their dispatch of a messenger into the city (v. 19) out of the fear that if they were discovered they might be forced to return to the religion (milla) of their people (v. 20). The Qur’an’s own rhetorical purposes for evoking the legend of the “Companions of the Cave” are shaped in these verses. There is one detail in particular, mentioned in the context of God’s care for the youths, which has given a distinctive mark to the Islamic rendition of the legend. The Qur’an says that while God would turn the “sleeping” youths over to the right and the left, “their dog was stretching its paws on the threshold” (v. 18). There is no mention of the dog in the pre-Islamic, Syriac tradition of the “Youths of Ephesus.” But there is mention in Jacob of Serugh’s mêmrâ of a guardian for their “sleeping” members. The text of recension I says that having taken their spirits to 127 SIDNEY GRIFFITH heaven, the Lord “left a watcher (‘îrâ) to be the guardian of their limbs.”81 Normally in the milieu of Aramaic one would think of such a “watcher” as an angel,82 an understanding which is made explicit in recension II of the mêmrâ where the text says the Lord “dispatched a ‘watcher’ to go down to guard their limbs.”83 It may not be too far a leap to suppose that it was this “watcher” of the Syriac tradition which became the watch-dog of the Arabic Qur’an,84 albeit that such an interpretation, with at least one notable exception,85 is never to be found in the Islamic commentary tradition.86 Neither watch-dogs nor guard-dogs seem to have had a place in the entry under “dog” (kalb) in the Arabic lexicon of the Bedouin Arabs; the nomads prized hunting dogs.87 But watch-dogs do appear in the Syriac lexicon of the Arameans, among whom both shepherds and shepherding imagery have a high literary profile.88 In fact, in Jacob of Serugh’s mêmrâ on the “Youths of Ephesus,” the mention of the “watcher” who served as their guard occurs in just such a context, featuring shepherd imagery, in the youths’ prayer to their Lord. The text says They went up the mountain; they entered the cave and stayed there. They called out to the Lord in a doleful voice and spoke thus, “We beseech you, Good Shepherd, who has chosen His servants, guard your flock from this wolf who thirsts for blood.” The Lord saw the faith of the blessed lambs, and He came to give a good wage for their recompense. He took their spirits and brought them up to heaven, and He left a watcher to be the guardian of their limbs. (Guidi, Testi Orientali Inediti, 1, 19–20, ## 55–6189) In spite of its allure, it is probably a temptation to be avoided to suppose that the watch-dog in the Arabic Qur’an is a conceptual “Syriacism.” The step down from guardian angel to watch-dog is no doubt too steep a step to imagine the Arabic-speaking Christians in the Qur’an’s audience readily to have taken in translating the legend. As for the Qur’an’s own imagery, in spite of the later Islamic interpretive tradition, which obviously struggles with the appearance of the dog in the narrative, the watch dog with its paws spread on the cave’s threshold remains right on the mark from the perspective of the pastoral metaphors evoked in the Syriac tradition. The second narrative phase (18:21–6) Allah tells Muhammad that while the purpose of the narrative of the “Companions of the Cave” was to make known to people that “God’s promise is true and there is no doubt about the hour” (v. 21) people have nevertheless been caught up in disagreements about them. The disagreements began already at the 128 CHRISTIAN LORE AND THE ARABIC QUR’ AN time of the discovery of the risen youths, who were soon to sleep again. According to the Qur’an some said, “Build a building over them,” but “those who prevailed said ‘we shall build a place of prayer (masjid) over them’ ” (v. 21). The Syriac tradition too preserves a recollection of differing opinions about what to do with the bodies of the companions once they went to sleep for good after the discovery of their momentary reawakening. At first the emperor Theodosius proposed to bring them back into Ephesus, where he would build a “sanctuary” (haykla) over them. But the youths requested to be left in their cave and the emperor acceded to their request.90 It is notable that in the Ecclesiastical History of Zacharias of Mitylene the narrative says of the emperor’s action, “He left them there to this very day; meanwhile a great sanctuary has been built over the cave for honor’s sake, and for a house of prayer (bayt slutâ), and for liturgy (teshmeshtâ) over their bodies.”91 One notes in passing the specification of the sanctuary as “a house of prayer,” a detail that may be echoed in the Qur’an’s mention of the masjid which “those who prevailed” would build over the bodies of the companions. Allah has specific instructions for Muhammad about other aspects of the disagreements surrounding the story of the “Companions of the Cave.” He tells the prophet not to get into arguments with people about the number of the companions (v. 22). In the Syriac tradition there is in fact some disagreement about the number. According to the mêmrâ of Jacob of Serugh and the Ecclesiastical History of John of Ephesus, the number is consistently eight, the youths’ leader and spokesman/emissary (Yamlîkâ) and his “seven companions,” while the Ecclesiastical History of Zacharias of Mitylene speaks of “their leader Akleides and his six companions.”92 In other early Christian language traditions the youths are usually called the “Seven Sleepers of Ephesus.” Noting a concern about the number of years the companions stayed in the cave, “three hundred years, nine to be added” (v. 25), Allah tells Muhammad to say, “Allah knows best how long they stayed” (v. 26). In the pre-Islamic Syriac texts there is in fact disagreement about the number of years the youths stayed asleep in the cave. For the most part the differences seem to come from the methods of computing the number of years which elapsed between the reigns of the emperors Decius (249–51) and Theodosius II (408–50). Recension I of Jacob of Serugh’s mêmrâ says of Decius, “According to the numbering and the reckoning of the Greeks, he was the king three hundred and seventy-two years ago,”93 but recension II says, “According to the numbers and the reckoning of the Greeks, Decius passed on three hundred and fifty years ago.”94 The Ecclesiastical History of John of Ephesus says that the coinage of Decius in the youths’ possession was current three hundred and seventy years ago;95 in the Ecclesiastical History of Zacharias of Mitylene, the bishop of Ephesus tells the youth Dionysius that Decius reigned “two hundred years ago, more or less.”96 With this reckoning, the bishop would seem to be “more or less” correct; the number of years which elapsed between the end of the reign of Decius (d. 251) and the last year of the reign of Theodosius (d. 450), when the youths were discovered, is roughly 199. 129 SIDNEY GRIFFITH So the Qur’an’s advice about quarrelling over the number of years the youths spent in the cave seems to be well taken. Conclusion Reading the Qur’an’s evocation of the legend of the “Companions of the Cave” against the background of the fuller narrative as we have it in the extant, pre-Islamic, Syriac tradition, and particularly in the mêmrâ of Jacob of Serugh, enables the scholar of Syriac to recognize the fidelity of the Qur’an’s reprise of a piece of Christian lore as it must have circulated orally among the Arabicspeaking, “Jacobite” Christians of Muhammad’s day in Arabia. One notices not only the Qur’an’s familiarity with details of the story and the different understandings of them, but also the way in which the Qur’an on the one hand removes the Christian frame of reference and on the other hand provides an Islamic, Qur’anic horizon within which the legend takes on a whole new hermeneutical significance. While it is beyond the range of this study to pursue the matter, the review of the Syriac texts discussed here, together with a rapid survey of the Islamic exegetical tradition, provides abundant evidence that early Muslim commentators on the Qur’an were also much indebted to these same Syriac sources for many of the details they included in their commentaries on the story of the “Companions of the Cave.” The debt is already evident in what may be one of the earliest of them, the Kitab al-mubtada’ of Muhammad Ibn Ishaq (d. c. 767).97 The story line of the “Companions of the Cave” as it appears in this prologue to the biography of Muhammad as it has been recovered from the works of later writers, largely from al-Tabari’s History and Commentary,98 owes an obvious debt to the Syriac account of the “Youths of Ephesus” as it appears in the Ecclesiastical History of Zacharias of Mitylene. This recognition highlights the need for serious studies in the future on the role of Syriac sources in early Islamic scholarship, how they came to the attention of the Arabic-speaking Muslims, and how Syriac themes, duly Islamicized, came to furnish important ingredients in classical Islamic literary culture. In the “Sectarian Milieu” in which the early Islamic accounts of “salvation history” were composed,99 contemporary Christian writers in Arabic were also alive to the need to include important moments in Islamic religious history in their own apologetically inspired accounts of events. This was especially the case when Islamic history offered an “Islamicized” version of an earlier Christian narrative. The story of the “Youths of Ephesus” who became the “Companions of the Cave” is a case in point. Already in the earliest recension of Eutychius of Alexandria’s (877–940) Annals,100 the author is concerned to replace the narrative into its Christian context; he distributes the relevant portions of the legend between his accounts of the reigns of the emperors Decius and Theodosius II. And then at the end, as if to correct the prevailing Islamic reckoning, that is, the Qur’an’s mention of 309 years, he writes, “From the time when the youths fled 130 CHRISTIAN LORE AND THE ARABIC QUR’ AN from King Decius and went to sleep in the cave, to the time in which they appeared again and died, one hundred and forty-nine years elapsed.”101 It would seem that much Christian lore in Syriac lies behind the Qur’an’s evocation of the Christian scriptures, the beliefs and practices of the churches, and their homiletic traditions, as they must have circulated among many Arabicspeaking Christians in the Qur’an’s original audience in the time of Muhammad. This study is meant to suggest and to illustrate, in a particular instance, some hermeneutical assumptions, grounded in a plausible historical and cultural scenario, which may usefully be applied to the discernment of “Syriacisms,” both lexical and thematic, in the Qur’an’s discourse. The Syro-Aramaic tradition is not the only source of Christian discourse present in the milieu of the Arabic Qur’an, but it is arguably the most important and most pervasive one. It is hoped that the hermeneutical assumptions applied in this one study might suggest ways to avoid some of the extremes of both the philological and the thematic methods of past inquiries into the presumed foreign influences on the Qur’an. One result of the application of these hermeneutical considerations to the study of the narrative of the “Companions of the Cave” which is striking is that it leaves the canonical Qur’an’s Arabic diction intact at the same time as it suggests ways to explain how certain grammatical or lexical anomalies in the text may have come about. What is more, this approach offers no threat to the Islamic exegesis of the Qur’an. Rather, it enhances our knowledge of the social, cultural and religious complexity of the Arabic-speaking audience addressed by the Arabic Qur’an and in the process it discloses the Qur’an’s own detailed awareness of the folklore of that audience’s Christian members, whose patristic and liturgical heritage was distinctly Syriac. Notes 1 See the review, in a broader context, in M.R. Zammit, A Comparative Lexical Study of Qur’anic Arabic, Leiden: Brill, 2002, esp. 51–63. 2 A. Mingana, “Syriac influence on the style of the Üur’an,” Bulletin of the John Rylands Library Manchester 11, 1927, 77–98. 3 FV, 20–1. 4 Ibid., 22. 5 See C. Luxenberg, Die syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der Koransprache, Berlin: Das Arabische Buch, 2000; 2nd rev. edn; Berlin: Schiler, 2004. 6 See S.H. Griffith, “The Qur’an in Arab Christian Texts: The Development of an Apologetical Argument; Abu Qurrah in the Maælis of al-Ma’mun,” Parole de l’Orient 24, 1999, 203–33. 7 Andrae first published the results of his researches in a series of three long articles: T. Andrae, “Der Ursprung des Islams und das Christentum,” Kyrkohistorisk Årsskrift 23, 1923, 149–206; 24, 1924, 213–92; 25, 1925, 45–112. Subsequently the articles were collected into the volume, Tor Andrae, Der Ursprung des Islams und das Christentum, Uppsala: Almqvist and Wiksells, 1926, which was then translated into French as T. Andrae, Les origines de l’Islam et le christianisme, trans. J. Roche, Paris: Adrien-Maisonneuve, 1955. In later works Andrae continued to appeal to Syriac 131 SIDNEY GRIFFITH 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 sources, most notably in T. Andrae, Mohammed, sein Leben und sein Glaube, Göttingen: Vandenhoeck and Ruprecht, 1932; English trans.: Mohammed: The Man and His Faith, trans. T. Menzel, New York: Charles Scribner’s Sons, 1936 and I Myrtenträdgården: Studier I Tidig Islamisk Mystik, Lund: Albert Bonniers Forlag, 1947; English trans.: In the Garden of Myrtles: Studies in Early Islamic Mysticism, Albany, NY: State University of New York Press, 1987. Andrae, Les origines de l’Islam, 68. Ibid. Ibid., 145, 160, 203. Ibid., 146. Ibid., 192, 199, 202. Ibid., 206, with a reference back to p. 24. Presumably a reference to al-Ma’idah (5):116. Presumably a reference to the Qur’an’s dictum, “They have disbelieved who say God is the Messiah, son of Mary,” al-Ma’idah (5):72. See Andrae, Les origines de l’Islam, 209–10. See also Andrae, Mohammed: The Man and His Faith, 91. See J. Bowman, “The debt of Islam to monophysite Syrian Christianity,” Nederlands Theologisch Tijdschrift 19, 1964/65, 177–201; also published in E.C.B. Mac Laurin, Essays in Honour of Griffithes Wheeler Thatcher (1863–1950), Sydney: Sydney University Press, 1967, 191–216. See Andrae, Les origines de l’Islam, esp. 145–61. Andrae used the old, uncritical Editio Romana of the works of Ephraem. See J.S. Assemani (ed.), Sancti Patris Nostri Ephraem Syri Opera Omnia quae exstant Graece, Syriace, Latine, 6 vols; Romae, 1732–46. He was apparently unaware of the problems of authenticity between the works of the so-called Ephraem Syrus and Ephraem Graecus; he referred to them indiscriminately. Ibid., 151–4, where, on p. 151, n. 4, Andrae attributes the original insight to Hubert Grimme, Muhammad, 3 vols: Münster im W.: Aschendorff, 1892–5, 2, 160, n. 9. This is Andrae’s version of a strophe from St Ephraem’s Syriac Hymni de Paradiso, 7:18, based on the text in the Editio Romana of his works, Assemani, Opera Omnia, 2, 563ff. See E. Beck, “Eine christliche Parallele zu den Paradiesesjungfrauen des Korans?” Orientalia Christiana Periodica 14, 1948, 398–405. Beck returned to the issue with some further observations in a later article, Edmund Beck, “Les Houris du Coran et Ephrem le Syrien,” MIDEO 6, 1959–61, 405–8. For Beck’s own understanding of Ephraem’s hymn, see Edmund Beck, Ephraems Hymnen über das Paradies, Studia Anselmiana, 26; Rome: Herder, 1951, 63–76, and E. Beck, Des heiligen Ephraem des Syrers Hymnen de Paradiso und Contra Julianum, CSCO 174 and 175; Louvain: Peeters, 1957. Beck’s version of the stanza quoted by Andrae is as follows: “Wer des Weines * in Klugheit sich enthielt, – dem eilen freudiger * die Weinstöcke des Paradieses entgegen – und jeder wird seine Trauben * ihm darreichen. – Lebter er auch noch jungfräulich, * dann führen sie ihn ein – in ihren reinen Schoss, * weil er als Asket – nicht gefallen ist in den Schoss * und in das Bett der Ehe.” Beck, Des heiligen Ephraem des Syrers Hymnen de Paradiso, 175, 7:18, 28. M.A.S. Abdel Haleem, The Qur’an: A New Translation, Oxford: Oxford University Press, 2004, 322–3, 345. Luxenberg, Die Syroaramäische Lesart, 226: “Wir werden es ihnen unter weissen, kristall(klaren) (Weintrauben) behaglich machen.” The Qur’an itself alludes to the “foreign” (a‘jami) speech of one whom some of Muhammad’s adversaries claim was his teacher (al-Nahl [16]:103) and it also defends the idiom of revelation be it “foreign” or Arabic (Fussilat [41]:44). 132 CHRISTIAN LORE AND THE ARABIC QUR’ AN 25 It is unlikely that this pre-Qur’anic, Christian Arabic would have been available in writing, although it is not impossible that a few literate individuals may have had notes or some other rudimentary texts for personal use. See Gregor Schoeler, Écrire et transmettre dans les débuts de l’islam, Paris: Presses Universitaires de France, 2002, esp. 26–9. 26 See S.H. Griffith, “ ‘Syriacisms’ in the Arabic Qur’an: Who were ‘Those who said that Allah is third of three, according to al-Ma’idah 73?’ ” in Meir M. Bar-Asher, Simon Hopkins, Sarah Stroumsa and Bruno Chiesa (eds.), A Word Fitly Spoken: Studies in Mediaeval Exegesis of the Hebrew Bible and the Qur’an;Presented to Haggai Ben-Shammai Jerusalem: The Ben-Zvi Institute, 2007, 83ff . 27 See especially L. Massignon, “Les ‘sept dormants’; apocalypse de l’islam,” Analecta Bollandiana 68, 1950, 245–60; idem, “Les sept dormants d’Ephèse (ahl al-kahf ) en islam et chrétienté,” Revue des Études Islamiques 12, 1954, 61–110; idem, Le culte liturgique et populaire des VII dormants, martyrs d’Ephèse (ahl al-kahf): trait d’union orient-occident entre l’islam et la chrétienté, Studia Missionalia, Rome: The Gregorian University, 1961, 62. The first and third of these items are republished in Louis Massignon, Opera Minora, ed. Y. Moubarac, 3 vols; Beirut: Dar al-Maaref, 1963, 3, 104–18 and 119–80. See also H. Kandler, Die Bedeutung der Siebenschläfer (Ashab al-kahf ) im Islam: Untersuchungen zu Legenden und Kult in Schrifttum, Religion und Voksglauben unter besonderer Berücksichtigung der SiebenschläferWallfahrt, Bochum: Universitätsverlag Dr N. Brockmeyer, 1994. 28 See F. Jourdan, La tradition des Sept Dormants: Une rencontre entre chrétiens et musulmans, Paris: Maisonneuve et Larose, 2001. 29 Regarding the translation of the term al-raqim as “inscription” see the discussion later. 30 See Theodor Nöldeke and Friedrich Schwally, GdQ1, 140–3, where sura 18 is assigned to the second Meccan period. 31 See, for example, Abdel Haleem, The Qur’an, 183, n. a. 32 See, for example, N. Robinson, Discovering the Qur’an: A Contemporary Approach to a Veiled Text, London: SCM, 1996, 54. 33 See, for example, J. Henninger, “Spuren christlichen Glaubenswahrheiten im Koran,” Neue Zeitschrift für Missionswissenschaft / Nouvelle Revue de Science Missionaire 1, 1945, 135–40, 304–14; 2, 1946, 56–65, 109–22, 289–304; 3, 1947, 128–40, 290–301; 4, 1948, 129–41, 284–93; 5, 1949, 127–40, 290–300; 6, 1950, 207–17, 284–97. See also C.H. Becker, Christianity and Islam, trans. H.J. Chaytor, New York and London: Harper, 1909; R. Bell, The Origin of Islam in its Christian Environment: The Gunning Lectures, Edinburgh University, 1925, London: Macmillan, 1926. 34 Some modern commentators have spoken of two versions of the story in these verses. See, for example, R. Bell, A Commentary on the Qur’an, ed. C.E. Bosworth and M.E.J. Richardson, Manchester: University of Manchester, 1991, 1, 483; R. Blachère, Le Coran; traduction nouvelle, Paris: Maisonneuve, 1947–50, 2, 328. 35 See the masterful survey of M. Huber, Die Wanderlegende von den Siebenschläfern: Eine literargeschichtliche Untersuchung, Leipzig: Harrassowitz, 1910. See also Massignon, “Les Sept Dormants . . . en islam et en chrétienté,” and “Le culte liturgique et populaire des VII Dormants.” It is interesting to note that because of its Christian/ Muslim, cross-cultural potential, the legend also found an important place in Goethe’s West-östlicher Divan; see David Bell, “Goethe’s ‘Siebenschläfer’ and the Heroes of the West-östlicher Divan,” Bulletin of the John Rylands University Library of Manchester 84, 2002, 67–84. 36 See esp. E. Honigmann, “Stephen of Ephesus (April 15, 448-October 29, 451) and the legend of the Seven Sleepers,” in E. Honigmann, Patristic Studies, Studi e Testi, 173; Vatican City: Biblioteca Apostolica Vaticana, 1953, 17, 125–68. See also Jourdan, La tradition des Sept Dormants, 40–50. 133 SIDNEY GRIFFITH 37 See I. Guidi, “Seven Sleepers,” in J. Hastings (ed.), Encyclopaedia of Religion and Ethics, New York: Charles Scribner’s Sons, 11, 1921, 428–30; Arthur Allgeier, “Der Ursprung des griechischen Siebenschläferlegende,” Byzantinische-neugriechische Jahrbücher 3, 1922, 311–31; “Seven Sleepers of Ephesus,” in F.L Cross and E.A. Livingstone (eds), The Oxford Dictionary of the Christian Church, 3rd edn; Oxford: Oxford University Press, 1997, 1489–90. The account of the legend in the Syriac version/epitome of the Ecclesiastical History of Zacharias of Mytilene mentions as an occasion of the miracle, controversies over the fate of the human body after death sparked by works of Origen. See E.W. Brooks (ed.), Historia Ecclesiastica Zachariae Rhetori vulgo Adscripta II, CSCO 84, Paris: E Typographeo Reipublicae, 1921, 114–15. See n. 47 later. 38 See A. Allgeier, “Untersuchungen zur syrischen Überlieferung der Siebenschläferlegende,” OC 4, 1915, 279–97; 5, 1915, 10–59; idem, Die westsyrische Überlieferung der Siebenschläferlegende, Leipzig: W. Drugulin, 1915; idem, “Die älteste Gestalt der Siebenschläferlegende,” OC 6, 1916, 1–43; 7–8, 1918, 33–87. 39 The texts are published in I. Guidi, Testi Orientali Inediti sopra I Sette Dormienti di Efeso, Reale Accademia dei Lincei, 282, 1884–5; Roma: Tipografia della R. Accademia dei Lincei, 1885, 18–29. One of the recensions is also published in H. Gismondi, Linguae Syriacae Grammatica et Chrestomathia cum Glossario, 4th edn; Rome: C. De Luigi, 1913, 45–53. 40 See T. Kollamparampil (intro. and trans.), Jacob of Serugh: Select Festal Homilies, Rome: Centre for Indian and Inter-religious Studies/Bangalore: Dharmaram Publications, 1997; idem, Salvation in Christ according to Jacob of Serugh: An Exegetico-theological Study on the Homilies of Jacob of Serugh (451–521 AD) on the Feasts of Our Lord, Bangalore: Dharmaram Publications, 2001; Tanios Bou Mansour, La théologie de Jacques de Saroug, Bibliothèque de l’Université Saint-Esprit, 16 and 40; Kaslik, Lebanon: L’Université Saint-Esprit, 1993 and 2000. 41 See D.S. Wallace-Hadrill, Christian Antioch: A Study of Early Christian Thought in the East, Cambridge: Cambridge University Press, 1982. 42 See J. Spencer Trimingham, Christianity among the Arabs in Pre-Islamic Times, London and New York: Longman, 1979; I. Shahid, The Martyrs of Najrân: New Documents. Subsidia Hagiographica 49. Bruxelles: Société des Bollandistes, 1971; idem, Byzantium and the Arabs in the Sixth Century, Washington, DC: Dumbarton Oaks, 1995–2002; R. Tardy, Najrân: Chrétiens d’Arabie avant l’islam, Beyrouth: Dar al-Mashriq, 1999. See also E.K. Fowden, The Barbarian Plain: Saint Sergius between Rome and Iran, Berkeley: University of California Press, 1999, esp. 25–6, where a mêmrâ of Jacob of Serugh is quoted in celebration of St Sergius’ role in bringing the faith to the Arabs at Rusafa. 43 R. Schröter, “Trostsschreiben Jacob von Saruj an die Himyaritischen Christen,” ZDMG 31, 1877, 360–405. 44 See S.P. Brock, “The ‘Nestorian’ Church: A Lamentable Misnomer,” Bulletin of the John Rylands University Library of Manchester 78, 1996, 23–35. 45 See A. Mingana, Narsai Doctoris Syri Homiliae et Carmina, Mosul: Fratrum Praedicatorum, 1905. 46 The account of the “Youths of Ephesus” from the Ecclesiastical History of Zacharias is published in CSCO 84, Paris: E Typographeo Reipublicae, 106–22 and in J. P. N. Land, Anecdota Syriaca, 3, Zachariae Episcopi Mitylenes aliorumque Scripta Historica Graece plerumque Deperdita; Leiden: Brill, 1870, 87–99. 47 See J.J. van Ginkel, John of Ephesus: A Monophysite Historian in Sixth-Century Byzantium, PhD dissertation; Groningen: Rijksuniversiteit Groningen, 1995. 48 See W. Witakowski, The Syriac Chronicle of Pseudo-Dionysius of Tel-Ma hre: A Study in the History of Historiography, Studia Semitica Upsaliensia 9, Stockholm: Almqvist and Wiksell, 1987. The account of the “Youths of Ephesus” from the 134 CHRISTIAN LORE AND THE ARABIC QUR’ AN 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 Ecclesiastical History of John of Ephesus is printed in Guidi, Testi Orientali Inediti, 35–44 and in P. Bedjan, Acta Martyrum et Sanctorum Syriace I, Paris: Otto Harrassowitz, 1890, 301–25. In this connection, see n. 38 earlier. In this connection one recalls the remarks of Ignazio Guidi about the Qur’an’s account of the “Companions of the Cave”: “Its source is undoubtedly oral tradition; it goes back probably to Christian monks, and from them directly or indirectly the story reached Muhammad . . . . A certain resemblance to the homily of James of Serugh, which must have been very well known and often repeated among the Syrian monks, confirms this supposition.” I. Guidi, “Seven Sleepers,” 429. See Jourdan, La tradition des sept dormants, 57. In addition to the MSS indicated by Jourdan, there is at least one other MS which includes a mêmrâ on the “Seven Sleepers” attributed to Jacob of Serugh; it is MS Mardin Orth. 139; which recension it represents remains unknown. See A. Vööbus, Handschriftliche Überlieferung der Memre-Dichtung des Ja‘qob von Serug, CSCO 344 and 345, 421 and 422, Louvain: Secrétariat du CSCO 1973 and 1980, 344, 71–2. There is also an anonymous mêmrâ (n. 203) on the “Seven Sleepers” in MS 156 at the monastery of St Mark in Jerusalem. See Vööbus, Handschriftliche Überlieferung der Memre-Dichtung des Ja‘qob von Serug, 344, 149 and n. 19. See Vööbus, Handschriftliche Überlieferung der Memre-Dichtung des Ja‘qob von Serug, 344, 51–2. See the rather full description of Jacob of Serugh’s mêmrâ on the “Youths of Ephesus” as contained in this MS in J.S. Assemani, Bibliotheca Orientalis Clementino-Vaticana I; Rome: Typis Sacrae Congregationis de Propaganda Fide, 1719, no. 221, 335–8. See A. Mai, Scriptorum Veterum Nova Collectio e Vaticanis Codicibus Edita, vol. 4; Romae: Typis Vaticanis, 1831, 410–11; S.E. Assemanus and J.S. Assemanus (eds), Bibliothecae Apostolicae Vaticanae Codicum Manuscriptorum Catalogus in Tres Partes Distributus, 3; Paris: Maisonneuve Frères, 1926, no. 10, 503–5. See Kollamparampil, Salvation in Christ, 33. Guidi, Testi Orientali Inediti, 2, 18; different wording in 2, 23. Artemis was the patron deity of Ephesus, whose festival is the setting in which St Paul found himself on his visit to the city as described in Acts 19:28–35. The memory of this passage was doubtless in Jacob of Serugh’s mind. Guidi, Testi Orientali Inediti, 1, 19, # 45. Elsewhere in recension I and in recension II the text speaks in the singular of “a lead tablet (lu hâ d’abarâ);” only here is the plural used. Guidi, Testi Orientali Inediti, 1, 20, # 75. Ibid., 1, 20, # 80. Ibid., 1, 20, # 85. This is the only one of the companions whose name Jacob mentions. At a later point in the narrative, Yamlîkâ speaks of “I and my seven companions.” 2, 22, # 158. In recension II the sophist estimates Yamlîkâ to be about fifteen years old and he says, “Decius passed away three hundred and fifty years ago.” 2, 28, ## 175 and 179. Ibid., 2, 22–3, ## 175–6. According to recension II, the bishop was the first one to take up the lead tablet and to read the names and exploits of the youths; Theodosius then follows suit. 2, 28, ## 195–6, 205. Ibid., 2, 23, # 179. According to recension II, Theodosius “took the mantle in which he was clothed and covered them; he left them and they lay down to sleep, the sleep of [eternal] rest.” 2, 29, ##, 105–6. See R. Tottoli, “Raqim,” in J.D. McAuliffe, EQ, 4, 351–2. See the brief but comprehensive review of understandings in Tottoli, “Raqim,” and in James A. Bellamy, “Al-Raqim or al-Ruqud? A note on Surah 18:9,” JAOS 111, 1991, 115–17. 135 SIDNEY GRIFFITH 68 Bellam y, “Al-Raqim or al-Ruqud?” 116. 69 See Luxenberg, Die syroaramäische Lesart, 66–7, where it is explained that in addition to mistaking dal for mim, as Bellamy had argued, a copyist must also have failed to understand the Syro-Aramaic orthographic possibility for ya’ to indicate a, a recognition which for Luxenberg removes the difficulty of alleging two mistaken characters in the same word, as Bellamy’s suggestion involved. 70 Ibid., 67. 71 See E.W. Lane, Arabic-English Lexicon, 2 vols; Cambridge: The Islamic Texts Society, 1984, 1, 1138–40. 72 Both moments involving the lead tablet(s) appear in all three pre-Islamic, Syriac texts of the legend. In the mêmrâ of Jacob of Serugh the text of recension 1 speaks of “tablets” only in regard to the first moment; in the second it speaks of a single “tablet”; the text of recension II speaks of a single tablet in both instances. In the histories of Zacharias of Mitylene and John of Ephesus the texts speak of “tablets” encased in caskets of brass, in both instances. 73 In recension II (p. 25, ## 80–1) there is only one lead tablet, as in the other passage to be quoted from recension I. 74 See T. Nöldeke, Compendious Syriac Grammar, trans. J. A. Crichton; Winona Lake, IN: Eisenbraun’s, 2001, 218, # 278. 75 Ibid., 70, # 110. 76 It is interesting to note the collocation of al-raqim with marqum in Lane, ArabicEnglish Lexicon, 1, 1139. 77 Guidi, Testi Orientali Inediti, 1, 19, # 54. 78 Ibid., 1, 19, ## 55–6. 79 Ibid., 1, 19, # 60, “[God] took their spirits and brought them up to heaven.” 80 Ibid., 1, 20, # 75, “The Lord willed to awaken the sons of light.” 81 Ibid., 1, 19–20, # 61. 82 This understanding is already evident in the Aramaic portions of the Bible; see, for example, Daniel 4:13, 17:23. As for Syriac, see W. Cramer, Die Engelvorstellung bei Ephräm dem Syrer, Orientalia Christiana Analecta 173, Rome: 1965; R. Murray, “The origin of Aramaic ‘îr, angel,” Orientalia 53, 1984, 303–17; idem, “Some themes and problems of early Syriac angelology,” V Symposium Syriacum 1988: Katholieke Universiteit, Leuven, 29–31 août 1988, Orientalia Christiana Analecta, 236, Rome: Pont. Institutum Studiorum Orientalium, 1990, 143–53. 83 Guidi, Testi Orientali Inediti, 2, 25, # 68. 84 The suggestion was made already in Huber, Die Wanderlegende von den Siebenschläfern: Eine literargeschichtliche Untersuchung, 240–2. 85 The Muslim historian, al-Ya‘qubi (d. 292/897), who recorded the story of the “Companions of the Cave” in his History, says that they were accompanied by a shepherd and the shepherd’s dog, whose name he gives as Qatmir. See Ibn Wadhih Qui Dicitur al-Ja‘qubi Historiae, ed. M.Th. Houtsma, Leiden: Brill, 1969, 1, 173. 86 It is interesting to note in passing the range of interpretations for the dog in the story of the “Companions of the Cave” in Islamic tradition. See B. Fudge, “Dog,” in McAuliffe, EQ, 1, 545–6. See also Kandler, Die Bedeutung der Siebenschläfer (Ashab al-kahf) im Islam, Exkurs II: Der Hund, 56–8. 87 See Lane, Arabic-English Lexicon, 2, 2624–7. 88 See R. Payne Smith, Thesaurus Syriacus, 2 vols.; Oxford: Clarendon Press, 1879–1901; reproduced in Hildesheim: Olms, 1981, 1, cols. 1741–4. 89 See also 2, 25, ## 60–7. 90 Ibid., 1, 23 ## 180–1. According to mêmrâ 2, Theodosius, “took the cloak that covered him and he covered them; he left them and they lay down to sleep the sleep of [eternal] rest” (p. 29, ## 210–1). In addition to the royal cloak, Zacharias of Mitylene and John of Ephesus both speak of “sheets/planks of gold” (qupsê d-dahbâ) 136 CHRISTIAN LORE AND THE ARABIC QUR’ AN 91 92 93 94 95 96 97 98 99 100 101 which the emperor ordered to be put underneath the bodies; there is no mention of writing on them. See Brooks, Historia Ecclesiastica Zachariae Rhetori vulgo Adscripta II, 121; Bedjan, Acta Martyrum et Sanctorum Syriace I, 1, 324. Brooks, Historia Ecclesiastica Zachariae Rhetori vulgo Adscripta II, 121–2. Ibid., 110. While Akleides, the son of the prefect of Ephesus, is the youths’ leader (rîshâ), their chosen chargé d’affaires (sa‘urâ), according to this source, was “the wise and swift youth” Dionysius (p. 111). Jacob of Serugh names only Yamlîkâ; both Zacharias of Mitylene and John of Ephesus name all of the companions, but the names are different on the two lists. Guidi, Testi Orientali Inediti, 1, 22, ## 154–5. The “reckoning of the Greeks” refers to the practice among the “Jacobite” Syrian Orthodox to compute dates from the beginning of the Seleucid era in Syrian history, that is, from the beginning of the reign in Antioch of Seleucus I Nicator (312–281 BC). Ibid., 2, 28, ## 179–80. Bedjan, Acta Martyrum et Sanctorum Syriace I, 1, 320. Brooks, Historia Ecclesiastica Zachariae Rhetori vulgo Adscripta II, CSCO 84, 120. The earlier remarks of Wahb ibn Munabbih (d. c. 110/728) show no traces of an awareness of the Syriac tradition. See Kandler, Die Bedeutung der Siebenschläfer (Ashab al-kahf) im Islam, 62. See G.D. Newby, The Making of the Last Prophet: A Reconstruction of the Earliest Biography of Muhammad, Columbia, SC: University of South Carolina Press, 1989, 212–23. See J. Wansbrough, The Sectarian Milieu: Content and Composition of Islamic Salvation History, London Oriental Series, 34; Oxford: Oxford University Press, 1978. See S.H. Griffith, “Apologetics and Historiography in the Annals of Eutychios of Alexandria: Christian Self-Definition in the World of Islam,” in R. Ebied and H. Teule (eds), Studies on the Christian Arabic Heritage in Honour of Father Prof. Dr. Samir Khalil Samir, S.I. at the Occasion of his Sixty-Fifth Birthday, Eastern Christian Studies 5, Leuven: Peeters, 2004, 65–89. M. Breydy, Das Annalenwerk des Eutychios von Alexandrien: ausgewählte Geschichten und Legenden kompiliert von Sa‘id ibn Batriq um 935 AD, CSCO 471, Louvain: Peeters, 1985, 88. In fact, the number of years was roughly 199. See n. 102 earlier. 137 Part 2 THE RELIGIOUS CONTEXT OF THE LATE ANTIQUE NEAR EAST 6 THE THEOLOGICAL CHRISTIAN INFLUENCE ON THE QUR’AN A reflection Samir Khalil Samir The title that was proposed to me is: “The Theological Christian Influence on the Qur’an: A Reflection.” Yet can one speak about Christian influences on the Qur’an? If so, to what extent? Can one specify exactly the magnitude and origin of these influences? The problem To begin with we must note that the very concept of influence is generally rejected by all of traditional Islam. The Qur’an cannot be subject to influences, since it comes directly from God and is in no way a human work. If it were a work attributable to Muhammad himself, it could be subject to influences. However, being a divine message brought down upon Muhammad, there is no other influence but that of God. By this fact alone the very question that we raise is already excluded by traditional Islamic thought. Yet if we consider the Qur’an as a literary document, then one has the right to pose the question in order to know if, and in what measure, there could have been influences from earlier literary documents. In addition, given the number of Biblical allusions that one finds in the Qur’an, and in light of what is known of the Christian milieu of Mecca and the Jewish milieu of Medina, it is normal to inquire into the Biblical (Jewish and Christian) influences that could have impacted the Qur’an. I note first of all that there are a certain number of suras that speak of Christ, the Virgin Mary mother of Jesus, John the Baptist, Zechariah, the apostles, and so on. Evidence would show, even without a long study, that there are indeed parallels of these pericopes in the Qur’anic text on one hand and the Gospel texts – canonical or apocryphal – on the other hand. It is evidently in these verses where one would eventually find the most Christian influences. However here I will leave aside these “historical” descriptions where 141 SAMIR KHALIL SAMIR the parallels are evident and well known (those of the infancy of Mary, the annunciation of John the Baptist, the annunciation to Mary, the infancy miracles of Jesus, and so on.), in order to examine other passages where the Christian parallels are not evident. God chose Adam, Noah, Abraham and the family of Imran (Q 3:33–4) I begin with verses 33–4 of sura 3 (The Family of ‘Imran) inna llaha stafa ’Adama wa-Nuhan wa-’ala Ibrahima wa-’ala ‘Imrana ‘ala l-‘alamina, dhurriyyatan ba‘duha min ba‘din. wa-llahu sami‘un ‘alim. God did choose Adam and Noah, the family of Abraham, and the family of ‘Imran above all people, Offspring, one of the other: And God heareth and knoweth all things (tr. Yusuf Ali).1 How many are the elect? The first important remark to be made is that this verse speaks of election (istafa). This term has a large Biblical resonance, particularly in Hebrew tradition, where Israel is the Elect of God, the “elect people” of God. However, the four proper names mentioned here are: Adam, Noah, Abraham, and the family of ‘Imran. If one considers the chronological succession, the order indicated is correct. And yet, if the three first names are certainly identifiable in the Bible, the fourth is problematic. Indeed, in the Qur’anic context, the expression Al ‘Imran refers to the family of Jesus through Mary. Thus we have a series of four names, which encompasses Adam, Noah, Abraham, and Jesus. These four figures have been elected. And yet this series is surprising, since the figure who is cited most often in the Qur’an is not Adam, Noah, Abraham or Jesus. Instead it is Moses, who by far exceeds all of the other prophets (being mentioned 138 times). And yet, as though by chance, Moses is not named here. How is this possible? And if he is not cited, it means then that this verse has no Jewish influence, and could not stem from Jews: it can only stem from a Christian milieu. Similarly, in light of the importance of Moses (who probably represents a model for Muhammad more than all of the other cited figures, being at once the leader of a people and the messenger of God) in the Qur’an, it seems to me impossible that the Qur’an would have omitted here Moses in this series of the Elect of God. The name ‘Imran In apocryphal Christian texts, the parents of Mary are called Anne and Yuwakim. In the Qur’an Anne is unknown – which should come as no surprise2 – and Yuwakim is called ‘Imran. Where does this name come from? 142 THE THEOLOGICAL CHRISTIAN INFLUENCE According to 1 Chronicles 5:29 (Hebrew), the children of ‘Amran are Aaron, Moses and Miriam (Mary); in Exodus 15:20, Miriam is called the “sister of Aaron.” Could the Qur’an have confused Mary the mother of Jesus with Mary the sister of Aaron? This seems to be confirmed by Qur’an 19:28, where Mary is called “O sister of Aaron,” and Qur’an 66:12, where Mary is called the daughter of ‘Imran.3 And yet we know that the fathers of the Church (e.g. Aphraates [fl. fourth century] and Gregory of Nyssa [d. 395], both of whom were well known to Syriac Christians) often compared the two Marys (of the Old and New Testaments) and that this connection found a place in popular Christian preaching. According to Joseph Henninger, this would explain the confusion.4 Meanwhile, Muslim commentators note this difficulty. They explain that ‘Imran father of Mary is different than ‘Imran father of Miriam. Meanwhile “sister of Aaron” indicates “descendent of relative of Aaron.”5 Thus we have two ‘Imrans: that of the Hebrew Bible, who is the father of Moses, Aaron and Miriam, and that of the Qur’an, who is the father of Aaron and Mary. It seems to me that, in the expression “Family of ‘Imran,” one should see the meeting of these two traditions, the Hebrew and the Arab, the Biblical and the Qur’anic. In other words, there would not be only four names of the elect (Adam, Noah, Abraham, and the family of Mary and Jesus), but in fact five, including Moses and his family before that of Jesus. An evident Christian influence If one accepts this working hypothesis, then we will find a schema that is typically Christian, namely “the Covenants of God with humanity”: the Covenant (berith) of Adam, expressed by what is known as the proto-evangelium (Genesis 3:15); the Covenant of Noah, symbolized by the rainbow; the covenant of Abraham, in which all of the nations of the earth will be blessed, given material form in the circumcision of men; the Covenant of Moses, marked by the tablets of the Law given on Mt Sinai; and finally the Covenant that the Christians call the “New Covenant” in Jesus. If this is the case, then it is necessary to understand the election (istifa’) of these men in the sense that they were chosen by God who contracted with them a Covenant. Thus we find anew the theme of God’s five Covenants with humanity, a classical theme with the Fathers of the Church, from Origen (d. ca. 254) to John Chrysostom (d. 407) and many other Fathers. We find again this theme with certain Arab Christian authors, for example with the little known Palestinian author from the beginning of the ninth century, Butrus al-Bayt Ra’si, that is, Peter the bishop of Capitolia in Jordan, author of a Kitab al-Burhan (falsely attributed to Eutychius of Alexandria).6 Commenting on the parable of the workers of the last hour (Matthew 20:1–15), in which there are five successive invitations to workers from the first to the last hour, our apologist 143 SAMIR KHALIL SAMIR interprets this along with patristic tradition, as a reference to the five Covenants of God with humanity. It is clear that the preaching (or call) of Christ and His disciples is the last and complete preaching. It is for all nations alike, and is not specially for one nation rather than another, in accordance with the parable, spoken by Christ in the Gospel, of the five calls whereby God called His servants to work for the kingdom of heaven in the world to come. ((pseudo-)Eutychius of Alexandria, The Book of the Demonstration (Kitab al-Burhan), trans. Watt, CSCO 193, § 366, 156) Thereafter he cites the text of the parable (Matthew 20:1–15) and gives the following explanation: The first calling whereby God called on men to act in obedience to Him was the call of Adam early at the beginning of the world. Then the second call was the call of Noah after the deluge. The third call was the call of Abraham. These three calls were by the law of nature. The fourth call was the call of Moses by the scriptural law of the Torah. The fifth call was the call of Christ and His apostles by the law of the Gospel. This call was not specially for the children of Israel, but was for all men alike, for the many idle nations to whom the book of the Torah had not been revealed and whom none of the prophets had hired. This is the call of the law of grace, for Christ gave [as grace] to the people of His call the full hire for their faith in Him; He did not reckon it to them by hours of the day. He began paying them before the first, and He did not listen to the murmuring of the envious, because of His power to do what He wills with His own. Each of these calls represents a Covenant from God and a public testament between Him and the people of that Covenant. ((pseudo-)Eutychius of Alexandria, The Book of the Demonstration (Kitab al-Burhan), trans. Watt, CSCO 193, § 367–8, 156–7) Naturally, Butrus does not mention Muhammad. Yet his very absence, and the fact that there are no more than five Covenants and that the Covenant of Christ is the only universal one, implies that the Covenant of Muhammad is excluded by the Bible.7 On the other hand, one is surprised to see that in the Qur’an Muhammad is not named among these elect. One expects him to be part of this series, since his name is al-Mustafa in Islamic tradition, the elect par excellence, and it is this very term (istafa) that is used in our verse. Yet Muhammad does not participate in this tradition of the Elect, and, according to the hypothesis which we are presenting, does not bring a Covenant. 144 THE THEOLOGICAL CHRISTIAN INFLUENCE From whom could such an enumeration come from if not the Christians? More precisely, if one accepts our interpretation of the expression “Al ‘Imran,” that it encompasses at once Moses and Jesus, then it could only come from a JudaeoChristian milieu, that is from Christians who have preserved Hebrew traditions. For Jews Al ‘Imran is none other than Moses (along with Aaron and Miriam). The fact that the Qur’anic tradition, on the contrary, attributes this name to the family of Mary and Jesus, shows that there is a Christian tradition here. For this reason I conclude that there is a Judaeo-Christian tradition here. Jesus creator (Q 3:49) It would be impossible here, due to space, to examine the miracles of Jesus mentioned in the Qur’an, notably in Q 3:49. Since they are sufficiently well known and since their relation with the Gospel texts is evident, I will be content to limit myself to the first miracle in order to single out two particular points. Aya ⫽ sign wa-rasulan ila bani Isra’ila anni qad ji’tukum bi-’ayatin min rabbikum And (appoint him) an apostle to the Children of Israel, (with this message): “I have come to you, with a Sign (aya) from your Lord.” The word aya is a keyword in Islamo-Christian revelation. One finds it seventyseven times in the New Testament, especially in the Gospel of John, in the Greek form sêmeion, and 287 times in the Qur’an (however somewhat rarely in the first Meccan surans). It signifies a prodigious sign that attests to the authenticity of the one who accomplishes it (it should not be translated with ‘alama, which signifies only “signal”). In Arabic the word designates the verses of the Qur’an or the Bible. This comes from the idea that every phrase of the Qur’an is a miraculous sign for those who truly believe, an idea that is connected with the dogma of the inimitability (i‘jaz) of the Qur’an. This word, aya (“sign”) is frequently discussed by the Arab lexicographers. It has no Arabic derivation and is evidently a borrowing from a Semitic language. It could be translated from Hebrew ôt (likewise translated in the Septuagint as sêmeion), but the specialists more often derive it from Syriac ata. The word aya is frequently attested in pre-Islamic poetry, for example in the Diwan of Imru’ al-Qays.8 If this is the case, then the probability that it entered Arabic through Syriac speaking Christians is more likely, as Jeffery explains While it is not impossible that the Arabs may have got the word from the Jews, it is more probable that it came to them from the Syriac-speaking Christians. (FV, 72) 145 SAMIR KHALIL SAMIR The verb khalaqa anni akhluqu lakum mina l-tini kahay’ati l-tayri, fa-anfukhu fihi fa-yakunu tayran, bi-idhni llah I make for you out of clay, as it were, the figure of a bird, and breathe into it, and it becomes a bird by God’s leave. This miracle is also mentioned in Q 5:110. It is found in the apocryphal Infancy Gospels.9 The verb khalaqa is found 180 times in the Qur’an and it is always translated, in various languages, with “to create.” With the exception of Q 20:17 (takhluquna ifkan ⫽ you invent a lie), it always designates the creative action of God. In 177 cases, the subject of the verb is God, while in the other two cases (3:49 and 5:110) it is Christ. Evidently this could only come from Christians; Muslim tradition, which could not uphold this meaning (the only one attested in the Qur’an), interprets it with the meaning of “to fashion, mold.” Meanwhile, the action of “breathing into” is, in the Bible as in the Qur’an, typical of the creative action of God. Thus the two verbs used in this verse both reflect the divine creative action, and not the human action of a potter, for example, thereby confirming the Christian origin of this verse. Christ the new Adam (Q 3:59) The Qur’anic text Inna mathala ‘Isa ‘inda llahi kamathali Adama, khalaqahu min turabin, thumma qala lahu “kun” fa-yakunu The similitude of Jesus before God is as that of Adam; He created him from dust, then said to him: “Be.” And he was. This verse establishes a parallel or a likeness (mathal) between Jesus and Adam. The question presented is whether the Qur’an specifies the content of this likeness. The most common response of the Muslim commentators, even today, is to say that this likeness is made explicit in the second part of this verse: God “created him from dust then said to him: ‘Be.’ And he was.” Yet this explanation does not agree with the Qur’an itself. Indeed, when the Qur’an says, God “created him from dust,” the personal pronoun could not refer to Christ, since he was not created from dust but from the breath of God in Mary. wa-llati ahsanat farjaha, fa-nafakhna fiha min ruhina, wa-ja‘alnaha wa-bnaha ayatan li-l-‘alamina ⫽ “And (remember) her who guarded her chastity: We breathed into her of Our spirit, and We made her and her son a sign for all peoples.” (Q 21:91) 146 THE THEOLOGICAL CHRISTIAN INFLUENCE Nevertheless, many Muslim readers curiously apply the pronoun to Christ. Yet this phrase can only refer to Adam alone. The first sura (chronologically speaking) of the Qur’an (96:1) affirms that Adam was created from dust when it speaks of ‘alaq, which should be understood in the sense of “sticky mud,” and not of “sperm” as many today think.10 On the other hand, the second part of the phrase (“then said to him: ‘Be.’ And he was.”) also can only be applied to Adam. For God created everything by his word (amr), as the Qur’an affirms: “Verily, when He intends a thing, His Command is, ‘be,’ and it is!” (36:82; cf. 3:47). In reality, this verse contains two ideas: one, the affirmation of a likeness between Christ and Adam (mathala ‘Isa ‘inda llah; yet of what this likeness consists the Qur’an does not say); and two, a report concerning Adam, that God created him from dust and his word. One can assume nevertheless that the likeness between the two, according to the Qur’an, is that like Jesus Adam was created without a father. This point is rather frequently mentioned in Muslim tradition until the present day. Muslim apologists of the Middle Ages, as those of the modern era (e.g. Shaykh Rashid Rida in his commentary on the Qur’an) emphasize even that Adam was superior to Jesus because he was born without a mother. Christian theologians (notably Elijah of Nisibis [d. 1056]) respond that in this Adam is not different than the first donkey or bull, since he is the first of his jins. But Jesus was born without a father when there were an infinity of possible fathers. According to them the eminence of Christ above Adam is thereby demonstrated. The Parallel according to Paul The parallel between Jesus and Adam is important to Paul. Thus in 1 Corinthians 15:22: “For as in Adam all die, even so in Christ shall all be made alive” (King James Version). Or again in Romans 5:12–21 Wherefore, as by one man sin entered into the world, and death by sin; and so death passed upon all men, for that all have sinned: (For until the law sin was in the world: but sin is not imputed when there is no law. Nevertheless death reigned from Adam to Moses, even over them that had not sinned after the similitude of Adam’s transgression, who is the figure of him that was to come. But not as the offence, so also is the free gift. For if through the offence of one many be dead, much more the grace of God, and the gift by grace, which is by one man, Jesus Christ, hath abounded unto many. And not as it was by one that sinned, so is the gift: for the judgment was by one to condemnation, but the free gift is of many offences unto justification. For if by one man’s offence death reigned by one; much more they which receive abundance of grace and of the gift of righteousness shall reign in life by one, Jesus Christ.) Therefore as by the offence of one judgment came upon all men to 147 SAMIR KHALIL SAMIR condemnation; even so by the righteousness of one the free gift came upon all men unto justification of life. For as by one man’s disobedience many were made sinners, so by the obedience of one shall many be made righteous. Moreover the law entered, that the offence might abound. But where sin abounded, grace did much more abound: That as sin hath reigned unto death, even so might grace reign through righteousness unto eternal life by Jesus Christ our Lord.11 This parallel is also frequently found in patristic literature: Christ is the new Adam, who liberates us from the sin of Adam and the death that followed from it, and so on. In reading the Qur’an, the reader has the sense of encountering a Christian (Pauline par excellence) topos, integrated into Qur’anic thought and Islamized dogmatically. The Eucharist (5:112–15)12 The Table (Q 5:112–14) (112) Idh qala al-hawariyyuna : ⰆYa ‘Isa bna Maryama, hal yastati’u rabbuka an yunazzila ‘alayna ma’idatan mina l-sama’?Ⰷ. Qala ttaqu llaha in kuntum mu’minina. (113) Qalu nuridu an na’kula minha wa-tatma’inna qulubuna wa-na ‘lama an qad sadaqtana wa-nakuna ‘alayha mina al-shahidina. (114) Qala ‘Isa bnu Maryama: allahumma rabbana, anzil ‘alayna ma’idatan mina l-sama’i takunu lana ‘idan li-awwalina wa-’akhirina wa-’ayatan minka. Wa-rzuqna wa-anta khayru l-raziqina. (112) Behold! the disciples, said: “O Jesus the son of Mary! can thy Lord send down to us a table set (with viands) from heaven?”13 Said Jesus: “Fear Allah, if ye have faith.”14 (113) They said: “We only wish to eat thereof and satisfy our hearts, and to know that thou hast indeed told us the truth; and that we ourselves may be witnesses to the miracle.” (114) Said Jesus the son of Mary: “O Allah our Lord! Send us from heaven a table set (with viands), that there may be for us – for the first and the last of us – a solemn festival and a sign from thee; and provide for our sustenance, for thou art the best Sustainer (of our needs).” Al-Ma’ida (Q 5:112, 114) We note first an interesting detail. This chapter of 120 verses has only four verses (112–15) on the Table, the ma’ida is only mentioned in verses 112 and 114, and nowhere else in the Qur’an. However, it is this most rare word and these verses that provide the title to this chapter, proof that they were felt to be important. 148 THE THEOLOGICAL CHRISTIAN INFLUENCE Western researchers agree that these verses refer to two groups of Gospel narratives: the Multiplication of Loaves and the Last Supper. Some of them also have said that the Table that descends from the sky at the demand of the apostles and upon the prayer of Jesus could be an allusion to the vision of Peter of a sheet descending from the sky (Acts 10:9–13); this does not seem apparent to me. The first philological remark is the fact that the term ma’ida is evidently a non-Arabic word borrowed from another language, as the derivation of the Arabic root myd is “pulled by the hair.” The relatively rare opinion that it derives from Persian is weak. The majority of researchers are agreed that the term is borrowed from Ethiopic; even more, it is the technical term used by the Christians of Ethiopia to signify the Last Supper. According to a report that I was not able to confirm, this term is already in the Ethiopic translation of the Bible made in the fourth century, on the basis of Hebrew and Syriac.15 Otherwise, there is unanimity between Muslims and non-Muslims on the fact that this sura is one of the last, if not the last, of the Qur’an. That is to say that it is certainly posterior to the return from Ethiopia of the last Meccan emigrants. This would support the fact that this would be a borrowing from Ethiopian Christianity. Feast (‘id) and sign (aya) (v. 114) The term ‘Id is found nowhere else in the Qur’an. According to unanimous scholarly opinion it is a borrowing from the Syriac ‘ida, which signifies “Feast” or “liturgical festival.”16 As for the word aya, “sign,” we have seen that it is also probably a borrowing from Syriac Christians. This ma’ida is thus defined by two terms: ‘id and aya, a “Feast” or “liturgical festival” and a “sign.” Is this not the most appropriate definition of the Eucharist of Christians, which is a festive celebration and a sacramental sign? Even more, it seems evident that in this passage we are dealing with a rather faithful description of Christian faith, otherwise not shared by Muslims. How could this be? li-awwalina wa-’akhirina (v. 114) The expression li-awwalina wa-’akhirina is also found nowhere else in the Qur’an. Even the words awwaluna and akhiruna are found nowhere else in the Qur’an. Meanwhile, awwal and akhir are never used in the Qur’an to designate a person. The expression apparently means “all, nobody excluded.” We find again here something that relates to the Eucharist and which, as far as I know, has not yet been noticed by researchers. We know that there are two traditions on the narrative of the institution of the Eucharist: one according to Matthew and Mark and the other according to Luke and Paul. According to Matthew–Mark we have: “This is the blood of my Covenant,17 which is shed for many,” while for Luke–Paul we have: “This cup is the new Covenant,18 in my blood, which is shed for you.”19 We see that while Luke–Paul have “for you,” 149 SAMIR KHALIL SAMIR ~ Matthew and Mark have “for many” (Matthew 26:28: ␲‡␳ì ␲␱␭␭␻ ␯; Mark 14:24: ~ ␷´ ␲‡␳ ␲␱␭␭␻ ␯ ⫽ pro multis, in the Latin version of Jerome). The ancient Roman liturgy integrates the two: “Qui pro vobis et pro multis effundetur in remissionem peccatorum.” Yet the expression “pro multis” is typical of the New Testament: we find it there ten other times.20 In these twelve passages, it does not signify “for many,” as though a certain number are excluded, but rather “for all.” This turn of phrase is well known to the exegetes, and we find a very clear allusion to it in the letter of John Paul II to the priests for Holy Thursday 2005. “Hoc est enim corpus meum quod pro vobis tradetur.” The body and the blood of Christ are given for the salvation of man, of the whole man and of all men. This salvation is integral and at the same time universal, because no one, unless he freely chooses, is excluded from the saving power of Christ’s blood: “qui pro vobis et pro multis effundetur.” It is a sacrifice offered for “many,” as the Biblical text says (Mark 14:24; Matthew 26:28; cf. Isaiah 53:11–12); this typical Semitic expression refers to the multitude who are saved by Christ, the one Redeemer, yet at the same time it implies the totality of human beings to whom salvation is offered: the Lord’s blood is “shed for you and for all,” as some translations legitimately make explicit. Christ’s flesh is truly given “for the life of the world” (John 6:51; cf. 1 John 2:2).21 Conclusion These few verses have been glossed by Muslim commentators with the help of various narratives, which one can find for example in the work of Roger Arnaldez22 or Michel Hayek.23 In my opinion, it cannot be truly understood except as a clear allusion to the Eucharist,24 which could only have Syriac or Ethiopic Christians as its source, and which remains veiled to Muslims. One is struck meanwhile by the number of words here that appear nowhere else in the Qur’an. “Provide for our sustenance” (v. 114) and the Lord’s Prayer Wa-rzuqna wa-anta khayru l-raziqina “Provide for our sustenance, for thou art the best Sustainer (of our needs)” Rizq, in Classical Arabic, is the food necessary for one day.25 Thus one thinks necessarily of the Lord’s Prayer: “Panem nostrum quotidianum da nobis hodie” ⫽ “ton arton emôn ton epi-ousion dos emin semeron” (Matthew 6:11) or “didou emin to kathemeran” (Luke 11:3) and in English “Give us today our daily bread.” 150 THE THEOLOGICAL CHRISTIAN INFLUENCE Yet we know that the text of the Lord’s Prayer was understood by various churches of the East as an allusion to the Eucharist, precisely due to the word epi-ousion, which can signify “daily” (as Jerome translated it in the Vulgate and as it is in the majority of western languages after him), or even “super-substantialis” as the Vetus Latina translates it, as well as the Arab Byzantine version (khubzana l-jawhari ⫽ our essential bread), or even “of tomorrow” as the Coptic Tradition understands it (cf. khubzana lladhi li-l-ghad a‘tina l-yawm ⫽ our bread for tomorrow give us today), along with that of the Ebionites. The last two interpretations evidently make an allusion to the Eucharistic bread. Thus this brief Qur’anic phrase, marvelous in its conciseness, probably invokes the Lord’s Prayer, which is always connected to the Eucharist in the tradition of the Eastern churches. The thematic unity of this verse thus becomes even more evident. The threat of God (v. 115) Qala llahu inni munazziluha ‘alaykum, fa-man yakfur ba‘du minkum, fa-inni u‘adhdhibuhu ‘adhaban la u‘adhdhibuhu ahadan mina l-‘alamina. Allah said: “I will send it down unto you: But if any of you after that resisteth faith, I will punish him with a penalty such as I have not inflicted on any one among all the peoples.” This terrifying phrase invokes 1 Corinthians 11:27–9, which speaks of the Eucharist in these terms: Wherefore whosoever shall eat this bread, and drink this cup of the Lord, unworthily, shall be guilty of the body and blood of the Lord. But let a man examine himself, and so let him eat of that bread, and drink of that cup. For he that eateth and drinketh unworthily, eateth and drinketh damnation to himself, not discerning the Lord’s body. Conclusion In summary, we have seen that the verses 112–15 give the name to the entire sura al-Ma’ida. This term, which is found nowhere else in the Qur’an other than this passage, is borrowed from Ethiopian and is the technical term that designates the Eucharistic table to Ethiopian Christians. This section of the Qur’an, mysterious for those who are not familiar with the New Testament and Christian theology, becomes on the contrary clear and profound in this light. This ma’ida is at a once festive celebration and sacramental sign. It is offered to all, to the first as to the last. It is the necessary bread for daily subsistence (rizq). The one who rejects this sign or does not recognize it will be punished terribly, an admonition made by Paul. 151 SAMIR KHALIL SAMIR The Qur’anic non-crucifixion (Qur’an 4:156–9)26 This Qur’anic text is the most important to be discussed, for it denies in a clear manner the crucifixion of Christ. The sura can be dated to the year 626. According to Muslim chronology it occupies the ninety-second place (according to most Orientalists the 102nd place). Three motifs of the punishment of Jews (vv. 156–7a) (156) Wa-bi-kufrihim wa-qawlihim ‘ala Maryam buhtanan ‘aziman, (157a) wa-qawlihim inna qatalna al-Masih ‘Isa ibn Maryam rasul Allah . . . (156) That they rejected Faith; that they uttered against Mary a grave false charge; (157a) That they said (in boast), “We killed Christ Jesus the son of Mary, the Apostle of God” According to this text, there are three motifs that led to the condemnation of the Jews: their rejection of faith, their charge against Mary (of having had relations with a man) and their pretension of having killed Christ. 1 2 3 The first accusation refers to the fact that the Jews did not recognize Jesus as the messiah, even if messiah is not a concept of great importance to the Qur’an. It seems to be simply a proper name. The second accusation is that of adultery. It relates to an earlier text (Q 19:27–8) wherein Mary comes to her family, carrying Jesus in her arms. They say to her “O Mary! truly an amazing thing hast thou brought! O sister of Aaron! Thy father was not a man of evil, nor thy mother a woman unchaste!” This is a theme developed widely in the apocryphal Gospels. The third accusation is for having said: “We killed Christ Jesus the son of Mary, the Apostle of God.” The polemic here is against the Jews, not the Christians. They killed him not, nor crucified him (v. 157b) The Qur’anic text continues: “but they killed him not, nor crucified him, but so it was made to appear to them (wa-lakin shubbiha lahum).” The translations are many and diverse here. The text, far from evident, is shrouded in mystery. Some commentators speak of another person who was substituted in the place of Jesus. According to the Qur’an, the Christians who affirm the death of Christ do so without proof and without certitude, being content to follow a hypothesis (zann): “Those who differ therein are full of doubts, with no (certain) knowledge, but only conjecture to follow, for of a surety (yaqinan) they killed him not.” The adverb yaqinan at the end of the phrase is surprising. One could translate it, as Muhammad Kamel Hussein does in his novel Qarya zalima (City of Wrong): “It is not certain that they killed him.” Others go further and suggest the translation: “It is certain that they did not kill him.” 152 THE THEOLOGICAL CHRISTIAN INFLUENCE God raised him up to Himself (v. 158) “Nay, God raised him up unto Himself (rafa‘ahu Allah ilayhi); and God is Exalted in Power, Wise.” The phrase “God raised him up to Himself ” is found only with Jesus in the Qur’an (cf. 3.55). One finds, it is true, in regard to the prophet Idris, that God “raised him to a lofty station” (Q 19:57; wa-rafa‘nahu makanan ‘aliyyan),27 but not “to Himself.” Meanwhile, the pious Muslim tradition regarding Muhammad’s nocturnal journey (with reference to Q 17:1, usually dated to the years 615–19), does not have him arrive to God Himself. There is probably a reminiscence here of the Ascension of Christ to the Father, such an important aspect of Christian theology. Yet here Christ is made to escape death. The ascension to Christ is so prominent that it is taken up in a Qur’anic text (3:55) that is dated to the year 630: “Behold! God said: ‘O Jesus! I will take thee (mutawaffîka) and raise thee to Myself (wa-rafi‘uka ilayya).’ ” All will believe in Jesus, who will be a witness against the infidels on the Day of Resurrection (v. 159) Verse 159a: “And there is none of the People of the Book but must believe in him before his death.” The expression “before his death” is not clear. Does the pronoun “his” refer to each person or to Jesus? It seems that the only meaning appropriate to the Qur’an (since indeed many have died without believing in Jesus) is “the death of Jesus.” Thus all of the Jews will believe in Jesus before he dies at the time when he has returned to earth. The verse continues: “and on the Day of Judgment he will be a witness against them.” We find this expression in the mouth of Jesus, in the last sura (chronologically speaking): “I was a witness over them whilst I dwelt amongst them” (5:117). Thus Jesus was a witness against them during his life, and he will be so again on the Day of Resurrection. Here there is probably an echo of the Christian theology which affirms that Christ will come at the end of time “to judge the living and the dead” (Nicene Creed). Thus the traditional Islamic position on Jesus, briefly summarized, is 1 2 3 4 5 6 Jesus was not crucified. Someone who looked like him was crucified in his place. He is therefore not dead, but was raised up to God. At the end of the world he will return to earth, fight the Antichrist and proclaim Islam as the true religion. He will proclaim the coming of the Hour of Judgment, (marry) and die. He will be raised on the day of the final resurrection. Origin of the theory of substitution All of the western commentaries conclude that the theory of the substitution of Christ on the Cross derives from Docetism, the heretical Christian current according 153 SAMIR KHALIL SAMIR to which Christ took on only human appearance but was not truly a man, and therefore could not be crucified. Docetism is already found in the beginning of the second century with Ignatius of Antioch (d. ca. 110). The apocryphal Acts of John likewise suggest the same doctrine. It is also suggested in a document of the Gnostic library of Nag‘ Hammadi, the Authentikos logos, which is datable to the second half of the second century. Yet the text most evocative of the Qur’anic passage is that of the Gnostic Judaeo-Christian Basilides, reported by Irenaeus of Lyon at the end of the second century. Basilides writes Those angels who occupy the lowest heaven, that, namely, which is visible to us, formed all the things which are in the world, and made allotments among themselves of the earth and of those nations which are upon it. The chief of them is he who is thought to be the God of the Jews; and inasmuch as he desired to render the other nations subject to his own people, that is, the Jews, all the other princes resisted and opposed him. Wherefore all other nations were at enmity with his nation. But the Father without birth and without name, perceiving that they (⫽ the Minds) would be destroyed, sent his own first-begotten Nous (he it is who is called Christ) to bestow deliverance on them that believe in him, from the power of those who made the world. He appeared, then, on earth as a man, to the nations of these powers, and wrought miracles. Wherefore he did not himself suffer death, but Simon, a certain man of Cyrene, being compelled, bore the cross in his stead; so that this latter being transfigured by him, that he might be thought to be Jesus, was crucified, through ignorance and error, while Jesus himself received the form of Simon, and, standing by, laughed at them. For since he was an incorporeal power, and the “Nous” (mind) of the unborn Father, he transfigured himself as he pleased, and thus ascended to him who had sent him (⫽ the Father), deriding them, inasmuch as he could not be laid hold of, and was invisible to all. Those, then, who know these things have been freed from the principalities who formed the world; so that it is not incumbent on us to confess him who was crucified (⫽ Simon), but him who came in the form of a man, and was thought to be crucified, and was called Jesus, and was sent by the Father, that by this dispensation he might destroy the works of the makers of the world. If any one, therefore, he declares, confesses the crucified, that man is still a slave, and under the power of those who formed our bodies; but he 154 THE THEOLOGICAL CHRISTIAN INFLUENCE who denies him has been freed from these beings, and is acquainted with the dispensation of the unborn Father. (Irenaeus, Against Heresies, Book 1, ch. 24, no. 4) This does not mean that the Gnostic theory was current among the Christians of Arabia, but simply that the idea of a substitute for Jesus on the Cross was present. Allah khayru l-makirin (3:54) And (the unbelievers) plotted and planned, and God too planned, and the best of planners is God. This text is to be understood in connection with the verses that follow and speak of the non-crucifixion of Christ. The impious plotted against God, but God was craftier than them and delivered Christ from the hands of those who sought to crucify him. This theme of God’s trick, which we find elsewhere in the Qur’an (8:30), seems strange, if not shocking. In fact, it is traditional with the fathers of the Church and frequent among the Christian Arab theologians (e.g. Butrus al-Bayt Ra’si in the ninth century and Sawirus Ibn al-Muqaffa‘ in the tenth century). The theme is likewise connected to the crucifixion of Christ but the meaning is different. For Christian theologians the theme is based on the idea that Satan deceived Adam in paradise by a trick, and that this led to his exclusion from paradise and the tendency of rebellion as the second nature of humans (original sin). God therefore decided to save Adam and his descendents. But God is just (‘adl Allah) and does not want to use coercive methods with Adam. He therefore used the same method of Satan, namely the trick (al-makr). He took on a human form, in Jesus Christ, to deceive Satan (who occasionally perceived that Jesus was not a human but the Messiah, as seen in the Gospel exorcism accounts). Above all he deceived him on the Cross when he took the form of a slave, something that is entirely improper for God. In this way he overcame Satan with his own tactic, the trick. It seems to me that this beautiful patristic theme is found again here in the Qur’an, in the same context, but in its cosmic theological dimension (which we find again, for example, in Revelation). Do not be excessive regarding Christ: he did not judge it improper to be a slave (4:171–2) Now we turn to the final verses of the sura of women (171) Ya ahla l-kitabi la taghlu fi dinikum wa-la taqulu ‘ala llahi illa l-haqqi: innama l-Masihu ‘Isa bnu Maryama rasulu llahi wa-kalimatuhu alqaha ila Maryama wa-ruhun minhu. Fa-’aminu bi-llahi wa-rusulihi wa-la taqulu 155 SAMIR KHALIL SAMIR thalathatun intahu khayran lakum! innama llahu ilahun wahidun subhanahu an yakuna lahu wa-ladun! lahu ma fi al-samawati wa-ma fi l-ardi wa-kafa bi-llahi wakilan. (172) Lan yastankifa l-Masihu an yakuna ‘abdan li-llahi wa-la l-mala’ikatu l-muqarrabuna. wa-man yastankif ‘an ‘ibadatihi wa-yastakbir fa-sayahshuruhum ilayhi jami‘an (171) O People of the Book! Commit no excesses in your religion: Nor say of God aught but the truth. Christ Jesus the son of Mary was (no more than) a messenger of God, and His Word, which He bestowed on Mary, and a spirit proceeding from Him: so believe in God and His messengers. Say not “Trinity” : desist: it will be better for you: for God is one God: Glory be to Him: (far exalted is He) above having a son. To Him belong all things in the heavens and on earth. And enough is God as a Disposer of affairs. (172) Christ disdaineth nor to serve and worship God, nor do the angels, those nearest (to God): those who disdain His worship and are arrogant, – He will gather them all together unto Himself to (answer). Commit no excesses! (v. 171) That which God says to the Christians is not to be excessive when speaking of Christ, to speak only the truth: that Christ is the Messenger of God (like Muhammad), the Word of God and the Spirit of God! This is an astonishing affirmation: at the same time that the Qur’anic text commands the Christians not to be excessive . . . it is excessive, if one accepts the meaning of its words. The formula “Word of God” (Kalimat Allah) is evidently taken from the prologue of John (John 1:1). It is surprising to find it in the Qur’an, for it does not correspond to that which is normally said of the messengers of God, who are all created by the word of God. It is only applied to Christ and does not correspond with the Qur’anic conception of the oneness of God. Evidently it does not mean that which John (and Christians after him) understand by the “Word of God,” but the Qur’an also does not furnish an explanation that matches with the rest of its theological approach. As for the formula “a Spirit (proceeding) from God,” which is transformed in Islamic (particularly Sufi) literature, into “Spirit of God,”28 it is no less surprising. It never appears in the New Testament. It is found, however, in Gnostic literature. This point is not without interest. We have already found, on the subject of the Crucifixion of Christ, a Gnostic Christian influence on the Qur’an. Christ does not judge it improper to lower himself (v. 172) This verse is very suggestive: “Christ – like those angels who are close (to God) – does not judge it improper to be a slave of God.” (Translation mine; The inaccurate translation of Yusuf ‘Ali – “Christ disdaineth nor to serve and worship God . . . .” – seems to reflect his apprehension with this verse.) 156 THE THEOLOGICAL CHRISTIAN INFLUENCE Yet adjusting one word is enough to bring out a phrase central to Christian faith and the message of Saint Paul in Phillipians 2:5–11 (especially verses 6–8). Let this mind be in you, which was also in Christ Jesus: Who, being in the form of God, thought it not robbery to be equal with God: But made himself of no reputation, and took upon him the form of a servant, and was made in the likeness of men: And being found in fashion as a man, he humbled himself, and became obedient unto death, even the death of the cross. Wherefore God also hath highly exalted him, and given him a name which is above every name: That at the name of Jesus every knee should bow, of things in heaven, and things in earth, and things under the earth; And that every tongue should confess that Jesus Christ is Lord, to the glory of God the Father. In the Qur’anic text, only one word makes the difference. If one had, in the phrase “Lan yastankifa l-Masihu an yakuna ‘abdan lillahi,” the verb yasira, for example, in the place of yakuna, we would have a most beautiful Arabic adaptation of the text of Paul: “Christ did not judge it improper of himself to become a slave.” Here, as elsewhere in the Qur’an when it comes to Christ and Christian doctrine, the Qur’anic text scrupulously respects the Christian affirmation, but juxtaposes it with Qur’anic dogma. Jesus and the hour (Q 43: 61) The Hour in the Qur’an In the Qur’an, the term sa‘a (“hour”) appears fortyeight times. When one finds it in the definite form (al-sa‘a), it indicates “the Hour of the final judgment.” We find this expression frequently (forty times) in the Qur’an, often as a threat made to unbelievers. This is indeed the same meaning that we find in the Gospels, even in the mouth of Christ, and it seems likely that there was a direct Christian influence, since the expression does not seem to belong in Jewish tradition. Only God has “knowledge of the Hour” Elsewhere in the Qur’an, it is normally God who has “knowledge of the Hour” (‘ilm al-sa‘a), as one finds in the following three verses: Verily the knowledge of the Hour is with God (alone). It is He Who sends down rain, and He Who knows what is in the wombs. Nor does any one know what it is that he will earn on the morrow: Nor does any one know in what land he is to die. Verily with God is full knowledge and He is acquainted (with all things) (31:34). 157 SAMIR KHALIL SAMIR To Him is referred the Knowledge of the Hour (of Judgment: He knows all): No date-fruit comes out of its sheath, nor does a female conceive (within her womb) nor bring forth the Day that (God) will propound to them the (question), “Where are the partners (ye attributed to Me)?” They will say, “We do assure thee not one of us can bear witness!” (41:47). And blessed is He to Whom belongs the dominion of the heavens and the earth, and all between them: with Him is the Knowledge of the Hour (of Judgment): and to Him shall ye be brought back (43:85). Yet this is also an affirmation of the New Testament and is even found in the very mouth of Christ.29 Does Christ have “knowledge of the Hour”? However, in one verse (Q 43:61), Christ is referred to have “having knowledge of the Hour,” or even more: “being the knowledge of the Hour” (“innahu la-‘ilmun li-l-sa‘a”). This formula is not very clear and most commentators avoid it. They prefer another reading, which does not exist in the canonical Cairo text: “innahu la-‘alamun li-l-sa‘a” ⫽ “He is the sign of the Hour.” The commentators explain that the return of Christ to earth (universally affirmed by Muslim tradition) will be the sign announcing the Day of Judgment. Thus ‘Abdallah Yusuf ‘Ali, in a note on this verse in his translation, writes This is understood to refer to the second coming of Jesus in the Last Days just before the Resurrection, when he will destroy the false doctrines that pass under his name, and prepare the way for the universal acceptance of Islam, the Gospel of Unity and Peace, the Straight Way of the Qurân. (The Qurân, 1337, n. 466230) Conclusion According to these two readings of the Qur’an Jesus is the “sign” (‘alam) and “knowledge” (‘ilm) of the Hour. However, he is not the knowledge of the Hour, which is reserved for God, as the Qur’an repeats, echoing the Gospels. Jesus is likewise not the judge of the Last Day, a position reserved for God. Here too we find an affirmation of the Gospel clearly expressed in a parable of Christ (Matthew 25), in which the Father judges all of the humanity, yet in relation to their response to Christ. He is the expected Messiah (al-mahdi al-muntazar) who will come to the earth, from the presence of God, as an announcer of the final judgment. In his role as Mahdi, he will accomplish his final mission: to purify the earth of all idols, superstition and erroneous beliefs. He will fight evil, the Antichrist, and defeat him. Finally he will proclaim the true religion: Islam. This is the role and the mission of Christ, by which he appears to be a perfect Muslim prophet. 158 THE THEOLOGICAL CHRISTIAN INFLUENCE Conclusion Many other points could be evoked that would greatly illuminate the possible influence of Christianity on the Qur’an: the name of God al-Rahman; the expression Sibghat Allah (2:138) over which the commentators stumble, the parallels between Jesus and the angels, and so on. Qur’anic proper names that reflect and a written Arabic Christian text Yet what seems even more significant to me are certain proper names that reflect a written Christian text. a b c According to Qur’an 61:6 Christ announces a prophet who will come after him and gives him the name Ahmad. In the face of Christian objections that they know no such prophet, Muslim tradition sought to explain this verse according to the theme of the Paraclete, the Consoler, which is found three times in the Gospel of John. This interpretation is found already in the early Muslim–Christian debates, such as that between Timothy (d. 823) and the caliph al-Mahdi in 164/781 (r. 158/775–169/785). In the Sira of Ibn Hisham (d. 218/833) we find the passages of John on the Paraclete cited, according to the Syro-Palestianian version, which he transliterates as . The graphic likeness of this to (in the accusative) probably served to reinforce this interpretation. The Qur’anic name John the Baptist ( ) is evidently a reading of (a form attested among Christian Arabs at least until the tenth century). The Qur’anic name of the prophet Isaiah , is evidently a reading of (Sha‘ya), still pronounced this way by Christian Arabs today. The Qur’an’s faithfulness to its Christian sources As I said in the Introduction, one can detect Christian influences in the Qur’an in its narrative aspects: the narratives of the infancy of Mary, the annunciation of John the Baptist, the annunciation to Mary, the infancy miracles of Jesus, and so on. The majority of these texts stem from the apocryphal infancy Gospels, some of them from the canonical Gospels. With all of these parallel texts, one is struck by the faithfulness of the Qur’an to its source. One usually has the sense that the author of the Qur’an (I begin with the hypothesis that this religious document has a human author) seeks to reproduce that which he has gathered as faithfully as possible, sometimes without understanding it, or at least while understanding it in his own manner. This is evident in regard to the Holy Spirit (Ruh al-Qudus), of Christ the Word of God, of his conception without human intervention, or the fact that Christ created living things (by shaping them and breathing into them), and so on. 159 SAMIR KHALIL SAMIR However, the principle of coherence meant that the Qur’anic text had to add certain things to mitigate the effect of its reports (for example the addition of bi-idhn Allah – “by God’s leave” after the mention of Christ’s miracles), just as much as it had to give other certain words another meaning (thus ruh – “spirit” – is often taken with the meaning of angel), or even (quite commonly) to leave a phrase ambiguous. What type of influence? Jewish influence on the Qur’an is quite evident on the level of Biblical reports, cult, juridical regulations and traditions of daily life (without mentioning prayer and fasting before the “reform” of 623–4). Christian influence is more evident on level of New Testament reports and certain theological themes. Yet Christian influence seems to me evident no less on a cultural level. This is evident from the written Christian influence on certain proper names in the Qur’an, as presented earlier. This affirmation is confirmed by the Arab tradition which unanimously reports a Syriac or Nabatean (thus Christian) influence on Arabic writing. Meanwhile, one could bring up numerous Qur’anic Arabic words of Syriac origin that relate to writing or culture. 31 Judaeo-Christian/Gnostic influence is another matter. How could it have been a factor in the historical context of the Qur’an? In particular, how could the Gnostic perspective on the death of Christ on the Cross have arrived in Arabia in the seventh century? We know unfortunately very little on the Christianity of Arabia during this period. It is possible that the famous formula attributed to Theodoret of Cyrrhus (d. 458 or 466): “Arabia haeresium ferax” (“Arabia bearer [or mother] of heresies”) could be a key to the problem, even if it is unclear of which “Arabia” he is speaking.32 The same Theodoret teaches us, in the fourth book of his Ecclesiastical History, that Arabia was often a place of exile, for example of Pelagius in 367,33 and of various orthodox Christians a bit later. He is precise enough to add that they were sent to the furthest (eschatiás)34 regions of Arabia, presumably Arabia deserta. Some scholars believe that the Elchasites were refugees in Arabia, and this is another track that could illuminate certain Qur’anic passages. I do not think that there were proper Judaeo-Christian communities left in Arabia, but JudaeoChristian oral traditions could certainly have survived there until the time of the Qur’an’s origins. I offer the following working hypothesis: when Jerusalem was conquered by the Persians in 614 in the era of the Khosrow II Parviz (r. 590–628), and re-conquered by the Byzantines in 629–30 in the era of Heraclius (r. 610–41), the Jews were expelled from there. Is it possible that Judaeo-Christians expelled with them found refuge in Arabia, far from the empire, and then transmitted or reinforced the idea that someone was substituted for Christ on the Cross? As far as the chronology of the Qur’an, this event corresponds perfectly to sura four, which is datable to this period. Of course, this is only a hypothesis. 160 THE THEOLOGICAL CHRISTIAN INFLUENCE In brief, there is no need to demonstrate that there was a Christian influence on the Qur’an, in as much as this is apparent from the evidence of a number of narratives. What is more interesting is research on Qur’anic material that does not explicitly address Christian themes, yet still reveals this influence. Meanwhile, it should be remembered that discussion of “influence” is not opposed to the dogmatic position that the Qur’an was revealed, if one might accept the idea that the word “revealed” need not exclude human activity. At the same time, further philological, literary and theological research might introduce new elements that will further illuminate this matter. Notes 1 According to the original Yusuf ‘Ali translation (The Holy Quran, [Lahore: Muhammad Ashraf, 1938]) and not the later editions (Beltsville, Maryland: Amana, 1989-present) that reflect Saudi sponsored modifications. 2 In fact, other than Mary, no other woman is named in the Qur’an! 3 Cf. C.F. Gerock, Versuch einer Darstellung der Christologie des Korans, Hamburg: Perthes, 1839, 23. 4 See J. Henninger, “Spuren christlicher Glaubenswahrheiten in Koran,” Neue Zeitschrift für Missionswissenschaft 1, 1945, 306–7. 5 Cf. M. Hayek, Le Christ de l’Islam, Paris: Seuil, 1959, 83; and especially R. Arnaldez, Jésus fils de Marie, prophète de l’Islam, Paris: Desclée, 1980, 33–8. 6 See (pseudo-)Eutychius of Alexandria, The Book of the Demonstration (Kitab al-Burhan), P. Cachia (ed.) CSCO 192 and 209, trans. W.M. Watt CSCO 193 and 210, Leuven: Peeters, 1960 and 1961. 7 In the same way Marqus b. al-Qunbar, a Coptic exegete of the late twelfth century rejects that there was a Covenant with Muhammad with an allegory based on the three wives of Abraham. On this author see S.K. Samir, “Vie et œuvre de Marc Ibn al-Qunbar,” Christianisme d’Égypte: Mélanges René-Georges Coquin, Cahiers de la Bibliothèque Copte 9 (Louvain: Peeters 1995), 123–58. 8 Poem 65, first verse, The Divans of the Six Ancient Arabic Poets: Ennabiga, ‘Antara, Tharafa, Zuhair, ‘Alqama and Imruulqais, W. Ahlwardt (ed.), London: Trübner, 1870, 160 (reprinted Osnabrück: Biblio, 1972). 9 Gospel of Thomas 2:2–4; Arabic Infancy Gospel 36:1–2 and 46:102; Armenian Infancy Gospel 18:2; Pseudo-Matthew 27. Blachère, in a note on this verse (p. 82 in the 1 volume edition: Paris: Masonneuve, 1966), writes: “Ce miracle est mentionné dans l’Évangile de l’Enfance où il se présente sous deux formes sensiblement différentes. Dans la première, Jésus joue avec des camarades et, ayant façonné des figurines d’argile représentant des animaux, notamment des oiseaux, ceux-ci sur son ordre s’envolent; se posent et picorent. Dans l’autre, Jésus, le jour du sabbat, façonne des figurines d’argile représentant des oiseaux; un pharisien arrive et veut les détruire; Jésus frappe des mains et les figurines s’envolent. Cette seconde version est celle qui se trouve chez Pseudo-Matthieu et dans l’Évangile de Thomas.” 10 Cf. Ch. Luxenberg, Die syro-aramäische Lesart des Koran. Ein Beitrag zur Entschlüsselung der Koransprache, Berlin: Schiler, 2004, 315–17. 11 See also 1 Corinthians 15:45–9. 12 For this section see also Durrah Al-Haddad, al-Injil fi-l-Qur’an, Jounieh: Librairie pauliste: 1982, 290–6. 13 Cf. John 6:51: “I am the living bread which came down from heaven.” 14 That is, “Do not tempt God!” 161 SAMIR KHALIL SAMIR 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 See FV, 255–6. See FV, 213. An expression borrowed from Exodus 24:8. An expression borrowed from Jeremiah 31:31. Luke 22:20 ⫽ 1 Corinthians 11:25 (the latter verse does not include “which is shed for you”). See Matthew 20:28 ⫽ Mark 10:45; Romans 5:15, 19; 12:4; 1 Corinthians 10:17; 12:2, 14, 20; Hebrews 9:28. See § 4 at: http://www.vatican.va/holy_father/john_paul_ii/letters/2005/documents/ hf_jp-ii_let_20050313_priests-holy-thursday_en.html Arnaldez, Jésus fils de Marie, prophète de l’Islam, 173–85. Cf. Le Christ de l’Islam, 220–2, where the opinions of various commentators on the Table are recorded. Cf. H. Michaud, Jésus selon le Coran, Neuchatel: Delachaux et Niestlé, 1960, 57. See A. de Biberstein Kazimirski, Dictionnaire arabe-français, Paris: Maisonneuve, 1860, 1, 855b: “Subsistance journalière, pain quotidien.” On this section, see S. Kh. Samir, ⰆLa crocifissione di Cristo nel CoranoⰇ, in: Piero Coda e Mariano Crociata (Edd.), Il crocifisso e le religioni. Compassione di Dio e sofferenza dell’uomo nelle religioni monoteiste (Roma: Città Nuova, 2002) 49–82. Idris seems to correspond with the Biblical Enoch; cf. Genesis 5:24 and Hebrews 11:5. In Sufi texts the exclamation “Ya Ruh Allah!” addressed to Jesus, is frequently found. See Matthew 24:36; Mark 13:32. ‘Abdallâh Yûsuf ‘Alî, The Qurân, Beirut, Dar al-‘Arabiyya, n.d. This is taken almost literally by L. Bonelli, Il Corano, Milan: Hoepli, 19972, 89, n. 3, in relation to Q 4:159; and also by Federico Peirone, Il Corano, introduzione traduzione e commento, Milan: Mondadori, 1980, 2, 682, n. 57. See S. Kh. Samir, “La culture arabe chrétienne ancienne en interaction avec la pensée arabe musulmane,” Islamochristiana 8, 1982, 1–35 (in Arabic), esp. 14. I could not find the precise reference to this phrase in Theodoret’s writings (which are not available in the original Greek) or elsewhere. I also could not consult R.W. Smith, “Arabia Haeresium Ferax?” A History of Christianity in the Transjordan to C.E. 395, Dissertation, Miami University (Ohio), 1994. In a letter of December 19, 2005, Prof. Theresia Hainthaler explained to me: “I have not yet found the expression with Theodoret (and also not with Epiphanius, Sozomenus, Hippolytus, and so on.). I do not expect to find it with Theodoret. I have also not encountered the expression in Harnack’s Missionsgeschichte. Augustine (De Haeresibus 83), referring to Eusebius of Caesarea (Historia Ecclesiastica 6:37), speaks of representatives of a heresy whom he names Arabici (because they are to be found in Arabia and the heresiarch is unknown).” In a letter of February 24, 2006, after a long series of references, she concludes, “Hence it appears that this expression does not stem from a patristic or early medieval Author.” Theodoret Kirchengeschichte, Léon Parmentier (ed.), Berlin: Akademie Verlag, 1954, chapter 13.3 ⫽ 233.13. Cf. The Ecclesiastical History, Dialogues, and Letters of Theodoret, trans. B. Jackson, A Select Library Nicene and Post-Nicene Fathers of the Christian Church 3 (Grand Rapids, MI: Eerdmans, 1979), chapter. 12, 115b-16a: “Him, too, Valens relegated to Arabia, the divine Meletius to Armenia, and Eusebius, that unflagging labourer in apostolic work, to Thrace.” Ibid., Parmentier (ed.), chapter 18.5 (⫽ p. 240. 18–20) ⫽ Jackson, chapter 15, p. 118b: “Thereupon Valens ordered that they were to be separated into pairs and sent in different directions, some to Thrace, some to the furthest regions of Arabia, and others to the towns of the Thebaid.” 162 7 MARY IN THE QUR’AN A reexamination of her presentation Suleiman A. Mourad If we assume that the verses of the Qur’an reflect the religious milieu of the prophet Muhammad, then the Qur’anic stories about Mary should help us reconstruct the type (or types) of Christianity the nascent religion came into contact with in early seventh-century-CE Arabia. Moreover, it should allow for a more concrete understanding of the nature of the religious movement that Muhammad initiated and how it fits in the context of the other two monotheistic traditions, Judaism and Christianity. In medieval Islam, the Qur’anic references to Christianity and Christian figures received the attention of a variety of scholars, ranging from the pro-Christian apologists to the anti-Christian polemicists, and these debates have continued in modern scholarship. Certain Qur’anic references to Mary have been cited to prove that Islam venerates her, to point to historical inaccuracies and errors in the Qur’anic text, or even to suggest that Muhammad came into contact with heretical forms of Christianity, about which we know very little (derived primarily from scant references in the text of the Qur’an). In order to establish the degree to which the Qur’anic Mary differs from the one presented in Christian sources, I will examine three issues: the way the Qur’an identifies Mary and traces her lineage, the stories of the Annunciation and the birth of Jesus, and the flight of Mary and Jesus. This study will investigate whether or not these stories have any peculiarities for which one cannot find parallels in Christian canonical and extracanonical sources. I will also explore the way some Muslim exegetes dealt with particular verses relating to Mary, and to what extent their preformed polemical biases against Christianity, as well as intra-Muslim competition, shaped the way they interpreted these verses. The identity of Mary One of the most contentious issues about the Qur’anic Mary is her identity and lineage. The relevant verses read Remember the words of ‘Imran’s wife. “Lord,” she said, “I dedicate to Your service that which is in my womb. Accept it from me. You alone 163 SULEIMAN A. MOURAD hear all and know all.” And when she was delivered of the child, she said: “Lord, I have given birth to a daughter” – God well knew of what she was delivered: the male is not like the female – “and I have called her Mary. Protect her and all her descendants from Satan, the Accursed One.” (Q 3:35–6)1 Sister of Aaron, your father was never a whore-monger, nor was your mother a harlot. (Q 19:28) And in Mary, Imran’s daughter, who preserved her chastity and into whose womb We breathed Our Spirit, who put her trust in the words of her Lord and His scriptures, and was truly devout. (Q 66:12) Verse 3:35 refers to Mary’s mother as wife of Amram (Arabic, ‘Imran) and verse 66:12 refers to her as Amram’s daughter; verse 19:28 refers to her as Aaron’s sister. At first glance, one might assume, as many modern scholars do, that the Qur’an is identifying Mary as the daughter of Amram and the sister of Aaron, which has led many to argue that Muhammad confused Mary the mother of Jesus with Miriam, the sister of Aaron and Moses, whose father was the biblical Amram.2 Medieval Muslim exegetes made similar assumptions. Some argued that the title of the third Chapter of the Qur’an, Al ‘Imran (Amram’s Family), is a reference to Mary’s immediate family. Al-Tabari (d. 310/922), for example, relates on the authority of Ibn Ishaq (d. 150/767), the famous narrator of the life of the prophet Muhammad, that in the context of verse 3:35, the woman identified as Amram’s wife is none other than Mary’s mother Hanna (Anna), and that the lineage of Amram, Mary’s father, is as follows: Amram son of Josiah son of Amos son of Manasseh son of Hezekiah son of Ahaz son of Jotham son of Eleazar son of Amaziah son of Joash son of Uzziah son of Joram son of Jehoshaphat son of Asaph son of Abijah son of Rehoboam son of Solomon son of David son of Jesse.3 (Al-Tabari, Jami’ al-bayan, 3, 2344) But this lineage is almost exactly the same as the lineage we find given to Joseph, Mary’s husband, in the Matthew 1:6–11. Joseph son of Jacob son of Matthan son of Eleazar son of . . . Josiah son of Amos son of Manasseh son of Hezekiah son of Ahaz son of Jotham son of Uzziah son of Joram son of Jehoshaphat son of Asaph son of Abijah son of Rehoboam son of Solomon son of David son of Jesse.5 In fact, al-Tabari also relates the same Gospel lineage of Joseph prior to the report he quotes from Ibn Ishaq.6 Interestingly, he also identifies Mary’s grandfather as Matthan (Mary daughter of Amram daughter of Matthan), who is the same grandfather he provides for Joseph (Joseph son of Jacob son of Matthan).7 But even in 164 MARY IN THE QUR’ AN some medieval Islamic sources, this Gospel lineage of Joseph is verbatim applied to Mary, for instance in Ta’rikh madinat Dimashq by Ibn ‘Asakir (d. 574/1176).8 Thus we have, on the one hand, confusion on the part of some medieval Muslim sources that mistook Joseph’s lineage for that of Mary. On the other hand, if we accept Ibn Ishaq’s report, then Mary becomes a descendent of David, in which case she cannot also be from an Aaronic lineage – yet it is on the basis of her Aaronic lineage that Mary could serve in the Temple. We know from such Christian sources as the Protevangelium of James that Mary’s father was named Joachim.9 Although these reports about Mary’s family have no real historical value,10 once they were introduced they became an accepted part of the way Christians and others identified her. Therefore, there are no grounds for arguing that Mary the daughter of Amram could have been a correct reference to Mary, the mother of Jesus. But could it be accurate in another sense: if the term “daughter” does not necessarily mean direct child, then it is very plausible that the reference is to Mary, the mother of Jesus.11 In Christian canonical sources, the term son does not always denote a direct child: in Matthew 1:1, for example, Jesus is identified as Jesus the Messiah, the son of David, the son of Abraham. Leaving aside the issue of Mary’s virginal conception, Jesus is not the direct son of David, nor for that matter of Abraham. Similarly, in Matthew 1:20, Joseph is identified as Joseph son of David, although his father’s name was Jacob. In Christian canonical texts, then, the term son denotes, besides direct parentage, one’s ancestral descent. Is the Qur’an following the same pattern by identifying an individual not only by his or her parents, but also by his or her ancestral lineage? In the Qur’an, too, the terms ibn and bint (and their derivatives) do not only mean “direct child,” but are also used in the sense of “descendents,” as in the cases of Bani Isra’il in verses 2:246, 3:49, and 5:72, which indicates the Israelites as a people, and not only Jacob’s direct children, and Bani Adam in verses 7:35, 17:70, and 36:60, which refers to all human beings, and not strictly Adam’s direct children. Similarly, the Qur’anic akh and ukht (and their derivatives) do not always indicate a sibling relationship. In twenty-eight cases, they refer either to a tribal relationship (e.g. verse 7:73: And to Thamud We sent their kinsman [akhahum] Salih), a religious bond (e.g. verse 3:103: He united your hearts, so that you are now brothers [ikhwanan] through His grace), or an ancestor/predecessor relationship (e.g. verse 7:38: As it enters, every community will curse the one that went before it [ukhtaha]). Thus one cannot argue on the basis of the most common meaning of the terms bint and ukht that their use in the Qur’an indicates only daughter and sister; clearly they are not limited to these two meanings. The expression sister of Aaron, moreover, occurs in the Qur’anic reference to the questioning of Mary in the Temple. It is especially appropriate in this context for the questioners, the Temple’s priests, to magnify Mary’s moral transgression (her pregnancy) by appealing to her ancestor Aaron, whose descendents are the only Israelites qualified to serve in the Temple, where Mary herself was raised. In other words, Mary as a descendent of Aaron is expected to keep the purity of the sanctuary, rather than defile it by supposedly committing the shameful act that 165 SULEIMAN A. MOURAD would lead to a pregnancy. Here too, there are no grounds on which to argue that the Qur’an is identifying Mary as literally the sister of Aaron. Medieval Muslim exegetes argued that Mary’s identification as the sister of Aaron was either an allegorical reference to Mary’s descent from the biblical Aaron, or a reference to a contemporary Aaron who was a relative of hers and was renowned for his piety; the latter would explain away any confusion that might arise in the reader’s mind as a result of the centuries that separate Aaron’s time from Mary’s. Al-Tabari, for instance, maintains, on the basis of a hadith attributed to the prophet Muhammad, that it must be a reference to a relative of Mary’s named Aaron.12 Al-Tabari also relates other opinions, which he does not agree with, that Aaron in verse 19:28 is a reference to the biblical Aaron, the brother of Moses, because Mary is identified here with her ancestor.13 Al-Tha‘labi (d. 425/1037) adds two more possible identifications: the Aaron of verse 19:28 was either Mary’s brother who became famous for his piety, or a man known for his immorality and decadence, to whom she is being compared as a result of her pregnancy.14 The references to both Amram and Aaron must, then, be taken allegorically. This leads me to argue that Amram of sura Al ‘Imran (the verse in question is 3:33: God exalted Adam and Noah, Abraham’s descendants and the descendants of ‘Imran above the nations) is the biblical Amram, father of Moses, Aaron, and Miriam, and the ancestor of Mary the mother of Jesus. Moreover, the reference to Mary’s mother as Amram’s wife is a reference to biblical Amram in the sense that Mary’s mother was married to a descendant of his. The annunciation and the birth of Jesus Modern western scholars have missed some subtle differences in the two Qur’anic stories of the Annunciation, Mary’s conception of Jesus, and her delivery, which provide us with clear clues as to their origin. The annunciation story in sura Al ‘Imran 3:42–9 derives from the extracanonical text the Protevangelium of James 11, whereas the version in sura Maryam 19:17–21 derives from the Gospel of Luke 1–2 or the corresponding sections in Tatian’s Diatessaron.15 Both the Gospel of Luke and sura Maryam start with the annunciation of John to Zechariah, followed by the annunciation of Jesus to Mary. Sura Al ‘Imran and the Protevangelium of James, however, report the annunciation of Jesus following the story of the annunciation of Mary and her upbringing in the Temple, which are not mentioned in either the Gospel of Luke or sura Maryam. Moreover, in sura Maryam and the Gospel of Luke, the angel tells Mary that she will conceive a boy; whereas in sura Al ‘Imran and the Protevangelium of James, Mary is told that she will conceive the word of God. These close similarities are proofs that here the Qur’an is borrowing canonical and extracanonical material that was used by mainstream Christians;16 the Protevangelium of James was heavily used in eastern Christian liturgical collections (in western Christianity, although the text was banned, the Protevangelium of James was reworked in the form of the Gospel of 166 MARY IN THE QUR’ AN Pseudo-Matthew). It is the most important noncanonical gospel to provide information about Mary’s life, thus inspiring the many frescos, mosaics and other artistic representations of her life;17 the best examples can still be seen in the frescos in the Chora (Kariye) Church in Istanbul.18 Other passages common to the Qur’an and a Christian extracanonical text, in this case the Gospel of Pseudo-Matthew, are the details relating to Mary’s delivering Jesus by a palm tree and the miracle that followed. The Qur’an relates Mary’s delivery in sura Maryam. The palm tree episode covers verses 19:22–6; it is held to be the only part of the conception and delivery story that has no known Christian origin. Thereupon she conceived him, and retired to a far-off place. And when she felt the throes of childbirth she lay down by the trunk of a palm-tree, crying: “Oh, would that I had died before this and passed into oblivion!” [And he (Jesus), from below her],19 cried out to her: “Do not despair. Your lord has provided a brook that runs at your feet, and if you shake the trunk of the palm-tree it will drop fresh ripe dates in your lap. Therefore eat and drink and rejoice. The story in the Gospel of Pseudo-Matthew 20, which was composed sometime between the middle of the sixth century and the end of the eighth century, is reported as follows: And it came to pass on the third day of their journey, while they were walking, that Mary was fatigued by the excessive heat of the sun in the desert; and, seeing a palm-tree she said to Joseph, “I should like to rest a little in the shade of this tree.” Joseph therefore led her quickly to the palm and made her dismount from her beast. And as Mary was sitting there, she looked up to the foliage of the palm and saw it full of fruit and said to Joseph, “I wish it were possible to get some of the fruit of this palm.” And Joseph said to her, “I am surprised that you say so, for you see how high the palm-tree is, and that you think of eating its fruit. I am thinking more of the want of water because the skins are now empty, and we have nothing with which to refresh ourselves and our cattle.” Then the child Jesus, reposing with a joyful countenance in the lap of his mother, said to the palm, “O tree, bend your branches and refresh my mother with your fruit.” And immediately at these words the palm bent its top down to the very feet of Mary; and they gathered from it fruit with which they all refreshed themselves. And after they had gathered all its fruit it remained bent down, waiting the order to rise from him who had commanded it to bend down. Then Jesus said to it, “Raise yourself, O palm, and be strong and be the companion of my trees which are in the paradise of my Father; and open from your roots a vein of water which is hidden in the earth and let the waters flow, so that we may 167 SULEIMAN A. MOURAD quench our thirst.” And it rose up immediately, and at its root there began to gush out a spring of water exceedingly clear and cool and sparkling. And when they saw the spring of water, they rejoiced greatly and were all satisfied, including their cattle and their beasts and they gave thanks to God. (Elliott (ed.), The Apocryphal New Testament, 95–620) The story as it appears in sura Maryam is obviously much shorter than the one in Pseudo-Matthew. Moreover, in sura Maryam the story takes place while Mary is in labor, and the setting is identified only as a remote place. In Pseudo-Matthew, Jesus has already been born, and the incident occurs during the flight to Egypt. It appears, then, that we are dealing here with two stories that have in common the theme of the palm tree miracle. My research has identified the source for both texts as the Greek myth of Leto’s labor and the birth of Apollo. Leto, who was desperately trying to hide herself from the angry Hera, sought the remote island of Delos. Aggrieved and distressed, she sat by a palm tree alongside the Inopos River and there delivered Apollo. There are several presentations of this myth in Hellenistic lore; the most famous is the version in the Hymn to Delos by Callimachus (d. ca. 240 BCE). So you spoke, and she let go the long pang of her wandering with a sigh and sat down by the stream Inopos, abounding then in waters sprung from the earth in the season when the Nile comes at its greatest, cascading down the Ethiopian plateau. She untied her belt, leaning backwards, her shoulders against the trunk of a palm tree, utterly exhausted, her skin glistering with sweat, and said, in a whisper almost, “Why, boy, why so hard on your mother? Here, darling, is your island, sailing on the sea. Be born, boy, be born and come, gently, from the womb.” (Callimachus, Hymn to Delos, 205–1421) This legend was widely known in the Greek, Hellenistic, and Roman worlds. Brief allusions to it are frequent in classical literature, as in Homer’s Odyssey,22 the Hymn to Apollo,23 Euripides’ Hecuba,24 Thucydides’ Peloponnesian War,25 Cicero’s Laws,26 and Pliny’s Natural History.27 The myth of Leto giving birth to Apollo by the trunk of a palm tree is also depicted on several Hellenistic vases, as well as on other artwork dating to as early as the sixth century BCE.28 A number of variants of this myth are recorded; for example Plutarch (d. after 119 CE) reports that Delos is a mountain, not an island, and that the olive and the palm are two springs, not trees. A little below the marshes stands the temple of Apollo Tegyraeus. Here, according to the story, the god was born; and the neighbouring mountain is called Delos, and at its base the river Melas ceases to be spread out, and behind the temple two springs burst forth with a wonderful flow of 168 MARY IN THE QUR’ AN sweet, copious, and cool water. One of these we call Palm, the other Olive, to the present day, for it was not between two trees, but between two fountains, that the goddess Leto was delivered of her children. (Plutarch [d. after 119 CE], Life of Pelopidas, 16.3–429) All the various Hellenistic and Latin variants of the original myth of Leto giving birth to Apollo by a palm tree reflect the borrowing and adaptation by groups who reshaped it for their own objectives and needs. Appropriations of ancient myths were common in the ancient world, and the early Christians were no exception. The palm-tree story that found its way to sura Maryam is a reworking of Leto’s labor. It is about a distressed pregnant woman (Leto/Mary) who seeks an isolated place (Delos/a remote spot), sits by the trunk of a palm tree next to a stream (Inopos/a brook), and delivers a holy child (Apollo/Jesus). It is nevertheless unlikely that the myth of Leto was the direct source for sura Maryam. As was aforementioned, the concise version found in the latter has two parts: Mary’s labor and delivery, and the miracle. We might therefore suspect that there was a stage when Leto’s myth was borrowed and applied to Mary. This would reflect an attempt by a Christian group, probably converts who had previously worshipped Leto and Apollo, to modify the story by replacing them with Mary and Jesus. After that, the appropriated story was appended with a miracle typical of Jesus: the dead palm tree providing fruit and water for Mary. Indeed, a possible Christian group among whom such a story could have been circulated is the Christian community of Najran, in West Arabia, who, according to Ibn Ishaq and al-Tabari, used to worship a palm tree before converting to Christianity.30 If that was the case, then adapting the Leto myth to Mary would have permitted them to keep part of their belief, yet give it a Christian guise.31 The flight of Mary and Jesus The Qur’anic story of the flight of Mary and Jesus to escape what is described in the gospels as the Massacre of the Innocents (Matthew 2:13–18), provides a vague description of the site to which they fled (Q. 23:50). We made the son of Mary and his mother a sign to mankind, and gave them a shelter on a rabwatin dhati qararin wa-ma’inin (peaceful hillside watered by a fresh spring). Verse 23:50 does not specify the location of the rabwa, and only a few Muslim historians have associated this passage with the flight to Egypt. For instance, al-Tabari says in his Ta’rikh that the rabwa is located in Egypt, and quotes the Qur’anic verse in question as a proof of the flight’s historicity.32 Similarly, al-Maqrizi (d. 845/1442 CE) locates the rabwa in the Egyptian town of Bahnasa, because, according to him, the Copts say so.33 169 SULEIMAN A. MOURAD The majority of Muslim historians and exegetes, however, place the rabwa in Syria (Bilad al-Sham). Muqatil b. Sulayman (d. 150/767) states in his Tafsir that the rabwa is the Ghuta, the fertile agricultural land east and south of Damascus.34 Al-Tabari, in his Qur’an commentary, quotes several traditions that were known to him from a number of early exegetes about the interpretation of this verse, as if to contradict what he says in his Ta’rikh. According to these different traditions, the rabwa refers to Ramla, Jerusalem, or Damascus.35 Later exegetes, such as al-Tha‘labi, al-Mawardi (d. 450/1058), Ibn ‘Atiyya al-Andalusi (d. 543/1147), and al-Suyuti (d. 911/1505), reproduce in their respective Tafsirs the range of choices that one finds in al-Tabari.36 Ibn ‘Atiyya even adds one more location, Bethlehem, justifying it on the basis that Jesus was born there,37 and al-Suyuti adds Alexandria.38 The two most popular candidates among medieval Muslim exegetes were probably Ramla and Damascus. In the case of Damascus, it is evidently because of its Ghuta: if there is any place on earth that can be described as dhati qararin wa-ma‘inin, it must be Damascus, with its lush and fertile Ghuta. As for Ramla, it is popular among some scholars because of a hadith, although a very suspicious one, quoted by Ibn ‘Asakir,39 Ibn Kathir (d. 774/1373),40 al-Suyuti,41 and others, in which the prophet Muhammad goes to check on a sick companion of his named al-Aqra‘ b. Shafiy al-‘Ikki. Al-Aqra‘ was worried that he might die of his sickness, and the prophet assured him that he would not, and that he would live and immigrate to Syria where he would die in the rabwa named in the Qur’an. Al-Aqra‘ ended up dying in Ramla; therefore, the rabwa must be Ramla. Some exegetes, however, objected to the interpretation that the rabwa is a reference to Ramla. They maintain that the place as described in the Qur’an has the characteristics of being qarar and ma‘in, unlike Ramla. Al-Tabari, for instance, who resided in Ramla for a short time while on his way to Egypt, says that the Qur’an speaks of an elevated flat place with running water, which cannot be the description of Ramla because the city does not have running water anywhere around it.42 Egypt was also rejected as a possible location of the rabwa by Ibn ‘Atiyya. For him, the Ghuta of Damascus is the perfect place to be described as dhati qararin wa-ma‘inin, and no story is known that confirms that Mary and Jesus went to Egypt.43 But the strangest interpretation of the location of rabwa is undoubtedly the one provided in Shi‘ite commentaries. For instance, the Twelver-Shi‘ite exegete al-Tabrisi (fl. sixth/twelfth century CE) quotes in his Tafsir traditions attributed to the Shi‘ite imams Muhammad al-Baqir (d. ca. 117/735 CE) and Ja‘far al-Sadiq (d. 148/765 CE) that identify the rabwa with the town of Hira near Kufa, Iraq.44 Al-Tabrisi, who was well acquainted with earlier Qur’an commentaries, certainly among them al-Tabari’s Tafsir, adds that qarar refers to the mosque of Kufa and ma‘in to the Euphrates River.45 Ibn ‘Asakir, whose favored location of the rabwa is in his hometown Damascus, also mentions other locations; one of these, 170 MARY IN THE QUR’ AN which he attributes to Ja’far al-Sadiq, identifies the rabwa as the town of Najaf, near Kufa.46 It is clear that connecting the rabwa to the area of Kufa is related to Shi‘ite propaganda about the religious merits of Kufa and its environs. After all, Kufa was the capital city of ‘Ali, and next to it are Najaf, where he was buried, and the holy plain of Karbala, where his son al-Husayn was killed in 61/680. A similar case can be made for the other sites identified for rabwa. Most early exegetes, especially al-Tabari and Ibn ‘Asakir, specify the chains of authorities (isnads) that passed each report from its presumed initial informant down to them. The traditions that locate the rabwa in Ramla include Syrian informants, some of whom lived in or around Ramla; the most notable of them is the traditionist Abu ‘Utba ‘Abbad b. ‘Abbad al-Khawwas al-Ramli (d. 185/801). The tradition that identifies it with Egypt includes Egyptian informants, such as the traditionists ‘Abdallah b. Lahi‘a al-Misri (d. 174/790) and al-Layth b. Sa‘d al-Misri (d. 175/792). The identification with Shi‘ite holy sites in Iraq include, as was seen earlier, Shi‘ite informants. Yet the informants who associate the rabwa with either Jerusalem or Damascus dot not always have clear regional affiliation with Syria, because the associations were either based on the town’s holiness – in the case of Jerusalem – or famed Ghuta (in the case of Damascus). Even those hadiths that were forged in the first place to boost the sanctity of a city over others became later the basis for many exegetes and historians to establish the location of the rabwa. What we see here is competition among certain medieval Muslim scholars to boost the sanctity ( fada‘il) of a particular region or town by associating with it major religious figures such as Mary and Jesus, which ultimately became essential components of later Muslim perceptions of the history of those places. With the exception of Egypt, the traditions about the location of the rabwa could not have had historical basis. In particular, it cannot be argued that there were Christians in Arabia who located the flight of the holy family in Jerusalem, Damascus, Ramla, Bethlehem, Kufa, or Najaf, and at the same time used in their version of scripture the Gospel of Luke or the Diatessaron, the Protevangelium of James, and the Gospel of Pseudo-Matthew, all of which clearly associate the flight of the holy family with Egypt.47 It is more likely that Muslim scholars felt at ease in contradicting Christian sources, especially given the overwhelming belief, propagated in anti-Christian Muslim polemics such as al-Jahiz’s hujaj al-nubuwwa and al-Radd ‘ala al-nasara, that the Christians had tampered with the text of the Gospel and distorted the message and life of Jesus.48 Accordingly, they could easily suggest alternatives to Christian traditions, which would not challenge their own credibility. In the case of the rabwa, the alternatives seem to have been dictated primarily by mere regional interests – though in a few cases they seem to represent pure speculation. The varying identifications of its location serve to fit certain regional claims, whether political or religious, between rival cities and areas during the medieval Islamic period. 171 SULEIMAN A. MOURAD Conclusion The Qur’anic stories about Mary, especially those examined in this study, do not reflect any peculiarities about her for which one fails to find parallels in Christian canonical and extracanonical sources. Her identification as Amram’s daughter and Aaron’s sister are meant to highlight her biblical heritage, and were not meant to inform the audience of the Qur’an about her direct father and brother. Some of the Qur’anic references to Mary and Jesus point to clear influence from canonical texts, particularly the Gospel of Luke or the quasi-canonical Diatessaron, as well as extra-canonical texts now considered apocryphal that were heavily used in the Near East before and around the time of the emergence of Islam, such as the Protevangelium of James. These influences could have been exercised via popular mediums, and not necessarily through direct textual borrowing. Therefore, if one assumes that the Qur’an does reflect the religious milieu of the prophet Muhammad and his movement, then they were in contact with Christian groups who were using the Gospel of Luke or the Diatessaron, and the Protevangelium of James, among other sources. This, in my opinion, points in one direction: these Christian groups must have observed a mainstream type of Christianity, and could not have been heretical Christians. Another conclusion that may be drawn from this study is that medieval Muslim exegetes, for the most part, did not necessarily exhibit any solid knowledge of Arabian Christianity, or the context of these Qur’anic stories. In other words, when we move outside of Arabia, the Qur’anic references to Mary, and to Jesus, were interpreted in accordance with the types of Christianity these medieval Muslim exegetes came into contact with, and that was primarily in Iraq, Syria, and Egypt. We cannot credit them with any solid knowledge of Arabian Christianity, and they had no idea about its history and beliefs. Some of these exegetes were also involved in polemics against Christian groups in their midst, or were influenced by Muslim polemicists, and subsequently read the Qur’anic references to Christianity, including those to Mary, from those perspectives. Their interpretations of these references varied in accordance with their preoccupation with polemics or apologetics, or their concern with establishing the sanctity of one region over another. Moreover, they were not eager to understand exactly the form of Christianity with which the prophet Muhammad came into contact, probably out of fear that such an investigation might imply their endorsement of a Christian influence on Muhammad, the Qur’an, and the religion of Islam, which gradually became more and more unacceptable in Islamic religious discourse. Notes 1 All English translations of the Qur’an are those of N.J. Dawood (translator), The Koran (Penguin). 2 This issue is so diffused in medieval and modern scholarship on the Qur’an that, for instance, N.J. Dawood felt obliged to comment on verse 19:28 by dismissing any 172 MARY IN THE QUR’ AN 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 possible confusion in the Qur’anic text between Miriam and Mary: see The Koran, 215. On this trend, see also N. Geagea, Mary in the Koran: A Meeting Point between Christianity and Islam, Trans. L.T. Fares (ed.), New York: Philosophical Library, 1984, 61–4; N. Robinson, Christ in Islam and Christianity, Albany, NY: State University of New York Press, 1991, 63. For examples of scholarship in which such accusations are produced, see H. Michaud, Jésus selon le Coran, Neuchâtel: Éditions Delachaux et Niestlé, 1960, 17; D. Sidersky, Les origines des légendes musulmanes dans le Coran et dans les vies des prophètes, Paris: Geuthner, 1933, 141. See also idem, The History of al-Tabari. Volume IV: The Ancient Kingdoms, M. Perlmann (ed.), Albany, NY: State University of New York Press, 1987, 102–3. Jami‘ al-bayan ‘an ta’wil ’ay al-Qur’an, Muhammad ‘Ali Baydun (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1992. The order is reversed for the sake of comparison. All selections from the New Testament are taken from The HarperCollins Study Bible (NRSV). See al-Tabari, The History of al-Tabari. Volume IV, 103. See al-Tabari, The History of al-Tabari. Volume IV, 102–3. See Ibn ‘Asakir, Ta’rikh madinat Dimashq, ‘Umar al-‘Amrawi and ‘Ali Shiri (eds), Beirut: Dar al-Fikr, 1995–2001, 70, 75. See The Apocryphal New Testament, J.K. Elliott (ed.), Oxford: Clarendon Press, 1993, 57; New Testament Apocrypha, Wilhelm Schneemelcher (ed.), trans. R.M. Wilson, Cambridge: Clarke, 1991, 1, 426–7. See Elliott, The Apocryphal New Testament, 51. On this see also J. Henninger, “Spuren christlicher Glaubenswahrheiten in Koran,” Neue Zeitschrift für Missionswissenschaft 1, 1945, 306–7. Al-Tabari, Jami‘ al-bayan, 8, 336. Al-Tabari, Jami‘ al-bayan, 8, 335–6. Al-Tha‘labi, Al-Kashf wa-l-bayan fi tafsir al-qur’an, Abu Muhammad Ibn ‘Ashur (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 2004, 4, 174; see also al-Tabrisi, Majma‘ al-bayan f i tafsir al-qur’an, Hashim al-Mahallati (ed.), Beirut: Dar Ihya’ al-Turath al-‘Arabi, 1992, 6, 662. For the purposes of my argument, it does not matter whether the source of the annunciation story in sura Maryam was the Gospel of Luke or the Diatessaron. My thesis is that the Qur’anic stories about Mary and Jesus derive from a variety of Christian scriptural texts. For a more elaborate discussion of this, see S. Mourad, “On the Qur’anic stories about Mary and Jesus,” Bulletin of the Royal Institute for Inter-Faith Studies 1 (1999) 2, 13–24. See R.F. Hock, The Infancy Gospels of James and Thomas, Santa Rosa, CA: Polebridge Press, 1995, 27–8. See ⁄. Akvit, The Museum of Chora Mosaics and Frescoes, Istanbul: Akvit Kültür Turizm Sanat Ajans Ltd., 2002. Text between brackets is my emendation to Dawood’s translation. Cf. Schneemelcher, New Testament Apocrypha, 1, 463. The translation is that of Frank Nisetich in The Poems of Callimachus, Oxford: Oxford University Press, 2001, 45. Homer (fl. ninth or eight century BCE), Odyssey, 6, 163–4. Hymn to Apollo, 115–19. Euripides (d. 406 BCE), Hecuba, 455–61. Thucydides (d. after 404 BCE), Peloponnesian War, 3, 104. Cicero (d. 43 BCE), Laws, 1, 1. Pliny (d. 79 CE), Natural History, 16. 89. See Lexicon Iconographicum, 6, 2, 130 (no. 10), explanation 6, 1, 258 (no. 10); also the example from the fourth century BCE in 6, 2, 130 (no. 6), explanation 6, 1, 258 (no. 6). 173 SULEIMAN A. MOURAD 29 The translation is that of Bernadotte Perrin in Plutarch’s Lives, vol. 5 (Loeb Classical Library, 1917). 30 For the earliest account of the practice of worshipping a palm tree by the people of Najran before their conversion to Christianity, see Ibn Isaq, al-Sira al-nabawiyya, Mustafa al-Saqqa et al. (eds), Beirut: Dar al-Khayr, 1990, 1, 29. The same account is repeated in al-Tabari, Ta’rikh al-rusul wa-l-muluk, M. J. De Goeje et al.(eds), Leiden: Brill, 1879–1901, 1, 922; English translation, The History of al-Tabari. Volume V: The Sasanid, the Byzantines, the Lakhmids, and Yemen, trans. C.E. Bosworth, Albany, NY: State University of New York Press, 1999, 198. 31 For a more detailed discussion of this, see S. Mourad “From Hellenism to Christianity and Islam: The Origin of the Palm-tree Story concerning Mary and Jesus in the Gospel of Pseudo-Matthew and the Qur’an,” OC 86, 2002, 206–16. 32 Al-Tabari, Ta’rikh, 1, 729; The History of al-Tabari. Volume IV, 116. 33 Al-Maqrizi, Kitab al-Mawa‘iz wa-l-i‘tibar fi dhikr al-khitat wa-l-athar, n.e., Cairo: Maktabat al-Thaqafa al-Diniyya, 1987, 1, 237. 34 Muqatil b. Sulayman, Tafsir Muqatil b. Sulayman, ‘Abdallah Muhammad Shahata (ed.), Cairo: al-Hay’a al-Misriyya al-‘Amma li-l-Kitab, 1979–89, 3, 158. 35 Al-Tabari, Jami‘ al-bayan, 9, 217–20. 36 Al-Tha‘labi, Al-Kashf wa-l-bayan, 4, 327; al-Mawardi, Tafsir al-Mawardi, al-Sayyid b. ‘Abd al-Maqsud b. ‘Abd al-Rahim (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, n.d., 4, 55–6; Ibn ‘Atiyya al-Andalusi, al-Muharrar al-wajiz f i tafsir al-kitab al-‘aziz, ‘Abd al-Salam Muhammad (ed.), Beirut: Dar al-Kutub al-‘Ilmiyya, 1993, 4, 145–6; al-Suyuti, al-Durr al-manthur fi l-tafsir bi-l-ma’thur, Beirut: Dar al-Kutub al-‘Ilmiyya, 2004, 7, 17–19. 37 Ibn ‘Atiyya al-Andalusi, al-Muharrar al-wajiz, 4, 145. 38 Al-Suyuti al-Durr al-manthur, 7, 18. 39 Ibn ‘Asakir, Ta‘rikh madinat Dimashq, 1, 209–12. 40 Ibn Kathir, Tafsir al-qur’an al-‘azim, n.e. Beirut: Dar al-Jil, 1990, 3, 239. 41 Al-Suyuti, al-Durr al-manthur, 7, 18. 42 Al-Tabari, Jami‘ al-bayan, 9, 219. 43 Ibn ‘Atiyya al-Andalusi, al-Muharrar al-wajiz, 4, 145–6. 44 Al-Tabrisi, Majma‘ al-bayan, 7, 144–5. 45 Al-Tabrisi, Majma‘ al-bayan fi, 7, 145. Interestingly, the earlier Twelver-Shi‘ite exegete al-Tusi (d. 460/1068), whose Tafsir was one of the principal sources that al-Tabrisi depended upon, does not mention these reports: see al-Tusi, al-Tibyan fi tafsir al-Qur’an, Ahmad Qasir al-‘Amili (ed.), al-Najaf: Matba‘at al-Nu‘man, 1962, 7, 373. 46 Ibn ‘Asakir, Ta’rikh madinat Dimashq, 1, 212–13. 47 I am not suggesting that the flight is necessarily historical. 48 On this, see also G.S. Reynolds, A Muslim Theologian in the Sectarian Milieu: ‘Abd al-Jabbar and the Critique of Christian Origins, Leiden: Brill, 2004, 83ff. 174 8 THE ALEXANDER LEGEND IN THE QUR’AN 18:83–102 Kevin van Bladel In 1889 E.A. Wallis Budge edited a few Syriac texts about Alexander the Great including the Syriac version of the Alexander Romance of Pseudo-Callisthenes. Among these was the first edition of a Syriac work called Neshanâ dîleh d-Aleksandrôs, roughly “The Glorious Deeds of Alexander,” extant in the same five manuscripts as the Syriac Alexander Romance.1 Though often discussed in the context of the Alexander Romance tradition, and clearly inspired by traditions about Alexander’s conquests like the Romance, this Neshanâ is nevertheless an entirely different work with its own history and a different story to tell (to be dealt with later in detail). Budge named it “A Christian Legend Concerning Alexander” to distinguish it from the Alexander Romance itself. Recent scholarship has shortened this name to “the Alexander Legend” to distinguish it from the Alexander Romance. I follow this convention here. The next year (1890), Theodor Nöldeke published his study of the Alexander Romance, much of which was based on the Syriac version newly available in Budge’s edition. In this he also devoted a few pages to the Alexander Legend, arguing that it was in fact the source for an episode in the Qur’an, specifically the Qur’anic story of Dhu l-Qarnayn (Q 18:83–102).2 He stated that the Alexander Legend must have been transmitted orally to Muhammad along with the other ancient biblical and traditional stories circulating in the environment of Mecca.3 To prove this relationship Nöldeke indicated a few specific, important elements of the story of Alexander’s journeys appearing in both the Syriac Alexander Legend and the Qur’an. In the century since then, his discovery seems to have become almost forgotten in Qur’anic studies. For example, the recent Encyclopaedia of the Qur’an article “Alexander” does not even mention the Syriac Alexander Legend or Nöldeke’s thesis on the matter, though there could be no more appropriate place for it.4 Moreover, some recent scholarship has brought considerable confusion into the study of Alexander stories in relation to the Qur’an.5 The subject therefore deserves to be revisited. As I hope to show, it still has much more to offer than even Nöldeke expected. 175 KEVIN VAN BLADEL The present investigation will first show that Nöldeke was basically correct in his view: the Qur’an 18:83–102 is a retelling of the story found in this particular Syriac text. But that is just the beginning of the matter. Recent publications by scholars of Syriac and Greek apocalyptic texts of the early seventh century, especially several articles by G.J. Reinink, offer a precise understanding of the context in which this Syriac Alexander Legend was composed and its political and religious purposes in that context. These studies make it possible to shed new light on the use of the Alexander Legend’s story in the Qur’an and on the concerns of Muhammad’s community. Furthermore, once the affiliation of the Arabic and Syriac texts is established and the character of that affiliation is identified, it is possible to demonstrate (perhaps unexpectedly) the reliability of the traditional lexicography as well as the soundness of the Arabic text of this Qur’anic passage. All of these matters will be discussed later. I am deliberately avoiding entering into a discussion of other texts related to the Syriac Alexander Legend identified by previous scholars. Traces of the ancient story of Gilgamesh are found in the Syriac Alexander Legend and in Q 18:83–102. That these traces appear in both is unsurprising since both tell essentially the same story. But some scholars have argued that the passage immediately preceding the Dhu l-Qarnayn episode in the Qur’an, a story about Moses (Q 18:60–82), also contains different traces of the Gilgamesh story. This is a matter of decades-long controversy and it deserves further special studies of its own.6 Since two adjacent episodes in Qur’an 18 seem to contain material derived from the Gilgamesh story, modern scholars have tended to search for a single source common to both of them. Medieval Qur’an commentaries associated the two episodes together, too, though it seems for different reasons, with the result that the Qur’an commentaries are dragged into the modern confusion. I will also avoid discussing other texts in Syriac and in Greek that draw material from the Syriac Alexander Legend. One of these is the so-called Song of Alexander (also called Alexander Poem in modern scholarship), falsely ascribed to Jacob of Serugh (d. 521). It was composed several years after and in reaction to the prose Alexander Legend, but the story it relates is considerably different from that in the Alexander Legend and does not exactly match those in the Qur’anic tale of Dhu l-Qarnayn. What is most confusing for modern scholars is that still more traces of the Gilgamesh story, different from those in the Alexander Legend, are found in this Song of Alexander, but these are similar to the traces of Gilgamesh allegedly found in Q 18:60–82, the Moses story just mentioned. The coincidence has never been adequately explained, particularly since recourse must be had to a poorly documented late, probably oral tradition of the Gilgamesh story. Then there are other later seventh-century Christian apocalypses, such as the De fine mundi of Pseudo-Ephraem and the influential Apocalypse of Pseudo-Methodius. These also drew upon the Alexander Legend, evidently a widely known text in the seventh century.7 The bewildering interrelationships of all these traditions have made it difficult for scholars to arrive at a consensus about them. But the reason that I am avoiding 176 ALEXANDER LEGEND IN THE QUR’AN discussing all these related texts here is that they are irrelevant to the thesis that the Dhu l-Qarnayn episode in the Qur’an is derived from or retells the story found in the Syriac Alexander Legend. The account in Q 18:83–102 does not precisely match a story found anywhere other than in this one text, but previous attempts to deal with the problem have become confused by discussing all of the aforementioned traditions together. For example, one recent account, published twice, posits complicated interrelationships between the two episodes in Qur’an 18:60–102, the Qur’an commentaries, the Alexander Romance tradition, and the Song of Alexander, unfortunately just causing further misunderstanding and omitting almost any account of the crucially relevant Alexander Legend.8 In the present article, however, only two main problems are to be discussed: the relationship between the Alexander Legend and Qur’an 18:83–102 and the historical context of this relationship. The Syriac and the Arabic texts compared To prove convincingly an affiliation between this passage of the Qur’an and the Syriac Alexander Legend, a close comparison is required, closer at least than the brief treatment that Nöldeke gave to it. Since the relevant Arabic text, Qur’an 18:83–102, amounts to only twenty verses, they can all be given here in translation. 83. And they are asking you about the Two-Horned One (Dhu l-Qarnayn). Say: I will relate for you a glorious record (dhikr) about him. 84. We granted him power in the earth and gave him a heavenly course (sabab)9 out of every thing. 85. So he followed a heavenly course 86. until, when he reached the place of the sun’s setting, he found it setting in a fetid spring and he found by it a people. We said, “O Two-Horned One, either you will punish (them) or do them a favor.” 87. He said, “Whoever does wrong, we will punish him, and then he will be sent back to his Lord and He will punish them in an unknown way.” 88. “And whoever believes and acts righteously, he will have the best reward and we will declare ease for him by our command.” 89. Then he followed a heavenly course 90. until, when he arrived at the sun’s rising place, he found it rising over a people for whom We did not make a shelter beneath it. 91. Thus We knew everything that he encountered. 92. Then he followed a heavenly course 93. until, when he arrived between two barriers, 177 KEVIN VAN BLADEL he found outside them a people who could scarcely understand speech. 94. They said, “O Two-Horned One, Yajuj and Majuj are destroying the land. Shall we make a payment to you on the condition that you make a barrier between us and them?” 95. He said, “The power my Lord has given me is better, so, help me, with strength, that I may make a barricade between you and them.” 96. “Bring me blocks of iron.” Eventually, when he had leveled it off with the two clifftops, he said, “Blow.” Eventually, when he had made it a fire, he said, “Bring me brass that I can pour upon it.” 97. Thus they could not surmount it and they could not break through it. 98. He said, “This is a mercy from my Lord. When His promise comes, my Lord will make it a heap of earth and my Lord’s promise is true.” 99. And We shall leave them on that day surging like waves10 against each other and the horn will be blown and We shall gather them all together 100. and We shall truly show Gehenna that day to the unbelievers 101. whose eyes were covered from recollecting Me, nor could they hear. 102. Do those who disbelieve plan to take My servants under Me as protectors? We have prepared Gehenna as a guest-house for the unbelievers! For the purposes of this study, this can be divided into five parts. 1 2 3 4 5 an introduction to Dhu l-Qarnayn, the Two-Horned One (83–4), his journey to the sun’s setting and his punishment of unjust people there (85–8), his journey to the sun’s rising place where the people have no shelter from the sun (89–91), his journey to a place threatened by Yajuj and Majuj where he is asked to build a protective wall between two mountains, culminating in his uttering a brief prophecy (92–8), and finally God’s first-person warning of the events to come (99–102). The Syriac Legend of Alexander is quite a bit longer, twenty-one pages of Syriac text in the edition. A summary of the story, including its relevant details, here follows, showing how each of the five parts of the Qur’anic story finds a match in the Syriac text. Readers with insufficient knowledge of Syriac may find Budge’s English translation to be helpful but should be warned that it strays into error on some important points. The story of the Neshanâ begins when King Alexander summons his court to ask them about the outer edges of the world, for he wishes to go to see what surrounds it. His advisors warn him that there is a fetid sea, Oceanos (Ôqyanôs),11 178 ALEXANDER LEGEND IN THE QUR’AN like pus, surrounding the earth, and that to touch those waters is death. Alexander is undeterred and wishes to go on this quest. He prays to God, whom he addresses as the one who put horns upon his head, for power over the entire earth, and he promises God to obey the Messiah should he arrive during his lifetime or, if not, to put his own throne in Jerusalem for the Messiah to sit upon when he does come. This in essence matches Q 18:83–4, part one earlier, where God gives the two-horned one power over the entire earth. On the way, he stops in Egypt where he borrows seven thousand Egyptian workers of brass and iron from the king of Egypt to accompany his huge army. Then they set sail for four months and twelve days until they reach a distant land. Alexander asks the people there if they have any prisoners condemned to death in their prisons, and he asks that those evil-doers (‘abday-bîmê) be brought to him. He takes the prisoners and sends them into the fetid sea in order to test the potency of the poisonous waters. All the evil-doers die, so Alexander, realizing how deadly it is, gives up his attempt to cross the water. Instead he goes to a place of bright water, up to the Window of the Heavens that the sun enters when it sets, where there is a conduit of some kind leading through the heavens toward the place where the sun rises in the east. Though the text is completely vague here in its description of spaces, apparently Alexander follows the sun through its course to the east during the night but “descends” (nahet) at the mountain called Great Mûsas.12 His troops go with him. We are also told that when the sun rises in the eastern land, the ground becomes so hot that to touch it is to be burnt alive, so that people living there flee the rising sun to hide in caves and in the water of the sea. Alexander’s journeys west and east match Q 18:85–91, parts two and three earlier, exactly in many specific details and in fact make some sense of the cryptic Qur’anic story (though the Syriac leaves the specifics of his itinerary here fairly murky). We next find Alexander traveling at the headwaters of the Euphrates and the Tigris, where he and his armies stop at locales given very specific place-names. This specificity has rightly been taken as due to the Syriac author’s personal familiarity with the upper Tigris region, probably his homeland.13 Yet Alexander continues northwards into mountains, evidently the Caucasus, until he comes to a place under Persian rule where there is a narrow pass. The locals complain about the savage Huns who live on the opposite side of the pass. The names of their kings are listed to him, the first two of which are Gog and Magog. Alexander is treated to a vivid description of the barbarism of the Huns. Among the gruesome details it is reported that their cries are more terrible than those of a lion. The Huns have no qualms in killing babies and pregnant women. In short, they do not know civilization but only brutality. The people complain to Alexander that these savages raid with impunity and they hope his dominion will be established. After he satisfies his anthropological and geographical curiosity about the far northern peoples, Alexander asks the locals if they want a favor, and they answer that they would follow his command. So he suggests building a wall of brass and iron to hold out the Huns. Together they accomplish the task with the help of the 179 KEVIN VAN BLADEL Egyptian metalworkers. This account matches Q 18:92–8, part four earlier, in precise detail. The next part of the story is crucial to dating the text. Alexander puts an inscription on the gate containing a prophecy for events to follow his lifetime. These events are given precise dates. First he says that after 826 years, the Huns will break through the gate and go by the pass above the Haloras River14 to plunder the lands. Then after 940 years, there will come a time of sin and unprecedented worldwide war. “The Lord will gather together the kings and their hosts,” he will give a signal to break down the wall, and the armies of the Huns, Persians, and Arabs will “fall upon each other.”15 So many troops will pass through the breach in the wall that the passage will actually be worn wider by the spear-points going through. “The earth shall melt through the blood and dung of men.”16 Then the kingdom of the Romans will enter this terrible war and they will conquer all, up to the edges of the heavens. In closing, Alexander cites the prophet Jeremiah, 1:14, “And evil shall be opened from the north upon all the inhabitants of the earth.” Clearly this corresponds closely with Q 18:99–102, the fifth and last part of the story of Dhu l-Qarnayn. There are still some details and a conclusion to the story in the Syriac text that have no corresponding part in the Qur’an. When Alexander comes into conflict with the King of Persia, called Tûbarlaq, then, with the help of the Lord, who appears on the chariot of the Seraphim along with the angelic host, Alexander’s armies are inspired to conquer the king of Persia. When he is captured, the Persian king Tûbarlaq promises to give Alexander tribute for fifteen years in return for a restoration of the borders. But Tûbarlaq’s diviners predict that at the end of the world, the Romans will kill the king of Persia and will lay waste to Babylon and Assyria. Tûbarlaq himself puts the prophecy in writing for Alexander, saying that the Romans will conquer the entire world and rule it all before handing power over to the returning Messiah. The Alexander Legend finally comes to an end with the remark that at the end of Alexander’s life, he establishes his silver throne in Jerusalem just as he had promised. This last episode is not reflected in the Qur’anic story, but it has proven important in recent scholarship in assigning a date to the Syriac text (to be discussed later). Precise correspondences between the two texts Many of the correspondences between the Syriac and the Arabic stories are so obvious that they do not need special attention. Simply relating both stories together establishes their extraordinary similarity. However, some correspondences require emphasis and further comment. Alexander is twice said in the Syriac to have been granted horns on his head by God. Once it is in a prayer that he himself utters, referring to his horns, and the second time we are told that they were horns of iron.17 Though Alexander had been portrayed with horns as early as his own time, here one finds the epithet Dhu l-Qarnayn, the Horned One, as one element matching the present Syriac text.18 180 ALEXANDER LEGEND IN THE QUR’AN When Alexander came to the people in the west, he tested the efficacy of the deadly, fetid waters with the lives of convicts. This passage helps to explain the option given, for no apparent reason, by God to Dhu l-Qarnayn in the Qur’an: either to punish the people or to do them a kindness. Dhu l-Qarnayn says he will punish only wrongdoers (man zalama), who are like the prisoners sentenced to death in the Syriac text, described there as evil-doers (‘abday-bîmê). The Syriac text has Alexander travel from that point, near to where the sun sets, in the direction of the place where the sun rises, just as does Dhu l-Qarnayn in the Qur’an. The sun does not exactly set in the fetid water, but more vaguely nearby. And it is only this Syriac text that explains the meaning of Q 18:90, where the otherwise unknown eastern people who have no cover from the sun are mentioned. On his third journey, the people who can hardly understand speech are explained by the Syriac text as “Huns,” here a generic term for Central Asian pastoralists, who appeared to the residents of the Middle East as savages. Their allegedly bestial barbarism is explained at length in the Syriac. The Qur’anic text saying that they “could scarcely understand speech” together with reference by name to Gog and Magog makes sense only in the context of this Syriac tale. Dhu l-Qarnayn’s ability to build a wall of iron and brass is explained in the Syriac story by his being accompanied by seven thousand Egyptian “workers in brass and iron,” precisely the same metals. In both texts our hero builds the wall at a place between two mountains in order to fend off savages. Though the tradition of Alexander’s wall holding off the Huns is an ancient one going back at least to Josephus (d. ca 100), who specifies that the gates were of iron, nevertheless the details of the Arabic account are all matched only by this Syriac Alexander Legend.19 Most importantly, in both texts the hero issues a prophecy upon completing the fortification foretelling the end of the world in a time of great battles among nations. Thus, quite strikingly, almost every element of this short Qur’anic tale finds a more explicit and detailed counterpart in the Syriac Alexander Legend. In both texts the related events are given in precisely the same order. Already earlier several cases of specific words that are exact matches between the Syriac and the Arabic were indicated. The water at the place where the sun sets is “fetid” in both texts, a perfect coincidence of two uncommon synonyms (Syriac saryâ, Arabic hami’a). Also, the wall that Alexander builds is made specifically of iron and brass in both texts. We are told in the Syriac that God will “gather together the kings and their hosts,” which finds a nearly perfect match in Q 18:99: “the horn will be blown and we shall gather them together.”20 The proper names of Yajuj and Majuj are not uniquely matched by this Syriac text (where they appear as Agôg21 and Magôg), for their tradition is derived from the books of Ezekiel and the Apocalypse of John, but they do still count as specific word correspondences between the Syriac and Arabic texts in question here. In the Qur’an God is characterized as saying, “We shall leave them on that day surging like waves against each other,” wa-tarakna ba‘dahum yawma’idhin yamuju fi ba‘din, while the Syriac says similarly “and kingdoms will fall upon each other,” w-naplan malkwata hda ‘al hda. 181 KEVIN VAN BLADEL The title of the Syriac work is “Neshanâ of Alexander.” The word neshanâ means “glory” or “victory” but was often used to refer to a narrative account of a person’s heroic acts.22 In Q 18:83 God is portrayed as commanding Muhammad to say that he will recite a dhikr about the Dhu l-Qarnayn. Dhikr in Arabic has most of the same connotations as Syriac neshanâ: it refers to glory or good repute but it also can refer to an account remembered about someone. Could the word dhikr in Q 18:83 be a translation of the very title of the Syriac Alexander Legend? It is a tempting consideration, but there are a few other instances in the Qur’an where a dhikr of a person is related without any apparent reference to a written work.23 The translation of sabab (pl. asbab), occurring in Q 18:84, 85, 89, and 92 as “heavenly course” requires some explanation. These are conventionally translated merely as the “ways” that Dhu l-Qarnayn is made to follow, since among the many meanings of sabab in Arabic are prominently “means” and “ways of access.” However, Arabic lexicographers and much other evidence attest to the early use of the word to mean in particular heavenly courses, specifically cords leading to heaven along which a human might travel: asbab al-sama’, “ways to heaven” or “sky-cords.”24 In fact this is probably the only meaning of the word occurring in the Qur’an, appearing in four other places.25 Nor are these isolated cases of such a usage in Arabic. For example, it is also attested in the poetry of al-A‘sha (d. 625), an exact contemporary of Muhammad, where the phrase wa-ruqqita asbaba l-sama’i bi-sullam, “and were you to be brought up the gateways of heaven by a flight of steps,” is found with the synonymous, variant reading abwab al-sama’ “gates of heaven.”26 Thus, the translation given earlier, though unconventional, is not only suitable but likely. In the case of Dhu l-Qarnayn’s tale, it matches the window of heaven (kawwteh da-mmayyâ)27 through which the sun passes on its course, and which Alexander follows, in the Syriac Alexander Legend. The remaining problem is then to account for the third “way” mentioned in Q 18:92, the northward path that is not connected with any course of a heavenly body in the Alexander Legend. Here one may excuse the Arabic as following the pattern of the earlier journeys. The matter is bound up with the problem of how these heavenly courses were imagined, something I treat in detail elsewhere.28 If there were a closer correspondence of the Syriac and Arabic, it would be possible to argue that one was just a much modified translation of the other. As it is, however, the correspondences shown earlier are still so exact that it is obvious in comparison that the two texts are at least connected very closely. They relate the same story in precisely the same order of events using many of the same particular details. Every part of the Qur’anic passage has its counterpart in the Syriac, except that in the Qur’an the story is told through the first-person account of God. Also, as explained earlier, the Qur’an does not include the last part of the Alexander Legend, in which Alexander defeats the Persian emperor Tûbarlaq, who writes his own prophecy down for Alexander and gives it to him, to the effect that the Romans would one day decisively defeat the Persians, establishing a worldwide Christian rule that would remain until the return of the Messiah. 182 ALEXANDER LEGEND IN THE QUR’AN The Qur’anic account puts more emphasis on the coming end of things and God’s judgment and, not surprisingly, does not mention any expectation of universal Christian empire for the Romans. Dating and contextualizing the Syriac Alexander Legend At this point I think there can be no doubt whatsoever of the affiliation between the Qur’anic passage and the Syriac Alexander Legend. The question now becomes how to specify that affiliation. Here we will be assisted by finding a date and historical context for the Syriac text. Fortunately G.J. Reinink has devoted many articles to the problems posed by this Alexander Legend and related texts which have succeeded in determining definitively where, why, and when the Alexander Legend was written. I employ his detailed studies extensively in what follows, and the reader is urged to pursue them for further information that can be used to assign a date to this Syriac work.29 This section may seem to be a bit of an excursus, but it is crucially important to contextualize the Syriac text in order to relate it to the Qur’an. The Alexander Legend is an apocalyptic text in which the ancient Alexander is portrayed as presenting a prophecy written long ago for events to come, which were intended to be understood by the audience at the real time of authorship as referring to events leading up to and including their own time. This is how many texts of the apocalyptic genre work. Thus the date of composition for such apocalypses can often be found by locating the latest point at which events allegedly predicted match actual historical events. Where the events “predicted” diverge from history, there one usually can find the date of the composition. The message of the apocalypse for its own time is not just in the events it describes, but rather in the way it describes these events and the future that it expects to unfold given what has occurred. In the Syriac Legend, Alexander’s prophecy, written on the wall he himself erected, gives two dates marking the invasion of Central Asian nomads, called Huns, whose penetration of the great wall and arrival at the headwaters of the Tigris are portentous events to be taken as signs of the final battles preceding Christ’s return and the end of time. Alexander specifies how many years must elapse before these events take place. Already Nöldeke in 1890 calculated the dates according to the Seleucid Era (beginning 1 October 312 BCE) normally followed in Syriac tradition, also called the Era of the Greeks and, importantly, the Era of the Alexander.30 The first of the two dates is thus converted from 826 years later to 514–15 CE, precisely the time of the invasion of the nomadic Sabirs who entered Syria and Anatolia.31 Evidently this invasion, which holds no importance in the narrative, serves just as a key for the contemporary audience of the text that they can use to verify the accuracy of the second, more elaborate prophecy, associated with a later date. In any case, no scholar after Nöldeke has disputed the calculation of this first dating, as far as I have seen.32 183 KEVIN VAN BLADEL The second of the two dates, 940 years after Alexander, which marks the time of the final war preceding the Messiah’s return according to the prophecy, is converted likewise to 628–9 CE. The message of the prophecy actually concerns events around this date, which coincides with the end of a long and extremely difficult war between the Persians and the Romans (603–30) during which Jerusalem was devastated, the relic of the True Cross stolen from that city, and the Persians conquered Syria, Palestine, and Egypt, occupying Anatolia, too, and they even besieged Constantinople itself in 626 CE in concert with the Avars, who invaded from the north. The Byzantine remainder of the Roman Empire was only barely saved from the Persian onslaught by the emperor Heraclius’ daring campaign through Armenia, ending in the winter of 627–8 with a surprise invasion into Mesopotamia and damaging raids on the rich estates around Ctesiphon. In these invasions the Türks joined the Byzantines in raids south of the Caucasus at Heraclius’ invitation and afterwards continued to make war on Persian territory in Transcaucasia, plundering until 630. The Byzantine invasion of Mesopotamia led the Persian nobles to remove their King of Kings, Khosro II, from power in February of 628 and to negotiate for peace.33 Persian forces occupying former Byzantine territory withdrew to Persia in 629, and early in 630 Heraclius personally returned the relic of the True Cross to Jerusalem in a formal celebration. (Just a few months before Heraclius’ arrival in Jerusalem, tradition tells us, the inhabitants of Mecca surrendered peacefully to Muhammad and submitted to his government.) Given the date that Alexander’s prophecy signals, 628–9 CE, it must be referring to the devastating wars of that time and their successful end for the Romans. Reinink has shown that the Alexander Legend demonstrates, through its prophecy and its use of Alexander to prefigure the emperor Heraclius, detailed knowledge of the events of that war and its resolution with the restoration of the earlier borders, a peace treaty, and a final reference to Jerusalem. Using this information, too much to repeat entirely here, he has persuasively argued that the Alexander Legend was composed just after 628, perhaps in 630, the year in which Heraclius restored the cross to Jerusalem.34 In the course of the war, while the Byzantines were very hard pressed by the Persians, Heraclius resorted to highly religious propaganda in order to rally his allies and to improve Roman morale. This propaganda has received recent scholarly attention.35 Likewise Heraclius’ attempts to eradicate the schisms in the Church after the war are well known. Reinink considers Alexander Legend to be a piece of pro-Heraclian postwar propaganda designed to promote the emperor’s political cause not long after the war’s end, re-establishing Roman rule over provinces that had been under Persian power for well over a decade and trying to overcome the schismatic Christological differences dividing his Chalcedonian court from the monophysites of the provinces recently recovered from the Persians. His thesis is that the Syriac Legend of Alexander was composed “shortly after 628” (i.e. in 629 or 630) by an inhabitant of Amida or Edessa, or some other place near to those, in support of 184 ALEXANDER LEGEND IN THE QUR’AN Heraclius.36 He argues that the monophysite Syrians were the primary audience (although it is possible that the story was intended also to win over monophysites of other nations such as Arabs).37 Heraclius’ visit to Edessa in late 629 might have been an occasion for its composition. It is also possible that the text was written a few months later when Heraclius restored the cross to Jerusalem.38 The specific details in the Alexander Legend that reflect this historical context are numerous. But unlike the Qur’anic story of Dhu l-Qarnayn, the Syriac text ends with the Persian king’s own prophecy containing what Reinink has characterized rightly as a message of Byzantine Imperial eschatology: the prediction that one Byzantine emperor will soon establish a worldwide Christian rule which will be followed by the return of the Messiah.39 This was intended to counter the belief, widely held at the time as many sources show, that the total destruction of the Roman/Byzantine Empire and even the end of the world were imminent. As Reinink sums it up, the author of the Alexander Legend wants to demonstrate the special place of the Greek-Roman empire, the fourth empire of the Daniel Apocalypse, in God’s history of salvation, from the very beginning of the empire until the end of times, when the empire will acquire world dominion. He created an Alexander-Heraclius typology, in which the image of Alexander is highly determined by Byzantine imperial ideology, so that his contemporaries would recognize in Heraclius a new Alexander, who, just like the founder of the empire, departed to the east at the head of his army and combated and defeated the Persians. (G.J. Reinink, “Heraclius, the New Alexander. Apocalyptic Prophecies during the Reign of Heraclius,” 26) By now it should be amply clear that the Alexander Legend is the product of a very specific, identifiable historical and cultural environment, the end of a devastating war widely believed to carry eschatological implications, ending with Heraclius’ campaign in 628 and in 629 with the final withdrawal of the Persian armies. This needs to be held in mind when the relationship between this text and the Qur’an is considered. If this is the message of the Alexander Legend, what is the point in having Alexander make his journeys west, then east, then north, then return south? The answer is clearer when one imagines a map of his itinerary. In effect Alexander’s travels make a sign of the cross over the whole world. This symbol seems to have been overlooked by other commentators, but I believe it was intended by the author of the Alexander Legend. The sign of the cross was the emblem of victory for the Christian empire, and the prophecies in the Legend indicate the imminent universal rule of the Christian empire. One may even speculate that this crossshaped itinerary was intended symbolically to refer to Heraclius’ return of the 185 KEVIN VAN BLADEL relic of the True Cross early in 630 to Jerusalem, the city where Alexander places his throne at the end of the Alexander Legend. Alexander’s journeys describe the symbol of Christian Roman power across the entire world, which it will come to rule in its entirety according to the prophecy. But what is the point of having Alexander build the Wall of Gog and Magog? According to Greek and Latin traditions from the first century CE onward, Alexander was indeed credited with building gates in the Caucasus to keep out invaders. These gates, described by many ancient Greek and Latin authors, were usually identified as located at the pass of Darial in the middle of the Caucasus (Arabic Bab al-Lan).40 However, in the seventh century, just around the time of the Syriac Alexander Legend, confusion arose concerning the location of Alexander’s fortified pass. It now came to be identified with a gated wall situated on the Caspian coast that had been built more recently (Arabic Bab al-Abwab). By the mid-sixth century, the waters of the Caspian had receded considerably on their western shore, exposing a wide pass of land around the eastern end of the Caucasus.41 The Sasanian shahs constructed a very large wall (or series of walls) with a great gate in order to block this coastal gap as a defense against northerners who might otherwise easily raid Iran, Anatolia, and Mesopotamia. The scholarly literature documenting the existence and history of these walls through archaeology and written sources is enormous.42 The town Darband eventually came to be at these walls at the Caspian, presumably at first just a garrison town, eventually a very important site. Its Persian name, meaning “Door-Bolt,” indicates its original purpose. Seventh century sources mention these fortifications a number of times. For example, the Armenian historian called Sebeos, writing in the 680s, called it “the Gate of the Huns.”43 But the displacement of Alexander’s gate from Darial to the wall at Darband does not appear unambiguously in the sources until the Frankish Latin chronicler known as Fredegarius (wr. ca 660), in his report on the year 627, described Alexander’s gates as having been built over the Caspian Sea (super mare Cespium [sic]), saying that these are the gates that Heraclius opened to admit the savage nations living beyond them.44 From this time onward, Alexander’s Caspian gates were widely thought to be those at Darband. What caused this confusion to be held generally between Latin and Arabic tradition? It seems that the Syriac Alexander Legend may have prompted it. While it may have intended the pass at Darial (though the geographical expertise of the author is subject to doubt), the invasions of the Türks through the wall at Darband in 626–30 must have forced the association of Alexander’s walls with that route. In the early twentieth century Russian scholars discovered a number of Pahlavi inscriptions on the old wall at Darband, dated variously at first but with a final, general consensus to the sixth century.45 Thus the author of the Syriac Legend of Alexander was using common lore that would be readily understood by its audience: Alexander was thought to have built a real wall with a gate that was known to the inhabitants of the Caucasus region and indeed was famous far and wide, a wall that bore inscriptions. It is easy to see how one of these inscriptions might have been thought to have been carved there by Alexander. 186 ALEXANDER LEGEND IN THE QUR’AN The Alexander Legend’s account identifies the people beyond the wall, the “Huns,” as Gog and Magog. These names originally come from Hebrew scripture. They are associated by Ezekiel 38–9 with northern, invading nations, serving as God’s punishment, and then later by the Revelation of John 20 with final turmoil just before the ultimate redemption. Gog and Magog are, in short, an eschatological motif: they are northern nations whose invasion heralds the end of time.46 In the sixth century Andreas of Caesarea had made this association clear in his commentary on the Revelation of John, identifying Gog and Magog with the Huns, and in doing so he was following the sources going back at least to Josephus (d. ca 100).47 Thus the Alexander Legend combines two traditions (1) Alexander’s building of a wall in the Caucasus to hold out Huns and (2) the identification of Huns, a generic term for all Central Asian peoples, with Gog and Magog, thereby associating Alexander with the end of time and giving him an occasion to make eschatological prophecies. Alexander’s wall also explains why the Huns (Gog and Magog) cannot invade at just any time; they have to surmount the wall first. But when that wall is breached, that will be a sign of the approaching end. Once these traditions were combined, it was now easy to link Heraclius both with the world conquering Alexander, who similarly defeated the Persian emperor, and with the end of time. As already stated, in his final campaigns against the Persians, Alexander’s former enemies, Heraclius actually did enlist the help of Inner Asian peoples, the Kök Türks, in his war against the Persians (626–7) – they are called variously in the sources Türks and Khazars, being perhaps Khazars under Kök Türk rule, though the specific tribal or ethnic identity of these invaders is a subject of very long debate – and afterward these Türks fiercely raided Caucasian Albania, Georgia, and Armenia until 630.48 One wonders whether Heraclius or his supporters promoted the idea that his Türk allies, summoned from the north, were the people of Gog and Magog come to punish the Persians. The Türk invasions are known from the Greek chronicle of Theophanes49 and in some detail from a compilatory seventh-century source used by the Armenian History of the Caucasian Albanians (Patmut’iwn A„uanic’) by Movses Dasxuranc‘i. As it says, “During this period (Heraclius) . . . summoned the army to help him breach the great Mount Caucasus which shut off the lands of the north-east, and to open up the gates of #‘o„ay [i.e. the gates at Darband] so as to let through many barbarian tribes and by their means to conquer the king of Persia, the proud Xosrov.”50 Fredegarius, as mentioned, also states that Heraclius opened these gates. Thus the devastating raids of the terrifying “Huns” – “predicted” in the Alexander Legend – also match the Türk campaigns in the years 626–7 (alongside Heraclius) and 628–30 (independently), and inhabitants of Caucasian Albania and Iberia, Armenia, and the neighboring lands such as Mesopotamia and Syria were surely well aware of them. Moreover, Greek and Armenian sources show that these real invasions of Türk warriors in the early seventh century were actually interpreted then in 187 KEVIN VAN BLADEL apocalyptic terms and associated with the eschatological motif of Gog and Magog. J. Howard-Johnston has dubbed the source of historical information on these Caucasian campaigns used by the Armenian Movses Dasxuranc’i as the 682 History (because its reports end with the year 682 and we do not know its original name).51 This source describes the nomadic invaders in horrific terms in connection with the joint Byzantine-Türk siege of the Caucasian Albanian capital, Partaw (Arabic: Bardha‘a). They are depicted as ugly savages, like merciless wolves, who kill regardless of the victim’s age or sex.52 The Syriac Alexander Legend describes the Huns in quite similar terms, also stressing their readiness to kill women and children and their bestial nature.53 The way in which the Armenian source describes these wars between the Byzantines, the Türks, and the Persians gives yet another example for how people really did expect the end of time during or soon after these wars. The 682 History focuses its attention on the events around the Caucasian Albanian capital of Partaw, but first it begins with a special prologue to the description of these invasions, which are characterized as part of not just local but the universal calamities (i tiezerakan haruacoc‘s) prophesied by Jesus in the Gospels about the times of tribulations (i ∆amanaki 3‘ar3‘aranac‘n).54 This understanding is based in the 682 History explicitly on quotations of Jesus’ prophecies selected from Matthew 24 and Luke 21:5–28. The full prophecy of Jesus in Matthew, not cited in its entirety by the Armenian historian, indicated particularly that a siege of Jerusalem would be one of the signs of the end (Luke 21:20). This would be accompanied by signs in the heavens and confusion among nations before the final redemption. All of this helps to contextualize the role of the Huns in the Syriac Alexander Legend, who are to be identified with the Türks and their invasions into the Caucasus region from 626 to 630. To sum up, the Alexander Legend is seen to reflect many specific events and cultural tendencies of the period around 628–9, the year it indicates as a time of wars between many nations beginning with the breaking of Alexander’s wall by the Huns. Out of these wars the Roman Empire would emerge victorious, some time after which the Roman Empire would permanently overthrow the Persians and establish a universal Christian empire. It is best understood, following Reinink, as a piece of propaganda composed by someone sympathetic to the need of Heraclius around 630, immediately after almost thirty years of demoralizing war and unprecedented military loss, to help in reconsolidating quickly the loyalties of the regained territories of the empire and their monophysite inhabitants. The success or popularity of the Alexander Legend is indicated in that it was used by at least three more apocalypses, the so-called Song of Alexander attributed falsely to Jacob of Serugh (composed just a few years later but before the Arab conquest, between 630 and 636),55 the Syriac apocalypse De fine mundi attributed falsely to Ephraem (composed sometime between 640–83),56 and the Apocalypse of Pseudo-Methodius (composed around 692, quite possibly in reaction to the building of the Dome of the Rock).57 The Alexander Legend was evidently well known in the early seventh century. 188 ALEXANDER LEGEND IN THE QUR’AN The relationship between the Alexander Legend and Qur’an 18:83–102 To return to the main question, the extremely close correspondences between the Syriac Alexander Legend and Qur’an 18:83–102, reviewed earlier, must mean that the two texts are related. On the one hand, there is a Syriac text the date of which is almost certain, about 629–30 CE, and the historical context and political meaning of which is known fairly precisely (as just explained); on the other, we have a passage from the Qur’an, an Arabic compilation the precise dates and historical circumstances of which are debated by historians, but which tradition has understood to be collected into its current form during the caliphate of ‘Uthman (644–56) or at least after Muhammad’s death (632). It is possible to approach the problem of affiliation between the two systematically. The two texts must be related. That is the only explanation for their point-for-point correspondence. In that case there are three reasonable possibilities: (1) the Syriac takes its account from the Qur’an, or (2) the two texts share a common source, or (3) the Qur’an uses the account found in the Syriac. Could the Syriac text have its source in the Qur’an? If this were the case, then the Syriac text would have to be seen as a highly expanded version of the Qur’anic account, which would then need to be understood as an attempt to explain the cryptic Qur’anic story with rationalizations drawn from stories about Alexander. However, the Syriac text contains no references to the Arabic language the type of which one might expect to find if its purpose was to explain an Arabic text, and it is impossible to see why a Syriac apocalypse written around 630 would be drawing on an Arabic tradition some years before the Arab conquests, when the community at Mecca was far from well known outside Arabia. Moreover, the very specific political message of the Alexander Legend would not make any sense in this scenario. This possibility must therefore be discounted. Could the two texts share a common source? This also becomes practically impossible for some of the same reasons. The Syriac Alexander Legend was written to support Heraclius by indicating the author’s belief in the significance of events leading up to 629 AD, events supposed to be foreshadowing the establishment of a Christian world empire and the coming of the Messiah. Yet relating Dhu l-Qarnayn’s first prophecy of the end times is also the very purpose of the story in the Qur’an: the prediction of God’s actions at the time of judgment using an ancient voice of great authority. As already explained, the war between Byzantium and Ctesiphon went very badly for the Byzantines until the very end, prompting an intense bout of political and religious propaganda to boost the desperate war effort and to consolidate allegiances after the victory. Reinink has shown that this Syriac text, given its contents, must be understood as pro-Heraclian propaganda belonging to this milieu, dated to 629–30. If Alexander’s prophecy was composed just for this purpose at this time, then the correspondence between the Syriac and the Arabic, which contains the same prophecy reworded, cannot be 189 KEVIN VAN BLADEL due to an earlier, shared source.58 Put differently, the only way to posit a common source is to assume that everything held in common between the Qur’anic account and the Syriac Alexander Legend could have been written for and would have made sense in an earlier context. In light of the detailed contextualization given earlier, and in light of G.J. Reinink’s work referred to earlier as well, this becomes impossible. Stephen Gero implied in one article that since the text comes from this date (629 CE or later), it cannot be regarded as a source of the Qur’an. He does not explain in detail but I take the implication to be that such a date of composition is too late for it to have reached the human agents who related the Qur’an.59 But to me this seems to be the only real possibility because the others are invalid, as just explained. The Qur’anic account must draw from the Syriac account, if not directly then by oral report. Since the Qur’an is using the material found in this Syriac text, a text composed for a very specific context in contemporary politics and loaded with particular religious meaning, this gives historians an important opportunity to understand the religion of Muhammad and his early followers without relying entirely on later tradition. Before considering the significance of this further, it is important to ask how the text could have been known in Arabic and under what circumstances. The transmission of the story from the Syriac text into Arabic How could a Syriac text composed in northern Mesopotamia in 629–30 CE or just about that time have been transmitted to an Arab audience in Medina or Mecca so that it could become relevant enough to the followers of Muhammad to warrant a Qur’anic pronouncement upon it? Such a transmission would have been quite possible in the circumstances around 628–30 CE and soon after. Contemporary records in Greek, Syriac, Armenian and Arabic (poetry) repeatedly note the involvement of Arabs as troops and scouts on both Roman and Persian sides during and at the end of the great war of 603–30, and the Syriac Alexander Legend itself mentions Arabs as one of the nations involved in the last wars.60 Indeed, the Alexander Legend is likely to have been circulated widely if it was part of the Byzantine rallying cry after the war in the face of great losses and as a tool of Heraclius for rebuilding his subjects’ loyalty to the idea of a universal Christian empire undivided by schism. If it was aimed particularly at monophysites, as Reinink also proposed, then one would expect it to have been deliberately spread among the monophysite Arabs of the Ghassanid phylarchate, some of Heraclius’ close allies.61 It is even possible that Muhammad’s own followers heard the story of the Alexander Legend, for example during their raid on Mu’ta, around the southeast end of the Dead Sea (probably September 629) just a few months after the Persian withdrawal from Roman territory and a few months before Heraclius’ triumphant return of the cross to Jerusalem.62 190 ALEXANDER LEGEND IN THE QUR’AN Yet one is left wondering exactly how apocalyptic works were disseminated during these decades. Since they are full of political significance for a particular period of time, one expects that they would have been published and promoted actively by their authors. In any case, one can hardly doubt that this text was widely known. An indication of that was aforementioned: the Alexander Legend provoked a monophysite response in Syriac within a few years, one more cynical about the durability of Heraclius’ kingdom, and information connected with the Alexander Legend was known as far away as Gaul a few decades later (on which, see the discussion about Fredegarius, later).63 Nor is it difficult to suggest motives for Muhammad or his followers to have paid attention to this apocalypse. Even with the extraordinary skepticism over the early records of Islam prevailing today, no one disputes that Muhammad’s movement was based on the belief in prophets. The Qur’an contains many references to the prophets of the past. The Syriac Alexander Legend presents Alexander the Two-Horned as just such a prophet. Moreover, Alexander’s prophecy clearly indicates that final wars heralding the end of the world were taking place. Many in the community that followed Muhammad seem to have shared this apocalyptic sentiment with others in the contemporary Middle East.64 However, the Qur’anic account leaves out all mention of the Roman Empire’s inevitable, universal, Christian victory before the return of the Messiah, an important aspect of the last section of the Legend. Instead it focuses on and culminates in Dhu l-Qarnyan’s prophetic warning that God’s judgment will come in a time of wars between great armies. Evidently that was the message of the story that was most meaningful to the adaptor of the Arabic account, and the elements that make the story sensible as Byzantine propaganda are omitted completely in the Arabic. One may even suppose the words of Q 18:83, “And they are asking you about the Two-Horned One (Dhu l-Qarnayn). Say: I will relate for you a glorious record (dhikr) about him,” to be a true reflection of the environment in which the Syriac Alexander Legend was circulating. Here was an apocalypse widely known and certainly currently relevant. Perhaps Muhammad’s followers or others in the vicinity wanted an explanation of this apocalypse from him, and so they were given an account of it, adapted to make it appropriate to their movement. It may also be possible to see reflections of the prophecy of the Alexander Legend in surat al-Rum (Q 30:1–6), where the war between the Persians and Romans is referred to, but the Romans are said to be destined to conquer, at least according to the preferred reading of early Qur’anic exegetes.65 In short, there are many indicators that the Alexander Legend could easily have reached the community at Medina or Mecca and that, when it did, it would have been meaningful to them. There is no reason to doubt this possibility, and the relationship between the Syriac and Arabic texts determined earlier requires one to suppose that the Alexander Legend was in fact transmitted somehow. However, the precise time at which the story of Dhu l-Qarnayn entered the Qur’an – in Muhammad’s last years, or later – is still undecided. 191 KEVIN VAN BLADEL Floods of nations and the prophecy of Jesus There is one more point related to the Qur’anic retelling of this Syriac text that deserves attention. While it is widely known that Jesus was and is regarded as a prophet by Muslims, since he is so designated in the Qur’an (19:30), there is little discussion of just what Jesus was supposed to be a prophet of. It is often overlooked that Jesus was thought even by Christians to be prophesying nothing less than the end of the world (as in Matthew 24 and Luke 21:5–28), and that this would be preceded by a siege of Jerusalem (Luke 21:20). The sack of Jerusalem by the Persians in 614 therefore shocked Christian contemporaries especially because it seemed to indicate that the end the world and the return of the Messiah were near according to the very words of Jesus. Other signs predicted by Jesus preceding the end would be seen in the heavens, and there would be “distress of nations confused by the roaring of the sea and the waves” (Luke 21:25).66 Contemporary sources show that witnesses to the great war of 603–30 saw the fulfillment of Jesus’ words in it. Most important here is the account of the Türk invasion of Caucasian Albania used by the Armenian author Movses Dasxurants‘i, the 682 History. Before describing how the Türks broke through the Wall at Darband, this source adapts the prophecies appearing in Matthew 24:6–7, 29 and Luke 21:25 in its prologue, paraphrasing them, saying that there would be “confusion of nations like the confusion of the waves of the sea” x˝ovut‘iwnk‘ ∆o„ov7rdoc‘ orpes aleac‘ covu x˝oveloy.67 Then it goes on to describe the events of the wars, using allusions to these paraphrased words of Jesus’ prophecy in order to prove that the prophecies were fulfilled. For this purpose, the Türks are likened explicitly to overwhelming waves, the waves of confusion among nations in Jesus’ predictions: “Then gradually the waves moved on against us,” apa takaw mar∆ein alik‘n 7nddem mer.68 After raiding Caucasian Albania the Türks turned west: “the floods (u„xn) rose and rushed over the land of Georgia.”69 Even Khosro II “rose up like a raging torrent” when he set out to war against them70 and we are reminded of the image again with the phrase, “waves of invaders.”71 Thus the author of the 682 History takes every opportunity to show that Jesus’ prophecy was being fulfilled in the invasions of the Türks and the wars of this period in general. This image of nations as waves was also used specifically to describe the Türks’ overwhelming of the walls at Darband. The 682 History does not connect Alexander with these walls, but it does say that near #‘o„ (a town near the gate at Darband) were magnificent walls which the kings of Persia had built at great expense, bleeding their country and recruiting architects and procuring many different materials for the construction of the wonderful works with which they blocked [the passes] between Mount Caucasus and the eastern sea [the Caspian]. When the universal wrath confronting us 192 ALEXANDER LEGEND IN THE QUR’AN all came, however, the waves of the sea flooded over and struck it down and destroyed it to its foundations at the very outset. (Movses Dasxuranc‘i, trans. Dowsett, History of the Caucasian Albanians, 83) These waves are not real waves of the Caspian but rather attackers from the north. The text immediately next describes the physical appearance of the Türk invaders, portrayed as monstrous, whom Heraclius had invited to war as his allies. It is striking that this author, a resident of Caucasian Albania, the territory immediately south of these walls, reports that the Türks actually destroyed the wall (i himanc‘ tapaleal, “demolished it to the foundations”),72 just as Alexander’s Syriac prophecy in the Legend said that they would be destroyed. The Byzantine chronicler Theophanes also states that, at the beginning of their invasions, “the Khazars broke through the Gates of the Caspian” (diarrhéxantes tàs Kaspías púlas).73 Just so, in the Qur’an (18:98), Dhu l-Qarnayn prophesies that God will make the barrier a heap of earth at the time of his promise, the final judgment (fa-idha ja’a wa‘du rabbi ja‘alahu dakka’a). But the difference in the Armenian source is that in it the breaking of the wall by the Türks was identified as part of the fulfillment of Jesus’ words. There are not many other surviving reports about these Türk invaders and their passage through the wall. That is why it is especially striking that one of the few other authors to mention it, the contemporary Frankish chronicler known as Fredegarius (wr. ca 660) describes the gates as having been built out of bronze (aereas) by Alexander propter inundacione gentium sevissemorum (sic), “on account of the surging wave of most savage nations.” Here again the invaders are described as a surging wave, an inundacio of nations, held back by the gates that, Fredegarius goes on immediately to say, Heraclius himself ordered to be opened: easdem portas Aeraglius aperire precipit (sic).74 In light of the description of the Armenian 682 History, which was explicitly connected with the prophecies of Jesus, it seems likely that Fredegarius was drawing from a source that made a similar allusion to the waves of nations, paraphrasing Jesus’ prophecy in Luke 21:25. Moreover, this Latin chronicle’s association of Heraclius with the opening of the gates of Alexander that held back savages brings together most of the parts of the Alexander Legend. It is in fact the earliest known association of Alexander specifically with the wall at Darband (and not the wall at the Darial pass or another, unspecified place, as in the Alexander Legend).75 What is missing in Fredegarius is reference to Gog and Magog. But in his confusion that chronicle’s author bizarrely thinks that the Hagarene Saracens were admitted by this gate, not the Türks. This implies that he identified the Arabs as the people of Gog and Magog, though it is not explicitly stated. Now, the description of the Hun invaders as waves is not found in the Syriac Alexander Legend. However, as shown earlier, God is portrayed as saying in the Qur’an 18:99, “And We shall leave them on that day surging like waves (yamuju) against each other” when the wall holding back Gog and Magog is demolished. 193 KEVIN VAN BLADEL It is tempting therefore to think that the Syriac Alexander Legend was associated consciously at some stage of the transmission into Arabic with an explanation or oral commentary including reference to Jesus’ prophecies of the end of the world, since the near-contemporary source of Movses Dasxuranc‘i, the 682 History, shows that the reference to waves in Jesus’ prophecy was taken to refer to the invasions of the Türks, identified elsewhere by contemporaries as the eschatological peoples of Gog and Magog, and their involvement in a war of many nations. Fredegarius’ chronicle also describes them as waves. If this hypothesis is correct, the word yamuju, “surging like waves,” in the Qur’an, is essentially a verbal echo of Luke 21:25 (sunokhè ethnôn en aporíai ékhous thalásses kaì sálou), the “distress of nations confused by the roaring of the sea and the waves,”76 Jesus’ prophecy of what would happen before the final redemption. It suggests that Movses Dasxuranc‘i’s source (the 682 History) and Fredegarius’ source were not the only ones to consider Jesus’ prophecies to be fulfilled in these wars. Muhammad’s earliest followers may have understood the story of Dhu l-Qarnayn not just as the prophecy of the imminent end made by Alexander, regarded as a pious, ancient world-conqueror, but also as an allusion to the prophet Jesus’ similar warning of the end times, now very near, which they expected as seriously as other inhabitants of the region, when nations did indeed crash together, as it might have appeared on a field of battle, like waves of the sea. The language of Q 18:83–102 Now that the continuity of tradition between the Syriac Alexander Legend of Alexander and the Qur’anic passage in question (Q 18:83–102) is established, it is possible to draw some new conclusions about the language of the Qur’an here. Though controversy has been aroused by the recent attempt to find Syriac or Aramaic words in the Qur’an where they had not been part of the traditional reading, now one can see that where the Qur’an is definitely reinterpreting a Syriac text, not a single Syriac word is found, but rather there are true Arabic equivalents of Syriac words.77 Q 18:83–102 is a distinctively Arabic text and in no way is it Syriac. Thus it is clear that Qur’anic tradition and, in particular, the traditional Islamic lexicography of the Arabic words in this passage prove to be quite reliable. A high number of exact parallels of meaning between the Syriac and the Arabic (though the Arabic passage is short) come to light while reading the Qur’anic text in a way that accords very closely with the traditional Muslim interpretation – interpretation of the words themselves, that is, the lexicography, and not the explanatory commentary or tafsir. Whatever problems one finds in the grammar and script of the Qur’an, it is quite clear that the words and basic meanings of this passage of the Qur’an have been understood by Islamic tradition correctly. In a sense this Syriac Alexander Legend vindicates the reliability of some basic, traditional claims about this Qur’anic passage, providing means to verify Arabic tradition. 194 ALEXANDER LEGEND IN THE QUR’AN Conclusions The main conclusion reached here is that a Syriac text quite current and important in the last years of Muhammad’s life was adapted for twenty verses of the Qur’an. This is not entirely new, since Nöldeke made a similar argument in 1890. Nor is it surprising, since the Qur’an relates many other well known ancient stories in its own way to deliver its own message, as Muslims generally accept. However, it is now shown beyond any reasonable doubt that this is the case for a text contemporary with Muhammad. Moreover, what is most important for our understanding of the adaptation of the Alexander Legend in the Qur’an is not the fact of the borrowing but rather the way in which the particular religious and political message associated with the Alexander Legend was used, truncated, and altered for new purposes. This is not a sweeping theory about the formation of the Qur’an, for it only concerns one small portion of a text agreed upon by almost all to have been compiled from different oral and written materials collected together after the death of the prophet. This theory makes no claims about the text of the Qur’an as a whole, but it nevertheless requires that the Syriac Alexander Legend be taken into account by any theory attempting to account for the whole Qur’an. It is only in studying the Qur’an as a text in its own historical context, which historians of the Qur’an have neglected to a surprising extent in their overdependence on later Arabic sources for the history of the seventh century, that it will become comprehensible to the historian and to those truly concerned with understanding its inimitable history. The findings of this article may be summarized as follows. The Syriac Alexander Legend, written in 629–30 as religious and political propaganda in favor of Heraclius after a devastating war, puts forth two prophecies: one about the impending end of the world in a war of all nations, the other predicting that Roman, Christian rule would come over the entire earth before the Messiah’s return. This text was evidently well known soon after is publication since several other texts written in the seventh century react to or include material derived from it. The Arabic, Qur’anic account of Dhu l-Qarnayn also repeats this story, but includes only the first of its two prophecies, along with the narrative of Alexander’s journeys. If Muhammad himself did speak Q 18:83–102, then it may well have been his response to questions concerning the publication of these prophecies (“They ask you about the Two-Horned One. Say. . . .”). Whatever the precise circumstances of the Arabic composition were, its primary message is that God’s judgment is very much imminent. The reference to contemporary wars reflects the notion, widely held around this time, that the violence and strife of this period were indeed an indicator of the rapidly approaching end of the world. It is not surprising that a community of Arabs whose religion was based on a belief in prophecy would find the contents of this story meaningful, since it put a prophecy supporting the apocalyptic sentiments that they shared, designed for their troubled times, into the mouth of an ancient and respected world-conqueror. 195 KEVIN VAN BLADEL What is striking is that the strongly pro-Roman element, appearing especially in the second prophecy of the Alexander Legend, is completely omitted, though many details of the other parts of the story are included. Surely this omission also reflects some attitude in the community of Muhammad. Finally, though it depends (probably through oral report) on a Syriac work for its content, Q 18:83–102 shows no hint of Syriac vocabulary. It is an entirely Arabic text likely to have been first uttered in the early seventh century. The extraordinary correspondences between the Syriac and the Arabic vindicate the early Muslim understanding of the meaning of the words in this text, but not their exegesis of it.78 Approaching the Qur’an by contextualizing it in the milieu of the early seventh century clearly has much to offer, but it is surprising to find how disconnected the field of Qur’anic research is from other historical studies on the same period and region, with some notable recent exceptions. It seems now that the future of Qur’anic studies lies not within the discipline construed as Islamic studies alone but rather that many major historical problems of the Qur’an will be solved by historians of Late Antiquity, whose approaches to the first century of Islam are proving more successful than the various apologetic and polemical approaches that predominate in the modern study of early Islam. That is perhaps to be expected, since scholars in the field of Islamic studies are largely concerned with later tradition and has generally (though not in every case) failed to find adequate tools for approaching the Qur’an in its original context, the early seventh century. Yet almost every primary source used in the present study was published more than fifty years ago, many of them more than a century ago. Scholars of Islamic studies have brought historical–epistemological problems – which are problems particularly when they confine themselves to late sources – so prominently to the foreground that it is nearly impossible to read the texts themselves, while the general abandonment of the basic preliminary tools of historical scholarship – the philological methods used to establish text that can then serve as objects of historical research – are sorely neglected. But Qur’anic studies now require scholars trained in Greek and Syriac, not to mention other forms of Aramaic, and even Armenian, Ethiopic and other languages, as much as in Arabic.79 With the great surge in research and publication on Late Antiquity, the very context into which Islam came, answers to the pressing theoretical questions as well as to some of the historical ones also may at last appear.80 Notes 1 E.A. Wallis Budge, The History of Alexander the Great, being the Syriac Version of the Pseudo-Callisthenes, Cambridge: Cambridge University Press, 1889. 2 Th. Nöldeke, “Beiträge zur Geschichte des Alexanderromans,” Denkschriften der Kaiserlichen Akademie der Wissenschaften in Wien, philosophisch-historische Klasse, 38, 1890, 5, 27–33. 3 Nöldeke, “Beiträge zur Geschichte des Alexanderromans,” 32, “Wie andre Geschichten so hat Muhammed natürlich auch diese auf mündlichen Wege erhalten.” 196 ALEXANDER LEGEND IN THE QUR’AN 4 J. Renard, “Alexander,” in EQ, vol. 1, 61–2. The article “Gog and Magog” in the same encyclopaedia (Keith Lewinstein, vol. 2, 331–3) at least mentions the Syriac Alexander Legend and cites Nöldeke’s work, but it does not refer to Nöldeke’s thesis that this text was a source of the Qur’an, rather citing the Alexander Legend with extraordinary understatement inconclusively as one of a few “suggestive parallels.” 5 See note 8. 6 Like Gilgamesh, the Alexander of this Legend travels to the edges of the world where he found a sea of deadly waters, the touch of which meant death. Also like Gilgamesh, this Alexander journeys through the passage through which the sun passes every night, entering it at sunset and emerging at the eastern end at sunrise. Both Gilgamesh and Alexander follow the sun’s nightly course just after it sets, apparently having to pass through before the sun comes around again and catches them. In both stories, the sun’s passage is associated with a mountain, Mamu in the Akkadian and Mûsas in the Syriac, evidently related names. Thus there seems to have been an oral tradition of the Gilgamesh story that became associated at an unknown time with Alexander. Moreover, the Syriac Song of Alexander, written in reaction to the Alexander Legend, and a Talmudic account of Alexander contain more material derived from the Gilgamesh tradition (such as the search for the water of life). This points to the existence of a late antique Aramaic oral tradition of Gilgamesh in which the name Alexander replaced the more ancient hero’s name. References to literature on connections between Alexander, Gilgamesh, and the Qur’an are collected in Brannon Wheeler, Moses in the Quran and Islamic Exegesis, London and New York: RoutledgeCurzon, 2002, 10–37 (plus endnotes), though the conclusions reached there are sometimes doubtful (see note 8). 7 G.J. Reinink, “Heraclius, the new Alexander: Apocalyptic prophecies during the reign of Heraclius,” in G.J. Reinink and B.H. Stolte (eds), The Reign of Heraclius (610–641): Crisis and Confrontation, Groningen Studies in Cultural Change 2, Leuven: Peeters, 2002, 81–94; idem, “Alexander the Great in the Seventh-Century Syriac ‘Apocalyptic’ Texts,” Byzantinorossika 2, 2003, 150–78 (this journal is available online at http://byzantinorossica.org.ru/byzantinorossika.html where it can be viewed with DjVu software, which is, at the time of this writing, available elsewhere online as a free download); now out of date but full of useful information is F. Pfister, “Alexander der Große in Offenbarungen der Griechen, Juden, Mohammedaner und Christen,” in F. Pfister, Kleine Schriften zum Alexanderroman, Beiträge zur Klassischen Philologie 61, 1976, 301–37. 8 B. Wheeler, “Moses or Alexander? Q 18:60–5 in early Islamic exegesis,” Journal of Near Eastern Studies 57.3, 1998, 191–215 and idem, Moses, 10–36. Wheeler does not address directly Nöldeke’s hypothesis of the relationship of the Alexander Legend to Q 18:83–102, which is the subject of the present paper, though he does refer in his notes to Nöldeke’s work (“Moses or Alexander?” 201, n. 52; Moses, 138, n. 55 to chapter 1). This strikes me as an unfortunate oversight. While this is not the place to redraw Wheeler’s charts showing the supposed interrelationships of these texts, a few critical remarks are in order to guide the reader. In discussing the Qur’an, its commentaries, three different texts about Alexander (the Legend, the Song, and different recensions of the Romance), and then also the Talmudic story of Alexander, Wheeler has overlooked a good deal of relevant published research (e.g. see later in this note) but has almost completely avoided getting into the details of the texts that could be used to establish their real interrelationships. To take just one of the problematic conclusions as an example, his charts of affiliations (Wheeler, “Moses or Alexander?” 202–3; Moses, 17, 19) argue that the Babylonian Talmud is a source of the Christian Song of Alexander, which is extremely unlikely. He argues, without foundation, that when Qur’an commentators refer to extra-Qur’anic traditions, it becomes 197 KEVIN VAN BLADEL 9 10 11 12 impossible for the Qur’an to refer to the same extra-Qur’anic traditions; the Qur’an itself is cleared of relying on the same ancient traditions (Moses, 28–9). This and other problematic schemata aside, Wheeler has not included the Legend of Alexander in his chart of affiliations, but only the Song of Alexander, which has been shown not actually to be by Jacob of Serugh, as Wheeler seems to think: “Moses or Alexander?” 201; Moses, 17; following Nöldeke, actually, but missing much of the subsequent scholarship: for example, A. Baumstark, Geschichte der Syrischen Literatur, Bonn: A. Markus und E. Weber, 1922, 191; K. Czeglédy, “Monographs on Syriac and Muhammadan sources in the literary remains of M. Kmoskó,” Acta Orientalia Academiae Scientiarum Hungaricae 4, 1955, (19–90) 35–6; G.J. Reinink, “Ps.-Methodius: A concept of history in response to Islam,” in A. Cameron and L.I. Conrad (eds) The Byzantine and Early Islamic Near East I: Problems in the Literary Source Material, Studies in Late Antiquity and Early Islam 1, Princeton, NJ: Darwin, 1992, (149–87) 167 n. 73; S. Gero, “The legend of Alexander the Great in the Christian orient,” Bulletin of the John Rylands Library 75, 1993, 3–9, 7; and above all the introduction accompanying the standard edition of the Song of Alexander itself: Das syrische Alexanderlied. Die Drei Rezensionen, CSCO 454 (edition)-455 (translation), Scriptores Syri 195–6, Trans. G.J. Reinink (ed.), Louvain: Peeters, 1983. Compare Wheeler’s reference to “the brief so-called Legend of Alexander, which is often said to be a prose version of Jacob of Serugh’s (Song) . . .” (Wheeler, Moses, 17, no references given) with Reinink’s statement: “No scholar has seriously considered the possibility that the legend is dependent on the (Song)” (Reinink, “Alexander the Great,” 153). Not even Budge, who first edited the Legend, thought that it was a prose version of the Song; rather he supposed that they shared a common source (Budge, History of Alexander, lxxvii). As Reinink has shown, the Song of Alexander is to some degree a reaction to the Alexander Legend composed not many years after the latter, probably between 630 and 640 CE (Reinink, “Alexander the Great,” 152–5 and 165–8). On the translation of sabab as “heavenly course,” see my discussion later in the article. The verb yamuju here means to move as waves move. The reference to the armies moving like waves becomes important in what follows. Budge, The History of Alexander, text 256, line 12, trans. 145. Here Budge has misunderstood the passage leading to a nonsensical translation (The History of Alexander, text 260–1, trans. 148): “And when the sun enters the window of heaven, he straightway bows down and makes obeisance before God his Creator; and he travels and descends the whole night through the heavens, until at length he finds himself where he rises. And Alexander looked toward the west, and he found a mountain that descends, and its name was ‘the great Mûsâs;’ and [the troops] descended it and came out upon Mount Ülaudiâ.” The passage should rather be understood as follows: “And when the sun entered the window of heaven, he (Alexander) immediately bowed down and made obeisance before God his Creator, and he traveled and descended the whole night in the heavens, until at length he came and found himself where it (the sun) rises. He saw the land of the setting sun and found a mountain where he descended, named Great Mûsas, and they (the troops) descended and arrived with him. And they went forth to Mount Qlawdiyâ (Claudia).” The role of Mount Ararat, called Great (Mec) Masis in Armenian, in this story goes back to the very ancient times. At some unknown point it was identified as the mount Mamu (ma-a-mu) of Tablet IX of the Gilgamesh epic, where Gilgamesh finds a way into the passage through which the sun enters at nightfall. The later Arabic rendering of the story found in an Adalusian manuscript (on which more later) renders the name of the mountain as al-Judi, an Arabic name for another, smaller mountain at the northern end of Mesopotamia called Ararad. On the confusion about these mountains see M. Streck, “Djudi,” EI2, vol. 2, 573b-4a. On the various mountains 198 ALEXANDER LEGEND IN THE QUR’AN 13 14 15 16 17 18 19 20 21 22 23 24 25 known in Armenian as Masis, see J.R. Russell, “Armeno Iranica,” in Papers in Honour of Professor Marcy Boyce, Acta Iranica 25, Homages et opera minora vol. XI, Leiden: Brill, 1985 (447–58) 456. For the occurrence of al-Judi in a later Arabic translation of the Alexander Legend (discussed in the present article), see E. García Gómez, Un texto árabe occidental de la Leyenda de Alejandro, Madrid: Instituto de Valencia de Don Juan, 1929, Arabic edition 50, l. 24. K. Czeglédy, “The Syriac Legend Concerning Alexander the Great,” Acta Orientalia Academiae Scientiarum Hungaricae 7, 1957, (231–49) 245, who seeks to localize the author by reference to the rivers Haloras and Kallath. Reinink, “Entstehung,” 280, and idem, “Alexander the Great,” 161, suggested that the author’s home was Edessa or Amida. The Haloras (Arabic Haluras, Armenian Olor) is a high tributary of the Eastern Tigris, upstream north of Amida. Just beyond its head is a pass leading down from the Arsanias river, so that by this way one could cross between the Armenian valleys and northern Mesopotamia. Budge, The History of Alexander, text 269–270, trans. 154–5. Idem, The History of Alexander, text 270, trans. 155. Idem, The History of Alexander, trans. 146 and 156; text 257 l. 14 and 272 l. 11. It is well known that already in his own time Alexander was portrayed with horns according to the iconography of the Egyptian god Ammon. (A.R. Anderson, “Alexander’s horns,” Transactions and Proceedings of the American Philological Association 58, 1927, [100–22] 101.) But the problem here is not to illustrate the entire history of this image, something already investigated in detail by others, but rather to show the proximate source of the information used in the Qur’an. Josephus, De Bello Iudaico, 7.7.4. The ancient traditions on Alexander’s wall, its iron gates, and its location are treated amply by A.R. Anderson, “Alexander’s Horns,” 109–10 and especially in “Alexander at the Caspian gates,” Transactions and Proceedings of the American Philological Association 59, 1928, 130–63. The wall is discussed further elsewhere in the present article. In the Syriac Alexander Legend and in the Qur’an God gathers the peoples who will fight (Syriac nkannem, Arabic fa-jama‘nahum jam’an). In the earlier tradition of Revelation of John 20:7–9 it is Satan who “gathers” (sunagageîn) Gog and Magog to fight. It is likely that one should emend the text of the Qur’an from Yajuj (y’jwj) to Ajuj (’jwj) on the basis of the Syriac source combined with the attestation of the form Ajuj in Arabic recorded by al-Zamakhshari, Fakhr al-Din al-Razi, and Ibn Hajar (E. van Donzel and C. Ott, “Yadjudj wa-Madjudj,” EI2, vol. 11, 231a-3b). An unintended [y] may be easily read in that position (before initial alif ) by mistake in either Syriac or Arabic script. Many examples of this usage are found in W. Wright, Catalogue of the Syriac Manuscripts in the British Museum, London: Trustees of the British Museum, 1870–1872, vol. 3 (e.g. on 1090b, 1113a, 1127b). For example, surat Maryam (Q 19:2) is headed, “a dhikr of the mercy of your lord on his servant Zakariyya.” Here, dhikr clearly means “record” or “account.” Cf. surat Taha (Q 20:99), where the word apparently refers to accounts of former (Biblical) times. E.W. Lane, Arabic-English Lexicon, Book 1, part 4, 1285c, entry “sabab.” My study of Qur’anic sabab will appear elsewhere. The meaning “heavenly courses” is explicit in Q 40:37, asbab al-samawat, where Pharaoh wants to vie with Moses in reaching the asbab, the courses, of the heavens to behold Moses’ God. It also appears in Q 38:10, fa-l-yartaqu fi l-asbab, where God challenges those who vie with his allmastery to reach heaven by ascending by asbab; in Q 22:15, fa-l-yamdud bi-sababin ila l-sama’i thumma l-yaqta‘, which has been taken by many to refer to stringing a 199 KEVIN VAN BLADEL 26 27 28 29 30 noose from the roof of a house; it appears rather to be a challenge to ascend to the heavens by the extraordinary means of a heavenly course, but it is doomed to failure, as in the previous example; and in Q 2:166, where God says that the asbab (the heavenly cords) will collapse on judgment day. For this list of occurrences I used Arne A. Ambros, A Concise Dictionary of Koranic Arabic, Wiesbaden: Reichert, 2004, 126–7. K. van Bladel, “Heavenly cords and prophetic authority in the Qur’an and its Late Antique context,” Bulletin of the School of Oriental and African Studies 70, 2007, 223–247; Lane, Arabic-English Lexicon, article “thamanun,” Book 1, part 1, 355c-356a. Asbab in this verse is preferred as the lectio difficilior. Budge, History of Alexander, text 260, trans. 148. The cosmology of the Alexander Legend is very similar to that found in the Book of Enoch (1 Enoch) and a few of the elements of Alexander’s experiences of the far edges of the world are just like those encountered by Enoch. Like Alexander, Enoch visits the four corners of the earth. 1 Enoch 17–36 tells this story, in which Enoch sees the prison (1 Enoch 18:14, 21:10: Greek desmotérion, Ethiopic beta moq7h) for fallen angels and a place of punishment of the souls of sinners (1 Enoch 22:13, Greek hamartoloí, Ethiopic xat7’an) in a far western place. This is quite like the prison that Alexander draws those whom he sends into the deadly waters to test them. In his vision, it is the winds that serve as the pillars of heaven over the earth (1 Enoch 18:3: Ethiopic a‘7mada samay; Greek “foundation of heaven” steréoma toû ouranoû), pillars calling to mind perhaps the Arabic asbab al-samawat under discussion. Enoch also finds “gates of heaven” in the north (1 Enoch 34–6: Ethiopic x7wax7wa samay). In the portion of the work known as “The Book of the Heavenly Luminaries,” 1 Enoch 75–82, Enoch sees the gates of heaven (1 Enoch 72:2ff.: again Ethiopic x7wax7wa samay, sing. xox7t) and the windows (1 Enoch 72:3ff. Ethiopic maskot, pl. masak7w) to their right and left. The sun, moon, stars, and winds pass through these gates. (Words cited here are taken from the Ethiopic and the surviving Greek portions of 1 Enoch: Das Buch Henoch. Äthiopischer Text, J. Flemming (ed.), Leipzig: J.C. Heinrichs’sche Buchhandlung, 1902 and Apocalypsis Henochi Graece, M. Black (ed.), Leiden: Brill, 1970). It is likely that the author of the Alexander Legend knew the story of Enoch or shared its cosmology. Gates of Heaven, abwab al-sama’, are also mentioned in Q 7:40 and 54:11, where in the former case they seem to be portals leading to the Garden (al-Janna) and in the latter case they are the hatches through which rains come to earth in Noah’s story, reflecting the “windows of heaven” in Genesis 7:11 (Hebrew ÷rubbôt hammamayim). G.J. Reinink, “Die Entstehung der syrischen Alexanderlegende als politisch-religiöse Propagandaschrift für Herakleios’ Kirchenpolitik,” in C. Laga, J.A. Munitiz, and L. Van Rompay (eds), After Chalcedon. Studies in Theology and Church History offered to Professor Albert van Roey for his Seventieth Birthday, Orientalia Lovaniensia Analecta 18, Leuven: Peeters 1985, 263–81; “Ps.-Methodius,” 1992, 149–87; “Pseudo-Ephraems ‘Rede über das Ende’ und die syrische eschatologische Literatur des siebenten Jahrhunderts,” Aram 5, 1993, 437–63; “Alexandre et le dernier empereur du monde: les développements du concept de la royauté chrétienne dans les sources syriaques du septième siècle,” in L. Harf-Lancner, C. Kappler et F. Suard (eds), Alexandre le Grand dans les littératures occidentales et proche-orientales. Actes du colloque de Paris, 27–29 novembre 1997, Nanterre: Centre des sciences de la littérature de l’Université Paris X, 1999, 149–59; “Heraclius, the New Alexander;” “Alexander the Great,” 2003, 150–78. The Era of the Greeks began in 1 October 312 BC according to the Julian Era. The dates can be converted by subtracting 312–311 from the Common Era year. On the use of this era in Syriac see P. Ludger Bernhard, Die Chronologie der syrischen Handschriften, Verzeichnis der orientalischen Handschriften in Deutschland, Supplementband 14, Wiesbaden: Steiner, 1969, 110–2. One will notice that the Era of 200 ALEXANDER LEGEND IN THE QUR’AN 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 the Greeks, frequently called in Syriac tradition the Era of Alexander, does not actually correspond with the death of Alexander. The assumption of the author of the Syriac Alexander Legend, that the Era of Alexander began with Alexander’s death, is a mistake easy to make. He wanted only to signal the dates of his prophecies with an era in common use. The little that is known of the general early history and ethnic affiliation of the Sabirs is summarized by Peter B. Golden, An Introduction to the History of the Turkic Peoples: Ethnogenesis and State-Formation in Medieval and Early Modern Eurasia and the Middle East, Wiesbaden: Harrassowitz, 1992, 104–6. It is approved by, for example, Czeglédy, “The Syriac Legend,” 245. Earlier scholarship has used the signal of the year 628–9 CE in Alexander’s prophecy to date the text in different ways. Reinink, “Alexander the Great,” 2003, shows that the text was composed in 630 or just before that time. Reinink, “Alexander the Great.” For example, J. Howard-Johnston, “Heraclius’ Persian campaigns and the revival of the East Roman Empire, 622–630,” War in History 6, 1999, (1–44) 26–40. W.E. Kaegi, Heraclius, Emperor of Byzantium, Cambridge: Cambridge University Press, 2003. Reinink, “Alexander the Great,” 160–2. Idem, 163–4. Idem, “Die Entstehung” and “Alexander the Great.” Ibid. A.R. Anderson, “Alexander’s Horns,” 109–10. Idem, “Alexander at the Caspian Gates” gives an exhaustive account of the ancient and modern confusion over precisely where these gates and the pass that they blocked were located. They were at least since the first century CE mistakenly thought by many classical authors to have been located at the pass of Darial in the middle of the Caucasus. Later, around the seventh century, this site was confused with the pass at Darband along the Caspian Sea. Robert H. Hewsen, Armenia: A Historical Atlas, Chicago: University of Chicago Press, 2001, 89; cf. maps 56 and 57 on 66–7. Extensive bibliographies are found in E. Kettenhofen, “Darband,” Encyclopaedia Iranica, 7, 13–19; G.G. Gamzatov, “Da∫estan i: Cultural relations with Persia,” Encyclopaedia Iranica, 6, 568–75; D.M. Dunlop, “Bab al-Abwab,” EI 2, 1, 835. Hewsen, Armenia, 85, 90–1, contains maps of these walls and in particular a detailed close-up map of the wall at Darband. For a list of frontier walls built by the Sasanians: H. Mahamedi, “Wall [sic] as a system of frontier defense during the Sasanid period,” in T. Daryaee and M. Omidsalar (eds), The Spirit of Wisdom: Essays in Memory of Ahmed Tafazzoli, Costa Mesa: Mazda, 2004, 145–59; this study makes almost no use of relevant Greek, Armenian, and other sources. The Armenian History Attributed to Sebeos, trans. and comment. R.W. Thomson and J. Howard-Johnston, Liverpool: Liverpool University Press, 1999, 1, trans. 148. Chronicarum quae dicuntur Fredegarii Scholastici libri IV. Cum Continuationibus, Monumenta Germaniae Historica, Scriptorum Rerum Merovingicarum tomus II, Bruno Krusch (ed.), Hannover: Hahn, 1888, (1–214) esp. 153. Erich Kettenhofen, “Darband,” Encyclopaedia Iranica, 7, 13–19, 16. In so characterizing these names I paraphrase Reinink, “Heraclius, the New Alexander,” 2002, 85. On the tradition of these names in Arabic, see E. van Donzel and C. Ott, “Yadjudj wa-Madjudj,” EI2, 11, 231a-33b, and K. Lewinstein, “Gog and Magog,” EQ, 2, 331–3. Andreas, Commentary on Revelation, J. Schmid (ed.) in Studien zur Geschichte des griechischen Apokalypse-Texte I. Der Apokalypse-kommentar des Andreas von Kaisareia, Munich: Zink, 1955, 223 (kephalaion 63): eînai dè taûtá tines mèn Skuthikà éthne nomízousin huperbóreia, háper kaloûmen Ounniká, páses epigeíou basileías, hos horômen, poluanthropóterá te kaì polemikótera. “Some people think that these 201 KEVIN VAN BLADEL 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 (scil. Gog and Magog) are the Scythian, Hyperborean nations, which we call Hunnic, both most populous and most warlike, as we see, of the entire earthly kingdom.” Cf. Josephus, Judaean Antiquities, 1.6.1. For the complicated debate about the identity of the leader of the Türks (or their Khazar subordinates) in these invasions see A. Bombaci, “Qui était Jebu Xak’an?” Turcica 2, 1970, 7–24. See also P. Golden, Khazar Studies, Budapest: Akadémiai Kiadó, 1980, 49–51. Theophanes, Chronographia, C. de Boor (ed.), Hildesheim: Olms, 1963.The Chronicle of Theophanes the Confessor, trans. and comment. Cyril Mango and Roger Scott, Oxford: Clarendon, 1997, 446–8. See also the modern synthesis of W. Kaegi, Heraclius, 142–6. History of the Caucasian Albanians, trans. C.J.F. Dowsett, London: Oxford University Press, 1961, 87. J. Howard-Johnston, “Armenian Historians of Heraclius: An Examination of the Aims, Sources, and Working-Methods of Sebeos and Movses Daskhurantsi,” in G.J. Reinink and B.H. Stolte (eds), The Reign of Heraclius, 41–62. The earliest known citation of the work is by Anania Mokac‘i, writing some time after 958, by which time the History of the Caucasian Albanians had a reputation of its own (Dowsett, History of the Caucasian Albanians, xv-xvi). Now that two palimpsest texts written in the Caucasian Albanian (A„uan) language have been discovered in the Sinai and deciphered, proving that there was at least an ecclesiastical literary tradition in this language, it is possible to wonder whether the description of these invasions of Caucasian Albania was originally composed in the local literary language before being translated into Armenian. It is noteworthy that the anonymous author of these passages states that he came from the village of Ka„ankatuk‘ (Dowsett, History of the Caucasian Albanians, 84; this is the reason the compiler Movses Dasxuranc‘i is sometimes mistakenly called Ka„ankuac‘i), located very near Partaw, the capital of Caucasian Albania (see the map of Hewsen, Armenia, 41). For more on the discovery and decipherment of the new Caucasian Albanian texts by Zaza Aleksidze, see the internet site armazi.unifrankfurt.de, following the link “Albanica.” Dowsett, History of the Caucasian Albanians, 83–4 (Book 2, chapter 11). Budge, History of Alexander, text 263–5, trans. 150–1. Dowsett, History of the Caucasian Albanians, 75 (Book 2, chapter 9). Edition: Movses “Ka„ankatuac‘i”, Patmut’iwn A„uanic‘ Amxarhi, M. Emin (ed.), Tbilisi, n.p., 1912 (reprint of M. Emin’s 1860 Moscow edition), ed. 144. Reinink, “Alexander the Great,” 165. Idem, “Pseudo-Ephraems ‘Rede über das Ende’,” 455–63. Idem, “Pseudo-Methodius,” and “Alexander the Great,” 171–7. G. Reinink’s (“Alexander the Great,” 2003, 152) more general remarks on the interrelationships of early eight century apocalypses are worth repeating: “The postulating of some older ‘common source,’ which is supposedly lost today, does not always form a satisfactory explanation of the differences between these texts and especially not, if we should completely ignore the specific literary and historical conditions under which each of these works came into being, conditions which may have led to certain reinterpretations, adaptations and modifications of the existing tradition.” Stephen Gero, “The legend of Alexander the Great,” 7. Budge, History of Alexander, text 155, trans. 270 l. 1. On Ghassanid monophysitism and their connections with Heraclius, see I. Shahid, Byzantium and the Arabs in the Sixth Century, Washington, D.C.: Dumbarton Oaks Research Library and Collection, 1995, vol. 2. On the raid at Mu’ta see W. Kaegi, Byzantium and the Early Islamic Conquests, Cambridge: Cambridge University Press, 1992, 171–4. Reinink, “Alexander the Great,” 165–8. 202 ALEXANDER LEGEND IN THE QUR’AN 64 F. Donner, Narratives of Islamic Origins: The Beginnings of Islamic Historical Writing, Princeton: Darwin, 1998, 228–9, includes a basic bibliography for early Islamic apocalypticism. On apocalyptic feeling in other sources from this period, see Reinink, “Heraclius, the New Alexander,” 81–3. 65 N.M. El Cheikh, Byzantium Viewed by the Arabs, Cambridge, MA: Harvard University Press, 24–33. 66 New Oxford Annotated Bible, 2nd edn. rev. 67 Dowsett, History of the Caucasian Albanians, 75; Tbilisi edn, 144. Cf. the Gospel text of Luke 21:25 edited by Beda O. Künzle, Das altarmenische Evangelium, New York: Lang, 1984, 1, 205: i yahe„ barba˝oy ibrew covu ew x˝ovut‘ean. 68 Tbilisi edn 154, text, Dowsett, History of the Caucasian Albanians, 84. 69 Tbilisi edn 156, Dowsett, History of the Caucasian Albanians, 85. 70 Dowsett, History of the Caucasian Albanians, 88. 71 Idem, 89. 72 Idem, History of the Caucasian Albanians, 83, Tbilisi edn, 153, l. 26. 73 Theophanes, Chronographia, 315–16. 74 Chronicarum quae dicuntur Fredagarii, 153. 75 Anderson, “Alexander at the Caspian Gates,” 135. 76 New Oxford Annotated Bible, 2nd edn. rev. 77 The controversial thesis was published under the name Christoph Luxenberg, Die Syro-aramäische Lesart des Koran, Berlin: Das arabische Buch, 2000. Detailed review of this work can be found in other articles in the present volume. 78 For evidence that there was once a more substantial knowledge of the Alexander Legend in Arabic, at least as early as the mid-ninth century, including some investigation of the Qur’an commentaries on Q 18:83–102, see K. van Bladel, “The Syriac sources of the early Arabic narratives of Alexander,” in H.P. Ray (ed.), Memory as History: The Legacy of Alexander in South Asia, New Delhi: Aryan International, 2007, (54–75) 64–67. See also the important study of the Alexander Romance in Arabic by F.C.W. Aerts, Alexander Magnus Arabicus: Zeven eeuwen Arabische Alexandertraditie: van Pseudo-Callisthenes tot Suri, Dissertation, University of Leiden, 2003. 79 This was made amply clear by Robert G. Hoyland, Seeing Islam as Others Saw It: A Survey and Evaluation of Christian, Jewish, and Zoroastrian Writings on Early Islam, Princeton, NJ: Darwin, 1997. 80 I thank Alexander Treiger for commenting on this paper in discussion with me at various points in its composition and for finding a number of errors in a draft version. All the views expressed herein and any remaining errors are solely my responsibility. 203 9 BEYOND SINGLE WORDS Ma’ida – Shaytan – jibt and taghut. Mechanisms of transmission into the Ethiopic (G7‘7z) Bible and the Qur’anic text Manfred Kropp This is a workshop report on ongoing research; bibliographical references have been reduced to a minimum. The aim of the report for this conference is to present the first salient results of project, “Ethiopic influence on the Qur’an and Early Islam,” that I have undertaken since 2002. It developed out of the larger frame of the Thesaurus linguae aethiopicae, meant as a collection of digital Ethiopic texts and presented on-line to the scientific community.1 The Ethiopic influence on the Qur’an has been recognized for some time but the treatment of the question has suffered several drawbacks. Other aspects of Qur’anic research, including the influence of Syriac/Aramaic language, monophysite-Jacobite or Nestorian theology and arguments, and the special question of Jewish ideas, have been demonstrated and treated at length. After the fast progress of South Arabian/Sabaean studies, the influence of Old South Arabian religion and culture were discussed to some extent. Yet besides these studies the question of Ethiopic influence has been treated by and large only from the point of view of philology, more precisely in form of the discussion of possible loan words from Ethiopic in (Classical) Arabic and thence into Qur’an. The two fundamental studies remain those by T. Nöldeke in 1910 and the materials contained in A. Jeffery’s 1938 book.2 Commentary on possible theological influence from the Ethiopian side on Muhammad’s views and teachings remained vague and casual, perhaps due to the rather marginal importance and relevance of Ethiopian Christianity in the framework of scientific research on Christian Oriental churches and theologies. Now it is evident that the loan words are the best and clearest indicators of influence. But even these have not been really studied exhaustively; many questions have been left open, even in the magistereal study of Nöldeke and those of his followers, up to the recent compilation of the results of these studies in Leslau’s Comparative Dictionary of 1987. 204 BEYOND SINGLE WORDS The question has to be studied starting from certain Ethiopic loan words in the text of the Qur’an. Thereafter the facts have to be scrutinized: were these words already known in the contemporary Arabic of Muhammad’s time or are they innovations of the Qur’anic text and message? Certainly, this crucial question cannot always be answered, since our documentation of pre-Islamic Arabic is meager and the authenticity of supposedly pre-Islamic literature (i.e. Jahili poetry) is doubtful. Thus the research of Ethiopic influence beyond single words must nevertheless start with specifically religious words and concepts. Even if we cannot tell whether these words were already current in pre-Islamic Arabia before Muhammad’s time and before his message, these terms form a special layer of the vocabulary expressing religious concepts in the core of that message. If we can say and decide that they are definitely Ethiopic, and not, let us say, Aramaic, Syriac and so on, or at least that they have not arrived directly from these spheres, then we will have an important statement about possible religious and cultural influence. In this volume Böwering rightly points out that still today we cannot precisely identify textual parallels in the Qur’an and Jewish or Christian texts. Instead we have only loose or vague allusions which point to a rich oral transmission to Muhammad of the concepts and contents of foreign texts, not to a direct contact by reading or reciting those texts. Thus it would be very important to identify a specific channel, which has not yet been rightly defined, of this oral tradition. I need not argue for the importance of Ethiopian, especially Aksumite, history, the knowledge of which must serve as a background for any discussion of Ethiopian influence on pre-Islamic Arabia, Northern Arabia and Early Islam. As for the political, religious and cultural history of the Horn of Africa, Ethiopia from attested historical times – second millennium BC in Egyptian sources – has its own history and plays an important role in the commercial and cultural exchange between Africa, the Indian Subcontinent, the Arabian Peninsula and the Mediterranean world. Needless to say, this region, I mean the Red Sea, had and has its solid ties to the neighboring African countries and regions. Thence it acquired its specific products, animal hides, ivory, musk, gold and so on, and sent them to the civilized regions to the North. For a lucky moment in history, starting perhaps from the third century BC and lasting till the eighth century AD, this commercial and cultural exchange allowed for the emergence of an empire, a widespread urban civilization on the Ethiopian plateau. For almost a millennium, Aksum participated rather actively in the political, cultural and religious life of the oecumene. Linguistically the medium of this culture were Semitic languages, Classical Ethiopic, G7‘7z, attested from the fourth century BC at the latest, if not earlier. Amharic, Tigrinya, Tigre and so on were to follow later. These languages were strongly influenced by underlying Cushitic, which acted as an adstratum or a substratum. The script also, the South Semitic script (e.g. in the Sabaean script) underwent a fundamental change: the writing direction changed to left-to-right. At the same time that the Bible was translated a pseudo-syllabary was introduced, 205 MANFRED KROPP perhaps under Indian influence: each letter received seven slightly different forms which now expressed regularly a consonant and a following vowel. The Christianization of Ethiopia and the subsequent translation of the Bible and other theological and secular literature – mostly from Greek – marked a deep change in the civilization and culture of Ethiopia. Whatever the relationship of the first Ethiopian states on the plateau and the South Arabian states may have been – whether there were colonies or independent Ethiopian states assimilating elements of a neighboring culture or society in South Arabia – in the third century AD Ethiopia was exerting military and political force in Yemen. It established garrisons and stable settlements in the Tihama. Probably the Ethiopians never left Yemen again until the time of Abr7ha (Ar. Abraha), the Ethiopian but independent king of Yemen of the middle of the sixth century, and the subsequent Persian invasion, some decades before Islam. Some mosques in the Tihama till today show architectural features of Christian Ethiopic churches. But this Christian influence in Arabia is not without rivals. The peninsula sees the establishment of Jewish communities from the Northwest to the South. Yemen becomes a Jewish kingdom in the fifth century which is the direct cause for further Ethiopic invasions in the region – as told in the famous story of the martyrs of Najran. Thus it comes as no surprise to find all kinds of Ethiopic linguistic and conceptual influence in the text of the Qur’an. When it comes to the question of Ethiopic influence one has first to consider an important fact: Ethiopia is first Christianized by a Lebanese man, a merchant’s son of Tyre /Sur. The Ethiopian church formally and juridically depended on the Patriarchate of Alexandria, but most of the important and influential missionaries came from Syria and Palestine. The same region also heavily influenced adjacent Arabia: as a result it can be difficult to decide if a given word or a given influence came directly from Syria–Palestine or via Ethiopia to Arabia. Refined research on the history of the words, the context where they appear for the first time, their phonetic shape and semantics should be done in order to decide these questions. The following report will give the examples of three words where this has been done. But first a last word to this introduction: The Muslim tradition has it that Muhammad understood and spoke even Ethiopic.3 This may well be a topos of (Muslim) hagiographical literature, as I. Goldziher pointed out,4 but it may reflect Ethiopia’s significant historical role in Islamic origins. Ma’ida and Shaytan The first results of this ongoing research have already been published and may thus be simply summarized: Ma’ida in the Qur’an (Q 5, 112; 114) is a clear borrowing from Ethiopic ma’7d(d)7 with the meaning “table; dishes; banquet.” As usual for the words for “table” in these languages it is a foreign word, having no etymological explanation 206 BEYOND SINGLE WORDS in Semitic. The etymon is most probably (vulgar) Latin or Koiné Greek mágida “table; plate of wood, silver and so on,” only rarely attested in literature. The translators of the Ethiopic Bible replaced consequently tarapeza by ma’ida thus following the general guidelines for early Christian Bible translators: to use a popular, generally understood language, not necessarily a literary one. This is naturally even more the case when there was yet no literary language, when it had to be created exactly by Bible translation. For the case of Ethiopic and Ge‘ez, however, we have to acknowledge an already existing literary tradition, that seen the pagan Aksumite inscriptions. The plausible conclusion is that mágida was a currently used, “popular” word in G7‘7z at the time, part of a lexical layer of loanwords that resulted from the multifold contacts and exchanges between Ethiopia and the Mediterranean world in Late Antiquity. The Qur’an received the word al-ma’ida as part of a narrative motif, namely the passage of surat al-Ma’ida (5:111–15) that has led to so much guessing about the Biblical passage behind it (the “Our Father” or the Last Supper or Acts 10:9 the vision of St Peter?).5 At the end of this paper I will present a passage taken from an Ethiopic homily which “breathes” the atmosphere of the Qur’anic text. Shaytan proves to be a more complex but equally illustrative example of a Qur’anic loanword. Let us start with the fact that there is a genuine Arabic root √SHTN. with the general meaning “fetching water (from a well) by means (of a bucket and) a rope;” Shaytan, “rope,” then means in metaphorical use “snake, serpent” – from where later on the link to the “devil” was made – and was used as a proper name among pre-Islamic Arabs. These words originally have nothing to do with Qur’anic Shaytan “devil; Satan.” The interesting and illuminating problem is the phonetic shape as Shaytan. The meaning and lastly the word is certainly taken from the well-known “Satan” as present in nearly all the languages sharing a common (religious and linguistic) heritage with the Hebrew Bible. One could easily propose – and it would be a plausible proposal – to explain it as a kind of phonetic and popular etymological assimilation and adaptation into Arabic.6 But there is the fact of Ethiopic Shaytan in the Ethiopic Bible which precedes the Qur’an, and, moreover, the fact that al-Shaytan al-rajim is clearly not the “stoned devil” but “the cursed one” from r7gum in Ethiopic. As in the case of ma’ida the Ethiopic translator(s) found satanas in their Greek Vorlage and put regularly Shaytan. The conclusion in parallel to the above proposal is that this form was common, current, popularly used and understood by the people of the time. How can we explain this form by phonetic shifts? This requires several and admittedly (as of today) hypothetical steps. It starts with the phenomenon of strong imala (a → e and lastly → i) in spoken languages in Palestine.7 As a second step such long vowel tends to become a diphthong under accent.8 The final assumption is that the missionaries and translators coming from Syria preferred again this – hypothetical as of now – popular form to literary “Satan,” unless this form was already received and used by the Ethiopian people at the time.9 207 MANFRED KROPP New examples The word pair jibt/taghut in the Qur’an versus g7bt/ta‘ot in the Ethiopic Bible The question of Ethiopic influence on the Qur’an has been neglected to some extent as has been said. The question of the two words jibt/taghut is a slight exception to this statement. The fact that Muslim commentators and lexicographers recognized jibt nearly unanimously as Ethiopic, giving it normally the meaning of “soothsayer” or something similar, led modern (Western) scholars to conclude the same. The meaning there was given according to its actual meaning in the Ethiopic Bible10 – amal7ktä g7bt “the new and foreign gods” or similar as will be explained later. How it came to stand in a pair in the Qur’an with the second hotly disputed loanword taghut “idol(s)” has not been really explained, nor has the precise origin and development of the particular Qur’anic and Arabic form taghut in comparison to (Palestinian–)Aramaic ta‘utha and Ethiopic ta’ot. Modern, well-organized and technical research disciplines, numismatics for example, have adopted the rational usage of shortcutting the older scientific discussion and literature by giving one bibliographic reference where an exhaustive bibliography can be found, and taking the knowledge of the foregoing arguments for granted: “ . . . für ältere Auffassungen siehe . . . .” and then starting right away with the argument on the schedule. I do not say that Oriental philology has reached this state of organization and internal coherence but for the sake of shortness refer for the whole argument and its treatment in earlier scientific discussion to Wahib Atallah, “Jibt et taghut dans le Coran,” Arabica, 17. 1970, 69–82.11 After this rapid shortcut I must only add that I do not share Wahib Atallah’s new thesis that jibt goes back to qibt “Egyptian” and taghut has to do with the Egyptian deity Thot, for phonetic reasons and because I doubt the proposed meanings of these words in the Qur’an. Thus I come right away to the exposition of what I have to say on these two words, according to the following guidelines. After a look into the Ethiopic Bible text and its relationship to its Greek Vorlage and – in the broader perspective – to their common Vorlage in the Hebrew Bible I will see if the results provide an adequate explanation of the Qur’anic facts. We will see that the two words are definitely linked in usage and meaning already in the Ethiopic Bible to such a degree that they are virtually synonymous. Furthermore we will see how the manifold, complicated and diverse terminology in the Hebrew Bible for the details and utensils of religious cults, especially heathen cults, has been translated first in the LXX in Greek, and subsequently with considerable freedom into Ethiopic, after the application of all the different translation techniques at the disposal of the Ethiopic translator(s). There is no doubt about the – isolated – meaning of the Ethiopic word g7bt. Derived from √WGB, rarely used as a finite verb, the adjective/noun means “new; invented; casual; sudden; unforeseen.” It is mostly used as an adverb.12 208 BEYOND SINGLE WORDS In the Ethiopic Bible it renders in the first place Greek prosfatos “fresh, new,” for example, Sir 9:10: i-t7wd7g ‘arkä-kä zä-t7kat, 7smä i-y7käww7nä-kkä käma-hu ‘arkä g7bt; “Do not forsake your friend of old, because the new one will not help you like him.”13 Referring to god(s) and deities the next relevant passage is the oft-cited Psalm 81(80), 9(10): “You shall have no foreign god among you, you shall not bow down to an alien god!” where Hebrew el zar, el nekar corresponds to Greek theos prosfatos, theos allotrios and Ethiopic amlakä g7bt, amlak näkkir. The semantic shift from “new” to “foreign, alien” is clear. Typical for the stereotype and repetitive polemical utterings in the Hebrew OT against alien gods, foreign cults and new idols, which require a sophisticated terminology for description, is Deuteronomy 32:17: “They sacrificed to demons which are not God – gods they had not known, gods that recently appeared, gods your fathers did not fear;” Greek prosfatoi is rendered and even explained in Ethiopic amal7ktä g7bt 7llä y7gäbb7ru wä-i-y7bäqwqw7‘u “new alien gods which they recently themselves had fabricated and are of no use ⫽ without power.”14 This addition makes clear that the translator did not only translate the specific passage in question, but also had a clear idea of the category to which all these “stelae, idols, altars, images, new and alien gods”15 belong in the Biblical polemics of the partisans of Yahweh. That is, he had an idea of idols in general, if you think of the material representation and fabrication, which was rendered by ta‘ot (still to be discussed) and of “new, foreign, alien deities,” if you think of the theological concept, which was rendered by amal7ktä g7bt. Deuteronomy 7:5 is another nice example of the developed terminology not to be discussed in detail here. But we have to treat the question of interchangeability between the two aforesaid categories; in fact, Paralipomena 2 ⫽ 2 Chronicles 23:17 and 24:18 may serve as an illustration where Greek idola, regularly translated by ta‘ot in Ethiopic, is rendered by amal7ktä g7bt “new, alien gods.”16 Thus given evidence for the interchangeability of the two terms in the Ethiopic Bible (both meaning “vain, new, foreign gods and their respective idols”) we have to have a look at the word ta‘ot. The meaning thereof is clearly attested in many passages in the OT as “idol.”17 The origin and etymology of the word is clear and has long been recognized: Aramaic ta‘u(tha) is the widely attested grammatical form.18 The specific meaning “idol,” derived from the general sense “error” is only attested in Western (Jewish) Palestinian Aramaic.19 Finally we venture into the Qur’anic text and the attestations of jibt and taghut there. The communicative situation of the Qur’anic passages in question is remarkably similar to that in the Bible. The text attacks polemically either people who are worshipping idols instead of God, or people who after having received a part of the revelation (nasib min al-kitab) are turning nevertheless toward idolatry. Jibt is a hapax legomenon and comes in pair with taghut in Q 4 (al-Nisa’):51: a-lam tara ila lladhina utu nasiban mina l-kitabi yu’minuna bi-l-jibt wa-ltaghut?! “Don’t you look at these people who, having received a share of the revelation, believe (nevertheless) in new and alien gods and in the idols?!”20 The second element of the Ethiopic expression amal7ktä g7bt “gods recent” has 209 MANFRED KROPP been understood as a kind of proper name or specification which may stand alone, just as taghut does, without the general regent “gods.” There remains a word to be said about the Arabic form taghut, with the ghayn. This certainly has been triggered by the affinity with the homophonous Arabic root √TGHY widely attested in the Qur’an, meaning “to oppress, to be a tyrant.” And a good number of later commentators made speculations and guesswork about the meaning of taghut derived from this root. After the discussion of the Ethiopic origin this is not my topic and problem. But my problem is, naturally, why Muhammad and the “original” reading or recitation of the Qur’an did not have ta‘ut rather than taghut? It is difficult for the moment to find decisive arguments for one possibility. But the Qur’anic text has proved to be well informed about other Ethiopic linguistic details (see earlier). Moreover, at the moment I am hesitant to draw conclusions about the authenticity of the canonical reading and recitation of the Qur’an. Regarding the word ma’ida: Qur’anic reflection of an Ethiopic homily? This last paragraph is meant only to give an impression of the “atmosphere” of the emerging research on the larger context of Qur’anic passages where Ethiopic loan words appear. The story told in Q 5 (al-Ma’ida):111–15 has not found a convincing parallel in the Gospels and related texts. 111. And when I inspired the disciples, (saying): Believe in Me and in My messenger, they said: We believe. Bear witness that we have surrendered (unto Thee) “we are muslims.” 112. When the disciples said: O Jesus, son of Mary! Is thy Lord able to send down for us a table spread with food from heaven? He said: Observe your duty to Allah, if ye are true believers. 113. (They said:) We wish to eat thereof, that we may satisfy our hearts and know that thou hast spoken truth to us, and that thereof we may be witnesses. 114. Jesus, son of Mary, said: O Allah, Lord of us! Send down for us a table spread with food from heaven, that it may be a feast for us, for the first of us and for the last of us, and a sign from Thee. Give us sustenance, for Thou art the Best of Sustainers. 115. Allah said: Lo! I send it down for you. And whoso disbelieveth of you afterward, him surely will I punish with a punishment wherewith I have not punished any of (My) creatures. The comments of Western scholars on this passage are manifold on one side, convergent on the other: they propose different passages from the NT as possible sources – usually the “Our Father” or the Last Supper or Acts 10, the vision of St Peter – but they agree on emphasizing either the ignorance of the Arab Christians at Muhammad’s time, or Muhammad’s misunderstandings of what he had been told or otherwise received from the Christian message.21 Could it be that 210 BEYOND SINGLE WORDS they were too focused on Biblical texts alone, or the extrabiblical Jewish and Christian traditions and texts to the exclusion of the Ethiopic heritage? I am fully aware that the Ethiopic text I am proposing as having “Qur’anic atmosphere” is far from offering literal or otherwise direct parallels to surat al-Ma’ida, 111–15. But in both texts emphasis is laid on the common table: of the Saints in the Ethiopic text, of the disciples of Jesus in the Qur’an. And the table (and the light descending on it) is in both texts a miracle and a feast for the believers. The passages are taken from the homily of a bishop of Aksum on one of the “Syrian saints” of the Ethiopic church who lived and evangelized, as the tradition has it, in Ethiopia in the fifth century AD. The homily, which is translated later, should have been written not too long thereafter. But first two last remarks: One, the Ethiopic text has been edited, but not translated, now more than one hundred years ago. I had the occasion above to make a general remark on the state of Near Eastern philology and history – certainly due to the relatively small number of “laborers in the vineyard,” especially so when Christian Near Eastern subjects are concerned. How often has this text been read by specialists in these more than one hundred years? Two, I want to repeat that the research on Ethiopic influence on the Qur’an and early Islam “beyond the single words” has just begun. Notwithstanding this very tentative beginning, it seems a promising field. The Homily of John, Bishop of Aksum, in Honor of Gärima.22 (The nine saints from the Byzantine empire have come to Ethiopia, share a house there, celebrate common prayers and dedicate their lives to preaching and works of charity). Every time they came together at the table (ma’7d(d)7), lights descend on them shining like the sun; they tasted food only at dawn in the morning and on the Sabbath, on Christmas, the day of His birth, and on the day of Epiphany; on the day of the miracle of Qana they did not taste food except the Eucharist (flesh and blood of Christ), and on Easter and Pentecost they ate only three grapes of wine. (p. 153, line 117ff) Angels descend to tell them to disband, that each of them might search his own region to preach; the angels are welcomed in the saints’ house. And when they had completed the washing of the feet of the angels and had drunk the remaining water they sat for meal and the light appeared as usual. When they had finished the table (ma’7d(d)7) they heard a voice coming and calling on them: . . . Angels were visiting them, they had visions of our Lord, until they were capable of resurrecting the dead, healing the lepers by their prayer, as well as making the blind see and the 211 MANFRED KROPP lame walk and the deaf hear. Thus they performed many miracles and the lights as usual kept descending from heaven on their table (ma’7d(d)7). Thus they stayed for 13 years. (p. 157, line 228ff) Then a monk came to them named Malkiyanos who did not know any other work than women’s hairdressing. He made his living out of that. He did not sit at their table and he ate his bread alone. Thus they rebuked him, but thereafter these lights departed from them and shone above that monk. The saints then asked: “Where are our lights that used to come down (from heaven) on our table (ma’7d(d)7)? Why are they now hidden from us?” The monk answered them: “My masters, today this light is shining upon me!” At this moment they understood and said: “That happens because we have rebuked this monk!” They embraced each other, wept bitterly for 40 days and nights . . . and then they departed separately, each one in his new region. (p. 159, line 295ff) Conclusion Let us briefly summarize the results of the research project hitherto achieved: – – – The Ethiopic Bible and the Qur’an have in common a layer of mostly religious terminology. Some of this terminology is of foreign origin in both languages. When of Aramaic origin then Palestinian Aramaic (to the exclusion of other branches of Aramaic) seems to be a prominent donor language. The Ethiopic Bible text precedes chronologically the Qur’an.23 The translator(s) of the Ethiopic Bible use(s) different techniques for rendering specific terminology, in most cases religious terminology, including a b c neologisms and or explanatory translations with genuine Ethiopic words; simple transliterations of the foreign – mostly Greek – words (some of which become current words in the literary tradition afterward); foreign words which are not identical with those in the Vorlage. The presumption is that these words were already spoken, popular ones, part of a lexical layer of Old Ethiopic (G7‘7z) which borrowed from the international languages in the Mediterranean and the Red Sea worlds of late Antiquity (Greek and Aramaic). As in the concrete translation process at least two partners certainly participated – Christian missionaries having arrived from outside and cultivated native Ethiopian speakers. The fact that, according to the Ethiopic tradition most of these missionaries (monks and clerics) came from “Syria,” is of importance (see earlier Palestinian Aramaic). Thus one may speak of cultural and linguistic contact and exchange in a channel of oral transmission. 212 BEYOND SINGLE WORDS – – – Some of these foreign words are to be found later in the Qur’an and can be defined as Qur’anic innovations, that is, they were not present in Northern Arabic in Muhammad’s time, at least as far as we can tell. Cross checking is only possible via the only preserved attestations of pre-Islamic Northern Arabic, the very rare inscriptions and pre-Islamic Arabia poetry (the latter being highly disputed in its authenticity). On the basis of phonetic shape and semantic peculiarities it is probable that some of these Ethiopic words – even when ultimately of Aramaic origin – came into Arabic and the Qur’an via direct and possibly oral transmission: Ethiopian merchants, mercenaries, slaves and, why not, priests, or – perhaps most importantly – Muslim Arab muhajirun of the first hijra to Ethiopia. This last assumption is corroborated by the fact that the respective Qur’anic passages are all (late) Medinan ones. These words are not isolated lexical items, not isolated borrowings. The context of Ethiopic words in the Qur’an demonstrates that they are part of transmitted material. This material has to be scrutinized and taken into account carefully. Future research along the guidelines of this working hypothesis on the respective vocabulary and passages in the Ethiopic Bible and the Qur’an will probably yield new results which can enable us to draw a more detailed, precise and correct picture of Ethiopic influence in general on the Qur’an and early Islam. This, it is hoped, will do justice to this important but rather neglected field of linguistic, religious, political, cultural, and economic exchange between both shores of the Red Sea. Notes 1 This larger enterprise has suffered considerable delay, caused first by technical difficulties. These are being slowly overcome, by the general progress of technology suitable for philology and languages written in non-Latin script, and with the good service of Mr Reinhard Hiß, the technical specialist for this project. The second reason was my somewhat unforeseen and unplanned transfer from Mainz University to the Orient-Institute in Beirut to serve as a director for now more than six years. The third reason is the behavior of colleagues at Mainz University during my absence. This has led me to the conclusion to organize and research only individual projects, without personal or financial resources from third parties – the specifics and peculiarity of studies in the field of humanities allowing. 2 T. Nöldeke, “Lehnwörter in und aus dem Äthiopischen,” Neue Beiträge zur Semitischen Sprachwissenschaft, Strassburg: Trübner, 1910, 31–66; FV. 3 Cf. FV, 12–14. To cite one of the more critical scholars as to the possibility and probability of direct (religious) Ethiopic influence on Muhammad and the Qur’an: “There are many loanwords in Arabic aside from those in the Koran . . . and something is known in regard to their origin. S. Fraenkel, Die aramäischen Fremdwörter im Arabischen, 210–16, in discussing the numerous Arabic words of Ethiopic origin dealing with ships and shipping, showed that these are a partial fruit of the long period during which the Arabs and Abyssinians were associated . . . in charge of the traffic 213 MANFRED KROPP 4 5 6 7 8 9 10 round the Red Sea.” C.C. Torrey, The Jewish Foundation of Islam, New York: Ktav Publishing House, 1967 (New edition of the Hilda Stich Stroock Lectures at the Jewish Institute of Religion, 1933 and introduction by F. Rosenthal), 53. The project described in some examples in this paper is a contribution to this undecided question. The fact of the first hijra to Ethiopia and its reasons is discussed by Frants Buhl (Das Leben Muhammads, Darmstadt: Wissenschaftliche Buchgesellschaft, 1961, 171–4). He married a woman who had lived in Ethiopia for a long time, precisely one of these first muhajirun. An anecdote of the later Medinan time of Muhammad tells that emigrants having come back from Ethiopia came with their children to the prophet. One of their girls embraced him and showed him nice yellow Ethiopian shirt. Muhammad exclaimed, joking: “sana, sana” (what can be read as “shanna(y), shanna(y)” also, meaning in Ethiopic: “nice, nice.”). The Arabic translation given in the text of Ibn Sa’d, K. al-Tabaqat, E. Sachau (ed.), Leiden: Brill, 1905–21, 4, 72 is “abli wa-akhlaqi” but the better version is in the edition: Beirut: Dar al-Kutub al-‘Ilmiyya, 1982, 4, 69, “abli wa-akhlifi,” meaning: “wear this cloth until it is worn, and then have another (one as nice as this)” ⫽ “I wish you may have always a cloth as nice as this one” (this expression in a slight different wording is known still today in vernacular Arabic). This anecdote is not only a good hint that Muhammad was able to converse and joke with children, as Watt has it (Muhammad at Medina, Oxford: Oxford University Press, 1956, 323 and note 5; idem: Muhammad. Prophet and Statesman, Oxford: Oxford University Press, 1961, 230). It is a hint in the Muslim tradition that Muhammad knew at least some Ethiopic words. The thorny question of a “revisionist” view of contested Muslim tradition can be retained, even if a radically skeptical view cannot be discussed here. I am fully aware of the fact that I am using part of this tradition to support arguments developed and supported from other sources. In my case, I am making points about the history of the Qur’anic text based on linguistic and philological methods; the results obtained have their own scientific value. A necessary comparison with the historical tradition, be it Muslim or otherwise, be it concordant or contradictory, is a second step which should not be confused with the first one. The independence of text-critical work (linguistic or philological) as a methodology is an essential condition if such work is to contradict historical tradition. A subsequent scientific discussion must then work out syntheses – or simply candidly acknowledge contradictions. I. Goldziher, “Linguistisches aus der Literatur der muhammedanischen Mystik,” ZDMG 26, 1872, (764–85) 770. For the full discussion see M. Kropp, “Viele fremde Tische, und noch einer im Koran: Zur Etymologie von äthiopisch ma’7d(d)7 und arabisch ma’ida/mayda,” OC 87, 2003, 140–3. A recent proposal was to interpret the ya’ as a mater lectionis for the long vowel a and thus to read right away “Satan.” For pre-Islamic Arabic in the region (Syria) it is proven by the existence of the form Ibrahim instead of Abraham; cf. Muhammad Abu l-Faraj al-’Ushsh, “Kitabat ‘Arabiyya ghayr manshura fi jabal Usays,” Al-Abhath, 17, 1964, 227–316; here no. 85 p. 302. Such phonetic laws and peculiarities seem to cling to the geographic region, across the time and change of languages. Western, Palestinian Aramaic is highly likely to share these features. Lebanese Arabic offers a lot of examples: hada “this” → hayda; a prominent one is berut “wells” → Bayrut (the name of the city). For the full discussion see M. Kropp, “Der äthiopische Satan ⫽ Shaytan und seine koranischen Ausläufer; mit einer Bemerkung über verbales Steinigen,” OC 89, 2005, 93–102. One has to add here a note on the character of Muslim research into the text and the history of the Qur’an. They tell us near to nothing about the historical origins of the 214 BEYOND SINGLE WORDS 11 12 13 14 15 16 17 18 19 20 21 text they try to explain – excepting the genre of the “legends of the prophets” which includes material from the Jewish and Christian tradition. For the rest – in our case what is needed is a simple investigation into Old Ethiopic and its Bible, something totally feasible to Muslim scholars – there is only guessing. Reading through the enormous material gathered by Muslim scholarship from early to modern times one soon gets the impression of a radical isolation of this scholarship from that which preceded it. Certainly, for the Islamic period this material forms a point of reference which has to be studied carefully for the history of Islam itself. However, the history of the origins of the Qur’anic and Islamic message is a totally different matter. That is perhaps not really doing justice to the contributions to the problem found in: R. Dvořak, “Über die Fremdwörter im Korân,” Kaiserliche Akademie der Wissenschaften. Phil.-Hist. Classe. Sitzungsberichte 109, 1, 1885, 481–562. T. Nöldeke, “Lehnwörter in und aus dem Äthiopischen,” 31–66. FV. But a future publication of the results of the project “Ethiopic Influence on the Qur’an” will have the necessary space to do full justice to 150 years of scientific research. Cf. A. Dillmann, Lexicon linguae aethiopicae (henceforth DL), New York: Ungar, 1955 (Reprint of Leipzig, 1865), 938; W. Leslau, Comparative Dictionary of Ge‘ez (Classical Ethiopic) (henceforth LCD), Wiesbaden: Harrassowitz, 1987, 608a. I am citing the Ethiopic text according the partial edition of the OT in G7‘7z with the And7mta-commentary: Mäsah7ftä Sälomon wä-Sirak. Addis Abeba: T7nfa’e Masatämiya D7r7jjs, 1988 a.m. ⫽ 1995; Reprint of the edition 1917 a.m. ⫽ 1924. Cf. A. Dillmann, Octateuchus Aethiopicus (⫽ Biblia Veteris Testamenti Aethiopica 1), Lipsiae: Vogelii, 1853–97, 157, apparatus criticus to Deuteromomy. To cite all the relevant terminology in Hebrew, for example, asherah, ‘asab, pesel and so on, Greek idola, glypta, cheiropoietoi and so on, is relegated to a future extensive study of this terminology in the framework of a study on its translations from Hebrew to Greek to Ethiopic. It is important to see first, that the Ethiopic translators dispose of separate, specific terms such as gelfo (glypta) (LCD 190a), m7sl “image” (LCD 365b), g7brä 7dawi-hon “work of their hands,” m7hramatä ‘om (asherah) “grove” (LCD 62a; DL 994) and so on. But thinking in categories he takes the freedom to add a general explanation to such terms in the text, or to translate a specific directly by a categorical one. This means that in the two earlier cited results the categorical terms (amlakä) g7bt and ta‘ot prevail in the texts. I am using here the Ethiopic text in the partial edition of the OT Mäshafä nägäft, Asmära: Mawbärä Hawaryat F7re Haymanot, 1974 a.m. ⫽ 1981. Cf. for example, Isaiah, 10:11; DL 1243 with further attestations; LCD 584a. But note that the Syriac word is different: ta‘yutha. Cf. (1) M. Sokoloff, A Dictionary of Jewish Palestinian Aramaic, (⫽ Dictionaries of Talmud, Midrash and Targum. (2), Ramat Gan (Israel): Bar Ilan University Press, 1990, 227b; M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic, (⫽ Dictionaries of Talmud, Midrash and Targum. (3), Ramat Gan (Israel); Baltimore: Bar Ilan University Press; John Hopkins University Press, 2002, 509a. It is certainly due to the very restricted and lacuna-filled documentation of Christian Palestinian Aramaic that the word in the same meaning is not (yet) attested there. Nevertheless, in this case as well, our hypothesis of oral transmission from Palestinian Aramaic to Ethiopic works. The other passages where taghut figures alone are partly parallel to this text; in every one of them the meaning “idols” for taghut fits: Q 2:256, 257; 4:76; 16:36; 39:17. To cite one of the more frequently used reference works: “5,112–115 Der Abschnitt über den Tisch (ma’ida), den die Jünger Jesu aus dem Himmel erbitten, bezieht sich offensichtlich auf die Einsetzung des Abendmahls. Vielleicht ist er aber auch durch die Vision des Petrus in Apostelgeschichte 10,10 ff beeinflußt. Jedenfalls ist der Sachverhalt mangelhaft erfaßt und weithin mißverstanden. Siehe W. Rudolph, Die Abhängigkeit des Qorans von Judentum und Christentum, Stuttgart 1992, S. 81 f.” 215 MANFRED KROPP R. Paret, Der Koran. Kommentar und Konkordanz, Stuttgart (etc.): Kohlhammer, 1971, 133. Yet Rudolph does not really widen the horizon, except for the mention of the thesis of Ebionite influence. 22 C. Conti Rossini, “L’omelia di Yohannes, vescovo d’Aksum in onore di Garimâ,” Actes du Congrès international des Orientalistes, section sémitique, Paris, 1898, 139–77. 23 By Bible text I mean not strictly the text of the Bible only, but rather the rich religious literature – homilies, lives of saints and so on – that developed quickly in Ethiopia after Christianization and the translation of the Bible. Biblical citations, for example, appear very soon in secular texts, such as royal inscriptions. 216 10 NASCENT ISLAM IN THE SEVENTH CENTURY SYRIAC SOURCES Abdul-Massih Saadi Introduction The Arab invasions of the seventh century marked the beginning of a dramatic change in the heartland of Eastern Christianity. The Arabs’ style until that time had been to overrun and pillage the landscape, and then, just as quickly, to withdraw to their desert. At this time, however, things were different. They called their new invasion :hijra, that is, Immigration,1 and the Syriac people called them :Mhaggrayê, that is, Immigrants. In the present paper I will ask how the Syriac Christians responded when the Mhaggrayê settled in this conquered land. How did they view the Mhaggrayê historically, religiously, and ethnically in the seventh century? The diverse Syriac Christians had varying opinions about the Mhaggrayê. However, their common description of the Mhaggrayê warrants attention and consideration. In this paper, except for comparative hints, I will avoid the Byzantine Christian sources. Byzantine Christians were attacked and lost their land to the conquering Muslims. Having faced this experience, Byzantine authors tend to be polemical. Additionally, I will avoid Arabic, Muslim authors not simply because they were victors who wrote in justification of their conquests, but in actual fact because there is no contemporary Arabic writing concerning the events in the seventh century. Arabic chronologies and other writings are not available before the late eighth century. Thus, I will limit myself to the Syriac writers, who were neither declared enemies nor outward friends of the new conquerors. The diverse Syriac Christians By the seventh century, Syriac Christians were ecclesiastically divided into four main groups. There were Chalcedonians (or Melkites) and Non-Chalcedonians. Each of these two groups was further subdivided into two groups. The Chalcedonians were either monotheletes or dyotheletes while the non-Chalcedonians were either “Nestorians” or “Jacobites.” The Syriac, Chalcedonian Christians, 217 ABDUL-MASSIH SAADI who had been at something of an advantage under the Byzantine rule because of their ecclesiastical affiliation with the dominant Byzantine emperor,2 grieved their loss and expressed anger at the invasion. The Non-Chalcedonian Christians, however, lost no advantage and were saved of religious and doctrinal persecution – the Jacobites had been oppressed by the Byzantines over doctrinal issues, and many Nestorians had undergone persecution by the Persian Zoroastrians. Thus, they viewed the advance of the Mhaggrayê from the standpoint of their own ecclesiastical affiliations, their relations with the previous political authorities, and their relations with the Arabs, whom they referred to as the Tayyayê. Historical orientation Before the invasions Syriac writers refer to Arabs in various contexts and with various appellations: ‘Arabayê, Tayyayê, Sons of Hagar and Ishmaelites. With the seventh century advance of the Arabs, however, additional names were employed, like Tayyayê of Muhammad, but most often: Mhaggrayê. The unprecedented name, Mhaggrayê, is provocative because it provides the greatest evidence for their self-identification as immigrants (muhajirun in Arabic). In other words, the name Immigrants (muhajirun ) implies that the Arabs had arrived to stake a claim on, occupy, and then inherit the land. The Syriac writers, reporting and repeating what they were hearing rather than inventing a historical event, merely Syriacized this native Arabic name. An early story circulated among the Syriac writers, also present in the contemporary history of Sebeos,3 that reflects the impression of Christian writers that the Arabs had come with the intention to stay. According to the story, when Muhammad visited Palestine, he admired the monotheism of the Jews and the fertility of the land, which “had been given to them (Jews) as a result of their belief in one God.” The story continues that when Muhammad returned to his tribesmen, declared: “If you listen to me, abandon these vain gods and confess the one God, then to you too will God give a land flowing with milk and honey.”4 That is to say, according to the Syriac writers, the Arab tribesmen misinterpreted and exploited a Jewish tradition to legalize wars and conquer lands. Meanwhile, the Syriac writers employed secular or political terms to address the leaders of Mhagrayye. They frequently called the Arab caliphs Kings (melkê ), the governors as princes (amirê ), rulers (shalitê , or rishê , or rishanê ).5 In a clearly ethnic approach to the Mhaggrayê as a group, the seventh century Nestorian John of Phenek (d. 690s) writes, “Among them (Arabs), there are many Christians, some of whom are from the heretics, others from us.”6 Concerning Muhammad, for the most part, he is described as “the first king of the Mhaggrayê,”7 but occasionally he is called Guide (mhaddyana )8 or Teacher (tera’a ),9 or Leader (mdabrana ),10 or the great ruler (shalita rabba ).11 None of the Syriac writers ascribed to Muhammad or the 218 NASCENT ISLAM IN SYRIAC SOURCES caliphs any religious title, instead, they regarded them as ethnic rulers, and they perceived their conquest as a temporary invasion sent by God in punishment. The invasion as God’s punishment The four groups agreed in principle on this basic precept: that the invasion was sent by God Himself in punishment for their sins. On Christmas Eve of 639, one year after the conquest of Palestine, the Melkite Patriarch, Sophronius of Jerusalem exhorted his congregation to repent so that God’s punishment may be removed, namely, the occupation of the Ishmaelites. He continued, “Through repentance, we shall blunten the Ishmaelite sword and break the Hagarene bow, and see Bethlehem again.”12 John of Phenek (690s) viewed the Mhaggrayê as a people sent by God to punish Christians on account of heresies, but also for laxity of faith of his own community.13 John also sees the Arab invasion as God’s punishment of the sinful Empires, namely the Byzantine and the Persian. He writes that “God called the Arabs from the end of the earth, to destroy through them a sinful kingdom and to humiliate through them the proud spirit of the Persians.”14 Often, however, Syriac authors blame one another for having brought down God’s anger. Jacobite authors report that because of the persecution of the Byzantines, the God of vengeance sent the Ishmaelites, the most insignificant of the peoples of the earth, from the land of the South; in this way we were saved from the tyrannical rule of the Byzantine.15 The Chalcedonian, monothelite author16 of the Syriac Life of Maximus attributed the sweeping invasion of the Arabs (Tayyayê) to all the lands and islands where Maximus’ heresy (dyotheletism) was present, a result of God’s wrath.17 Conversely, the dyothelete Anastasios considers the Arab victories as God’s punishment for the emperor Constans II’s pro-monothelite belief.18 The common assumption among all Syriac groups that the advance of Mhaggrayê was a divine act of punishment implies that they all expected a quick end, or some kind of closure of the Mhaggrayê’s dominance. Writing in last decade of the seventh century, John of Phenek stated, “From that time on the kingdom of the Tayyayê was no longer firmly established.”19 This was not merely conjecture. In fact, all sources that date to the late seventh century saw in the “First and Second Civil Wars” among Arab political and tribal factions a sign of their total destruction.20 The Mhaggrayê’s religious orientation The Syriac writers were the first people to report about and eventually engage with the Mhaggrayê on religious matters. The earliest Syriac document, dated to 644, reports a religious colloquium between the Emir of the Mhaggrayê and the Syrian Patriarch, John of Sedreh. The document refers to Mhaggrayê as 219 ABDUL-MASSIH SAADI having accepted the Torah just as the Jews and Samaritans.21 Moreover, the document refers to some learned Jews who were with the Emir of Mhaggrayê and scrutinized the Christians’ quotations of the Scriptures.22 Although a good portion of the discussion between the Emir of the Mhaggrayê is about the scriptures, it never refers to the Qur’an, a possible indication that the Qur’an was not yet in circulation. One of the earliest Syriac apocalypses in Islamic times, pseudonymously attributed to Ephraem, refers to the Mhaggrayê in religious terms as “the offspring of Hagar, handmaid of Sarah, who holds the covenant of Abraham, the husband of Sarah and Hagar.”23 Again, in reference to the Arabs’ religious practice, the seventh century Syriac Brief Chronicle sees nothing unusual for the Arabs (Tayyayê) to worship at the Dome (qubta ) of Abraham – a reference to the Ka’ba – since they had been doing so from the ancient days to pay homage to the father of their nation/ community (i.e. Ishmael).24 Moreover, the Brief Chronicle argues that the Arabs (Tayyayê) changed the name of the city of Yathrib to Medina after the name of the fourth son of Abraham, Midian.25 In a direct description of the Mhaggrayê’s religious orientation, John of Phenek describes the Mhaggrayê as worshippers of One God in accordance with the Old Law (Old Testament).26 John continues that the Mhaggrayê follow the instructions of Muhammad, who became their instructor (tara’a/mhadyana ) and inflicted the death penalty on any person violating his instructions.27 Later chronicles associated the building of the Dome of the Rock on the site of the temple of Solomon with the eschatological rebuilding of the temple by the Jews.28 None of the Syriac sources describe Mhaggrayê as Jews, but they perceive them as having a monotheistic belief with a Jewish precedent. Conclusion Although the Syriac writers did not intend to write a history of early Islam per se, what they did write is vital for shedding more light on that controversial period of events. In addition, religiously speaking, the Syriac writers believed early Muslims (Mhaggrayê) to be the descendants of Abraham through Ishmael and Hagar, people confessing the One God, who were brought to the region to punish the heretics, and to cause the faithful to repent. In other words, they were seen as a people with a divine task, but they were also to be banished upon the completion of that task. However, there is no clear indication that the Syriac writers recognized or realized the birth of a new religion called Islam, a term that they never employed. Notes 1 P. Crone, “The first-century concept of higra,” Arabica 3, 1994, 352–87. 2 Due to their affiliation with the ecclesiastical tradition of the emperor (Chalcedonian), their Syriac rivals called them :melkites, which means followers of the king. 220 NASCENT ISLAM IN SYRIAC SOURCES 3 Sebeos’ History, trans. R. Bedrosian, New York: Sources of the Armenian Tradition, 1985. 4 Sebeos’ History, 30, 123; Bar Hebraeus, Chronicon Syriacum, P. Bedjan (ed.), Paris: Maisonneuve, 1890, 97; Michael the Syrian, Chronique de Michel le Syrien Patriarche Jacobite d’Antioche (1166–1199), ed. J.B. Chabot, Paris: Leroux, 1899–1924, 2, 403 (ed.); 4, 405 (trans.); Chronicon ad annum 1234, CSCO 81, ed. J.B. Chabot, Louvain: Peeters, 1920, 227–8. The Syriac writers are here drawing on Dionysius of Tellmahre. 5 All these terms were used by Syriac writers even before Islam. John of Ephesus, Historiae Ecclesiasticae III, ed. E. Brook, Paris: E Typographeo Reipublicae, 1935, 80, 181, 217, 221, 224; Ahudemmeh, “Histories d’Ahoudemmeh et de Marouta,” Patrologia Orientalis 3, ed. F. Nau, Paris: Firmin–Didot, 1909, 25. For the seventh century, see Isho’yahb, Liber Epistularum, CSCO 11, ed. R. Duval, Paris: E Typographeo Reipublicae, 1904–5, 226. 6 A. Mingana, Sources Syriaques, Leipzig: Harrassowitz, 1908, 147*. 7 See the list of Arab “kings” in J.P.N. Land, Anecdota Syriaca, Leiden: Brill, 1868, 2, 11 of addenda; French translation by F. Nau in “Un colloque du patriarche Jean avec l’émir des Agaréens,” Journal asiatique 11, 1915, 225–79. 8 Mingana, Sources Syriaques, 146*. 9 Mingana, Sources Syriaques, 146*. 10 Chronica Minora, CSCO 1, ed. I. Guidi, Leipzig: Harrassowitz, 1903, 30. 11 Liber Epistularum, 226. 12 H. Usener, “Weihnachtspredigt des Sophronios,” Rheinisches Museum für Philologie 41, 1886, 508, 515. 13 Mingana, Sources Syriaques, 145*, 148* (text); 174*, 177* (trans.). 14 Ibid., 141*–142*. 15 Chronicon ad annum 1234, 236–7; Chronique de Michel le Syrien, 2, 412–13 (ed.); 4, 410 (trans.). 16 The author identifies himself as “George of Rish Ayna, the disciple of Sophronius.” 17 S. Brock, “An early Syriac life of Maximus the Confessor,” Analecta Bollandiana 91, 1973, (reprint: London: Variorum Reprints, 1984, XII), 312–13 (text); 318 (trans.): . . 18 W. Kaegi, “Initial Byzantine reactions to the Arab conquest,” Church History 38, 1969, 142. 19 Mingana, Sources Syriaques, 155*. 20 Ps.-Ephraem; Ps-Methodius; (John the Lesser) see The Gospel of the Twelve Apostles, ed. J.R. Harris, Cambridge: Cambridge University Press, 1900, 15–21 (text; 34–9 (trans.); For text and translation of John of Phenke see book XV, A. Mingana, Sources Syriaques, 143*–179*; see also S. Brock, “North Mesopotamia in the Late Seventh Century: Book XV of John of Penkaye’s Ris Melle,” JSAI 9, 1989, 51–75 (reprint: Studies in Syriac Christianity, London: Variorum, 1992) on the historical background of the Arab Civil Wars, see G.R. Hawting, The First Dynasty of Islam: The Umayyad Caliphate AD 661–750, London: Routledge, 1989, 51–3. 21 A.-M. Saadi, “The letter of John of Sedreh: A new perspective on nascent Islam,” Karmo 1, 1998, 1, 18–31 (Arabic and Syriac); 1, 1999, 2, 46–64 (English). Theodore Bar Koni (d. 792) sees them as “believing as the Jews.” Theodore Bar Konai, Theodorus Bar Koni Liber Scholiorum, CSCO 55, ed. A. Scher, Paris: E Typographeo Reipublicae, 1910, 235. Michael the Syrian wrote: “Muhammad attached himself to the faith of the Jews, because it pleased him” ( ). Chronique de Michel le Syrien, 2, 405; See also Anonymi Auctoris Chronicon ad Annum Christi 1234 Pertinens, CSCO 81 (text) 109 (Latin translation) 354 (French translation), ed. J.B. Chabot, Paris: E Typographeo Reipublicae, 1920, 227–8. 221 ABDUL-MASSIH SAADI 22 Saadi, “The letter of John of Sedreh: A new perspective on nascent Islam,” Karmo 1.2, 52. 23 Ephraem, Des Heiligen Ephraem des Syrers Sermones III, CSCO 320, ed. E. Beck, Louvain: Peeters, 1972, 61. 24 Chronica Minora, CSCO 1, 38 (ed.); Chronica Minora, CSCO 2, 31 (trans.). 25 Ibid. On Midian see Genesis 25:2. 26 Mingana, Sources Syriaques, 146*. 27 Mingana, Sources Syriaques, 146*. 28 Chronique de Michel le Syrien, 2, 431 (ed.), 4, 421 (trans.); Chronicon ad annum 1234, 260. 222 Part 3 CRITICAL STUDY OF THE QUR’AN AND THE MUSLIM EXEGETICAL TRADITION 11 NOTES ON MEDIEVAL AND MODERN EMENDATIONS OF THE QUR’AN Devin J. Stewart The publication of Christoph Luxenberg’s Die syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der Qur’ansprache, in which he argues that attention to Syriac Christian vocabulary, syntax, calques, and texts enables one to “decipher” the Qur’an, thus revealing the secrets of its difficult and obscure passages, has provoked strong reactions, both positive and negative.1 A celebratory review by Robert R. Phenix Jr. and Cornelia B. Horn in Hugoye, a Syriac studies journal, touts this work as a major breakthrough in Qur’anic criticism – the first serious attempt, in their estimation, to apply historical and textual criticism to the Qur’an. Not in the history of commentary on the Qur’an has a work like this been produced. Similar works can only be found in the body of text-critical scholarship on the Bible. From its method to its conclusions on the language and content of the Qur’an, Luxenberg’s study has freed scholars from the problematic tradition of the Islamic commentators. Whether or not Luxenberg is correct in every detail, with one book he has brought exegetical scholarship of the Qur’an to the “critical turn” that biblical commentators took more than a century ago. This work demonstrates to all exegetes of the Qur’an the power of the scientific method of philology and its value in producing a clearer text of the Qur’an. Scholars of the first rank will now be forced to question the assumption that, from a philological perspective, the Islamic tradition is mostly reliable, as though it were immune to the human error that pervades the transmission of every written artifact. If biblical scholarship is any indication, the future of Qur’anic studies is more or less decided by this work.2 The reviewers in this case may know something about the history of Biblical criticism but are apparently completely uninformed about the history of Qur’anic studies. Such a sweeping claim – and one should point out that Luxenberg himself 225 DEVIN J. STEWART makes less exaggerated, though certainly large, claims for his work – is simply nonsense. The “critical turn” in Western scholarship on the Qur’an occurred over a century ago; most scholars in the field would agree that the critical watershed was Nöldeke’s Geschichte des Qorans, published in 1860.3 Contrary to the views of these over-zealous reviewers, no stretch of the imagination allows one to see in Luxenberg’s work the introduction of a startling new critical paradigm into Qur’anic studies. One might even argue that critical scholarship on the Qur’an had actually begun many centuries earlier, within the Islamic tradition itself. The efforts of medieval Muslim scholars to label the Qur’an’s 114 suras as belonging to the Meccan or Medinan periods of the Prophet Muhammad’s mission, to determine their historical order, and to record the circumstances in response to which specific passages were revealed all represent historical criticism of a sort. Phenix and Horn seem to be utterly unaware of either tradition of scholarship. They summarize Luxenberg’s book carefully for the English reading public but cannot situate it adequately within the tradition of scholarship on the Qur’an, nor are they able to judge the verisimilitude of Luxenberg’s proposals about how to read the Qur’anic text. Luxenberg’s work makes a number of valid points concerning the text of the Qur’an, though in attempting to place them within the framework of a claimed “decipherment” of Qur’anic language, he presents them somewhat clumsily and in several cases fails to document and situate them properly in the history of Qur’anic criticism. It is a secret neither to modern scholars nor to medieval Muslim investigators of the text that the Qur’an contains vocabulary of Aramaic or Syriac origin.4 A particularly prominent example is the word used to designate Mount Sinai, tur (Q 2:63, 93; 4:154; 19:52; 20:80; 23:20; 28:29, 46; 52:1; 95:2) which derives from Syriac/Aramaic tur “mountain,” as opposed to Arabic jabal and Hebrew har. It is also evident that the Qur’an has close connections with Christian and Jewish material, probably through both oral and written sources. In this regard, Luxenberg’s work is simply one among many studies that investigate the Christian background of the Qur’an and stress the Qur’an’s debt to Christian tradition.5 Many have argued that the Qur’an is somewhat more indebted to Jewish tradition, and failure to recognize or account for this may be cited as a major shortcoming of Luxenberg’s work.6 The claim that the Qur’anic text draws on specific Christian literary sources written in Syriac is also not new. Christian canonical and non-canonical texts identified as probable sources of Biblical material in the Qur’an include the Gospels of Luke and John, the Diatessaron of Tatian, the Pseudo-Gospel of Matthew, the Life of Adam and Eve, the Seven Sleepers of Ephesus, and others. Such texts circulated in many different languages, including Greek, Latin, Armenian, Georgian, and others, but the versions from which the Qur’anic material is most likely to have derived were, in several cases, Syriac. In addition, Luxenberg’s claim that Qur’anic language was influenced by Syriac grammatical constructions is not all that far-fetched, though most reviewers have denied the possibility altogether. Since it is known that certain passages of the Qur’an are related to texts in Syriac/Aramaic and Hebrew, it is 226 NOTES ON EMENDATIONS OF THE QUR’AN quite likely that some features of the text reflect the syntax and other linguistic features of those underlying texts, as is commonly observed in translations of sacred texts, such as the Septuagint, the Latin Vulgate, the Greek of the New Testament, or the King James Bible. Thus, Mingana suggested that not only key words and religious terms in the Qur’an are derived from Syriac, but also that the syntax of certain passages exhibits underlying Syriac influence.7 Luxenberg may be correct, for example, in attributing to the influence of an underlying language the construction found in the phrases wa-qatta‘nahum ithnatay ‘ashrata asbatan umaman (7:160) and thalath mi’at sinina wa-zdadu tis‘an (18:25), both of which violate classical Arabic grammar (and even dialectal usage) in that they use a plural noun after a number greater than ten.8 Thus, it is possible to accept some of Luxenberg’s general points as plausible. Little is new or revolutionary, however, as the author would claim. The problem with many of Luxenberg’s supposed Syriacisms is that many are well within the normal range of attested grammatical possibilities within Arabic itself. Despite these elements of truth, Luxenberg identifies as the basis for the theories he proposes the idea that the language in which the Qur’an was recorded was a mixed language, an Arabic/Aramaic creole of sorts, spoken in Western Arabia by a community subject to intense Christian influence, is certainly wrong. Critics are correct to point out that Luxenberg offers no historical evidence to corroborate his claim that Mecca was actually an “Aramaic” settlement. The language of the Qur’an was in all probability neither the high language of pre-Islamic poetry in particular nor simply the spoken Hijazi dialect of the time. It was a high variety of Arabic, literary in a sense that includes oral literature, closely related to the Arabic language variety used at the time in speeches and in texts with religious content, such as prayers, omens, divinations, and the like. Indignation is the word that most readily describes the majority of reactions to Luxenberg’s study on the part of scholars in Arabic and Islamic studies in countless statements on electronic discussion lists and elsewhere. Luxenberg’s audacity has rankled for three reasons primarily: he makes a radical proposal about the early history of Islam, he is not a professional scholar of Islam, and he emends the text of the Qur’an. The following remarks focus on the issue of emendation. Four prominent scholars in the field have reviewed Luxenberg’s work: François de Blois, Claude Gilliot, Simon Hopkins, and Federico Corriente. These reviews differ in tone, though none is as celebratory as that of Phenix and Horn. De Blois is most caustic, Hopkins harsh but more restrained, Corriente critical but somewhat sympathetic, and Gilliot the most sympathetic of the lot.9 The reviews make a number of valid points, but even taken together have addressed only a small fraction of Luxenberg’s specific proposals. Hopkins, for example, characterizes as outlandish Luxenberg’s claims that Mecca was originally an Aramaic colony whose inhabitants spoke a “mixed language” composed of elements of Aramaic (Syriac) and Arabic, that the Qur’an was composed in this mixed language, and that one-fourth of the text remains undeciphered. He accuses Luxenberg of 227 DEVIN J. STEWART reckless methodology, wayward philology, and exegetical caprice. He faults Luxenberg for not citing a number of important earlier studies in the field, including works by Goldziher, Aryeh Levin on imala, and Ullman’s dictionary of classical Arabic (WKAS).10 These are all valid points, but one must guard against over-reaction. One senses that such statements are in part an expression of the need to guard professional turf. Many scholars in the field are bothered by the fact that the author is an amateur, not a professional scholar with proper credentials in Arabic and Islamic studies. It bears keeping in mind, however, that amateurs can produce important results, and that the work should be judged on its own merits rather than being dismissed summarily. After all, Michael Ventris was an amateur, yet his brilliant decipherment of Linear B indeed represented a revolution in the study of ancient Greek, and Schliemann, for all his eccentricities, did make fantastic discoveries at Troy and Mycenaea. Another source of indignation, however, is simply the feeling that to emend the text of the Qur’an is an act of enormous hubris. Lurking behind this sentiment is a sensitivity to popular Islamic reverence for the Qur’an and attendant claims that it is God’s eternal speech, that its continuous oral transmission has prevented its textual corruption or modification, that it does not include significant variants, that there are no significant disputes as to its integral text, and that it contains no contradiction. Such attitudes, while they accurately reflect what Muslim youth are taught at home or at the local mosque, do not hold up when examined in the light either of western critical scholarship or of traditional Islamic scholarship on the Qur’an, which involves a great deal of sophisticated, detailed, and insightful philological and historical criticism of the Qur’anic text. In particular, the reviewers have not shown awareness of the extent to which emendation has played a role in traditional or modern scholarship on the Qur’an. Critical reviews of Luxenberg’s work have not been entirely satisfactory. Certainly, they make valid points. Luxenberg is to be faulted for not taking adequate account of relevant scholarship to date, and in some cases it is clear that he has ignored or merely pushed aside certain widely accepted results without presenting any cogent argument for doing so. He may also be faulted for following his indicated procedures for detecting the “Syro-aramaic” reading of the text in a rather mechanical fashion. For some reason, both he and his reviewers Phenix and Horn take this as a positive point, thinking that it shows philological rigor—when the truth is far from it. However, reviews of his work have been somewhat disappointing in that they focus primarily on general methodological issues and do not say enough about the actual textual emendations and interpretations the work presents. All told, Luxenberg proposes hundreds of emendations to the text of the Qur’an, and the reviews address only a handful. The most detailed review, with regard to emendation, is that of Corriente. The majority of Luxenberg’s proposed emendations thus remain untested; scholars have not responded to them directly. Several investigators of the Qur’an have argued convincingly that the Qur’an is not immune to human error and that the oral tradition of Qur’anic recitation has not miraculously preserved the text from corruption.11 They also observe that the 228 NOTES ON EMENDATIONS OF THE QUR’AN variant readings of the Qur’an preserved in Islamic tradition provide evidence of textual corruption and copyists’ errors and that medieval Muslim scholars proposed insightful emendations.12 It was once supposed that Islamic tradition was continuous enough that the meaning of the Qur’an for the early generations of Muslims would be readily retrievable from the sources, but it has been argued that the traditional commentaries exhibit blind spots regarding the meaning of certain important texts, suggesting that there was a significant rupture in early Islamic history during which an understanding of many passages was simply lost.13 The interest in recovering the original form, sources, and meaning of the text has apparently dissipated since the major works of Nöldeke and Goldziher, and other scholars. Some scholars of Qur’anic studies have sidestepped this particular philological project by focusing instead on Qur’anic exegesis, seen as embodying what Qur’an texts meant historically for Muslims.14 While this is an intriguing topic in itself, it must be recognized as a type of reader-response criticism that can complement, but not replace, investigation of the meaning of the Qur’anic text in its original context. The Qur’an is open to the same types of copyists’ errors and problems of transmission that occur in other works handed down by humans, including sacred texts. The state of the text itself demands emendation, and, in the absence of early manuscripts, conjectural emendation must play an important role in this process. The common argument that an uninterrupted and completely reliable oral transmission has miraculously preserved the text of the Qur’an from such errors falls flat. The tradition of Qur’anic recitation can be shown to ignore or run roughshod over many discernible or retrievable features of the text, particularly with regard to rhyme, that must represent the oldest stage of its performance. In addition, while many of the variants recognized as legitimate within Islamic tradition may plausibly have arisen through oral transmission, many others cannot, being based on graphic and not phonic, resemblance. One may also point out Qur’anic passages where the received text does not make sufficient sense and an apt emendation can provide a superior reading. Emendation is an operation subject to two probabilities: the probability that the text as it stands is corrupt and the probability that the suggested emendation is correct. The second probability obviously depends on the first: if the textus receptus is unlikely to be corrupt or can be shown to be correct as is, no amount of ingenuity can discover a proper emendation. If the text as it stands indeed appears to be corrupt, one searches for an emendation that involves minimal or explainable changes to the rasm or ductus, fits the immediate context, concurs with parallel passages, if any, and gives an improved reading. The assessment of these probabilities can draw on considerations of syntax, rhyme, rhythm, lexicon, stylistics, etymology, or comparison with similar passages, but inevitably remains somewhat subjective. In the end, to borrow a phrase from Mu‘tazili epistemology, the test of validity comes down to sukun al-nafs “ease of mind,” an internal mental satisfaction that signals that the emendation in question has produced a likely and improved text. Rather than being entirely arbitrary and subjective, this 229 DEVIN J. STEWART mental state must be based on an extensive knowledge of Arabic and specifically Qur’anic style and relevant linguistic features. Nöldeke in Geschichte des Qorans and Goldziher in Die Richtungen der islamischen Koranauslegung both discussed traditional Islamic emendations of the Qur’anic text.15 One of their main sources was section forty-one – “On Knowing the Correct Syntactic Construal of the Text” – of al-Suyuti’s (d. 911/1505) major work on Qur’anic philology, al-Itqan fi ‘ulum al-Qur’an.16 In this case as in others, al-Suyuti was more a compiler than an innovator; al-Itqan is such an important work precisely because it summarizes the results of hundreds of important monographs produced over the course of the preceding six or seven centuries.17 Examples of emendations to the Qur’anic text proposed by earlier authorities that al-Suyuti cites include the following: wa-lladhina yu’tuna ma ataw wa-qulubihim wajilatun annahum ila rabbihim raji‘un (23:60) “Those who give that which they give while their hearts are afraid that they are about to return to their Lord.” ⬎wa-lladhina ya’tuna ma ataw wa-qulubihim wajilatun annahum ila rabbihim raji‘un “Those who approach that which they approach with their hearts afraid that they are about to return to their Lord.”18 Al-Suyuti reports the suggestion that the form IV verb yu’tuna, ataw “to give” should be read here as form I ya’tuna, ataw “to come, approach.” This emendation arguably produces an improved reading of the text. It makes more sense that the people in question – God-fearing believers – approach death or judgment with fear rather than give something away as a result of their fear. It is not clear what the object of the verb “to give” would be, whereas the ellipsis in the case of “to approach” is explicable. The change in the text is minimal in this emendation. ya ayyuha lladhina amanu la tadkhulu buyutan ghayra buyutikum hatta tasta’nisu wa-tusallimu ‘ala ahliha (24:27) “Oh you who believe! Do not enter houses other than your own without first seeking familiarity and greeting their inhabitants.” ⬎ya ayyuha lladhina amanu la tadkhulu buyutan ghayra buyutikum hatta tasta’dhinu wa-tusallimu ‘ala ahliha “Oh you who believe! Do not enter houses other than your own without first seeking permission and greeting their inhabitants.” This emendation is attributed to Ibn Abi Hatim (al-Razi, d. 327/938), who is supposed to have remarked, “In my opinion, this is an instance where the copyists erred” (fima ahsibu mimma akhta’at bihi l-kuttab).19 Here, the emendation tasta’dhinu “to seek permission” is preferable to the accepted text, “seeking familiarity” because one would logically seek permission before 230 NOTES ON EMENDATIONS OF THE QUR’AN entering; “familiarity” would only occur at a later stage, when one had already been invited in. a-fa-lam yay’asi lladhina amanu an law yasha’u Llahu la-hada l-nasa jami‘an (13:31) “Do not those who believe despair that God, had he willed, could have guided all mankind?” ⬎a-fa-lam yatabayyani lladhina amanu an law yasha’u Llahu la-hada l-nasa jami‘an “Are not those who believe clearly aware that God, had he willed, could have guided all mankind?” This emendation is attributed to the Companion Ibn ‘Abbas (‘Abdallah, d. 68/687), who is supposed to have remarked about it, “I think the scribe wrote it when he was sleepy.”20 It appears odd that the believers portrayed in this passage would be described as despairing. This contradicts the usual logic of Qur’anic narrative, which generally has a predictable moral compass. Good and belief are rewarded, while evil and unbelief are punished, and God rarely leaves his devotees in any doubt about where they stand in the balance, engendering in them great doubt or confusion. The proposed emendation thus produces a superior reading. It could easily have arisen because the ductus (rasm) of y.y.s, unpointed, closely resembles that of y.t.b.y.n. wa-qada rabbuka alla ta‘budu illa iyyahu wa-bi l-walidayni ihsanan. (17:23) “Your (sing.) Lord has decreed that you worship only Him and (that you show) kindness to parents.” ⬎wa-wassa rabbuka alla ta‘budu illa iyyahu wa-bi l-walidayni ihsanan “Your (sing.) Lord has charged that you worship only Him and (that you show) kindness to parents.” This emendation is also attributed to Ibn ‘Abbas.21 Here, the verb qada “He decreed” seems out of place, too deterministic to describe God’s relationship to a good deed on the part of humans for which they should expect a reward. The context requires that God impose a requirement or obligation that the believers must fulfill rather than compel or coerce them to perform a righteous act. The reading wassa “charged, advised” is preferable, and is confirmed by parallel texts, such as the following: . . . wa-laqad wassayna lladhina utu l-Kitaba min qablikum wa-iyyakum an ittaqu Llaha . . . (4:131) “ . . . And We have advised those who were given the Scripture before you, and you (as well): ‘Fear God.’ . . .” In this last example, the verb wassa indicates that the believers are advised, not compelled, to fear God, suggesting that the emendation is correct. The two readings differ only slightly in terms of the ductus: d.a. resembles s.a. in unpointed text exactly, 231 DEVIN J. STEWART and a waw accidentally run into the following letter could easily have been misread as qaf. Other traditional emendations appear in the variant readings of the Qur’an. The following variant is interesting in that it cannot have arisen through oral transmission, because the resemblance is graphic and not phonic. The reading given in most contemporary copies of the Qur’an, based on the reading of Hafs, is as follows: ini l-hukmu illa li-Llahi yaqussu l-haqqa wa-huwa khayru l-fasilin (6:57) “The verdict belongs to God alone. He relates the truth, and He is the best of decisive judges.” The variant reading is as follows: ⬎ini l-hukmu illa li-Llahi yaqdi bi-l-haqqi wa-huwa khayru l-fasilin “The verdict belongs to God alone. He judges according to the truth, and He is the best of decisive judges.” The verb yaqussu means “he relates, tells, narrates a story” and is used in this fashion in several other passages of the Qur’an, most often with the preposition ‘ala. Horovitz points out that the verb qassa, yaqussu is used in the Qur’an to report a past event in every case except this one.22 It does not collocate well with the direct object al-haqq, for one does not generally say “to narrate the truth.” In this case, it seems that the reading yaqdi bi-l-haqq “to judge fairly,” preserved in other readings, is preferable. The verse has to do with judgment or issuing a verdict, as is clear from the use of the noun hukm “verdict”: in il-hukmu illa li-Llah “There is no verdict except God’s” or “(The right) to issue a verdict belongs to God alone.” The phrase yaqdi bi-l-haqqi fits this context perfectly and is clearly a better reading. It is easy to explain yaqussu l-haqqa as a scribal error for yaqdi bi-l-haqq. The connection with judgment is confirmed by the end of the verse, wa-huwa khayru l-fasilin “and He is the best of decisive judges.” It is also confirmed by parallel passages elsewhere in the Qur’anic text. The verse Q 40:20 reads wa-Llahu yaqdi bi-l-haqqi wa-lladhina yad‘una min dunihi la yaqduna bi-shay’in “God judges by the truth, and those whom they call to other than Him do not judge by anything.” It seems that Jalal al-Din al-Suyuti, who authored this section of the famous Tafsir al-Jalalayn, endorsed this opinion as well, because he mentions this variant reading prominently in his commentary while regularly ignoring variant readings with regard to other verses.23 It is reported that ‘Ali b. Abi Talib proposed the following emendation of the Qur’an. The report is included in al-Tabari’s famous Tafsir and is discussed by Goldziher and Bellamy.24 As it stands now, the relevant passage, a description of paradise, reads: fi sidrin makhdud wa-talhin mandud wa- zillin mamdud 232 NOTES ON EMENDATIONS OF THE QUR’AN wa-m’in maskub . . . (56:28–31) “Among thornless lote trees, ranged bananas, spreading shade, and gushing water . . .” ⬎⬎fi sidrin makhdud wa-tal’in mandud wa-zillin mamdud wa-ma’in maskub . . . “Among thornless lote trees, ranged date clusters, spreading shade, and gushing water” It seems odd that bananas are mentioned here, for they do not occur elsewhere in the Qur’an. ‘Ali b. Abi Talib (d. 40/661) supposedly proposed an emendation, stating that this text should be read instead wa-tal’in mandud “ranged date clusters.” Goldziher gives the translation Blütenschichten. A reference to date clusters would make more sense, because dates and date-palms appear fairly often in the Qur’anic text and would have been an important feature of the environment in which the Qur’an was revealed. This is confirmed by the parallel passage wa-l-nakhla basiqatin laha tal‘un nadid “And (We have sent down) lofty date-palms that have ranged clusters” (50:10). The use in the latter verse of the adjective nadid, cognate and synonymous with mandud, suggests the equivalence of talh and tal‘. ‘Ali is claimed to have said that although this was the correct meaning of the text, the accepted reading should not be changed: inna l-Qur’ana la yuhaju wa-la yuhawwalu. It is thus clear that emendation was far from unheard of as a method in traditional Islamic scholarship. In my assessment, all of the proposed emendations described here have a high probability of being correct. They fit the existing script quite closely and produce superior readings more in keeping with the immediate context and with Qur’anic style. Only a handful of scholars in Qur’anic studies since Nöldeke and Goldziher have devoted serious attention to emendation of the text. Among them the most prominent is undoubtedly James A. Bellamy, who, in seven articles dating from 1973 to 2002, proposed 29 emendations of the Qur’anic text as well as dozens of emendations of the muqatta‘at, or “mysterious letters” of the Qur’an.25 These emendations are based on detailed consideration of the form of the rasm or ductus of the received Qur’anic text and the types of errors that are likely to have occurred at the hands of copyists. In most cases, however, they remain untested or critically examined. The scant attention that they have received in publications in the field may be a function of the slow pace of scholarship in Qur’anic studies and Islamic studies in general. Alternatively, one senses that this topic confronts in a tangible fashion widely held Muslim beliefs about the Qur’an, and so 233 DEVIN J. STEWART produces a certain unease with taking up a line of criticism that potentially questions the basic assumptions of Muslims or offends their sensibilities. Bellamy suggests that the well-known “mysterious letters” al-huruf al-muqatta‘at that occur at the beginning of twenty-nine suras of the Qur’an are various abbreviations of the basmalah.26 This proposal has provoked some response in the field, having been criticized by Alford Welch in the article “al-Üur’an” in the second edition of the Encyclopaedia of Islam. Welch points out that Bellamy’s proposal requires too many distinct emendations and does not explain the relationship of the mysterious letters to their immediate contexts – in many cases the following verses refer to writing, the Qur’an, the revelation, and so on, and often exhibit rhyme or near-rhyme with the following verses.27 For example, Q 20, surat Ta-ha, begins as follows: ta-ha ma anzalna ‘alayka l-qur’ana li-tashqa illa tadhkiratan li-man yakhsha ⬎bi-smi Llahi l-rahmani l-rahim ma anzalna ‘alayka l-qur’ana li-tashqa illa tadhkiratan li-man yakhsha To change ta-ha to bi-smi Llahi l-rahmani l-rahim would ruin the rhyme in -a that was clearly intended. Considerations of rhythm and the evidence of other “occult” speech genres also support this hypothesis. It is well known that other phrases involving numbers or letters of the alphabet often serve to introduce rhythmical texts. This is seen in popular chants that begin with alphabetic combinations or counting: A–B–C, 1–2–3, 2–4–6–8, and so on. Formulae that introduce magic spells, charms, or other texts connected with the supernatural often involve a combination of letters – or forms derived from letters – in a rhyming chant, such as abracadabra, based in some fashion on the opening of the Latin alphabet: a–b–c–d; Egyptian Arabic ingi bingi fing; Persian ajji majji la tarajji, and so on. Such openings establish not only the rhyme but also the rhythm that will serve as the structural basis for the following text. These considerations strengthen the suggestion that the mysterious letters are an integral part of the text, and I agree with Welch’s assessment that the mysterious letters – unlike the basmalah – form an integral part of the sura, and were not added to the text at a later stage. In this case, Bellamy’s proposed emendations, based on the assumption that the mysterious letters were added at a later stage, must be wrong. In contrast, the following emendation proposed by Bellamy appears correct to a very high degree of probability. innakum wa-ma ta‘buduna min duni Llahi hasabu jahannam (21:98) “You and those whom you worship instead of God are the gravel of Hell.” 234 NOTES ON EMENDATIONS OF THE QUR’AN ⬎innakum wa-ma ta‘buduna min duni Llahi hatabu jahannam “You and those whom you worship instead of God are the firewood of Hell.” The standard reading hasab “gravel” is unlikely, for the connection of gravel with hellfire is less than obvious. The emendation is quite simple as far as the script is concerned. It is plausible that the omission of the single vertical stroke of the ta’ caused the letter to be mistaken for a sad. The proposed emended word, hatab “firewood,” makes much more sense in context, and furthermore it is supported by parallel passages, such as Q 72:15: wa-amma l-qasituna fa-kanu li-jahannama hataba “Those who are unjust are indeed firewood for Hell.” The emendation appears to be hinted at by al-Suyuti’s claim in the section on foreign vocabulary in the Qur’an that hasab is Zinjiyya for hatab.28 This suggestion may derive ultimately not from any detailed knowledge of African languages, but from general knowledge of the speech of Africans in the Middle East, who may have pronounced t as s, or, more likely, the combination – st- or -st- as- ss- or -ss-. In any case, al-Suyuti’s proposal regarding this etymon suggests some recognition that the ordinary meaning of hasab does not fit the context as well as hatab does. Another of Bellamy’s proposed emendations which has a good chance of being correct is the following: wa-truki l-bahra rahwan (44:23) [God to Moses at the sea] “and leave the sea gaping wide (?).” ⬎wa-nzili l-bahra rahwan “and descend into the sea at an easy pace.” This emendation is supported primarily by the narrative context. In this statement, God addresses Moses at the sea, before the Hebrews enter the water and cross to the other side. Since they have not yet entered the sea, it is therefore quite odd that the verse would command Moses utruk “leave.” Furthermore, rahwan, which ordinarily means “slowly,” does not seem to fit the text as it stands. The emendation does not involve a radical change in the original script, collocates well with rahwan, and resolves the problem of the narrative’s logical progression. A distinct improvement on the text as is, the emendation is a highly probable emendation. The form of the name applied to Jesus in the Qur’an, ‘Isa, has long puzzled scholars, for one would expect Yasu‘ as the Arabic rendition of Hebrew or Aramaic/Syriac Ye(ho)shua‘. Yasu‘ is, in fact, the form regularly used by Arabicspeaking Christians. In a 2001 article, Bellamy proposed to explain all occasions of the proper noun ‘Isa as corruptions of an original masiyya, derived ultimately from Gr. Messias. He withdrew this interpretation the following year, suggesting instead that ‘Isa results from a corruption of Masih, itself derived from Yasu‘. He suggests that this was both a copyist’s error and an attempt to avoid the rare verb sa‘a, yasu‘u, which has an obscene sense.29 Horovitz suggested some time ago 235 DEVIN J. STEWART that the form of the name ‘Isa is intended to parallel that of Musa. The existence of other such rhyming pairs of Biblical personal names in the Qur’anic text tends to support this hypothesis. ‘Isa “Jesus” may be paired with Musa “Moses,” just as the form of Jalut “Goliath” matches that of Talut “Saul,” and Harun “Aaron” matches Qarun “Korah.” Similar pairs are the names of the angels Harut and Marut, as well as Yajuj “Gog” and Majuj “Magog.” Though they are not named in the text of the Qur’an itself, another important rhyming pair in Islamic tradition is that of Cain and Abel, Qabil and Habil. Only in the case of Gog and Magog does the rhyme go back to the Biblical versions of the names, and even in that case, the Arabic versions increase the morphological parallelism between the two. In the other cases, one of the pair of names has been modified quite radically in order to match the other. The equivalent of the ordinary Hebrew form of Cain would be rendered something like Qa’in in Arabic, Korah would be Qurah, and so on. It is therefore not farfetched to suggest that ‘Isa is a modified version of Jesus’ name the intent of which is to match Musa quite closely. This, among other things, suggests that these Qur’anic forms of Biblical proper names derive from an oral tradition in which the names occurred frequently in pairs. To my mind, this entirely plausible explanation rules out Bellamy’s suggestion. Bellamy connects Tuwa, the name of the sacred valley where Moses encounters the burning bush, with the toponym Gilgal mentioned in connection with Joshua 5.15. He suggests that, like Gilgal, the Qur’anic toponym, voweled Tawa instead of Tuwa, meant “he rolled” and has something to do with prostrating to God. This explanation is to my mind quite unlikely. The Biblical toponym is clearly referring to a different time, place, and context in Biblical history, and so, while it might be typologically similar, is not related in any discernible manner. In addition, tawa, which means “to fold, be bent” does not fit the context. Bellamy is correct in pointing out the oddity of this toponym, which does not relate obviously to any toponym in the Biblical account of Moses. I would suggest, alternatively, that Tuwa is related to Tur, the Aramaic/Syriac word for mountain used in the Qur’an to refer to Mount Sinai, either alone – al-Tur ⫽ the Mount (Q 2:63, 93; 4:159; 52:1) – or with a complement, as in Tur Sina’/Sinin (Q 23:20; 95:2). Horovitz already points out that Tuwa is identical with the phrase janib al-tur al-ayman, “the right side of Mount (Sinai),” the place where Moses received his calling at the burning bush (Q 19:52; 28:29, 46; 20:80).30 The deformation Tuwa ⬍ Tur would be similar to that which occurs in the name of Mount Sinai for the sake of rhyme in 95:3: wa-Turi Sinin ⬍ Sina’/Sayna’. That this is likely is suggested by the fact that the term Tuwa occurs only twice in the Qur’an, and both times in end-rhyme position: inni ana rabbuka fa-khla‘ na‘layka innaka bi-l-wadi l-muqaddasi Tuwa “I am your Lord, so remove your sandals, for you are in the sacred valley, Tuwa” (12:20); and idh nadahu rabbuhu bi-l-wadi l-muqaddasi Tuwa “When his Lord called to him in the sacred valley, Tuwa” (79:16). In both cases, the word Tuwa occurs in passages where the verses exhibit end-rhyme in –a, and other verses identify the place in question in the narrative – where Moses witnessed the burning bush – as the right side of Mount Sinai, so it appears 236 NOTES ON EMENDATIONS OF THE QUR’AN plausible that Tuwa would be a distorted form of Tur. This appears a much more likely explanation than that of Bellamy, for the connection with Moses is corroborated by many passages in the Qur’an, while a putative connection with Joshua is not. In sum, Luxenberg is not alone in proposing emendations of the Qur’anic text. Many emendations are reported in medieval works of Qur’anic philology, and a few modern scholars such as Bellamy have carried out significant work in this particular area. The results are varied. Many of the medieval emendations presented by al-Suyuti seem plausible and in some cases very probable. Many of Bellamy’s proposed emendations are highly unlikely, but a few are excellent proposals that provide improved readings. How does Luxenberg compare with his predecessors in the project of Qur’anic emendation? All told, Luxenberg proposes hundreds of emendations to the text of the Qur’an. Some Syriac “readings” of the text do not involve any emendation; others do. Many of his emendations of the text, however, do not involve any recourse to Syriac vocabulary or constructions, or do so only by a considerable stretch of the imagination. I have gone through Luxenberg’s text, in both editions, carefully. With regard to the emendations themselves, the two editions of Luxenberg’s work do not differ significantly; the second edition involves mostly cosmetic changes, being in a neater format with improved type-setting. It has more extensive footnotes. However, Luxenberg has apparently not removed, changed, or added any substantive claims to the work in the second edition. Here I will merely summarize and provide a few examples of an analysis which I hope to present elsewhere in greater detail. Most of Luxenberg’s emendations are implausible and often demonstrably wrong. A small number of proposed emendations is likely or merits further consideration. In several of his emendations, Luxenberg makes use of the fact that final alif in Syriac/Aramaic represents the definite article, suggesting that this usage may explain several Qur’anic texts. What is normally interpreted as an accusative singular indefinite nominal ending -an (in final position -a) in Arabic, he interprets as the Syriac singular definite article -a or the plural definite article -e. One example is the following: hal yastawiyani mathalan (11:24; 39:29). Luxenberg suggests that this represents the Syriac plural mathle, so that the phrase would be equivalent to hal yastawiyani l-mathalani “Are the two examples equal?”31 Both Corriente and De Blois criticize this suggestion, pointing out that hal yastawiyani l-mathalani is incorrect Arabic; if the dual agent follows, the verb should be singular: hal yastawi l-mathalani.32 There is another very important argument against this suggestion, and it is telling of Luxenberg’s method. He often fails to take adequate account of alternative explanations for the construction in the text he proposes to emend. In this case, the accusative ending in the Arabic text as it stands can be explained as an accusative of specification (tamyiz), so that the phrase means, hal yastawiyani mathalan “Are the two equal as an example?” That this is not strange is suggested by many other passages of the Qur’an which must be construed as accusatives of specification . . . wa-sa’a sabila (4:22; 17:32) 237 DEVIN J. STEWART “It is evil as a way.”; fa-sa’a qarina (4:38) “He is evil as a companion.”; wa-sa’at masira (4:97, 115) “It was evil as a destination.”; kabura maqtan ‘inda Llahi (40:35; 61:3) “It was great in terms of hate in God’s eyes.”; kaburat kalimatan takhruju min afwahihim (18:5) “It was a terrible as a word which came forth from their mouths.”; wa-hasuna ula’ika rafiqa (4:69) “Those are excellent as companions.” It is particularly clear that the accusative of specification is required where there is another word which must itself be the agent in the passage, as in wa-sa’a lahum yawmu l-qiyamati himlan (20:101) “It will be evil as a burden for them on the Day of Resurrection.” Since the accusative of specification is quite common in the Qur’an, there is no reason to resort to an unlikely explanation to provide an interpretation of the text earlier; it is perfectly intelligible as it is. In another example using this principle, Luxenberg suggests that the indefinite plural accusative sujjadan (2:58; 4:154; 7:161; 12:100; 16:48; 17:107; 19:58; 20:70; 25:64; 32:15; 48:29), equivalent to sajidin “prostrating” in meaning, is actually the Syriac plural sagde of equivalent meaning.33 Again, this explanation is unnecessary; it would only be required if the form sujjadan were unattested or did not fit the context. In all cases, however, the context requires a plural indefinite accusative. In addition, the form fu‘‘al is not rare or unattested in Arabic. It occurs in the Qur’an itself in rukka’ (2:125; 22:26; 48:29). In one case, however, Luxenberg applies this sort of explanation to a text that is a bit difficult to construe as it stands: innani hadani rabbi ila siratin mustaqimin dinan qayyiman millata Ibrahima hanifan wa-ma kana min al-mushrikin (6:161) “God guided me to a straight path, as a right religion, as the religious tradition of Abraham hanifan (?), and he was not one of the idolaters.” The text is awkward since it contains too many accusatives, and the syntactic function of each is not clear. In this case, Luxenberg’s suggestion that the final –a in hanifan, to be read as hanifa, represents the Syriac definite article would alleviate this problem somewhat by allowing hanifan to act as an adjective describing Abraham, thus reducing by one the number of disparate accusatives. In other words, the combination Ibrahima hanifan that occurs repeatedly would actually be equivalent to the Arabic Ibrahim al-hanif.34 This has some chance of providing an improved reading. It is helped by the fact that hanif may itself derive from Syriac hanpa, pl. hanpe, meaning “pagan,” a point that Luxenberg stresses and was suggested Nöldeke and others.35 The term hanif is somewhat problematic in the Qur’anic text; its literal meaning is disputed, and it seems to be used in two distinct syntactical modes. Altogether, it appears twelve times in the text, ten times in the singular (2:135; 3:67, 95; 4:125; 6:79, 161; 10:105; 16:120, 123; 30:30) and twice in the plural, hunafa’ (22:31; 98:5). In general the term is closely associated with Abraham (2:135; 3:67, 95; 4:125; 6:79, 161; 16:120, 123; 22:31) and contrasts with shirk “polytheism.” Scholars have seen this term as referring both to Islam and to the pre-Jewish, pre-Christian monotheism of Abraham. The term in particular stresses the identity of the two faiths.36 In some instances, the word hanif clearly occurs as a predicate noun or adjective to Abraham, as in ma kana Ibrahimu 238 NOTES ON EMENDATIONS OF THE QUR’AN yahudiyyan wa-la nasraniyyan wa-lakin kana hanifan musliman “Abraham was neither a Jew nor a Christian, but was (a) hanif (and) a Muslim.” (3:67); inna Ibrahima kana ummatan qanitan li-Llahi hanifan wa-lam yakun min al-mushrikin “Abraham was an ummah(?), obedient to God and (a) hanif; he was not one of the polytheists” (16:120). Luxenberg’s interpretation of hanif as an adjective to Abraham is thus not entirely farfetched. However, in the majority of its occurrences, hanif does not appear as a predicate adjective but as an accusative of circumstance (hal), describing someone’s condition while performing some other task. This is clear in the cases where hanif appears in the plural, hunafa’, and does not occur in connection with Abraham: 22:30–31 . . . fa-jtanibu l-rijsa min al-awthani wa-jtanibu qawla l-zur * hunafa’a li-Llahi ghayra mushrikina bihi “So shun (pl.) the filth of idols, and shun (pl.) lying speech, * Turning yourselves exclusively to God, an not ascribing partners to Him . . .” (22:30–31); and wa-ma umiru illa li-ya‘budu llaha mukhlisina lahu l-dina hunafa’a (98:5) “They were only commanded to worship God, devoting their faith completely to Him, and turning to Him exclusively.” In both these passages, hunafa’ is an accusative of circumstance and is plural because it refers to the plural agents of the verb occurring previously in the passage. Moreover, the passages suggest that hunafa’a is parallel in meaning to the other phrases that occur as accusatives of circumstance, ghayra mushrikina bihi and mukhlisina lahu d-dina. This usage occurs with the singular, and again, without explicit mention of Abraham, in the following passages: inni wajjahtu wajhi li-lladhi fatara l-samawati wa-l-arda hanifan (6:79); wa-an aqim wajhaka li-l-dini hanifan wa-la takunanna min al-mushrikin (10:105); fa-aqim wajhaka li-l-dini hanifan fitrati Llahi llati fatara l-nasa ‘alayha (30:30). In these passages, the singular hanifan may be construed as accusatives of circumstance referring to the first person singular pronoun implied in the verb wajjahtu in the first text, and the second person pronoun contained in the imperative aqim in the last two texts. However, in four passages, the term hanif occurs in contexts where it immediately follows the proper noun Abraham, and while it may be construed as an accusative of circumstance, it also looks as if it could be intended to serve as an adjective to Abraham. The problem with the latter interpretation, though, is that ordinarily a definite article would be required; that is, one would expect Ibrahima l-hanifi “Abraham the one who turns to God exclusively.” The four texts are the following: man ahsanu dinan mimman aslama wajhahu li-Llahi wa-huwa muhsinun wa-ttaba‘a millata Ibrahima hanifan “Who is better in terms of religion than he who surrenders his face to God, while doing good, and followed the religious tradition of Abraham, devoting himself exclusively?” (4:125); qul innani hadani rabbi ila siratin mustaqimin dinan qayyiman millata Ibrahima hanifan (6:161); wa-qalu kunu hudan aw nasara tahtadu qul bal millata Ibrahima hanifan wa-ma kana min al-mushrikin (2:135); sadaqa llahu fa-ttabi’u millata Ibrahima hanifan “Say: ‘God has spoken the truth, so follow the religion of Abraham hanifan” (3:95). If one interprets hanifan as an accusative of circumstance, it may be understood to refer to the singular in the first two passages, but in the second two passages, 239 DEVIN J. STEWART including the verse Luxenberg addresses, one would expect the plural hunafa’a instead, because of the plurals kunu and tahtadu in the third passage and the plural imperative fa-ttabi’u in the fourth. Is this a case where the text is corrupt? Ought the text to read hunafa’a instead of hanifan? The reading hunaf a’ would seem to be preferable, as it accords better with classical grammar, but a somewhat similar example of singular where one would expect a plural – in this case an accusative of specification – occurs at 4:69: . . . wa-hasuna ula’ika rafiqan “and those (men) are excellent as companion(s).” Nevertheless, in these cases, Luxenberg’s emendation appears to solve a problem, for it would allow hanifa, equivalent to al-hanif in Arabic, to serve as an adjective to Abraham rather than an accusative of circumstance. This not only removes the problem of having a singular appear where one would expect a plural form, but also takes care of the awkward separation of hanifan from the singular pronoun it would have to modify as an accusative of circumstance in 6:161: qul innani hadani rabbi ila siratin mustaqimin dinan qayyiman millata Ibrahima hanifan. It is worth noting, in possible support of Luxenberg’s argument, that the term hanif never occurs with the Arabic definite article in the Qur’an. Many of Luxenberg’s emendations do not depend on a Syriac reading of the text. Observing that the word Qur’an itself is on occasion written q.r.y.n in the rasm of the Qur’anic text, he concludes that the -y- may often represent -a-. In addition, this being the case, the possibility that an original -a- has been written as –y- and then misread as b/n/t/th also arises, these consonants being indistinguishable in unpointed text.37 Nöldeke had already pointed out that y frequently represents –a- in the Qur’anic text, but usually as a final letter or before attached prepositions, as in tahaha (t.h.y.h.a), and so on. He stated that medial y only represented –a- in the foreign borrowing al-Tawrah (a.l.t.w.r.y.h.) “the Torah.”38 Thus the suggestion that medial y may represent –a- more frequently than hitherto realized is one of Luxenberg’s tangible contributions. However, he connects this result rather unnecessarily and even illogically with Syriac. His presentation of this point is confused, for he attempts to argue that this orthographic feature is a logical consequence of the fact that the term Qur’an derives from Syriac qeryana “the Lesson” an etymology many had already accepted as convincing.39 One should point out that there is no necessary connection between the etymological derivation of Qur’an and the fact that -y- in Qur’anic script may sometimes represent -a-. The problem with arguing Syriac influence on Qur’anic orthography here is that the -y- in the Syriac does not actually represent -a-, but simply the consonant -y-, and this contradicts the point Luxenberg is trying to make. A number of Luxenberg’s emendations are based on the idea that y represents an underlying –a-. As is well known, the Sura of the Cave (Q 18) includes a version of the Christian tale of the Seven Sleepers of Ephesus, in which a group of noble youths refuse to perform pagan rituals at the request of the Roman Emperor Decius. They retire to a cave, where God miraculously saves them from persecution and preserves them in sleep for several centuries’ time. As Bellamy notes, the term al-raqim, occurring in the verse am hasibta anna ashaba l-kahfi 240 NOTES ON EMENDATIONS OF THE QUR’AN wa-l-raqimi kanu min ayatina ‘ajaba (18:9), has long puzzled commentators.40 Torrey suggested that this was a corruption of the name Decius (d.q.y.s ) read r.q.y.m.41 Horovitz criticized Torrey’s proposal as failing to explain the origin of the definite article in al-raqim.42 Jeffery, like Horovitz, pointed out that this emendation does not explain the appearance of the definite article and also dismissed Torrey’s interpretation on the grounds that the corrupt reading only works in Hebrew script, where samech resembles mem, but not in Syriac. In the end, he suggests that al-raqim is most likely a place-name.43 Bellamy suggested removing the conjunction wa- “and” and reading al-raqim as the plural adjective al-ruqud “the sleepers” that is, understanding the text to refer to “the sleeping companions of the cave.”44 Finally, Luxenberg suggests the emendation al-ruqad “sleep” for al-raqim, reading y as a and emending m to d, thus rendering the phrase “the Companions of the Cave and sleep.”45 Luxenberg’s emendation has the advantage of preserving the conjunction that Bellamy had to explain away, but remains quite close to Bellamy’s suggested reading in meaning. These emendations are all unnecessary, because there is a perfectly sensible explanation for the appearance of the term al-raqim. Horovitz already pointed out that the most likely explanation is that al-raqim refers to lead tablets (loha d-barta) outside the cave on which the seven sleepers of the story recorded their names. He further connects the term with the phrase kitab marqum that occurs elsewhere in the Qur’an (83:9, 20). Indeed, al-Farra’ (d. 207/822) had already mentioned the interpretation that al-raqim was a stone, iron, or lead tablet.46 Sidney Griffith’s contribution confirms this interpretation through the examination of several Syriac versions of the tale, arguing that al-raqim indeed refers to a plaque (loh) that was erected at the site of the cave to commemorate the miraculous occurrence.47 The term raqim is therefore in all likelihood correct as it is, and the proposed emendations are quite improbable. Another example of an emendation based on the principle that y – and consequently b/t/th/n as well – may represent –a- is the following. wa-yawma yunadihim ayna shuraka’i qalu adhannaka ma minna min shahid (41:47) “On the day when He calls to them, “Where are My partners?” they will answer, “We hereby announce (to) You: there is no witness among us.” wa-yawma yunadihim ayna shuraka’i qalu idh-dhaka ma minna min shahid “On the day when He calls to them, “Where are My partners?,” they will answer at that time, “There is no witness among us.” Luxenberg emends adhannaka “we announce (to?) you” (in the rasm, a.dh.n.k) to idh-dhaka “at that time” (hypothetically a.dh.y.k).48 This emendation, unlike the preceding one, enjoys some probability of being correct. First, the text as it stands is problematic precisely on account of this particular word. It does not make sense 241 DEVIN J. STEWART for the beings undergoing interrogation on the Day of Judgment to “inform” God of anything, it being understood that God already knows everything. Instead, He presents them with the record of their past deeds. In addition, it appears odd that they state that they are informing Him of something rather than simply announcing it tout court. There is some probability, therefore, that the text is corrupt. The proposed emendation also fits the context. Other terms meaning “at that time,” such as yawma’idhin, occur frequently in Qur’anic depictions of the end-time, and such a phrase here produces an intelligible statement that fits logically in the narrative. Again, one notes here that the emendation has absolutely nothing to do with Syriac: idh-dhaka is neither a Syriac etymon nor a borrowed or calqued expression, but simply Arabic. An example of an unlikely emendation Luxenberg proposes is the following: lisanu lladhi yulhiduna ilayhi a’jamiyun wa-hadha lisanun ‘arabiyun mubin (16:103) ⬎ lisanu lladhi y.lj.zuna ilayhi a’jamiyun wa-hadha lisanun ‘arabiyun mubin In this passage, Luxenberg claims that the verb yulhiduna does not fit the context and suggests that the text should be read y.lj.zuna, interpreted as a reflection of Syriac lgez, meaning here, “they hint at.”49 However, the Arabic equivalent of Syriac lgez is actually alghaza, with gh instead of j, also meaning “to hint at.” Luxenberg’s argument is based in part on the similarity in form of the consonants h and j in Arabic script, so this is something of a problem for his interpretation. Furthermore, the text as it stands is not likely to be corrupt, since the verb yulhiduna is perfectly comprehensible in this context. The verb alhada, yulhidu originally means “to miss the mark,” especially in archery, in contrast to asaba, yusibu “to hit the mark.” In this context it means that the Prophet’s detractors incorrectly identify or point out a supposed teacher of the Prophet. This interpretation is corroborated by the use of the same verb in verse 7:180, which refers to the unbelievers’ reprehensible misdirection of God’s divine epithets in prayer to pagan gods: wa-li-Llahi l-asma’u l-husna fa-d‘uhu biha wa-dharu lladhina yulhiduna fi asma’ihi sa-yujzawna ma kanu ya‘malun (Q 7:180) “God has the Very Beautiful Names, so pray to Him with them, and leave those who direct His names incorrectly – They will receive recompense for what they have done.” In both 7:180 and 16:103, yulhidu means to aim or fire at the wrong target or to direct something in the wrong direction. There is no need to invoke the meaning “to hint at.” Luxenberg’s most notorious readings and emendations are those that have to do with the houris, the fantastic female companions of paradise. He suggests that the term hur is related to Syriac hewara, hewarta meaning “white” and that ‘in is related to Syriac ‘ayna, meaning not simply “eye” but also appearance, color, brilliance, shine.” Then, he emends the phrase wa-zawwajnahum bi-hurin ‘in, literally, “We will marry them to women with large eyes having a marked contrast 242 NOTES ON EMENDATIONS OF THE QUR’AN between dark and light,” to wa-rawwahnahum bi-hurin ‘in, which he translates as “We will let them rest comfortably under white, crystal(-clear) grapes.”50 In favor of this interpretation, he argues that it would be odd for the eyes of beautiful women companions be described as “white,” the meaning he assigns to Arabic ahwar, hawra’, hur. However, this is not correct, for the term hawar refers to a very particular beauty trait, a strong contrast between the whites of the eyes and the dark irises and pupils. Bell’s translation “dark-eyed,” which Luxenberg criticizes, is actually a fair approximation, given the difficulty of rendering the term concisely in English.51 The existence of such a specific term should not strike one as odd; a parallel term, lama, with the associated adjectival forms alma, lamya’, and lumy, refers to a similar beauty trait, a dark color of lips and gums that contrasts with the light color of the teeth. Luxenberg’s other objection, that the Qur’anic descriptions of paradise assign to male denizens of paradise two sets of wives, both their earthly wives and these heavenly maidens, presents a genuine hermeneutical problem. Even in this case, though, the earthly wives are mentioned much less frequently (e.g. 36:56; 43:70), so that it would seem hazardous to attempt to explain away the many references to hur on those grounds. In order for this interpretation to work, Luxenberg must emend or reinterpret a host of other words in the passages that describe the houris, and also make many of the same emendations in disparate passages.52 The fact that he must emend so many individual items in texts that are not adjacent or connected by any physical pattern calls the entire complex of emendations into question. One can imagine, for example, a scenario in which physical damage to a text, such as a hole or blotch in a page, a tear in one or several adjacent pages, or the loss of a leaf or more of a text would require many individual emendations to restore the text to a form closer to the original. However, there is no evidence of such a situation here. The only remaining cause of such varied and numerous corruptions is an intentional reworking of the text by a later scribe for ideological reasons. While this is not entirely impossible, one would have to present a very strong argument for supposing it, and Luxenberg does not. Again, very importantly, Luxenberg does not take into account the alternative explanations for the text as it stands. His initial reason for seeking alternatives to the standard interpretation seems to be that the forms of the words hur, “houris,” and ‘in, “wide-eyed,” are odd in Arabic. They are not. The plural form of adjectives of the form af‘al (masc.), fa‘la’ (fem.) is fu‘l, and this adjectival form is not rare. The plural of ahwar (masc.), hawra’ (fem.) meaning “of marked contrast between the blacks and whites of one’s eyes” is hur, and the plural of a‘yan, ‘ayna’ (fem.) “large-eyed” is ‘in, just as the plural of a‘ma, ‘amya’ (“blind”) aswad, sawda’ (“black”) a‘war, ‘awra’ (“one-eyed”), ahwal, hawla’ (“cross-eyed”), and abyad, bayda’ (“fair, white”) are ‘umy, sud, ‘ur, hul, and bid. Since these forms are easily explained, the need to emend them or reinterpret them must be justified in some other manner. In combination with this, the need to emend so many other individual items in order to go along with the re-reading of the female companions as grape-vines, so that references to their having 243 DEVIN J. STEWART modest gaze (qasiratu t-tarfi) are reinterpreted as references to pendant leaves, and the phrase wa-zawwajnahum bi-hurin ‘in “we paired them (or married them) with women with large, dark eyes” (44:54; 52:20) is emended to wa-rawwah nahum bi-hurin ‘in, which Luxenberg understands to mean “we let them have their repose (ausruhen) with white grape-vines.”53 All of these emendations and re-readings are unnecessary and extremely improbable. Moreover, the emendation of the houri-passages requires the emendation of an entire second set of passages, those referring to handsome male companions or servants (wildan, ghilman) who also appear in paradise, for the presence of handsome youths in paradise would appear to corroborate references to female companions and militate against interpreting the latter as references to grapes or grape vines.54 These emendations, too, are extremely improbable. There nevertheless remains the possibility, already suggested by Beck, that descriptions of the gardens of paradise in Christian writings, including those of Ephraim the Syrian, influenced Qur’anic texts on the same topic.55 One of Luxenberg’s emendations that appear plausible is the following. Surat al-Saffat (Q 37) includes a refrain that reads as follows in the textus receptus: wa-tarakna ‘alayhim fi l-akhirin “We left on them in the succeeding generations.” (37:78, 108, 119, 129) ⬎ wa-barakna ‘alayhim fi al-akhirin “We blessed them for posterity.”56 The same emendation was already suggested by J. Barth in 1916, though Luxenberg is apparently unaware of this.57 This emendation involves no change to the rasm of the text, but only a change in the pointing of one letter – from b to t. The assumption of a long –a- vowel as opposed to short -a- is a simple matter, for –a- is frequently not marked in the Qur’anic text. The emendation solves a problem in the text as is, because the use of the verb taraka “to leave” with ‘ala “on” is odd. Why would God say, “We left on them”? Where is the direct object? Pickthall’s translation, for example, translates Q 37:108 as “And [We] left for him among the later folk (the salutation):”. Rendering ‘ala, ordinarily meaning “on” or “against,” as “for” is quite strange. In addition, he interprets the direct object of the verb taraka as being the phrase that occurs in the next verse, “Peace be unto Noah among the peoples!” The verb baraka (“to bless”), however, regularly takes the preposition ‘ala, and it would make much more sense for God to bless Noah, Abraham, Moses, Aaron, and Elias than to leave something on them. That this emendation is correct is strongly suggested by verse 113 of the same sura, which reads, wa-barakna ‘alayhi wa-‘ala Ishaq wa-min dhurriyyatihim muhsinun wa-zalimun li-nafsihi mubin “And We blessed him [Ishmael or possibly Abraham] and Isaac. And of their seed are some who do good, and some who plainly wrong themselves.” The crucial point here is that the verb baraka is used prominently in a context closely parallel to that of the other verses cited earlier. 244 NOTES ON EMENDATIONS OF THE QUR’AN Emendation of the Qur’anic text tends to raise eyebrows both in Muslim communities and in academic circles devoted to the study of Islam. Part of the reason for this is a lack of knowledge of the textual history of the Qur’an, as recorded in traditional Islamic texts, and, more specifically, a lack of awareness of the role that emendation has played in traditional Islamic scholarship. It would be as much a mistake to eschew emendation as a profitable approach to the Qur’anic text, whether out of an urge to avoid offending believers or in keeping with a “post-philological” trend of scholarship that holds it is futile or uninteresting to search for original texts and meanings as it would be to adopt the doctrinal view that oral transmission has miraculously preserved the text of the Qur’an from corruption. Whatever the merits of Luxenberg’s work, reactions that have deprecated it simply because he dares to emend the Qur’anic text are disappointing. Scholars of the Qur’an, if they are worthy of the name, should be able to advance arguments regarding the probability or improbability of the specific emendations Luxenberg has proposed rather than simply crying foul. Very few of Luxenberg’s emendations are likely; out of scores of suggested emendations and readings of the text, only a handful are plausible. Nevertheless, scholars of Islamic studies should endeavor to prove them incorrect in a clear, specific, and responsible manner if scholarship is to attain real advances in the understanding of the Qur’anic text. Notes 1 Ch. Luxenberg, Die syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der Qur’ansprache (“The Syro-Aramaic Reading of the Koran: A Contribution to the Decipherment of Qur’anic Language”), Berlin: Das arabische Buch, 2000; 2nd revised edition: Berlin: Schiler, 2004. 2 R.R. Phenix and C.B. Horn, Review of Luxenberg, Die syro-aramäische Lesart des Koran, Hugoye Journal of Syriac Studies 6.1 (2003). 3 Nöldeke’s work far outstripped Gustav Weil’s short introduction to the Qur’an, published in 1844: Einleitung in den Koran, Bielefeld: Belhagen und Klasing, 1844. 4 See FV. 5 See W.St. Clair Tisdall, The Original Sources of the Qur’an, New York: E. S. Gorham, 1905, 136–211; W. Rudolph, Die Abhängigkeit des Qorans von Judentum und Christentum, Stuttgart: Kohlhammer, 1922; R. Bell, The Origin of Islam in Its Christian Environment, London: MacMillan, 1926; reprint: London: Frank Cass & Co., 1968; T. Andrae, “Der Ursprung des Islams und das Christentum,” Kyrkhistorisk arsskrift 23 (1923): 149–206; 24 (1924): 213–92; 25 (1925): 45–112; A. Mingana, “Syriac influence on the Style of the Koran,” Bulletin of the John Rylands Library 11, 1927, 77–98; K. Ahrens, “Christliches im Qoran,” ZDMG 84, 1930, 15–68, 148–90; E. Gräf, “Zu den christlichen Einflüssen im Koran,” ZDMG 111, 1962, 396–98; G. Parrinder, Jesus in the Qur’an, New York: Barnes and Noble, 1965; H. Räisänen, Das Koranische Jesusbild: Ein Beitrag zur Theologie des Korans, Helsinki: Missiologian ja Ekumeniikan, 1971; A. Schall, “Die Sichtung des Christlichen im Koran,” Mitteilungen und Forschungsbeiträge der Cusanus-Gesellschaft 9, 1971, 76–91; G. Lüling, Über den Ur-Qur’an: Ansätze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur’an, Erlangen: Lüling, 1974; C. Schedl, Muhammad 245 DEVIN J. STEWART 6 7 8 9 10 11 12 13 14 15 16 17 und Jesus, Vienna: Herder, 1978; idem, “Die 114 Suren des Koran und die 114 Logien Jesu im Thomas-Evangelium,” Der Islam 64, 1987, 261–4; T.J. O’Shaughnessy, The Koranic Concept of the Word of God, Rome: Pontificio Istituto Biblico, 1984. A. Geiger, Was hat Mohammed aus den Judenthume aufgenommen? Bonn: F. Baaden, 1833; revised edn. Leipzig: M. W. Kaufmann, 1902; T. Nöldeke, GdQ1; H. Hirschfeld, New Researches into the Composition and Exegesis of the Qoran, London: Royal Asiatic Society, 1902; Tisdall, The Original Sources of the Qur’an, 51–135; I. Schapiro, Die haggadischen Elemente im erzählenden Teil des Korans, Leipzig: Gustav Fock, 1907; J. Horovitz, Koranische Untersuchungen, Berlin: de Gruyter, 1926; C.C. Torrey, The Jewish Foundation of Islam, New York: Jewish Institute of Religion Press, 1933; H. Speyer, Die biblische Erzählungen im Qoran (Gräfenhainichen: Schulze, 1931; reprint Hildesheim: Olms, 1961, 1988). Mingana, “Syriac influence on the Style of the Koran,” Bulletin of the John Rylands Library, 11, 1927, 77–98. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn., 68–9. F. Corriente, “On a proposal for a Syro-Aramaic reading of the Qur’an,” Collectanea Christiana Orientalia 1, 2004, 305–14; S. Hopkins, Review of Luxenberg, Die syro-aramäische Lesart des Koran, JSAI 29, 2003, 377–80; C. Gilliot, “Langue et Coran: Une lecture syro-araméenne du Coran,” Arabica 50, 2003, 380–93; F. de Blois, Review of Luxenberg, Die syro-aramäische Lesart des Koran, Journal of Qur’anic Studies 5, 2003, 92–7. Hopkins, Review of Luxenberg, Die syro-aramäische Lesart des Koran, JSAI 29, 2003, 377–80. Nöldeke, GdQ1, 265–79; I. Goldziher, Die Richtungen der islamischen Koranauslegung. An der Universität Upsala gehaltene Olaus-Petri-vorlesungen, Leiden: Brill, 1920, 2–32; D.S. Margoliouth, “Textual variations of the Koran,” The Muslim World 15, 1925, 334–44; F. Rosenthal, “Some minor problems in the Qur’an,” The Joshua Starr Memorial Volume, n.e. New York: n.p. 1953, 68ff.; D.S. Powers, “The Islamic law of inheritance reconsidered: A new reading of Q 4:12B,” Studia Islamica 55, 1982, 61–94; J.A. Bellamy, “Some proposed emendations to the text of the Koran,” JAOS 113, 1993. 562–73. Bellamy, “Some Proposed Emendations,” 562–63. J. Burton, “The vowelling of Q 65,1,” JSS 29, 1984, 267–83; P. Crone, “Two legal problems bearing on the early history of the Qur’an,” JSAI 18, 1994, 1–37; D.S. Powers, Studies in Qur’an and Hadith: The Formation of the Islamic Law of Inheritance, Berkeley: University of California Press, 1986. See J.D. McAuliffe, Qur’anic Christians: An Analysis of Classical and Modern Exegesis, Cambridge: Cambridge University Press, 1991; A. Rippin, “The Qur’an as literature: Perils, pitfalls and prospects,” British Society for Middle Eastern Studies Bulletin 10, 1983, 38–47; idem, “Quranic studies, Part IV: Some methodological notes,” Method and Theory in the Study of Religion 9, 1997, 39–46; B. Wheeler, Moses in the Quran and Islamic Exegesis, London: Routledge/Curzon Press, 2002. Nöldeke, GdQ1, 236–42; Goldziher, Die Richtungen der islamischen Koranauslegung. An der Universität Upsala gehaltene Olaus-Petri-vorlesungen, 31–6. Jalal al-Din al-Suyuti, al-Itqan fi ‘ulum al-Qur’an, n.e., Beirut: Dar al-Kutub al-‘Ilmiyya, 1995, 1, 392–4. Kenneth Edward Nolin’s study of the sources of al-Itqan provides a useful survey of these works, but is unfortunately riddled with errors, including incorrect dates, incorrect authors’ names, and false identifications of works. See The Itqan and Its Sources: A Study of al-Itqan fi ‘ulum al-Qur’an, by Jalal al-Din al-Suyuti, with Special Reference to al-Burhan fi ‘ulum al-Qur’an by Badr al-Din al-Zarkashi, Doctoral dissertation, The Hartford Seminary Foundation, 1968. 246 NOTES ON EMENDATIONS OF THE QUR’AN 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 49 50 Al-Suyuti, al-Itqan fi ‘ulum al-Qur’an, 1, 392. Ibid., 1, 392–3. Ibid., 1, 393. Ibid., 1, 393. J. Horovitz, Koranische Untersuchungen, 7. Jalal al-Din al-Mahalli and Jalal al-Din al-Suyuti, Tafsir al-Jalalayn, ed. Marwan Siwar, Beirut: Dar al-Jil, 1995, 134. Al-Mahalli wrote the commentary on suras 1, 18–114, and the beginning of sura 2, while al-Suyuti completed the work by writing the commentary on suras 2–17. See Ch. Pellat, “al-Mahalli,” EI2, 5, 1223. al-Tabari, Jami‘ al-bayan, n.e., Cairo: al-Matbaha al-Maymaniyya, 1321–8/1903–10, 27, 93. Goldziher, Die Richtungen der islamischen Koranauslegung. An der Universität Upsala gehaltene Olaus-Petri-vorlesungen, 36; Bellamy, “Some Proposed Emendations,” 563. J.A. Bellamy, “The mysterious letters of the Qur’an: Old abbreviations of the Basmalah,” JAOS 93, 1973, 267–85; idem, “Al-Raqim or al-Ruqud?: A note on Surah 18:9,” JAOS 111, 1991, 115–7; idem, “Fa-Ummuhu Hawiyah: A note on Surah 101:9,” JAOS 112, 1992, 485–7; idem, “Some proposed emendations to the text of the Koran;” idem, “More proposed emendations to the Text of the Koran,” JAOS 116, 1996, 196–204; idem, “Textual criticism of the Koran (Presidential Address),” JAOS 121, 2001, 1–6; idem, “A further note on ‘Isa,” JAOS 122, 2002, 587–8. J.A. Bellamy, “The mysterious letters of the Qur’an,” 267–85; idem, “Some proposed emendations to the text of the Koran,” 562–73. A. Welch, “üur’an,” EI2, 5, 400–29, here pp. 413–4. Al-Suyuti, al-Itqan fi ‘ulum al-Qur’an, 1, 292. Bellamy, “Textual criticism of the Koran,” 6; idem, “A further note on ‘Isa,” 587–88. Horovitz, Koranische Untersuchungen, 125. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn, 52–3. Corriente, 378; De Blois, 93–4. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn, 70–1. Ibid., 62–6. Th. Nöldeke, Neue Beiträge zur semitischen Sprachwissenschaft, Strassburg: Trüber, 1910, 35; FV, 112–5. W.M. Watt, “Hanif,” EI2, 3, 165–6; U. Rubin, “Hanif,” EQ, 2, 402–4. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn, 81–7. Nöldeke, GdQ1, 252–55. FV, 233–4. J.A. Bellamy, “Al-Raqim or al-Ruqud?,” JAOS 111, 1991, 115–7. See also R. Paret, “Ashab al-kahf,” EI2, 1, 691. C.C. Torrey, “Three difficult passages in the Koran,” in T.W. Arnold and R.A. Nicholson (eds), A Volume of Oriental Studies Presented to Edward G. Browne ( . . . ) on his 60th Birthday, Cambridge: Cambridge University Press, 1922, 457–9. Josef Horovitz, Koranische Untersuchungen, 95. FV, 144. Bellamy, “Al-Raqim or al-Ruqud?,” 115–7. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn, 92–4. R. Tottoli, “Raqim,” EQ, 4, 351–52, citing al-Farra’, Ma’ani al-Qur’an, edn. Ahmad Yusuf Najati and Muhammad ‘Ali al-Najjar, Cairo: Dar al-Kutub al-Misriyya, 1955–72, 2, 134. See Griffith’s essay in this volume. Luxenberg, Die syro-aramäische Lesart des Koran, 2nd ed, 87–8. Ibid., 119–22. Ibid., 254–74. 247 DEVIN J. STEWART 51 52 53 54 55 Luxenberg, Die syro-aramäische Lesart des Koran, 2nd ed, 266. Ibid., 254–94. Ibid., 276–8, 258–61. Ibid., 295–303. E. Beck, “Eine christliche Parallelen zu den Paradiesjungfrauen des Koran?,” Orientalia Christiana Periodica 14, 1948, 398–405. 56 Luxenberg, Die syro-aramäische Lesart des Koran, 2nd edn, 170–1. 57 J. Barth, “Studien zur Kritik und Exegese des Qorans,” Der Islam 6, 1916, 113–48, esp. 124–5. 248 12 SYRIAC IN THE QUR’AN Classical Muslim theories Andrew Rippin It is important to remember that by no means are Christoph Luxenberg or even Alphonse Mingana the first people to contemplate the presence of Syriac in the Qur’an. Starting in the early centuries of Islam, Muslim exegetes frequently discussed various words which they considered to be of Syriac origin. Early Muslim writers were apparently aware of a language either still spoken in their midst or evident in texts called suryani or nabati – and they appear to have appealed to that knowledge to solve exegetical problems in the Qur’an. The basic thrust is the same as the one for Luxenberg1 and Mingana:2 if the text is problematic, then perhaps Syriac can solve the issue. Medieval Muslims took a similar basic approach. In this paper I will examine the use of Syriac as a tool for medieval Muslim exegetes and investigate the reasons why they felt it necessary to look to the foreign origin of certain words and why it might be that they chose Syriac in certain Qur’anic instances, as compared to Greek, Coptic or Hebrew, other popular “foreign languages” which are adduced in their commentaries.3 In order to accomplish this, I will first speak about the concept of foreign languages in the Qur’an and contrast the classical Muslim approach with that of the more recent scholarly attitude. I will then turn specifically to the area of Syriac, try to add some clarity to the concept itself, and then consider the Muslim use of the category. With that information at hand, we will then be in a better position to try to compare some of the approaches of modern scholarship with the classical, which is the underlying theme of this essay. A basic philological approach to the Qur’an might well suggest that looking for foreign vocabulary in the text is simply a “natural” thing to do; languages interact, they grow, they change due to the mingling of people in situations in which they employ different languages. The evidence is plain to see in the emergence of modern English and there is no reason not to think that Arabic would not be exactly the same. Scholarly philological observations regarding Arabic were certainly stimulated in the past especially by the particular form of proper names which were familiar from the Biblical tradition which are found in 249 ANDREW RIPPIN the Qur’an; they definitely suggested certain linguistic questions: how is it that Avraham became Ibrahim in Arabic? Or, Yitshaq, became Ishaq? Through what language vehicles did these names enter Arabic in order to adopt these forms? These can genuinely be said to be questions of scholarly curiosity. Also, considering the history of the scholarly interest in the Arabic language, the observation of foreign vocabulary might be said to have been certain to arise as well. The great European philologists of the seventeenth, eighteenth and especially nineteenth centuries were attracted to the study of Arabic because of its value to the understanding of Biblical Hebrew. There clearly was a sense that the isolation of the desert would have preserved the purity of the Semitic languages and that Arabic would be the key to understanding some of the puzzles of the Bible. This romantic thought clearly motivated a good deal of interest. Polyglot lexica started to emerge in the seventeenth century – a famous one was published by Castell in 1669 and it provided a comparison of Hebrew, Aramaic, Syriac, Samaritan, Ethiopic and Arabic along with some Persian as well. Thus, the examination of Arabic vocabulary in the context of Biblical studies alerted modern scholars early on to the fact that certain terms used in the Qur’an were, in fact, being used in a technical religious sense that was found in other Semitic languages. This created the complicated situation of a dual goal for this type of study of foreign vocabulary in Arabic. For one, there was the historical question related to language itself, of understanding how languages grow. For the other, there is the search for meaning where some sense of the “original” word in an etymological context is felt to have a bearing on how the Qur’an is to be understood. The pitfalls of this latter approach have been explored by people such as James Barr and I will not focus attention on that problem in this context except to say that I do think that one can gain some appreciation for the accomplishment of Islam and the Qur’an by seeing how vocabulary gets transformed: etymology cannot be viewed as a determinant of meaning, but it is revealing of transformation and inventiveness. There is, however, another dimension to these modern scholarly investigations. The polemical tinge to the study of foreign vocabulary is notable and this appears to arise out of a concern with the context in which Islam arose. That is, the concern is for the contextualizing of the Qur’an in a Christian or Jewish background (or Zoroastrian or Manichaean, for that matter), an exercise which often proves to be a reductive process of removing any sense of originality from Islam and attributing all the good (or bad) ideas to a previous religion. This is often placed within many studies (despite the cautions explicitly expressed) in a specific linguistic context, rather than speaking of an ethos of monotheism or a metaphorical universe, concepts which appear to make historians edgy because they are unspecific and the route of influence cannot be traced. Be that as it may, the reductive impulse may be seen in many of the modern writings on the topic. Mingana is the most famous in this regard in finding Syriac as a key to understanding the Qur’an such that it allows him to declare that “The Jewish influence of the religious vocabulary of the Qur’an is indeed negligible” and that Christianity is the source of all the religious inspiration of Islam.4 250 SYRIAC IN THE QUR’AN For classical Muslim scholars, on the other hand, the discussion of foreign languages in the Qur’an occurred in quite a different context, that of general theories about the presence of foreign vocabulary in the Qur’an. This was an area of some considerable controversy but it certainly may be asserted immediately that, contrary to modern scholarship, the thought of foreign languages, or Syriac specifically, did not, of course, imply any sense of a Christian substratum to the text of the Qur’an. To the ninth-century philologist Abu ‘Ubayda is ascribed the statement, “Whoever suggests there is anything other than the Arabic language in the Qur’an has made a serious charge against God.”5 This sentiment appears to have been widespread in the formative centuries of Islam. Abu ‘Ubayda clearly recognized the existence of a similarity between certain words in foreign languages and those in the Qur’an but his response to that observation is to state that the form of a word in one language can correspond (yuwafiqu) to its form in another and its meaning in one language can approach that of another language, whether that be between Arabic and Persian or some other language.6 Also in the early ninth century al-Shafi‘i suggested that no one knew the entire stock of Arabic vocabulary, so what might be thought of as “foreign” to one group of Arabs was, in fact, known to others. He says Of all tongues, that of the Arabs is the richest and the most extensive in vocabulary. Do we know any man except a prophet who apprehended all of it? However, no portion of it escapes everyone, so that there is always someone who knows it. Knowledge of this tongue to the Arabs is like the knowledge of the sunna to the jurists: We know of no one who possesses a knowledge of all the sunna without missing a portion of it . . . . In like manner is the knowledge concerning the tongue of the Arabs by the scholars and the public: No part of it will be missed by them all, nor should it be sought from other people; for no one can learn this tongue save he who has learned it from the Arabs. (Al-Shafi‘i, Risala, 27–87) At the same time, al-Shafi‘i admitted that there may be “in foreign tongues certain words, whether acquired or transmitted, which may be similar to those of the Arab tongue, just as some words in one foreign tongue may be similar to those in others, although these tongues are spoken in separate countries and are different and unrelated to one another despite the similarity of some of the words.”8 Thus, while similarities may exist, they are there simply by coincidence and not because of a relationship between the words. Also in the early ninth century Abu ‘Ubayd makes a historical argument: words of foreign origin are found in the Qur’an but they had entered into Arabic before the revelation of the Qur’an and are thus now to be considered Arabic. The usage of the Arabic words is deemed to be superior to that of other languages.9 In the tenth century, al-Tabari provided yet another angle to the problem, although 251 ANDREW RIPPIN the view may well not originate with him: words which appear to be foreign reflect a similarity between languages; this does not indicate anything about the historical origins of the words. Al-Tabari argues a position that suggests certainty in these matters cannot be obtained; we will never know for sure whether a word started in one language or another. He suggests that, of the person who says, “these words were originally Arabic, and then spread and became current in Persian,” or “they were originally Persian and then spread to the Arabs and were Arabized,” one should say that, We should deem this person to be unlearned, because the Arabs have no more right to claim that the origin of an expression lies with them rather than with the Persians than the Persians to claim the origin lies with them rather than the Arabs. The only certain fact is that the expression is employed with the same wording and the same meaning by two linguistic groups. (Al-Tabari, Jami‘ al-bayan, 1, 1510) Such arguments were used in a variety of apologetic settings especially when debating the merits of the Qur’an. Arguments over the inimitability of the Qur’an were reinforced by denying that any special words were introduced into Arabic by Muhammad. Ultimately, the point was a theological one, tied into conceptions of the nature of Arabic as a language and Islam as divine revelation. To admit that there were foreign words in the Qur’an that had been intentionally borrowed would be to undermine the meaning of the challenge put forth to the masters of Arabic speech to produce a chapter of text which was “like” the Qur’an. Especially in later centuries, the idea of “foreign” vocabulary was not denied by commentators. The twelfth-century al-Jawaliqi, for example, spoke plainly about “foreign words found in the speech of the ancient Arabs and employed in the Qur’an” without any cautious restrictions.11 In the late fifteenth-century, al-Suyuti took the incorporation of foreign languages within the Qur’an as a positive fact, a change in attitude which was the result, perhaps, of an increased emphasis on the universal appeal of Islam and certainly taken as a part of the argument for the excellent qualities of the text of the scripture.12 Underlying all of these discussions was the reality of the language of the Qur’an as it was observed by Muslims. As is already apparent, such observations were, in fact, encouraged or stimulated by certain attitudes: some argued that the inclusivity of the Qur’an was reinforced by the presence of foreign words: so, the more languages that could be found the better! The claim even emerges that there are expressions in the Qur’an from every language. It is frequently pointed out that, among the early Arab grammarians, lexicographers and exegetes, there was a substantial number who had a language other than Arabic either as their mother tongue or as the language of their religious upbringing. It has been argued that some knowledge was brought to the study of “loan words” in Arabic, a topic which certainly was of some interest both within 252 SYRIAC IN THE QUR’AN the exegesis of the Qur’an and in general lexicography. For example, a number of Persian words were identified, often correctly by the judgment of today’s scholarship, probably as a result of the scholars’ personal knowledge of the language. Another factor that may have led to this type of observation would be words which were known from other languages by the early scholars, the meaning of which in the Qur’an was such as to suggest a relationship between the Qur’anic usage and the foreign language; this may have occurred because the meaning of the Arabic root would not support such a usage: din as both “religion” and “day of reckoning” may be an example.13 Yet another factor which led to observations about foreign words was the rise of grammatical studies in Arabic which led to understandings about the actual form or pattern of Arabic words. This then allowed observations about the aberrance – by Arabic standards – of some words found in the Qur’an. Among these would be examples of difficult morphological structures and irregular phonetic features as found in words such as istabraq, Persian for “silk brocade,” found four times in the Qur’an (Q 18:31; 44:53; 55:54; 76:21); zanjabil, meaning “ginger” (Q 76:17); barzakh, meaning “barrier,” used three times (Q 23:100; 25:53; 55:20); firdaws, meaning “paradise” (Q 18:107; 23:11); and namariq, meaning “cushions” (Q 88:15). Another such mode of consideration would be words from barren roots – that is, words which have no verbal forms associated with them – such as tannur in the sense of “oven” (Q 11:40; 23:27); jibt, meaning “idol,” Q 4:51); and rahiq, meaning “wine” (Q 83:25). The isolation of these features as “aberrant” depended, of course, upon a set of criteria being established which could act to define Arabic as a linguistic structure as such. These criteria were developed by early grammarians such as the famous eighth-century figures Sibawayhi and al-Khalil who established, for example, the permissible morphological forms of Arabic words. As well, certain combinations of letters which could not occur in Arabic words were determined and that acted as another criterion. Among the observations cited in al-Suyuti, for example, is that a jim and a qaf cannot be found in the same word. Words which violate these rules are deemed to be “foreign.”14 Finally hapax legomena and other infrequently used words were also often included in lists of foreign words (even, it should be remarked, in some cases if the origin of the word does, in fact, seem to be Arabic in our perception today). Many languages are isolated by the classical grammarians and lexicographers as sources of Arabic words, among them Syriac. Syriac, referred to as suryani or nabati, appears to have been well-known as a spoken language according to anecdotes found in the works of Ibn Qutayba and Ibn Durayd, both living in the tenth century. The association of Syriac with Christianity is also clear in the work of the eleventh-century writer al-Biruni.15 But we need to be extremely careful with terminology here. Our use of the term Syriac in modern parlance is in itself a slippery one, illustrated by Mingana’s simple lumping of Aramaic and Palestinian Syriac along with Eastern Aramaic under the umbrella term Syriac.16 The use of the term Syro-Aramaic here might 253 ANDREW RIPPIN be preferable although one might be justified in wondering just what that term refers to and it does lead to the suspicion that we have a subtle slide from a linguistic to cultural or even geographical category. I would observe tangentially that it does seem to me that Mingana’s argument about the lack of Jewish influence is quite shaky if, indeed, his concept of Syriac is one which includes an inventory of Western Aramaic associated with the Jewish Targums. If Syriac for him includes Jewish Aramaic but Syriac as a term also implies Christianity for him, then the entire grounds for his argument seem decidedly uncertain. Overall, the discussion of the topic of just what we mean by these linguistic terms would be helped immensely by careful definition of terms. We need to be very careful abut projecting modern categorizations of language back into earlier times and expecting a direct correlation. The significance here goes much further. If we take Luxenberg’s work, even if using the notion of “Syro-Aramaic” in its German sense, the argument is significant: Luxenberg argues that the Qur’an developed within a Syriac or Syro-Aramaic culture, Arabic not being a written language at the time. Many of the people involved in the emergence of this early Islam were Arab Syriac Christians who brought their approach to scripture to the development of the Qur’an. If, as is, it seems to me, reasonably generally accepted, Arabic script grew out of the Nabataean script, then are we now claiming that this is a part of this Syro-Aramaic network? Are we talking about Iraq or Syria or Palestine when we talk of this cultural environment? Western or Eastern Aramaic? What are the textual sources that actually underlie the claims? Too often there seem to be simple appeals to the dictionaries of Payne-Smith or Brockelmann with little consideration of the geographical and historical context which is involved in the assumptions about the rise of the Qur’an in relationship to the Syriac sources. But more critical for my immediate purposes is the parallel question of what did Muslims think of by the term Syriac? And what knowledge did they have of the language? The issue, too, is complicated and somewhat difficult to reconstruct. First it would seem that, as I have mentioned, two words are used to refer to what might be considered Syriac: Suryaniyya and Nabatiyya. The latter is a vexed term as a glance at the Encyclopaedia of Islam entry will disclose.17 First, in Arabic usage, the term appears to be a homonym, referring to two groups, one of them inhabiting northern Arabia and the other Mesopotamia. How that homonymity came about is a disputed historical matter, although some point to the Jewish Targum of Genesis 25:13 and elsewhere which appears to gloss these two Nabataeans by associating the eldest son of Ishmael, Nebaioth, (Ne-bay-oth) a name associated with ancient Assyria, with the spelling Nabat in reference to the northern Arabian community, involving a gloss with ta’ becoming ta’. Be that as it may, the Nabataeans (as we call them today) appear to share a common culture with contemporary Arabic speakers (given their names) and they spoke a western Aramaic dialect very close to the language of the earliest Arabic inscriptions (but, even then, some people claim their language may have some historical connections to eastern Aramaic). So, these Nabatis are what we might think of as the 254 SYRIAC IN THE QUR’AN Nabataeans today, although not every Arabic author of classical times associates these people with Petra and the like, which certainly produces a continuing lack of clarity for scholars today. In the time prior to the rise of Islam, these people were still associated with being traders – for example, there is reference to a Nabataean souk in Medina in the pre-Islamic period according to al-Waqidi, using the word Nabati.18 However, it is, in fact, the other usage of Nabati in which we are apparently interested. The Mesopotamian group is of Aramaean origin, and spoke an eastern Aramaic dialect, close to Mandaean, from which Syriac was derived. The tenth-century historian al-Mas‘udi speaks of this group in relation to the Assyrians, saying “the inhabitants of Nineveh (were) of those whom we have called Nabataeans and Syriac-speaking people; they are,” he emphasizes, “of the same race and they speak the same language.”19 For Ibn Khaldun in the fourteenth century, the Nabataeans were the native inhabitants of Mesopotamia before the Islamic conquest of Iraq. Assyrians, Babylonians and Chaldaeans are called Nabataeans; they were renowned for their magical practices. Other writers make it apparent that this designation was not linguistic exclusively (if at all) but rather an ancient group of people distinguished by their agricultural practice, as opposed to pastoral or military life.20 Aharon Maman’s Comparative Semitic Philology in the Middle Ages21 indicates quite clearly that some additional clarity (or perhaps it is an additional level of confusion) can be gained by consulting Jewish linguistic sources (written in Hebrew or Arabic) stemming from the classical Islamic time period. For writers such as Saadya writing in Arabic, Aramaic is called Suryani, the word itself understood as a transformation of the Hebrew for Assyrian, a language which is now known linguistically as Akkadian. So, Aramaeans are called Suryanis; post-Biblical Aramaic is called Suryani by both Hebrew and Arabic Jewish writers. But, Babylonian grammarians writing in Hebrew often use al-Nabati to refer to the same thing as do those writing in Arabic. Just to give an example, the tenth century Iraqi Saadya Gaon, writing in Arabic, uses Suryani to refer to Aramaic in his treatment of Job 15:29 in order to draw a linguistic comparison between Hebrew and Aramaic; but he also uses Nabati in reference to Daniel 3:8 (where the word is translated in English as Chaldeans). For the eleventh century al-Fasi, Suryani means the Aramaic of the Bible specifically, as compared to that of the Targums which he denoted by his use of the word Targum itself for the language. In other words, Suryani is usually used by Jewish authors writing in Arabic to mean what we would call Aramaic; but they may also call it Nabati, seemingly less frequently. So, in general, we may be able to assert that, for the Arab Muslim writers, it would appear that Nabatiyya is an ancient form of Suryaniyya, and both languages are to be associated with peoples of Iraq. One of the famous manifestations of Syriac for Arabs in classical Islamic times was the Kitab al-filaha al-Nabatiyya, “The Book of Nabataean Agriculture,” of Ibn Wahshiyya, an agricultural treatise apparently translated from Syriac.22 While it is difficult to 255 ANDREW RIPPIN be certain, there is a case to be made, especially on the basis of the Judaeo-Arabic sources, for Nabati being what we might call Syriac and Suryani being Aramaic, the opposite of what has sometimes been assumed. But then it is also possible that both refer to Syriac. Even then, what that might precisely refer to in terms that we can relate to today is, to say the least, ambiguous. Al-Suyuti, who died in 1505, edited in several different versions lists of foreign words in the Qur’an. One of his works is called al-Mutawakkili fima warada fi-l-Qur’an bi-l-lughat. The treatise, named after the caliph al-Mutawakkili who died in 943/1536 who ordered the author to compile the work, is a list of Qur’anic words that are “to be found in the speech of the Ethiopians, the Persians or any other people other than the Arabs.”23 The list is composed of 108 words attributed to eleven languages and is organized according to language and, within that organization, according to the order of the Qur’an, which makes it particularly useful for my current purpose; it may not necessarily be the best presentation of the Muslim tradition on foreign vocabulary but it is, I believe, at least representative. In his section on Suryani, seventeen words are listed. He also has a section on Nabati with eighteen words listed and there is a remarkable cross-over between the list with at least six words considered to come from either language. Another list has nineteen Hebrew words, including two from the Suryani list. The vocabulary treated is as follows: sari meaning “river” (Q 19:24) taha meaning “O Man!” jannat ‘adn meaning “vineyard and grapes” tur meaning “mountain” hawn meaning “wise men” (Q 25:63) hayta laka meaning “come here” (Q 12:23) wa-lata meaning “and there is not” (Q 38:2) rahwan meaning “tranquil” sujjadan meaning “with uplifted heads” (Q 2:55; 4:154) qayyum meaning “one who does not slumber” asfar meaning “books” qummal meaning “fly, bee” shahr (no definition given) yamm meaning “sea” (Q 7:132) salawat meaning “synagogues” (Q 9:100; 22:41) darasta (no definition given) qintar meaning “bull’s hide full of gold or silver” Now then the critical question is why did exegetes think these particular words to be Syriac and why did they choose Syriac as the language to designate rather than something else? 256 SYRIAC IN THE QUR’AN In the case of apparent Arabic words which are classed as “foreign” words, the immediate suspicion must be that an exegetical problem led to the suggestion of the foreignness of the word, as Arthur Jeffery argued in his work, The Foreign Vocabulary of the Quran.24 The hermeneutical advantage is clear: if the word is foreign, then it is open to a far greater interpretational variation than if the word is to be taken as a common Arabic word. That may account for some of the words on the list. However, the question remains as to why Syriac was chosen as the language. Was it on the basis of knowledge or was, it, as Jeffery states on several occasions, “as a cloak for their ignorance” or that Muslims writers used the designation “for anything ancient, time honored, and consequently little understood” or to denote “a word was of the old learned tongues and so more or less unintelligible to the ordinary person”? In specifying which non-Arabic language a given word might be thought to originate from, it appears to me that Muslim exegetes incorporated two elements into their procedures: one, some knowledge of foreign languages and, two, typical Muslim exegetical tools. At times, the combination of these two elements must have resulted in what must have appeared even to the exegetes themselves as intuitively “wrong” designations. In a previous essay, I pointed out that certain languages seem to have cultural associations for classical Muslims, in the same way that we might say “it’s all Greek to me” or use French expressions in English that have a certain social status (e.g. RSVP). Overall, it may be noted that, while there appears to be a knowledge that the Jewish Bible was written in Hebrew, the language of the Biblical characters spoken of in the Qur’an does not seem to have been connected to Hebrew very often. In al-Suyuti’s al-Mutawakkili, as I indicated only nineteen words are cited as possibly being Hebrew, and seven of those are cited in a manner which clearly indicates that al-Suyuti did not consider these claims to have much support. This is odd: one might have thought that Muslims would have known the Hebrew Bible was written in Hebrew and thus would have assumed that the ancient characters would have spoken Hebrew and that would be a popular language to suggest as a source of the Arabic vocabulary. However, suggestions regarding other languages such as Aramaic, Syriac and Coptic are quite significant, even when words arise in the context of narratives about prophets of the ancient past. This suggests that the ideas surrounding the languages from which “foreign” words were thought to originate were dictated to some extent by the spoken foreign languages known to the Arabs, suggesting a very nonhistorical view of the world: that is, that the language a group of people was speaking in the present was the language they had always spoken. There seem to be other factors at play as well.25 Certain common Arabic words – tahta said to mean “within” rather than its normal “under” in Q 19:24 for just one example – are attributed to Coptic when the words take on meanings which are opposite to their common Arabic designation. This leads to the 257 ANDREW RIPPIN observation that perhaps Coptic played a cultural role as a language of deception for Arabic speakers; there may well be a larger social picture behind this of an image of Copts as deceptive in their dealings with Muslims and twisting the Arabic language to their own advantage. Likewise, the attribution of a number of words to Greek seems to convey certain cultural assumptions rather than linguistic knowledge. In this case it appears to be matters related to commerce and urban society. For example, the following words are commonly attributed to Greek: qist, “justice;” qistas, “scales;” sirat, “road;” and qintar, “hundred weight.” It is worthy of note that while, in a number of instances, modern philology agrees with the early Muslim thoughts on certain words being derived ultimately from Greek, that does not indicate necessary linguistic knowledge. The idea that these words come from Greek does not, in fact, account historically for the words in Arabic. In no instance is it likely that the word passed directly into Arabic from Greek. It is far more likely that Aramaic or Syriac was the conduit for the transmission of the Greek words. In a number of cases, Greek is not the ultimate source anyway; rather, the words are Latin and have moved into the Middle Eastern languages through their Hellenized forms during times of Greek administrative rule. The idea held by Muslim exegetes that the words are Greek, therefore, is unlikely to be the result of linguistic observation. Given the context of classical Arab understanding of Nabati and Suryani especially as reflected in various statements that are found in classical writers and the existence of the Book of Nabataean Agriculture, we might expect to see a focus on agriculture, water, magic and so forth in the words which are designated as coming from this language. It is tempting to make that generalization – seeing words like yamm, “sea,” sari, “river,” and jannat ‘adn, “Garden of Eden,” but really that would be on the basis of only a few words out of the overall collection. The other intuitive place we might think of would be with words associated with Christianity but there, too, we do rather come up short once again with only a few words being specifically Christian in context within the Qur’an. Ultimately, the answer to this question of why these words would be chosen to be thought of as Syriac would seem at best to lie in multiple contexts about which it is difficult to generalize: at least my imagination cannot quite make the leap. But some Muslims did think of these words as having a Syriac background or parallel and that is the interesting fact with which we are dealing. In order to bring this paper to a conclusion, then, let me return to where I started and bring us back to the modern day. There’s no doubt that our knowledge of the transmission of language is much better developed than that of classical Muslim writers. But, of course, no one argues that mere parallels between Syriac and Arabic prove the case – lots of words have linguistic parallels and both languages share much common vocabulary. Careful modern scholars have thus generally looked to technical terms, especially religious ones, including proper names, as the key to understanding the Syriac background. The example of Qur’an and qeryana is 258 SYRIAC IN THE QUR’AN an obvious case, as is the instance of many proper names. In-built here is a presumption of a setting embedded in Christianity that has conveyed terminology presumably not present in Arabic before, because it had no particular application and thus no need for development. But an issue remains here that puts much of this discussion in the same realm of speculative association as the treatises of the medieval scholars: we have little sense of a social context in which to see this linguistic transfer happening and thus the talk is of “Syriac Christianity” defined in quite a loose manner. This is even more inconvenient when one considers the Qur’an to have originated in the Hijaz: the question must be, how did all this vocabulary infiltrate? We embark on a speculative path that becomes closer and closer to my attempts at generalizations for medieval theories: in medieval times, there were perhaps cultural images of water, agriculture and magic that became associated with Syriac; today we associate camel caravans and wandering Christians with the cultural forces and base this on isolated hadith reports that give some credence to the speculation. A review of the assumptions of Jeffery’s Foreign Vocabulary of the Qur’an will quickly show how dated the notions of cultural interchange are in historiographical terms.26 For those like Luxenberg who wish to see an even stronger Syriac presence in the text, the need to account for the mechanism by which this happened and, especially, to place that in a context of what we can postulate about the emergence of the text of the Qur’an itself, is even more compelling. That is, all the philological knowledge in the world is not going to help unless that is combined with a historical picture related to the emergence of the text of the Qur’an. All these factors definitely need to come together. François de Blois has made this point well in his recent article on the religious vocabulary of Christianity and Islam although he uses his evidence as an argument against the “revisionist” stance without himself postulating a critical assessment of the history upon which he relies beyond the philological–linguistic – other than saying that one could “imagine a situation where there existed, presumably in Mecca, an isolated outpost of Nazoraean ‘Jewish Christianity.’ ”27 Only when we gaze back to the tenth century can we see any sense of real certainty. For medieval Muslims, the answers to many of these matters were easy: they knew about the origin of the Qur’an and that the presence of Syriac words in the text – if one accepted that there were any to begin with – was a part of the revelation of God’s mystery and knowledge, and this did not need to be accounted for on a human level, in the way in which we must today. And that surely is the only difference between the two approaches, the modern and the medieval: modern speculative theories or theories based upon assertion do not, in fact, take us anywhere beyond the medieval position. Faith in the historical record rivals faith in the divine. Certainly I think that as our knowledge of early Islam and the Qur’an evolves, the place of foreign vocabulary in the text of the Qur’an will be one element – a critical one, I do think – in helping us understand the emergence of the phenomenon which we know as the Qur’an. But the attempt to specify this 259 ANDREW RIPPIN outside the general flow of history within an overall sectarian milieu of the Near East will always remain speculative. The search thus is not for the “origin” of the text of the Qur’an: the text of the Qur’an as we have it is what we must deal with, but that must be viewed as one element within the Jewish and Christian midrash of the Near East, a trend which continues down until today as we participate in that very process ourselves. Surely here is the significance of Luxenberg: what does he tell us about who he is, about his times, about his religious perspective in relationship to the Qur’an? Surely those are the most interesting questions. Notes 1 Christoph Luxenberg, Die syro-aramäische Lesart des Koran. Ein Beitrag zur Entschlüsselung der Koransprache, Berlin: Das Arabische Buch, 2000. 2 Alphonse Mingana, “Syriac influence on the style of the Koran,” Bulletin of the John Rylands Library 11, 1927, 77–98, reprinted in Ibn Warraq, What the Koran really says. Language, Text, and Commentary, Amherst, NY: Prometheus Books, 2002, 171–92, to which reference is made hereafter. 3 I have also treated aspects of these topics previously in recent essays: Andrew Rippin, “The designation of ‘foreign’ languages in the exegesis of the Qur’an,” in J. D. McAuliffe, B. D. Walfish, J. W. Goering (eds), With Reverence for the Word. Medieval Scriptural Exegesis in Judaism, Christianity and Islam, Oxford: Oxford University Press, 2003, 437–44; Andrew Rippin, “Foreign vocabulary,” EQ, 2, 226–37. 4 Mingana, “Syriac influence on the style of the Koran,” in Ibn Warraq, What the Koran really says. Lauguage, Text, and Commentary, Amherst, NY: Prometheus Books, 2002, 180. 5 Abu ‘Ubayda, Majaz al-Qur’an, F. Sezgin (ed.), Cairo: al-Khaniji, 1954, 1, 17–18. 6 Ibid., I, 17. 7 Muhammad Sayyid Kilani (ed.), Cairo: Halabi, 1973. English trans. M. Khadduri, Al-Imam Muhammad ibn Idris al-Shafi‘i’s al-Risala, Baltimore: Johns Hopkins University Press, 1961, 88–9. 8 Ibid., 28; English trans., 90. 9 Claude Gilliot, Exégèse, langue, et théologie en Islam. L’exégèse coranique de Tabari (m. 311/923), Paris: Vrin, 1990, 98–9. 10 Jami‘ al-bayan ‘an ta’wil ay al-Qur’an, Shakir (ed.), Cairo, Dar al-Ma‘arif, 1954. English trans. J. Cooper, The Commentary on the Qur’an by Abu Ja‘far Muhammad b. Jarir al-Tabari, Oxford: Oxford University Press, 1987, 14. 11 Al-Jawaliqi, al-Mu‘arrab min al-kalam al-a‘jami ‘ala huruf al-mu‘jam, Ahmad Muhammad Shakir (ed.), Cairo: Matba‘at Dar al-Kutub, 1942, 3. 12 This, it seems to me, is the underlying point of the various works of al-Suyuti devoted to foreign words in the Qur’an; see “Foreign vocabulary,” EQ, 2, 231–2. 13 This is a topic which has been studied in some detail by Ramzi Baalbaki, “Early Arab lexicographers and their use of Semitic languages,” Berytus 31, 1983, 117–27, reprinted in his Grammarians and Grammatical Theory in the Medieval Arabic Tradition, Aldershot: Ashgate/Variorum, 2004. 14 Al-Suyuti, al-Muzhir fi ‘ulum al-lugha wa-anwa’iha, Muhammad Abu l-Fadl Ibrahim et al. (eds), Cairo: Dar Ihya’ al-Kutub al-‘Arabiyya, 1958, 1, 270. 15 Baalbaki, “Early Arab lexicographers and their use of Semitic languages,” Berytus 31, 1983, 118–20. 16 Mingana, “Syriac influence on the style of the Koran,” in Ibn Warraq, What the Koran really says. Language, Text, and Commentary, Amherst, NY: Prometheus Books, 2002, 174. 260 SYRIAC IN THE QUR’AN 17 18 19 20 21 22 23 24 25 26 27 “Nabatiya,” EI2. Ibid. Ibid. Ibid. Aharon Maman, Comparative Semitic Philology in the Middle Ages from Sa’adiah Gaon to Ibn Barun (10th-12th C.), Leiden: Brill, 2004; for what follows see especially 53–5: “Nomenclature for Aramaic.” See the recent analysis Jaakko Hämeen-Anttila, “The Nabatean Agriculture: Authenticity, textual history and analysis,” Zeitschrift für Geschichte der arabischislamischen Wissenschaften 15 (2002/3), 249–80. Al-Suyuti, al-Mutawakkili fima warada fi-l-Qur’an bi-l-lughat, Trans. W. Bell (ed.), Cairo: Nile Mission Press, 1926. Arthur Jeffery, The Foreign Vocabulary of the Quran, Baroda: Oriental Institute, 1938, 31–2. As I have pointed out in my entry on “Foreign Vocabulary” EQ. Jeffery, Foreign Vocabulary, for example, 38. François de Blois, “Nasrani and hanif: Studies on the religious vocabulary of Christianity and Islam,” BSOAS 65, 2002, 27. 261 BIBLIOGRAPHY Primary sources, Islamic Abu ‘Ubayda. Majaz al-Qur’an. M. Sezgin (ed.). Cairo: al-Khaniji, 1954. Al-Baghawi. Tafsir. Khalid ‘Abd al-Rahman al-‘Ak and Marwan Sawar (eds). Beirut: Dar al-Kutub al-‘Ilmiyya, 1992. Al-Baladhuri. Futuh al-buldan. ‘Abd Allah and ‘Umar Anis al-Tabba (eds). Beirut: Mu’assasat al-Ma’arif, 1377/1957 (2nd edn: 1987). Bal‘ami: al-Tabari (i.e. Bal‘ami). Muhammad, sceau des prophètes. Trans. H. Zotenberg. Paris: Sindbad, 1980. Al-Bayhaqi. Dala’il al-nubuwwa. ‘Abd al-Mu’ti Qal‘aji (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1405/1985. Al-Bukhari. Al-Jami‘ al-sahih. L. Krehl (ed.). Leiden: Brill, 1862–1908. The Divans of the Six Ancient Arabic Poets: Ennabiga, ‘Antara, Tharafa, Zuhair, ‘Alqama and Imruulqais. W. Ahlwardt (ed.). London: Trübner, 1870 (Reprint: Osnabrück: Biblio, 1972). Hud b. Muhakkam. Tafsir. Balhajj b. Sa‘id Sharifi (ed.). Beirut: Dar Gharb al-Islami, 1990. Ibn ‘Abd Rabbihi. Al-‘Iqd al-farid. ‘Abd al-Majid al-Tarhini (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1987. Ibn Abi l-Zamanin. Tafsir. Husayn b. ‘Ukasha and Muhammad b. Mustafa al-Kanz (eds). Cairo: al-Faruq al-Haditha li-l-Tiba‘a wa-l-Nashr, 2002. Ibn ‘Asakir. Ta’rikh madinat Dimashq. Abu Sa‘id ‘Umar al-‘Amrawi and ‘Ali Shiri (eds). Beirut: Dar al-Fikr, 1995–2001. Ibn ‘Atiyya al-Andalusi. Al-Muharrar al-wajiz fi tafsir al-kitab al-‘aziz. ‘Abd al-Salam Muhammad (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1993. Ibn Hajar. Fath al-bari bi-sharh Sahih al-Bukhari. ‘Abd al-‘Aziz b. ‘Abdallah Baz (ed.), numeration of the hadiths according to the ed. of Muhammad Fu’ad ‘Abd al-Baqi, under the direction of Muhibb al-Din Khatib. Cairo: Dar al-Kitab al-Jadid, 1390/1970 (Reprint: Beirut: Dar al-Ma’rifa, n.d.). —— Tahdhib al-tahdhib. N.e. Hyderabad: Da’irat al-Ma‘arif al-Nizamiyya, 1907–9 (Reprint: Beirut: Dar Sadir, n.d.). Ibn Hanbal. Al-Musnad. Muhammad al-Zahri al-Ghamrawi (ed.). Cairo: al-Matba‘a al-Maymaniyya, 1895. Ibn Hisham. Sirat rasul Allah. In F. Wüstenfeld (ed.). Das Leben Muhammeds nach Muhammad Ibn Ishaq. Göttingen: Dieterich, 1858–60. Trans. A. Guillaume. The Life of Muhammad. Oxford: Oxford University Press, 1955. 262 BIBLIOGRAPHY Ibn Ishaq. See Ibn Hisham. Ibn Kathir. Tafsir al-qur’an al-‘azim. N.e. Beirut: Dar al-Jil, 1990. Ibn al-Nadim. Fihrist. G. Flügel (ed.). Leipzig: Brockhaus, 1871. Ibn Rusta. Al-A‘laq al-nafisa. M.J. de Goeje (ed.). Leiden: Brill, 1892. Ibn Sa‘d. K. al-Tabaqat. E. Sachau (ed.). Leiden: Brill, 1905–21. Ibn Shabba. Ta’rikh al-Madina al-munawwara. Fahim Muhammad Shaltut (ed.). Jadda: Ahmad, 1399/1979. Al-Idrisi. Opus geographicum. A. Barbaci (ed.). Naples-Rome: Istituto Universitario di Napoli, 1982. Al-Isfahani, Abu l-Faraj. Kitab al-Aghani. N.e. Bulaq: n.p., 1285/1868. Al-Jawaliqi. al-Mu‘arrab min al-kalam al-a‘jami ‘ala huruf al-mu‘jam. Ahmad Muhammad Shakir (ed.). Cairo: Dar al-Kutub al-Misriyya, 1942. Al-Ka‘bi al-Balkhi. Qabul al-akhbar. Abu ‘Amr al-Husayni b. ‘Umar b. ‘Abd al-Rahim (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1421/2000. al-Mahalli, Jalal al-Din and Jalal al-Din al-Suyuti. Tafsir al-Jalalayn. Marwan Siwar (ed.). Beirut: Dar al-Jil, 1995. Al-Maqrizi. Kitab al-Mawa‘iz wa-l-i‘tibar fi dhikr al-khitat wa-l-athar. N.e. Cairo: Maktabat al-Thaqafa al-Diniyya, 1987. Al-Maturidi. Ayat wa-suwar min ta’wilat al-Qur’an. A. Vanlioælu and B. Topaloælu (eds). Istanbul: Acar Matbaacilik, 2003. —— Ta’wilat ahl al-sunna. Fatima Yusuf al-Khaymi (ed.). Beirut: Mu’assasat al-Risala, 1425/2004. Al-Mawardi. Tafsir al-Mawardi. al-Sayyid b. ‘Abd al-Maqsud b. ‘Abd al-Rahim (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, n.d. Mu‘jam al-qira’at al-qur’aniyya. ‘Abd al-‘Al Mukarram and Ahmad ‘Umar (eds). Kuwait: Dhat al-Salasil, 1403/1983. Muqatil b. Sulayman. Tafsir al-Qur’an. ‘Ali Muhammad Shahata (ed.). Cairo: Mu’assasat al-Halabi, 1980–9. Al-Nuwayri. Nihayat al-arab fi funun al-adab. N.e. Cairo: Dar al-Kutub al-Misriyya, 1923–98. Al-Qurtubi. Tafsir. Ahmad ‘Abd al-Halim al-Barduni (ed.). Cairo: Dar al-Kitab al-‘Arabi, 1952–67. Al-Salihi. Subul al-huda. ‘Adil Ahmad ‘Abd al-Mawjud and ‘Ali Muhammad Mu‘awwad (eds). Beirut: Dar al-Kutub al-‘Ilmiyya, 1414/1993. Al-Shafi‘i, Risala. Muhammad Sayyid Kilani (ed.). Cairo: Halabi, 1973. Trans. M. Khadduri, Al-Imam Muhammad ibn Idris al-Shafi‘i’s al-Risala. Baltimore, MD: Johns Hopkins University Press, 1961. Sources de la transmission manuscrite du texte coranique, I. Les manuscrits de style hiæazi, le manuscrit arabe 328. F. Déroche and S.N. Noseda (eds). Lesa, Italy: Fondazione F. Noja Noseda, 1998. Sources de la transmission manuscrite du texte coranique, I. Les manuscrits de style hiæazi, Le manuscrit Or. 2165 (f. 1 à 61). F. Déroche and S.N. Noseda (eds). Lesa, Italy: Fondazione F. Noja Noseda, 2001. Al-Suhayli. Al-Ta‘rif wa-l-i‘lam fi-ma ubhima fi l-Qur’an min al-asma’ al-a‘lam. ‘Ali Muhammad al-Naqrat (ed.). Tripoli: Kulliyat al-Da’wa al-Islamiyya wa-Lajnat al-Hifaz ‘ala l-Turath al-Islami, 1992. Al-Suyuti. Al-Mutawakkili fima warada fi-l-Qur’an bi-l-lughat. Trans. W. Bell (ed.). Cairo: Nile Mission Press, 1926. 263 BIBLIOGRAPHY Al-Suyuti. al-Muzhir fi ‘ulum al-lugha wa-anwa‘iha. Muhammad Abu l-Fadl Ibrahim (ed.) Cairo: Dar Ihya’ al-Kutub al-‘Arabiyya, 1958. —— Al-Itqan fi ‘ulum al-Qur’an. Muhammad Abu l-Fadl Ibrahim (ed.). Cairo: al-Hay’a al-Misriyya al-‘Amma li-l-Kitab, 1967 (2nd edn: 1974–5). —— al-Durr al-manthur fi l-tafsir al-ma’thur. N.e. Beirut: Dar al-Kutub al-‘Ilmiyya, 2004. Al-Tabari, Abu Ja‘far. Tafsir. Ahmad Sa‘id ‘Ali and Mustafa al-Saqqa (eds). Cairo: Mustafa al-Babi al-Halabi, 1954–7. —— Annales (Ta’rikh). M.J. de Goeje (ed.). Leiden: Brill, 1879–1901. Al-Tabrisi. Majma‘ al-bayan fi tafsir al-qur’an. Hashim al-Mahallati (ed.). Beirut: Dar Ihya’ al-Turath al-‘Arabi, 1992. Tafsir al-Jalalayn, See al-Mahalli, Jalal al-Din. Al-Tha‘labi. Al-Kashf wa-l-bayan ‘an tafsir al-Qur’an. Abu Muhammad Ibn ‘Ashur (ed.). Beirut: Dar Ihya’ al-Turath al-‘Arabi, 2002. Al-Tusi. Al-Tibyan fi tafsir al-Qur’an. Ahmad Qasir al-‘Amili (ed.). Al-Najaf: Matba‘at al-Nu‘man, 1962. Wahb b. Munabbih. Kitab al-Tijan. Hyderabad: Matba‘at Majlis Da’irat al-Ma‘arif al-‘Uthmaniyya, 1929. Yahya b. Sallam. Tafsir. Hind Shalabi (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1425/2004. Al-Ya‘qubi. Ibn Wadhih Qui Dicitur al-Ja‘qubi Historiae. M.Th. Houtsma (ed.). Leiden: Brill, 1969. Yaqut al-Hamawi. “Mu‘jam” al-buldan. F. Wüstenfeld (ed.). Göttingen: Brockhaus, 1866–73. Al-Zurqani. Sharh al-Mawahib al-laduniyya. Muhammad ‘Abd al-‘Aziz Khalidi (ed.). Beirut: Dar al-Kutub al-‘Ilmiyya, 1996. Primary sources, non-Islamic 1 Enoch: Das Buch Henoch. Äthiopischer Text. J. Flemming (ed.). Leipzig: J.C. Heinrichs’sche Buchhandlung, 1902. Ahudemmeh. “Histories d’Ahoudemmeh et de Marouta.” Patrologia Orientalis 3. F. Nau (ed.). Paris: Firmin-Didot, 1909. Das altarmenische Evangelium. B.O. Künzle (ed.). New York: Lang, 1984. Andreas of Caesarea. Studien zur Geschichte des griechischen Apokalypse-Texte I. Der Apokalypse-kommentar des Andreas von Kaisareia. J. Schmid (ed.). Munich: Zink, 1955. Anonymi Auctoris Chronicon ad Annum Christi 1234 Pertinens, CSCO 81 (text) 109 (Latin translation) 354 (French translation). J.-B. Chabot (ed.). Paris: E Typographeo Reipublicae, 1920. Apocalypsis Henochi Graece. M. Black (ed.). Leiden: Brill, 1970. Bar Hebraeus. Chronicon syriacum. P. Bedjan (ed.). Paris: Maisonneuve, 1890. —— Ta’rikh mukhtasar al-duwal. Antun Salihani (ed.). Beirut: al-Matba‘at al-Kathulikiyya li-l-Aba’ al-Yasu‘iyyin, 1890. Callimachus. The Poems of Callimachus. Trans. F. Nisetich. Oxford: Oxford University Press, 2001. Chronica Minora. CSCO 1. I. Guidi (ed.). Leipzig: Harrassowitz, 1903. Chronicarum quae dicuntur Fredegarii Scholastici libri IV. Cum Continuationibus, Monumenta Germaniae Historica, Scriptorum Rerum Merovingicarum tomus II. B. Krusch (ed.). Hannover: Hahn, 1888. 264 BIBLIOGRAPHY Chronicle of Siirt, Ch. 60. Patrologia Orientalis 13, Paris: Firim-Didot, 1919. Chronicon ad annum 1234. CSCO 81. J.B. Chabot (ed.). Louvain: Peeters, 1920. Ephraem. Des heiligen Ephraem des Syrers Hymnen de Paradiso und Contra Julianum, CSCO 174–5. E. Beck (ed. and trans.). Louvain: Peeters, 1957. —— Des Heiligen Ephraem des Syrers Sermones III. CSCO 320. E. Beck (ed.). Louvain: Peeters 1972. —— Sancti Patris Nostri Ephraem Syri Opera Omnia quae exstant Graece, Syriace, Latine. J.S. Assemani (ed.). Rome: n.p., 1732–46. Eutychius of Alexandria. The Book of the Demonstration (Kitab al-Burhan). P. Cachia (ed.). CSCO 192 and 209. W.M. Watt (ed.). CSCO 193 and 210. Leuven: Peeters, 1960 and 1961. The Gospel of the Twelve Apostles. J.R. Harris (ed.). Cambridge: Cambridge University Press, 1900. The History of Alexander the Great, being the Syriac Version of the Pseudo-Callisthenes. Trans. E.A. Wallis Budge. Cambridge: Cambridge University Press, 1889. Isho‘yahb. Liber Epistularum. CSCO 11. R. Duval (ed.). Paris: E Typographeo Reipublicae, 1904–5. Jacob of Serugh. Jacob of Serugh: Select Festal Homilies. Trans. T. Kollamparampil. Rome: Centre for Indian and Inter-religious Studies /Bangalore: Dharmaram Publications, 1997. John of Ephesus. Historiae Ecclesiasticae III. E. Brook (ed.). Paris: E Typographeo Reipublicae, 1935. Littmann, E. Semitic Inscriptions. Princeton University Archaeological Expeditions to Syria. Leiden: Brill, 1914–49. Mäsah7ftä Sälomon wä-Sirak. Addis Abeba: T7nfa’e Masatämiya D7r7jjt, 1917 a.m. ⫽ 1924 (Reprint: 1988 a.m. ⫽ 1995). Michael the Syrian. Chronique de Michel le Syrien Patriarche Jacobite d’Antioche (1166–1199). J.B. Chabot (ed.). Paris: Leroux, 1899–1924. Movses Dasxuranc‘i. Patmut‘iwn A„uanic‘ Amxarhi. M. Emin (ed.). Moscow, n.p. 1860 (Reprint: Tbilisi, n.p., 1912). Trans. History of the Caucasian Albanians. Trans. C.J.F. Dowsett. London: Oxford University Press, 1961. Narsai. Narsai Doctoris Syri Homiliae et Carmina. A. Mingana (ed.). Mosul: Fratrum Praedicatorum, 1905. New Testament Apocrypha. W. Schneemelcher (ed.). Trans. R. Wilson. Philadelphia: Westminster Press, 1963 (2nd edn: Cambridge: Clarke, 1991). Octateuchus Aethiopicus (⫽ Biblia Veteris Testamenti Aethiopica 1). Lipsiae: Vogelii, 1853–97. Photius. Bibliotheca. Trans. R. Henry (ed.). Paris: Société d’édition les belles lettres, 1959–77. Sebeos’ History. Trans. R. Bedrosian. New York: Sources of the Armenian Tradition, 1985. Sozomen. Historia Ecclesiastica. R. Hussey (ed.). Oxford: Academic Press, 1860. Das syrische Alexanderlied. Die Drei Rezensionen. CSCO 454 (edn)-455 (trans.). G.J. Reinink (ed. and trans.). Louvain: Peeters, 1983. Testi Orientali Inediti sopra i Sette Dormienti di Efeso. Reale Accademia dei Lincei 282, 1884–5. I. Guidi (ed.). Roma: Tipografia della R. Accademia dei Lincei, 1885. Un texto árabe occidental de la Leyenda de Alejandro. Trans. E. García Gómez. Madrid: Instituto de Valencia de Don Juan, 1929. Theodore Bar Konai. Theodorus Bar Koni Liber Scholiorum. CSCO 55. A. Scher (ed.). Paris: E Typographeo Reipublicae, 1910. 265 BIBLIOGRAPHY Theodore of Cyrrhus. Theodoret Kirchengeschichte. L. Parmentier (ed.). Berlin: Akademie Verlag, 1954. Trans. The Ecclesiastical History, Dialogues, and Letters of Theodoret. A Select Library Nicene and Post-Nicene Fathers of the Christian Church 3. Trans. B. Jackson. Grand Rapids, MI: Eerdmans, 1979. Theophanes. Chronicle. C. de Boor (ed.). Leipzig: Teuneri, 1883–5. Trans. The Chronicle of Theophanes the Confessor. Trans. C. Mango and R. Scott. Oxford: Clarendon, 1997. Zacharias of Mytilene. Historia Ecclesiastica Zachariae Rhetori vulgo Adscripta II. CSCO 84. E.W. Brooks (ed.). Paris: E Typographeo Reipublicae, 1921. Secondary sources Abbott, N. The Rise of the North Arabic Script and Its Üur’anic Development, with a Full Description of the Üuran Manuscripts in the Oriental Institute. Chicago: The University of Chicago Press, 1939. Aerts, F.C.W. Alexander Magnus Arabicus: Zeven eeuwen Arabische Alexandertraditie: van Pseudo-Callisthenes tot Suri. Dissertation, University of Leiden, 2003. Ahrens, K. “Christliches im Qoran. Eine Nachlese.” ZDMG 84, 1930, 15–68, 148–90. Allgeier, A. “Untersuchungen zur syrischen Überlieferung der Siebenschläferlegende.” OC 4, 1915, 279–97; 5, 1915, 10–59. —— Die westsyrische Überlieferung der Siebenschläferlegende. Leipzig: W. Drugulin, 1915. —— “Die älteste Gestalt der Siebenschläferlegende.” OC 6, 1916, 1–43; 7–8, 1918, 33–87. —— “Der Ursprung des griechischen Siebenschläferlegende.” Byzantinische-neugriechische Jahrbücher 3, 1922, 311–31. Altheim, F. and R. Stiehl, Die Araber in der alten Welt. Berlin: de Gruyter, 1964–9. Ambros, A. and S. Prochazka. Concise Dictionary of Koranic Arabic. Wiesbaden: Reichert, 2004. Anderson, A.R. “Alexander’s horns.” Transactions and Proceedings of the American Philological Association 58, 1927, 100–22. —— “Alexander at the Caspian gates.” Transactions and Proceedings of the American Philological Association 59, 1928, 130–63. Andrae, T. “Der Ursprung des Islams und das Christentum.” Kyrkshistorisk årsskrift 23, 1923, 149–206; 24, 1924, 213–25; 25, 1925, 45–112 (Reprint: Der Ursprung des Islams und das Christentum. Uppsala: Almqvist and Wiksells, 1926). Trans. Les origines de l’islam et le christianisme. Trans. J. Roche. Paris: Adrien-Maisonneuve, 1955. —— Mohammed, sein Leben und sein Glaube. Göttingen: Vandenhoeck and Ruprecht, 1932. Trans. Mohammed: The Man and His Faith. Trans. T. Menzel. New York: Charles Scribner’s Sons, 1936. —— I Myrtenträdgården: Studier I Tidig Islamisk Mystik. Lund: Albert Bonniers Forlag, 1947. Trans. In the Garden of Myrtles: Studies in Early Islamic Mysticism. Trans. B. Sharpe. Albany, NY: State University of New York Press, 1987. Arabic Literature to the End of the Umayyad Period. A.F.L. Beeston, T.M. Johnstone, R.B. Serjeant and G.B. Smith (eds). Cambridge: Cambridge University Press, 1983. Armstrong, K. Muhammad, San Francisco, CA: HarperCollins, 1992. Arnaldez, R. Jésus fils de Marie, prophète de l’Islam. Paris: Desclée, 1980. Athamina, K. “The tribal kings in pre-Islamic Arabia.” al-Qantara 19, 1998, 19–37. ‘Atwan, Husayn. Riwayat al-shamiyyin li-l-maghazi wa-l-siyar fi l-qarnayn al-awwal wa-l-thani li-l-hijra. Beirut: Dar al-Jil, 1986. 266 BIBLIOGRAPHY Al-A‘zami, Muhammad Mustafa. Maghazi rasul Allah. ‘Urwa b. al-Zubayr b. al-‘Awwam. Transmitted by Abu l-Aswad. Riyadh: Maktab al-Tarbiya al-‘Arabi li-Duwal al-Khalij, 1981. Azzi, J. See al-Hariri, Abu Musa. Baalbaki, R. “Early Arab lexicographers and their use of Semitic languages.” Berytus 31, 1983, 117–27 (Reprint: Grammarians and Grammatical Theory in the Medieval Arabic Tradition. Aldershot: Ashgate/Variorum, 2004). Al-Bakri, Muhammad Hamdi. Usul naqd al-nusus wa-nashr al-kutub, muhadarat al-mustashriq al-almani Birjstrasir. Cairo: Matba‘at Dar al-Kutub, 1969. Barclay, J. “Melkite Orthodox Syro-Byzantine manuscripts in Syriac and Palestinian Aramaic.” Liber Annuus 21, 1991, 205–19. Baumstark, A. Geschichte der Syrischen Literatur. Bonn: A. Markus and E. Weber, 1922. —— “Das Problem eines vorislamischen christlich-kirchlichen Schrifttums in arabischer Sprache.” Islamica 4, 1929–31, 562–75. —— “Arabische Übersetzung eines altsyrischen Evangelientextes und die Sure 21, 105 zitierte Psalmenübersetzung.” OC, 1934, 3. ser. 9, 31, 165–88. —— “Eine frühislamische und eine vorislamische arabische Evangelienübersetzung aus dem Syrischen.” Atti del XIX Congresso Internazionale degli Orientalisti (1935). Rome: Bardi, 1938, 682–4. Beck, E. “Der ‘uthmanische Kodex in der Koranlesung des zweiten Jahrhunderts.” Orientalia 14, 1945, 355–73. —— “ ‘Arabiyya, Sunna und ‘amma in der Koranlesung des zweiten Jahrhunderts.” Orientalia 15, 1946, 180–224. —— “Die Kodizesvarianten der Amsar.” Orientalia 16, 1947, 353–76. —— “Eine christliche Parallele zu den Paradiesesjungfrauen des Korans?” Orientalia Christiana Periodica 14, 1948, 398–405. —— “Les Houris du Coran et Ephrem le Syrien.” MIDEO 6, 1959–61, 405–8. Becker, C.H. Christianity and Islam. Trans. H.J. Chaytor. New York and London: Harper, 1909. Bedjan, P. Acta Martyrum et Sanctorum Syriace I. Paris: Otto Harrassowitz, 1890. Beeston, A.F.L. “Nemara and Faw.” BSOAS 42, 1979, 1–6. —— “Antecedents of classical Arabic verse.” Festschrift Ewald Wagner. Beirut: Steiner, 1994, 1, 234–43. Bell, R. The Origin of Islam in its Christian Environment. London: Macmillan, 1926. —— “Muhammad’s call.” The Muslim World 24, 1934, 13–19. —— “Muhammad’s visions.” The Muslim World 24, 1934, 145–54. —— The Qur’an. Translated with a Critical Re-arrangement of the Suras. Edinburgh: Edinburgh University Press, 1937 (Reprint: 1960). —— A Commentary on the Qur’an. C.E. Bosworth and M.E.J. Richardson (eds). Manchester: University of Manchester, 1991. Bellamy, J.A. “The mysterious letters of the Qur’an: Old abbreviations of the Basmalah.” JAOS 93, 1973, 267–85. —— “A new reading of the Namarah inscription.” JAOS 105, 1985, 31–48. —— “Two pre-Islamic Arabic inscriptions revised: Jabal Ramm and Umm al-Jimal.” JAOS 108, 1988, 369–78. —— “Al-Raqim or al-Ruqud? A note on Surah 18:9.” JAOS 111, 1991, 115–17. —— “Fa-Ummuhu Hawiyah: A note on Surah 101:9.” JAOS 112, 1992, 485–7. —— “Some proposed emendations to the text of the Koran.” JAOS 113, 1993, 562–73. —— “More proposed emendations to the text of the Koran.” JAOS 116, 1996, 196–204. 267 BIBLIOGRAPHY Bellamy, J.A. “Textual criticism of the Koran.” JAOS 121, 2001, 1–6. —— “A further note on ‘Isa.” JAOS 122, 2002, 587–8. Bergsträsser, G. “Plan eines Apparatus Criticus zum Qoran.” Sitzungsberichte der Bayerischen Akademie der Wissenschaften 7, January 1930, 3ff. (Reprint: Der Koran. R. Paret (ed.). Wege der Forschung 326, 389–97). —— “Über die Notwendigkeit und Möglichkeit einer kritischen Koranausgabe.” ZDMG 84, 1930, 82–3. —— “Koranlesung in Kairo.” Der Islam 20, 1932, 1–42. Birkeland, H. The Lord Guideth: Studies on Primitive Islam. Oslo: Nygaard, 1956. Blachère, R. Introduction au Coran. Paris: Maisonneuve, 1947. —— Le problème de Mahomet. Paris: PUF, 1952. Blau, J. “Über einige christlich-arabische Manuskripte aus dem 9. und 10. Jahrhundert.” Le Museon 75, 1962, 105–7. Bombaci, A. “Qui était Jebu Xak‘an?” Turcica 2, 1970, 7–24. Bou Mansour, T. La théologie de Jacques de Saroug. Bibliothèque de l’Université Saint-Esprit 16 and 40. Kaslik, Lebanon: L’Université Saint-Esprit, 1993 and 2000. Bousquet, G.-H. Remarques critiques sur le style et la syntaxe du Coran. Paris: Maisonneuve, 1953. Bowman, J. “The debt of Islam to monophysite Syrian Christianity.” Nederlands Theologisch Tijdschrift 19, 1964/65, 177–201 (Reprint: Essays in Honour of Griffithes Wheeler Thatcher (1863–1950). E.C.B. Mac Laurin [ed.]. Sydney: Sydney University Press, 1967, 191–216). Buhl, F. Das Leben Muhammads. Darmstadt: Wissenschaftliche Buchgesellschaft, 1961. Breydy, M. Das Annalenwerk des Eutychios von Alexandrien: ausgewählte Geschichten und Legenden kompiliert von Sa‘id ibn Batriq um 935 AD. CSCO 471. Louvain: Peeters, 1985. Briquel Chatonnet, F. M. Debié and A. Desreumaux (eds). Les inscriptions syriaques. Paris: Geuthner, 2004. Brock, S.P. “An early Syriac life of Maximus the Confessor.” Analecta Bollandiana 91, 1973, 299–346 (Reprint: London: Variorum Reprints, 1984, XII). —— “North Mesopotamia in the Late Seventh Century: Book XV of John of Penkaye’s Ris Melle.” JSAI 9, 1989, 51–75 (Reprint: Studies in Syriac Christianity, London: Variorum, 1992, II). —— “The ‘Nestorian’ Church: A Lamentable Misnomer.” Bulletin of the John Rylands University Library of Manchester 78, 1996, 23–35. Brockelmann, C. Geschichte der arabischen Litteratur. Leiden: Brill, 1937–49. Burton, J. “Those are the high-flying cranes.” JSS 15, 1970, 246–65. —— The Collection of the Qur’an. Cambridge: Cambridge University Press, 1977. —— “The vowelling of Q 65,1.” JSS 29, 1984, 267–83. —— “Linguistics errors in the Qur’an.” JSS 33, 1988, 181–196. —— The Sources of Islamic Law. Edinburgh: Edinburgh University Press, 1990. Busse, H. Die theologischen Beziehungen des Islam zu Judentum und Christentum. Darmstadt: Wissenschaftliche Buchgesellschaft, 1988. Calder, N. “Tafsir from Tabari to Ibn Kathir: Problems in the description of a genre, illustrated with reference to the story of Abraham.” In G.R. Hawting and A.A. Shareef (eds). Approaches to the Qur’an. London: Routledge, 1993, 101–40. Casanova, P. Muhammad et la fin du monde: Étude critique sur l’islam primitif. Paris: Geuthner, 1911–24. 268 BIBLIOGRAPHY Chwolsohn, D. Die Ssabier und der Ssabismus. St. Petersburg: Buchdruckerei der kaiserlichen Akademie der Wissenschaften, 1856. Conti Rossini, C. “L’omelia di Yohannes, vescovo d’Aksum in onore di Garimâ.” Actes du Congrès international des Orientalistes, section sémitique. Paris, n.p., 1898, 139–77. Contini, R. “Il Hawran preislamico ipotesi di storia linguistica.” Felix Ravenna 1, 1987, 2, 25–79. Cook, M. “The stemma of the regional codices of the Koran.” Graeco-Arabica 9–10, 2004, 89–104. Cramer, W. Die Engelvorstellung bei Ephräm dem Syrer. Orientalia Christiana Analecta 173. Rome: Pontificium Institutum Studiorum Orientalium, 1965. Crone, P. “The first-century concept of higra.” Arabica 3, 1994, 352–87. —— “Two legal problems bearing on the early history of the Qur’an.” JSAI 18, 1994, 1–37. Crone, P. and M. Cook. Hagarism. The Making of the Islamic World. Cambridge: Cambridge University Press, 1977. Cumont, F. Catalogue des sculptures et inscriptions antiques des musées royaux. Brussels: Vromant, 1913. Czeglédy, K. “Monographs on Syriac and Muhammadan Sources in the Literary Remains of M. Kmoskó.” Acta Orientalia Academiae Scientiarum Hungaricae 4, 1955, 19–90. —— “The Syriac legend concerning Alexander the Great.” Acta Orientalia Academiae Scientiarum Hungaricae 7, 1957, 231–49. Daniel, R.W. “P. Petra Inv. 10 and Its Arabic.” Atti del XXII Congresso Internazionale di Papirologia. Florence: Istituto papirologico G. Vitelli, 2001, 1, 331–41. de Blois, F. “The Sabians (Sabi’un) in pre-Islamic Arabia.” Acta Orientalia 56, 1995, 39–61. —— “Elchasai–Manes—Muhammad. Manichäismus und Islam in religionshistorischem Vergleich.” Der Islam 81, 2004, 31–48. —— “Nasrani and hanif: studies on the religious vocabulary of Christianity and Islam.” BSOAS 65, 2004, 1–30. de Prémare, A.-L. “Les textes musulmans dans leur environnement.” In Cl. Gilliot et T. Nagel (eds). “Les usages du Coran.” Arabica 47, 2000, 391–408. —— Les fondations de l’islam. Entre écriture et histoire. Paris: Seuil 2002. ——Aux origines du Coran. Question d’hier, approches d’aujourd’hui. Paris: Téraèdre, 2004. Diem, W. “Some glimpses at the rise and early development of the Arabic orthography.” Orientalia 45, 1976, 251–61. —— “Untersuchungen zur frühen Geschichte der arabischen Orthographie, I: Die Schreibung der Vokale.” Orientalia 48, 1979, 207–57, 227. Donner, F.M. Narratives of Islamic Origins: The Beginnings of Islamic Historical Writing. Princeton, NJ: Darwin Press, 1998. —— “From believers to Muslims: Confessional self-identity in the early Islamic community.” Al-Abhath 50–1, 2002–3, 9–53. —— “Qur’anic furqan,” JSS 52, 2007, 279–300. Dundes, A. Fables of the Ancients? Folklore in the Qur’an. Lanham, MD: Rowman & Littlefield, 2003. Die dunklen Anfänge: Neue Forschungen zur Entstehung und frühen Geschichte des Islam. K.-H. Ohlig and G.-R. Puin (eds). Berlin: Schiler, 2005. Dussaud, R. and F. Macler. Mission dans les Regions Désertiques de la Syrie Moyenne. Paris: Imprimerie Nationale, 1903. Dvořák, R. Ein Beitrag zur Frage über die Fremdwörter im Koran. Inaugural dissertation, München, 1884. 269 BIBLIOGRAPHY Dvořák, R. “Über die Fremdwörter im Korân.” Kaiserliche Akademie der Wissenschaften. Phil.-Hist. Classe. Sitzungsberichte 109, 1, 1885, 481–562. El Cheikh, N.M. Byzantium Viewed by the Arabs. Cambridge, MA: Harvard University Press, 2004. Fischer, A. “Eine Qoran-Interpolation.” In Carl Bezold (ed.). Orientalische Studien: Theodor Nöldeke zum 70. Geburtstag. Gieszen: Töpelmann 1906, 1, 33–55. —— “Grammatisch schwierige Schwur- und Beschwörungsformeln.” Der Islam 28, 1948, 1–105. Fleisch, H. Introduction à l’étude des langues sémitiques. Paris: Maisonneuve, 1947. Fleischer, H.L. “Ueber arabische Lexicographie und T-a‘âlibi’s Fiüh al-lu∫a.” Berichten für die Verhandlungen der k. sächs. Gesellschaft der W. Philol.-histor. Cl. 1854, 1–14 (Reprint: Kleinere Schriften 3. Osnabrück 1964, 152–66). Fowden, E.K. The Barbarian Plain. Saint Sergius between Rome and Iran. Berkeley, CA: University of California Press, 1999. Fraenkel, S. Die aramäischen Fremdwörter im Arabischen. Leiden: Brill 1886. Fück, J. Muhammad ibn Ishâq. Literarhistorische Untersuchungen. Dissertation, Frankfurt am Main, 1925. —— ’Arabiya. Untersuchungen zur arabischen Sprach- und Stilgeschichte. Berlin: Akademie-Verlag, 1950. Gaudefroy-Demombynes, M. Mahomet. Paris: Albin Michel, 1957. Geagea, N. Mary in the Koran: A Meeting Point between Christianity and Islam. Trans. L.T. Fares. New York: Philosophical Library, 1984. Geiger, A. Was hat Mohammed aus dem Judenthume aufgenommen. Bonn: F. Baaden, 1833 (Reprint: Berlin: Parerga, 2005). Gerock, C.F. Versuch einer Darstellung der Christologie des Korans. Hamburg: Perthes, 1839. Gilliot, C. Exégèse, langue, et théologie en Islam. Paris: Vrin, 1990. —— “Muhammad, le Coran et les ‘contraintes de l’histoire’.” In S. Wild (ed.). The Qur’an as Text. Leiden: Brill, 1996. —— “Les ‘informateurs’ juifs et chrétiens de Muhammad. Reprise d’un problème traité par Aloys Sprenger et Theodor Nöldeke.” JSAI 22, 1998, 84–126. —— “Un verset manquant du Coran ou réputé tel.” In M.T. Urvoy (ed.). En hommage au Père Jacques Jomier, O.P. Paris: Cerf, 2002, 73–100. —— “Le Coran, fruit d’un travail collectif?” In D. De Smet, G. de Callatay and J.M.F. Van Reeth (eds). Al-Kitab. La sacralité du texte dans le monde de l’Islam. Acta Orientalia Belgica, 3, 2004, 185–23. —— “L’embarras d’un exégète musulman face à un palimpseste. Maturidi et la sourate de l’Abondance (al-Kawthar, sourate 108).” In R. Arnzen and J. Thielmann (eds). Words, Texts and Concepts Cruising the Mediterranean Sea. Leuven: Peeters 2004, 33–69. —— “Les traditions sur la composition ou coordination du Coran (ta’lif al-qur’an).” In Cl. Gilliot und T. Nagel (eds). Das Prophetenhadit. Dimensionen einer islamischen Literaturgattung. Göttingen: Vandenhoeck & Ruprecht, 2005, 14–39. —— “Zur Herkunft der Gewährsmänner des Propheten.” In K.-H. Ohlig and G.-R. Puin (eds). Die dunklen Anfänge: Neue Forschungen zur Entstehung und frühen Geschichte des Islam. Berlin: Schiler, 2005, 148–69. —— “Une reconstruction critique du Coran ou comment en finir avec les merveilles de la lampe d’Aladin?” In M. Kropp (ed.). Results of Contemporary Research on the Qur’an: The Question of a Historico-Critical Text. Beirut: Orient-Institut, 2006, 33–137. 270 BIBLIOGRAPHY —— “Collecte ou mémorisation du Coran. Essai d’analyse d’un vocabulaire ambigu.” JSAI, forthcoming. Ginkel, J.J. van. John of Ephesus: A Monophysite Historian in Sixth-Century Byzantium. PhD. dissertation, Groningen, 1995. Goldziher, I. “Linguistisches aus der Literatur der muhammedanischen Mystik.” ZDMG 26, 1872, 764–85. —— “Neue Materialen zur Literatur des Überlieferungswesens bei den Muhammedanern.” ZDMG 50, 1896, 465–506. —— Muhammedanische Studien. Halle: Niemeyer, 1888–90. Trans. Introduction to Islamic Theology and Law (Vorlesungen über den Islam). Trans. A. and R. Hamori. Princeton, NJ: Princeton University Press, 1981. Graf, D.G. and M.J. Zwettler. “The North Arabian ‘Thamudic E’ Inscriptions from Uraynibah West.” Bulletin of the American School of Oriental Research 335, 2004, 53–89. Gräf, E. “Zu den christlichen Einflüssen im Koran.” al-Bahith 28, Festschrift Joseph Henninger zum 70 Geburtstag. 1976, 111–44. Graf, G. Geschichte der christlichen arabischen Literatur. Rome: Biblioteca Apostolica Vaticana, 1947. Graf von Bothmer, H.-C., K.-H. Ohlig and G.-R. Puin. “Neue Wege der Koranforschung.” Magazin Forschung 1, 1999, 33–46. Griffith, S. “The Gospel in Arabic: An inquiry into its appearance in the first Abbasid century.” OC 69, 1985, 126–67. —— “The Qur’an in Arab Christian texts: The development of an apologetical argument; Abu Qurrah in the Maælis of al-Ma’mun.” Parole de l’Orient 24, 1999, 203–33. —— “Apologetics and historiography in the Annals of Eutychios of Alexandria: Christian self-definition in the world of Islam.” In R. Ebied and H. Teule (eds). Studies on the Christian Arabic Heritage in Honour of Father Prof. Dr. Samir Khalil Samir, S.I. at the Occasion of his Sixty-Fifth Birthday. Eastern Christian Studies 5. Leuven: Peeters, 2004, 65–89. —— “ ‘Syriacisms’ in the Arabic Qur’an: Who were ‘Those who said that Allah is third of three’, according to al-Ma’idah 73?” In M.M. Bar-Asher, S. Hopkins, S. Stroumsa and B. Chiesa (eds). A Word Fitly Spoken: Studies in Mediaeval Exegesis of the Hebrew Bible and the Qur’an; Presented to Haggai Ben-Shammai. Jerusalem: The Ben-Zvi Institute, 2007, 83ff. Grohmann, A. Arabische Paläeographie II. Graz, Vienna: Böhlau, 1971. Gruendler, B. The Development of the Arabic Scripts from the Nabataean era to the first Islamic century. Atlanta, GA: Scholars Press, 1993. Guillame, A. “The version of the Gospels used in Medina.” Al-Andalus 15, 1950, 289–96. Al-Haddad, Durrah. Al-Injil fi-l-Qur’an. Jounieh: Librairie pauliste, 1982. Haddad, Muhammad b. ‘Ali al-Husayni. al-Suyuf al-sahiqa. Cairo: Matba‘at al-Ma‘ahid, 1344. Hämeen-Anttila, J. “The Nabatean Agriculture: Authenticity, textual history and analysis.” Zeitschrift für Geschichte der arabisch-islamischen Wissenschaften 15, 2002/3, 249–80. Hamidullah, M. “Two Christians of Pre-Islamic Mecca.” Journal of the Pakistan Historical Society 6, 1958, 97–103. al-Hariri, Abu Musa. Nabiy al-rahma. Beirut: Diyar ‘Aql, 1990. —— Qass wa-nabiy. Beirut: n.p. 1991. Trans. J. Azzi. Le Prêtre et le Prophète. Trans. M.S. Garnier. Paris: Maisonneuve et Larose, 2001. 271 BIBLIOGRAPHY Harnack, A. von. Lehrbuch der Dogmengeschichte. Zweiter Band: Die Entwicklung des kirchlichen Dogmas. Tübingen: Mohr, 1909. —— Die Mission und Ausbreitung des Christentums in den ersten drei Jahrhunderten. Leipzig: Hinrichs, 1915. Hawting, G.R. The First Dynasty of Islam: The Umayyad Caliphate AD 661–750. London: Routledge, 1989. —— The Idea of Idolatry and the Emergence of Islam. Cambridge: Cambridge University Press, 1999. Hayek, M. Le Christ de l’Islam. Paris: Seuil, 1959. Healey, J. and G. Rex Smith. “Jaussen-Savignac 17: The earliest dated Arabic document.” Atlal 12, 1989, 77–84. Hebbo, A. Die Fremdwörter in der arabischen Prophetenbiographie des Ibn Hischam. Frankfurt: Peter Lang, 1984. Henninger, J. Spuren christlicher Glaubenswahrheiten im Koran. Schöneck: Administration der Neuen Zeitschrift für Missionswissenschaft, 1951. Hirschfeld, H. New Researches into the Composition and Exegesis of the Qoran. London: Royal Asiatic Society, 1902. Hock, R.F. The Infancy Gospels of James and Thomas. Santa Rosa, CA: Polebridge Press, 1995. Honigmann, E. “Stephen of Ephesus (April 15, 448 – October 29, 451) and the Legend of the Seven Sleepers.” In E. Honigmann (ed.). Patristic Studies, Studi e Testi, 173. Vatican City: Biblioteca Apostolica Vaticana, 1953, 17, 125–68. Horovitz, J. “Jewish proper names and derivatives in the Koran.” Hebrew College Annual, 2, 1925, 145–227 (Reprint: Hildesheim: Olms, 1964). —— Koranische Untersuchungen. Berlin and Leipzig: de Gruyter, 1926. —— The Earliest Biographies of the Prophet and Their Authors. L. Conrad (ed.). Princeton, NJ: Darwin, 2002. Howard-Johnston, J. “Heraclius’ Persian campaigns and the revival of the East Roman Empire, 622–630.” War in History 6, 1999, 1–44. —— “Armenian Historians of Heraclius: An Examination of the Aims, Sources, and Working-Methods of Sebeos and Movses Daskhurantsi.” In G.J. Reinink and B.H. Stolte (eds). The Reign of Heraclius (610–641): Crisis and Confrontation. Groningen Studies in Cultural Change 2. Leuven: Peeters, 2002, 41–62. Hoyland, R.G. Seeing Islam as Others Saw It. A Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam. Princeton, NJ: Darwin, 1997. —— “Language and identity: The twin histories of Arabic and Aramaic.” Scripta Classica Israelica 23, 2004, 183–99. —— “Epigraphy and the emergence of Arab identity.” In P. Sijpestein, L. Sundelin, S. Torallas Tovar, and A. Zomeño (eds). From Andalusia to Khurasan: Documents from the Medieval Islamic World. Leiden: Brill, 2006, 219–41. Huber, M. Die Wanderlegende von den Siebenschläfern: Eine literargeschichtliche Untersuchung. Leipzig: Harrassowitz, 1910. Jaussen, P. and P. Savignac. Mission archéologique en Arabie. Paris: Leroux, 1909–22. Jeffery, A. “Progress in the study of the Qur’an text.” Moslem World 25, 1935, 4–16 (Reprint: Der Koran. R. Paret (ed.). Wege der Forschung 326, 398–412). —— Materials for the History of the Qur’an. The Old Codices. Leiden: Brill, 1937. —— The Foreign Vocabulary of the Qur’an. Baroda: Oriental Institute, 1938 (Reprint: Leiden: Brill, 2007). 272 BIBLIOGRAPHY —— “The textual history of the Qur’an.” Journal of the Middle East Society (Jerusalem) 1, 1947, 35–49. —— The Qur’an as Scripture. New York: Moore, 1952. —— “The present status of Qur’anic studies.” Middle East Institute: Report of Current Research, Spring 1957, 1–16. Jeffery, A. and I. Mendelsohn. “The orthography of the Samarqand Qur’an codex.” JAOS 3, 1942, 175–194. Jourdan, F. La tradition des Sept Dormants: Une rencontre entre chrétiens et musulmans. Paris: Maisonneuve et Larose, 2001. Kaegi, W.E. “Initial Byzantine reactions to the Arab conquest.” Church History 38, 1969, 139–49. —— Byzantium and the Early Islamic Conquests, Cambridge: Cambridge University Press, 1992. —— Heraclius, Emperor of Byzantium. Cambridge: Cambridge University Press, 2003. Kahle, P.E. The Cairo Geniza. Oxford: Oxford University Press, 1947 (2nd edn: 1959). —— “The Qur’an and the ‘Arabiya.’ ” In S. Löwinger and J. Somogyi (eds). Ignace Goldziher Memorial Volume. Budapest: n.p., 1948, 163–8. —— “The Arabic readers of the Koran.” JNES 8, 1949, 65–71. Kandler, H. Die Bedeutung der Siebenschläfer (Ashab al-kahf) im Islam: Untersuchungen zu Legenden und Kult in Schrifttum, Religion und Voksglauben unter besonderer Berücksichtigung der Siebenschläfer-Wallfahrt. Bochum: Universitätsverlag Dr. N. Brockmeyer, 1994. Khalil, S. See Samir, S.K. Knauf, E.A. “Bemerkungen zur frühen Geschichte der arabischen Orthographie I: eine übersehene frühhocharabische Inschrift vom Nordostrand des Toten Meeres.” Orientalia 53, 1984, 456–8. Kollamparampil, T. Salvation in Christ according to Jacob of Serugh: An Exegeticotheological Study on the Homilies of Jacob of Serugh (451–521 AD) on the Feasts of Our Lord. Bangalore: Dharmaram Publications, 2001. Der Koran. Ed. R. Paret. Wege der Forschung 326, Darmstadt: Wissenschaftliche Buchgesellschaft, 1975. Krauss, S. “Talmudische Nachrichten über Arabien.” ZDMG 70, 1916, 321–53. Kropp, M. “Iatromagie und der Beginn der arabischen Schriftsprache: die nabatäischarabische Inschrift von ‘Ayn ‘Abada.” Mélanges de l’Université Saint Joseph 55, 1997–98, 91–117. —— “Viele fremde Tische, und noch einer im Koran: Zur Etymologie von äthiopisch ma’7d(d)7 und arabisch ma’ida/mayday.” OC 87, 2003, 140–3. —— “Der äthiopische Satan ⫽ Maitan und seine koranischen Ausläufer mit einer Bemerkung über verbales Steinigen.” OC 89, 2005, 93–102. Lammens, H. “Le Chrétiens à la Mecque à la veille de l’Hégire,” Bulletin de l’institut français d’archéologie orientale, 14, 1918, 191–230. Lamy, T.J. “Profession de foi.” Actes du onzième congrès international des Orientalistes. Paris: Leroux, 1897, 117–32. Land, J.P.N. Anecdota Syriaca. Leiden: Brill, 1868. Lane, E.W. An Arabic-English Lexicon. London: Williams and Norgate, 1863–93 (2nd edn Cambridge: Islamic Texts Society Trust, 1984). Lecker, M. “Zayd b. Thabit, ‘A Jew with Two Sidelocks,’ Judaism and Literacy in Pre-Islamic Medina (Yathrib).” JNES 56, 1997, 259–73. 273 BIBLIOGRAPHY Leemhuis, F. “Ursprünge des Korans als Textus receptus.” In S. Wild and H. Schild (eds). Akten des 27. Deutschen Orientalistentages. Würzburg: Ergon, 2001. Lester, T. “What is the Koran?” The Atlantic Monthly 283, January 1999, 1, 43–56. Levine, B.A. “The various workings of the Aramaic legal tradition: Jews and Nabataeans in the Nahal Hever Archive.” In L.H. Schiffmann, E. Tov and J.C. Vanderkam (eds). The Dead Sea Scrolls Fifty Years after their Discovery. Jerusalem: Israel Exploration Society, 2000. Lietzmann, H. and H. Hilgenfeld. Das Leben des heiligen Symeon Stylites. Leipzig: Hinrichs, 1908. Littmann, E. Safaitic inscriptions. Leiden: Brill, 1943. Lüling, G. Über den Ur-Qur’an: Ansätze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur’an. Erlangen: Lüling, 1974 (2nd edn: Erlangen: Lüling, 1993); Recently translated and expanded as A Challenge to Islam for Reformation. Delhi: Molital Banarsidass, 2003. —— Der christliche Kult an der vorislamischen Kaaba. Erlangen: Lüling, 1977. —— Die Wiederentdeckung des Propheten Muhammad: eine Kritik am “christlichen” Abendland. Erlangen: Lüling, 1981. —— “Preconditions for the scholarly criticism of the Koran and Islam, with some autobiographical remarks.” Journal of Higher Criticism 3, Spring 1996, 1, 73–109. —— “A new paradigm for the rise of Islam and its consequences for a new paradigm of the history of Israel.” Journal of Higher Criticism 7, Spring 2000, 1, 23–53. Luxenberg, C. Die syro-aramäische Lesart des Koran. Berlin: Das arabische Buch, 2000 (2nd edn: Berlin: Schiler, 2002) Trans. The Syro-Aramaic Reading of the Koran: A Contribution to the Decoding of the Koran. Berlin: Schiler, 2007. —— “Nöel dans le Coran.” Enquétes sur l’Islam. Paris: Editions Desclée de Brouwer, 2004, 117–38. —— “Weihnachten im Koran.” In C. Burgmer (ed.). Streit um den Koran. Berlin: Schiler, 2005, 35–41. —— “Die frühe Islamgeschichte nach inschriftlichen und numismatischen Zeugnissen.” In K.-H. Ohlig and G.-R. Puin (eds). Die dunklen Anfänge: Neue Forschungen zur Entstehung und frühen Geschichte des Islam. Berlin: Schiler, 2005, 124–47. Madigan, D.A. The Qur’an’s Self-Image. Writing and Authority in Islamic Scripture. Princeton, NJ: Princeton University Press, 2002. Margoliouth, D.S. “On the origin and import of the names muslim and hanif.” JRAS 1903, 467–93. —— “Textual variations of the Koran.” The Muslim World 15, 1925, 334–44. —— “Some additions to Jeffery’s Foreign Vocabulary.” JRAS, 1939, 53–61. Macdonald, M.C.A. “Languages, scripts and the uses of writing among the Nabataeans.” In G. Markoe (ed.). Petra Rediscovered: Lost city of the Nabataeans. New York: Abrams, 2003, 37–56. Maman, A. Comparative Semitic Philology in the Middle Ages from Sa’adiah Gaon to Ibn Barun (10th–12th C.). Leiden: Brill, 2004. Massignon, L. “Les ‘sept dormants’; apocalypse de l’islam.” Analecta Bollandiana 68, 1950, 245–60 (Reprint: Opera Minora. Y. Moubarac (ed.). Beirut: Dar al-Maaref, 1963, 3, 104–18). —— “Les sept dormants d’Ephèse (ahl al-kahf ) en islam et chrétienté.” Revue des Études Islamiques 12, 1954, 61–110. —— Le culte liturgique et populaire des VII dormants, martyrs d’Ephèse (ahl al-kahf): trait d’union orient-occident entre l’islam et la chrétienté. Studia Missionalia. 274 BIBLIOGRAPHY Rome: The Gregorian University, 1961 (Reprint: Opera Minora. Y. Moubarac (ed.). Beirut: Dar al-Maaref, 1963, 3, 119–80). McAuliffe, J.D. Qur’anic Christians: An Analysis of Classical and Modern Exegesis. Cambridge: Cambridge University Press, 1991. Melchert, C. “Ibn Mujahid and the establishment of seven Qur’anic readings.” SI 91, 2000, 5–22. Michaud, H. Jésus selon le Coran, Neuchatel: Delachaux et Niestlé, 1960. Millar, F. “Il ruolo delle lingue semitiche nel vicino oriente tardo-romano.” Mediterraneo Antico 1, 1998, 71–94. Mingana, A. Sources Syriaques. Leipzig: Harrassowitz, 1908. —— “The transmission of the Koran.” Muslim World 7, 1917, 223–32, 402–14 (Reprint: Ibn Warraq (ed.). The Origins of the Koran: Classic Essays on Islam’s Holy Book. Amherst, NY: Prometheus, 1998). —— “Syriac influence on the style of the Kur’an.” Bulletin of the John Rylands Library 11, January 1928, 1, 77–98 (Reprint: Ibn Warraq (ed.). What the Koran Really Says: Language, Text and Commentary. Amherst, NY: Prometheus, 2002, 171–92). Mingana, A. and A.S. Lewis. Leaves from Three Ancient Qurâns, Possibly Pre-’Othmanic, with a List of their Variants. Cambridge: Cambridge University Press, 1914. Motzki, H. “The collection of the Qur’an. A reconsideration of Western views in light of recent methodological developments.” Der Islam 78, 2001, 1–34. Mourad, S. “On the Qur’anic stories about Mary and Jesus.” Bulletin of the Royal Institute for Inter-Faith Studies 1, 1999, 2, 13–24. —— “From Hellenism to Christianity and Islam: The origin of the palm-tree story concerning Mary and Jesus in the Gospel of Pseudo-Matthew and the Qur’an,” OC 86, 2002, 206–16. Mouterde, P. “Inscriptions en Syriaque dialectal à Kamed (Beq’a).” MUSJ 22, 1939, 73–106. Muir, Sir W. The Life of Mohammad from Original Sources. T.H. Weir (ed.). Edinburgh: Grant, 1912. Müller, C.D.G. Kirche und Mission unter den Arabern in vorislamischer Zeit. Tübingen: Mohr, 1967. Müller, W. “Das Altarabische und das klassische Arabisch.” In W. Fischer (ed.). Grundriss der arabischen Philologie I. Wiesbaden: Reichert, 1982, 17–36. —— “Eine sabäische Gesandtschaft in Ktesiphon und Seleukia.” Neue Ephemeris für Semitische Epigraphik 2, 1974, 155–65. Muranyi, M. “Neue Materialien zur tafsir-Forschung in der Moscheebibliothek von Qairawan.” In S. Wild (ed.). The Qur’an as Text. Leiden: Brill, 1996, 225–55. Murray, R. “The origin of Aramaic ‘îr, angel.” Orientalia 53, 1984, 303–17. —— “Some themes and problems of early Syriac angelology.” V Symposium Syriacum 1988: Katholieke Universiteit, Leuven, 29–31 août 1988. Orientalia Christiana Analecta 236. Rome: Pontificium Institutum Studiorum Orientalium, 1990, 143–53. Nagel, T. Medinensische Einschübe in mekkanischen Suren. Göttingen: Vandenhoeck and Ruprecht, 1995. Nau, F. “Un colloque du patriarche Jean avec l’émir des Agaréens.” Journal asiatique 11, 1915, 225–79. —— Les Arabes chrétiens de Mésopotamie et de Syrie du VIIe au VIIe siècle. Paris: Imprimerie nationale, 1933. 275 BIBLIOGRAPHY Negev, A. Personal Names in the Nabatean Realm. Jerusalem: Hebrew University, 1991. Neuwirth, A. Studien zur Komposition der mekkanischen Suren. Berlin: de Gruyter, 1981. —— “Qur’anic literary structure revisited.” In S. Leder (ed.). Story-Telling in the Framework of Non-Fictional Arabic Literature. Wiesbaden: Harrassowitz, 1998, 388–420. —— “Qur’an and history – a disputed relationship. Some reflections on Qur’anic history and history in the Qur’an.” Journal of Qur’anic Studies 5, 2003, 1–18. —— “Zur Archäologie einer Heiligen Schrift: Überlegungen zum Koran vor seiner Kompilation.” In C. Burgmer (ed.). Streit um den Koran, Berlin: Schiler, 2004, 82–97. Nevo, Y. “Towards a pre-history of Islam.” JSAI 17, 1994, 108–41. Nevo, Y. and J. Koren, Crossroads to Islam: The Origins of the Arab Religion and the Arab State. Amherst, NY: Prometheus, 2003. Newby, G.D. The Making of the Last Prophet: A Reconstruction of the Earliest Biography of Muhammad. Columbia, SC: University of South Carolina Press, 1989. Nicholson, R.A. A Literary History of the Arabs. Cambridge: Cambridge University Press, 1969. Nöldeke, T. “Hatte Muhammad christliche Lehrer?” ZDMG 12, 1858, 699–708. —— “Zur Topographie und Geschichte des Damascenischen Gebietes und der Haurângegend.” ZDMG 29, 1875, 419–44. —— “Beiträge zur Geschichte des Alexanderromans.” Denkschriften der Kaiserlichen Akademie der Wissenschaften in Wien, philosophisch-historische Klasse 38, 1890, 5, 27–33. —— Sketches from Eastern History. London: Black, 1892. ——Neue Beiträge zur semitischen Sprachwissenschaft, Strassburg: Trübner, 1910. Nöldeke, T, F. Schwally, G. Bergsträsser and O. Pretzl. Geschichte des Qorans. 1st edn: Göttingen: Verlag der Dieterichschen Buchhandlung, 1860. Complete edition with F. Schwally, G. Bergsträsser and O. Pretzl (including Über den Ursprung des Qorans, Die Sammlung des Qorans, Die Geschichte des Koran-texts): Hildesheim: G. Olms, 1970. O’Shaughnessy, T.J. The Koranic Concept of the Word of God. Rome: Pontificio Istituto Biblico, 1984. Owen, J. “Id∫am al-Kabir and history of Arabic language.” In W. Arnold and H. Bobzin (eds). “Sprich doch Aramäisch, wir verstehen es!” 60 Beiträge zur Semitistik für Otto Jastrow zum 60. Geburtstag. Wiesbaden: Harrassowitz, 2002, 503–20. Paret, R. Mohammed und der Koran. Stuttgart: Kohlhammer, 1957. ——Der Koran, Kommentar und Konkordanz. Stuttgart: Kohlhammer 1971. Parrinder, G. Jesus in the Qur’an. New York: Barnes and Noble, 1965. Pfister, F. “Alexander der Große in Offenbarungen der Griechen, Juden, Mohammedaner und Christen.” Kleine Schriften zum Alexanderroman, Beiträge zur Klassischen Philologie 61, 1976, 301–37. Powers, D.S. “The Islamic law of inheritance reconsidered: A new reading of Q 4:12B.” SI 55, 1982, 61–94. —— Studies in Qur’an and Hadith: The Formation of the Islamic Law of Inheritance. Berkeley, CA: University of California Press, 1986. Pretzl, O. “Die Fortführung des Apparatus Criticus zum Koran.” Sitzungsberichte der Bayerischen Akademie der Wissenschaften 2, 1934. —— “Bericht über den Stand des Koranunternehmens der Bayerischen Akademie der Wissenschaften in München.” Mitgliederversammlung der DMG, Halle, 3. Januar 1936. ZDMG 89, 1935, 20–1. 276 BIBLIOGRAPHY —— “Die Wissenschaft der Koranlesung.” Islamica 6, 1935–6, 1–47; 230–46; 290–331. —— “Aufgaben und Ziele der Koranforschung.” Actes du XXe Congrès International des Orientalistes, Bruxelles 5–10 Septembre 1938. Louvain: Bureaux du muséon, 1940, 328–9 (Reprint: Der Koran. R. Paret (ed.). Wege der Forschung 326, 411–12). Puin, G.-R. “Observations on early Qur’an manuscripts in San‘a’.” In S. Wild (ed.). The Qur’an as Text. Leiden: Brill, 1996, 107–11. —— “Die korrumpierte Tradition? Ein Gespräch mit Gerd-Rüdiger Puin.” In C. Burgmer (ed.). Streit um den Koran. Berlin: Schiler, 2004, 51–63. —— “Über die Bedeutung der ältesten Koran-fragmente aus Sanaa (Jemen) für die Orthographie-geschichte des Korans.” In H.-C. Graf von Bothmer, K.-H. Ohlig and G.-R. Puin, “Neue Wege der Koranforschung.” Magazin Forschung 1, 1999, 33–46. Al-Qadi, W. “The impact of the Qur’an on early Arabic literature: The case of ‘Abd al-Hamid’s epistolography.” In A.A. Shareef and G.R. Hawting (eds). Approaches to the Qur’an. London: Routledge, 1993, 285–313. —— “The religious foundation of late Umayyad ideology and practice.” Saber Religioso y Poder Político en el Islam. Madrid: Agencia Española de Cooperacíon Internacional, 1994, 231–73. Rabin, C. Ancient West-Arabian, London: Taylor’s Foreign Press, 1951. —— Qumran Studies. New York: Oxford University Press, 1957. Räisänen, H. Das koranische Jesusbild: Ein Beitrag zur Theologie des Korans. Helsinki: Missiologian ja Ekumeniikan, 1971. —— Marcion, Muhammad and the Mahatma. London: SCM, 1997. al-Rashid, Sa‘d. Kitabat islamiyya min Makka al-Mukarrama. Riyad: Markaz al-Malik Faysal li-l-Buhuth wa-l-Dirasat al-Islamiyya, 1995. Reinink, G.J. “Die Entstehung der syrischen Alexanderlegende als politisch-religiöse Propagandaschrift für Herakleios’ Kirchenpolitik.” In C. Laga, J.A. Munitiz, and L. Van Rompay (eds). After Chalcedon. Studies in Theology and Church History offered to Professor Albert van Roey for his Seventieth Birthday. Orientalia Lovaniensia Analecta 18. Leuven: Peeters, 1984, 263–81. —— “Ps.-Methodius: A concept of history in response to Islam.” In A. Cameron and L.I. Conrad (eds). The Byzantine and Early Islamic Near East I: Problems in the Literary Source Material. Studies in Late Antiquity and Early Islam 1. Princeton, NJ: Darwin, 1992, 149–87. —— “Pseudo-Ephraems ‘Rede über das Ende’ und die syrische eschatologische Literatur des siebenten Jahrhunderts.” Aram 5, 1993, 437–63. —— “Alexandre et le dernier empereur du monde: les développements du concept de la royauté chrétienne dans les sources syriaques du septième siècle.” In L. Harf-Lancner, C. Kappler et F. Suard (eds). Alexandre le Grand dans les littératures occidentales et proche-orientales. Actes du colloque de Paris, 27–29 novembre 1997. Nanterre: Centre des sciences de la littérature de l’université Paris X, 1999, 149–59. —— “Heraclius, the new Alexander: Apocalyptic prophecies during the reign of Heraclius.” In G.J. Reinink and B.H. Stolte (eds). The Reign of Heraclius (610–641): Crisis and Confrontation. Groningen Studies in Cultural Change 2. Leuven: Peeters, 2002, 81–94. —— “Alexander the Great in the Seventh-Century Syriac ‘Apocalyptic’ Texts.” Byzantinorossika 2, 2003, 150–78. Results of Contemporary Research on the Qur’an: The Question of a Historico-Critical Text. M. Kropp (ed.). Beirut: Orient-Institut, 2006. 277 BIBLIOGRAPHY Retsö, J. The Arabs in Antiquity. London: Routledge, 2003. Revell, E.J. “The diacritical dots and the development of the Arabic alphabet.” JSS 20, 1975, 178–90. Reynolds, G.S. A Muslim Theologian in the Sectarian Milieu: ‘Abd al-Jabbar and the Critique of Christian Origins. Leiden: Brill, 2004. Rippin, A. “The Qur’an as literature: Perils, pitfalls and prospects.” British Society for Middle Eastern Studies Bulletin 10, 1983, 38–47 (Reprint: The Qur’an and Its Interpretive Tradition. Aldershot: Ashgate/Variorum, 2001). —— “Literary analysis of Qur’an, Tafsir, and Sira: The methodologies of John Wansbrough.” In R.C. Martin (ed.). Approaches to Islam in Religious Studies. Tucson: University of Arizona Press, 1985, 151–63 (Reprint: The Qur’an and Its Interpretive Tradition. Aldershot: Ashgate/Variorum, 2001). —— Muslims: Their Religious Beliefs and Practices. London: Routledge 1990. —— “Quranic studies, Part IV: Some methodological notes.” Method and Theory in the Study of Religion 9, 1997, 39–46 (Reprint: The Qur’an and Its Interpretive Tradition. Aldershot: Ashgate/Variorum, 2001). —— The Qur’an: Formative Interpretation. Aldershot: Ashgate/Variorum, 1999. —— The Qur’an: Style and Contents. Aldershot: Ashgate/Variorum, 2001. —— “The designation of ‘foreign’ languages in the exegesis of the Qur’an.” In J.D. McAuliffe, B.D. Walfish, J.W. Goering (eds). With reverence for the word. Medieval scriptural exegesis in Judaism, Christianity and Islam. Oxford: Oxford University Press, 2003, 437–44. Robin, C. “Les graffites arabes islamiques écrits en caractères sudarabiques de Umm Laylà.” Semitica 26, 1976, 188–93. —— “Les plus anciens monuments de la langue arabe.” In idem (ed.). L’Arabie antique de Karib’il à Mahomet. Aix-en-Provence: Edisud, 1991, 113–25. —— “Les inscriptions de l’Arabie antique et les études arabes.” Arabica 48, 2001, 545–56. Robin, C. and M. Gorea, “Un réexamen de l’inscription arabe préislamique du Æabal Usays.” Arabica 49, 2002, 503–10. Robinson, N. Christ in Islam and Christianity. Albany, NY: State University of New York Press, 1991. —— Discovering the Qur’an: A Contemporary Approach to a Veiled Text, London: SCM, 1996. Roncaglia, M.P. “Éléments Ébionites et Elkésaites dans le Coran.” Proche Orient Chrétien 21, 1971, 101–126. Rosenthal, F. “Some minor problems in the Qur’an.” The Joshua Starr Memorial Volume. New York: n.p., 1953. Rubin, U. “Hanifiyya and Ka’ba – an inquiry into the Arabian pre-Islamic background of Din Ibrahim.” JSAI 13, 1990, 85–112. ——The Eye of the Beholder. Princeton, NJ: Darwin, 1995. Rudolph, W. Die Abhängigkeit des Qorans von Judentum und Christentum. Stuttgart: Kohlhammer, 1922. Saadi, A.-M. “The letter of John of Sedreh: A new perspective on nascent Islam.” Karmo 1, 1998, 1, 18–31 (Arabic and Syriac); 1, 1999, 2, 46–64 (English). Samir, S.K. “La culture arabe chrétienne ancienne en interaction avec la pensée arabe musulmane.” Islamochristiana 8, 1982, 1–35 (in Arabic). —— “Vie et œuvre de Marc Ibn al-Qunbar.” Christianisme d’Égypte: Mélanges RenéGeorges Coquin. Cahiers de la Bibliothèque Copte 9. Louvain: Peeters, 1995, 123–58. 278 BIBLIOGRAPHY al-Samuk, S.M. Die historischen Überlieferungen nach Ibn Ishaq. Eine synoptische Rekonstruktion. Inaugural-Dissertation, Frankfurt am Main, 1978. Schall, A. “Die Sichtung des Christlichen im Koran.” Mitteilungen und Forschungsbeiträge der Cusanus-Gesellschaft 9, 1971, 76–91. Schapiro, I. Die haggadischen Elemente im erzählenden Teil des Korans. Leipzig: Gustav Fock, 1907. Schedl, C. Muhammad und Jesus. Vienna: Herder, 1978. —— “Die 114 Suren des Koran und die 114 Logien Jesu im Thomas-Evangelium.” Der Islam 64, 1987, 261–4. Schlatter, A. “Die Entwicklung des jüdischen Christentums zum Islam.” Evangelisches Missionsmagazin 62, 1918, 251–64. Schneider, G. Apocryphe Kindheitsevangelien. Freiburg: Herder, 1995. Schoeler, G. Charakter und Authentie der muslimischen Überlieferungen über das Leben Muhammads. Berlin: De Gruyter, 1996. —— Écrire et transmettre dans les débuts de l’islam. Paris: Presses Universitaires de France, 2002. —— “Foundations of a new biography of Muhammad: The production and evaluation of the corpus of traditions from ‘Urwah b. al-Zubayr.” In H. Berg (ed.). Method and Theory in the Study of Islamic Origins. Leiden: Brill, 2003, 21–8. Schoeps, H.J. Theologie und Geschichte des Judenchristentums. Tübingen: Mohr, 1949. Schröter, R. “Trostschreiben Jacobs von Sarug an die himjaritischen Christen.” ZDMG 31, 1877, 361–405. Segal, J.B. “Arabs in Syriac literature before the rise of Islam.” JSAI 4, 1984, 89–124. Sellheim, R. “Muhammeds erstes Offenbarungserlebnis.” JSAI 10, 1987, 1–16. Semaan, K.I.H. “A linguistic view of the development of the Arabic writing system.” Wiener Zeitschrift für die Kunde des Morgenlandes 61, 1967, 22–40. Sezgin, G. Geschichte des arabischen Schrifttums. Leiden: Brill, 1967–2000. Sfar, M. Le Coran, la Bible et l’Orient ancien. Paris: Sfar, 1998. —— Le Coran est-il authentique? Paris: Sfar, 2000. Shahid, I. The Martyrs of Najrân: New Documents. Subsidia Hagiographica 49. Bruxelles: Société des Bollandistes, 1971. —— “Byzantium and the Arabs during the reign of Constantine: The Nemara inscription.” Byzantinische Forschungen 26, 2000, 73–124. —— Byzantium and the Arabs. Washington, DC: Dumbarton Oaks, 1984–2002. Sidersky, D. Les origines des légendes musulmanes dans le Coran et dans les vies des prophètes. Paris: Geuthner, 1933. Speyer, H. Die biblischen Erzählungen im Qoran. Gräfenhainichen: Schulze, 1931. (Reprint: Hildesheim: Olms, 1961). Spitaler, A. “Die Nichtkanonischen Koranlesarten und ihre Bedeutung für die arabische Sprachwissenschaft.” Actes du XXe Congrès International des Orientalistes, Bruxelles 5–10 Septembre 1938. Louvain: Bureaux du muséon, 1940, 314–5. (Reprint: Der Koran. R. Paret (ed.). Wege der Forschung 326). —— “Otto Pretzl.” ZDMG 96, 1942, 163–4. —— “Die Schreibung des Typus im Koran. Ein Beitrag zur Erklärung der koranische Orthographie.” Wiener Zeitschrift für die Kunde des Morgenlandes 56, 1960, 212–26. Sprenger, A. “Ueber eine Handschrift des ersten Bandes des Kitàb Tabaqàt al-kabyr vom Sekretär des Wàqidy.” ZDMG 3, 1849, 450–6. 279 BIBLIOGRAPHY Sprenger, A. “Foreign words occurring in the Qôran.” Journal of the Asiatic Society of Bengal 21, 1852, 109–14. —— “Mohammad’s Journey to Syria and Professor Fleischer’s Opinion Thereon” (with texts and translations from al-Tirmidhi and Ibn Ishâq). Journal of the Asiatic Society of Bengal 21, 1852, 576–92. —— “Mohammad’s Zusammenkunft mit dem Einsiedler Bahyrâ.” ZDMG 12, 1858, 238–49. —— “Ueber den Ursprung und die Bedeutung des arabischen Wortes Nâmûs.” ZDMG 13, 1859, 690–701. —— Das Leben und die Lehre des Mohammad. 2nd edn, Berlin: Nicolai, 1869. Streit um den Koran. Die Luxenberg-Debatte: Standpunkte und Hintergründe. C. Burgmer (ed.). Berlin: Schiler, 2004. Stülpnagel, J. von. ‘Urwa Ibn az-Zubair. Sein Leben und seine Bedeutung als Quelle frühislamischer Überlieferung. Dissertation, Tübingen, 1957. Al-Tahir, Salwa Mursi. Bidayat al-kitaba al-ta’rikhiyya ‘inda l-’Arab. Awwal sira fi l-islam: ‘Urwa b. al-Zubayr b. al-‘Awwam. Beirut: al-Mu’assasa al-‘Arabiyya li-l-Dirasat wa-l-Nashr, 1995. Tardy, R. Najrân: Chrétiens d’Arabie avant l’islam. Beirut: Dar al-Mashriq, 1999. al-Theeb, Suleiman. Nuqush Jabal Umm Jadhayidh al-Nabatiyya. Riyad: Maktabat al-Malik Fahd al-Wataniyya, 2002. Tisdall, W.St. Clair. The Original Sources of the Qur’an. New York: Gorham, 1905. Torrey, C.C. “Three difficult passages in the Koran.” In T.W. Arnold and R.A. Nicholson (eds). A Volume of Oriental Studies Presented to Edward G. Browne ( . . . ) on his 60th Birthday. Cambridge: Cambridge University Press, 1922, 457–9. —— The Jewish Foundation of Islam. New York: Jewish Institute of Religion, 1933. (2nd edn: New York: Ktav Publishing House, 1967). Trimingham, J.S. Christianity among the Arabs in Pre-Islamic Times. London: Longman, 1979. Usener, H. “Weihnachtspredigt des Sophronios.” Rheinisches Museum für Philologie 41, 1886, 500–16. al-‘Ushsh, Muhammad Abu l-Faraj. “Kitabat ‘arabiyya ghayr manshura fi jabal Usays.” Al-Abhath 17, 1964, 227–316. van Bladel, K. “The Syriac sources of the early Arabic narratives of Alexander.” In H.P. Ray (ed.). Memory as History: The Legacy of Alexander in South Asia. New Delhi: Aryan International, 2007, 54–75. Van Reeth, J.M.F. “L’Évangile du Prophète.” Acta Orientalia Belgica 2004, 156–74. —— “Le vignoble du paradis et le chemin qui y mène. La thèse de C. Luxenberg et les sources du Coran.” Arabica 50, 2006, 4. Versteegh, C.H.M. Arabic Grammar and Qur’anic Exegesis in Early Islam. Leiden: Brill, 1993. —— The Arabic Language. Edinburgh: Edinburgh University Press, 1997 (2nd edn 1997). Violet, B. “Ein zweisprachiges Psalmfragment aus Damascus.” Orientalische Literaturzeitung 4, 1901, 384–403, 425–441, 475–88. Vollers, K. Volksprache und Schriftsprache im alten Arabien. Strassburg: Trübner, 1906. Vööbus, A. Handschriftliche Überlieferung der Memre-Dichtung des Ja’qob von Serug. CSCO 344 and 345, 421 and 422. Louvain: Secrétariat du CSCO, 1973 and 1980. Wallace-Hadrill, D.S. Christian Antioch: A Study of Early Christian Thought in the East. Cambridge: Cambridge University Press, 1982. 280 BIBLIOGRAPHY Wansbrough, J. Qur’anic Studies: Sources and Methods of Scriptural Interpretation. Oxford: Oxford University Press, 1977 (2nd edn: Amherst, NY: Prometheus, 2004). —— The Sectarian Milieu: Content and Composition of Islamic Salvation History. Oxford: Oxford University Press, 1978. Watt, W.M. Muhammad at Mecca. Oxford: Oxford University Press, 1953. —— Muhammad at Medina. Oxford: Oxford University Press, 1956. —— Muhammad. Prophet and Statesman. Oxford: Oxford University Press, 1961. —— Bell’s Introduction to the Qur’an. Edinburgh: Edinburgh University Press, 1970. —— Muhammad’s Mecca. History in the Qur’an. Edinburgh: Edinburgh University Press, 1988. Weil, G. Einleitung in den Koran. Bielefeld: Belhagen und Klasing, 1844. Wheeler, B. “Moses or Alexander? Q 18:60–65 in early Islamic exegesis.” JNES 57.3, 1998, 191–215. —— Moses in the Quran and Islamic Exegesis. London and New York: Routledge Curzon, 2002. Witakowski, W. The Syriac Chronicle of Pseudo-Dionysius of Tel-Mahre: A Study in the History of Historiography. Studia Semitica Upsaliensia 9. Stockholm: Almqvist and Wiksell, 1987. Yardeni, A. “Notes on two unpublished Nabataean deeds from Nahal Hever – P. Yadin 2 and 3.” In L.H. Schiffmann, E. Tov and J.C. Vanderkam (eds). The Dead Sea Scrolls Fifty Years after their Discovery. Jerusalem: Israel Exploration Society, 2000, 836–51, 862–74. Zammit, M.R. A Comparative Lexical Study of Qur’anic Arabic. Leiden: Brill, 2002. Zwettler, M. The Oral Tradition of Classical Arabic Poetry. Columbus, OH: Ohio State University Press, 1978. —— “Imra’alqays, son of ‘Amr, king of . . . ?” In M. Mir (ed.). The Literary Heritage of Classical Islam. Princeton, NJ: Darwin, 1993, 3–37. 281 INDEX OF BIBLICAL VERSES Hebrew Bible/Old Testament Genesis 3:15 143 5:24 162 (n. 27) 7:11 200 (n. 28) 25:2 222 (n. 25) 25:13 254 Exodus 15:20 143 24:8 162 (n. 17) Deuteronomy 7:5 209 32:17 209 Joshua 5:15 236 1 Chronicles 5:29 143 2 Chronicles 23:17 209 24:18 209 Job 15:29 255 Psalms 37:29 87 (n. 60) 81:9 209 Isaiah 6:3 92 10:11 215 (n. 17) 53:11–12 150 Jeremiah 1:14 180 31:31 162 (n. 18) Ezekiel 38–9 187 Daniel 3:8 255 4:13 136 (n. 82) 17:23 136 (n. 82) New Testament Matthew 1:1 165 1:6–11 164 1:20 165 2:13–18 169 6:11 150 12:23 99 20:1–15 143, 144 20:28 162 (n. 20) 24 188, 192 24:36 162 (n. 29) 25 158 26:28 150 Mark 4:26–7 10:45 10:51 13:32 14:24 99 162 (n. 20) 10 162 (n. 29) 150 Luke 1–2 166 11:3 150 283 INDEX OF BIBLICAL VERSES 21:5–8 188, 192 21:20 188, 192 21:25 192, 193, 194, 203 (n. 67) 22:20 162 (n. 19) John 1:1 156 6:51 150, 161 (n. 13) 15:23–16:1 87 (n. 60) 15:25–6 104 (n. 33) 20:7–9 199 (n. 20) 20:16 10 Acts of the A postles 10 210 10:9 207 10:9–13 149 19:28–35 135 (n. 57) Romans 5:12–21 147 5:15 162 (n. 20) 5:19 162 (n. 20) 12:4 162 (n. 20) 1 Corinthians 10:17 162 (n. 20) 11:25 162 (n. 19) 11:27–9 151 12:2 162 (n. 20) 12:14 162 (n. 20) 12:20 162 (n. 20) 15:22 147 15:45–9 161 (n. 11) Hebrews 9:28 162 (n. 20) 11:5 162 (n. 27) 1 Peter 5:8–9 98 1 John 2:2 150 Revelation 20 187 20:7–9 199 (n. 20) 284 INDEX OF QUR’ANIC VERSES surat al-baqara (2) 55 256 58 238 62 81 63 226, 236 93 226, 236 111 81 113 81 120 81 125 238 135 81, 238, 239 136 85 (n. 24) 140 81, 85 (n. 24) 166 200 (n. 25) 256 215 (n. 20) 257 215 (n. 20) surat al ‘Imran (3) 19 50 (n. 61) 33 166 33–4 142 35–6 164 35–48 99 37 99 42–9 166 47 147 49 145, 146 54 155 55 153 59 146 67 81, 238, 239 84 85 (n. 24) 95 238, 239 surat al-nisa’ (4) 22 237 38 238 51 209, 253 69 238 76 215 (n. 20) 82 9 97 238 115 238 125 238, 239 131 231 154 226, 238, 256 156–9 152 158 153 159 153, 162 (n. 30), 236 163 85 (n. 24) 171–2 155, 156 surat al-ma’ida (5) 18 81 47 81 51 81 64 81 69 81 72 132 (n. 15) 73 115 82 81 110 146 112 206 112–15 148, 207, 210, 211, 215 (n. 21) 114 206 116 132 (n. 14) 117 153 surat al-an‘am (6) 57 232 79 238, 239 161 238, 239 surat al-a‘raf (7) 40 200 (n. 28) 132 256 285 INDEX OF QUR’ ANIC VERSES 160 161 180 227 238 242 surat al-anfal (8) 30 155 surat al-tawba (9) 30 81 72 49 (n. 36) 100 256 surat Yunus (10) 105 238, 239 surat Hud (11) 24 237 40 253 surat Yusuf (12) 2 63 20 236 23 256 100 238 surat al-ra‘d (13) 31 105 (n. 54), 231 37 63 surat Ibrahim (14) 4 63, 94 surat al-hijr (15) 43 81 surat al-nahl (16) 36 215 (n. 20) 48 238 103 15, 48 (n. 34), 63, 89, 90, 94, 95, 103 (ns. 20, 21), 132 (n. 24), 242 120 238, 239 123 238 surat al-isra’ (17) 1 163 23 231 32 237 107 238 surat al-kahf (18) 4 117, 118 5 238 9 117, 118, 125, 126, 241 9–26 117 10 118, 119 10–12 119, 127 10–20 118, 127 11 119 12 119 13–20 119 21–6 118, 119, 120, 128 25 227 26 117, 118 27 117, 118 31 253 60–5 197 (n. 8) 60–82 176, 177 83 182, 191 83–4 179 83–102 175, 176, 177, 178, 189, 194, 195, 196, 197 (n. 8), 203 (n. 78) 84 182 85–91 179 89 182 90 181 92 182 92–8 180 98 193 99 181, 193 99–102 180 107 253 surat Maryam (19) 2 199 (n. 23) 3 99 17–21 166 22–6 99 24 256, 257 27–8 152 28 143, 164 30 192 52 226, 236 57 153 58 238 97 63 surat ta ha (20) 1 234 17 146 70 238 80 226, 236 99 199 (n. 23) 101 238 286 INDEX OF QUR’ ANIC VERSES surat al-anbiya’ (21) 7 81 19 146 98 234 105 87 (n. 60), 105 (n. 46) surat al-hajj (22) 15 199 (n. 25) 17 81 26 238 30–1 239 31 238 41 87 (n. 56), 256 surat al-mu’minun (23) 11 253 20 226, 236 27 253 50 169 60 230 100 253 surat al-nur (24) 27 230 surat al-furqan (25) 4 103 (ns. 21, 22) 4–5 89, 90, 102 (n. 16) 5 103 (n. 24) 53 253 63 256 64 238 surat al-shu‘ara’ (26) 195 15, 48 (n. 34), 63, 94, 102 (n. 16) 196 102 (n. 15) surat al-qasas (28) 29 226, 236 46 226, 236 surat al-rum (30) 1–6 191 30 238, 239 surat Luqman (31) 34 157 surat al-sajda (32) 15 238 surat ya sin (36) 56 243 82 147 surat al-saffat (37) 78 244 103 17 108 244 119 244 129 244 surat sad (38) 2 256 10 199 (n. 25) surat al-zumar (39) 17 215 (n. 20) 29 237 surat ghafir (40) 20 232 35 238 37 199 (n. 25) surat ha mim (41) 10 49 (n. 43) 44 132 (n. 24) 47 158, 241 surat al-zukhruf (43) 61 157, 158 70 243 85 158 surat al-dukhan (44) 3 78 15 103 (n. 21) 23 235 53 253 54 100, 112, 113, 244 58 63 surat al-ahqaf (46) 12 63 surat al-fath (48) 29 99, 238 surat qaf (50) 10 233 surat al-tur (52) 1 226, 236 20 100, 112, 244 24 100 287 INDEX OF QUR’ ANIC VERSES surat al-najm (53) 19–20 84 (n. 8) 36–7 82 62 78 surat al-qamar (54) 11 200 (n. 28) surat al-rahman (55) 20 253 54 253 72 xii, 112 surat al-waqi‘a (56) 17 100 22 112 28–31 233 surat al-saff (61) 3 238 6 159 surat al-tahrim (66) 12 143, 164 surat al-qalam (68) 13 49 (n. 41) surat al-jinn (72) 15 235 surat al-insan (76) 17 253 19 21 100 253 surat al-nazi‘at (79) 16 236 surat al-mutaffifin (83) 9 241 20 241 25 253 surat al-a‘la (87) 19 82 surat al-ghashiyya (88) 15 253 surat al-tin (95) 2 226, 236 surat al-‘alaq (96) 1 147 1–5 9 19 78 surat al-qadr (or qadar, 97) 1 78 surat al-bayyina (98) 5 239 surat al-kawthar (108) 1 98 2 78 288 INDEX OF PEOPLE, PLACES AND SUBJECTS Aaron, priest, prophet 143, 145, 152, 164, 165, 166, 172, 236, 244 ‘Abdallah b. Ja‘far 93 ‘Abd al-Malik 14, 34, 35, 79, 100 Abel 236 Abraha 206 Abraham 7, 17, 22 (n. 38), 37–8, 76, 78, 81, 82, 86 (n. 54), 142–4, 161 (n. 7), 165, 214 (n. 7), 220, 238–40, 244 Abu Bakr, caliph 91 Abu Bakr al-Asamm 98 Abyssinia 112, 114, 213 (n. 3) Adam 52, 80, 142–4, 146–8, 155, 165 ‘Addas 90, 91 ahl al-kitab 81, 153, 156 Aksum 205, 211 Aleppo 49 (n. 50), 55, 59 Alexander the Great 19, 51, 175–203 Alexander Legend 175–7, 180–91, 193–6, 197 (ns. 4, 6, 8), 198 (n. 8), 199 (ns. 12, 20), 200 (n. 28), 201 (n. 30), 203 (n. 78) Alexander Romance 175, 177, 203 (n. 78) Alexandria 170, 206 ‘Ali b. Abi Talib 171, 232, 233 Ammon, god 119 (n. 18) Anatolia 183, 184, 186 Angels 10, 17, 118, 154, 156, 159, 200 (n. 28), 211, 236 Anne, mother of Mary 142, 164 Antioch 59, 137 (n. 93) Apocalyptic 176, 183, 185, 188, 191, 195, 203 (n. 64) Apocrypha (Biblical) 70, 83, 85 (n. 39), 86 (n. 52), 87 (n. 60), 112, 115, 141, 142, 146, 152, 154, 159, 172; Gospel of the Hebrews 80; Proto-Evangelium of James 165–6, 171–2; Pseudo-Matthew 166–8, 171, 174 Arabian Peninsula 70, 76, 77, 205, 206 Arabic language 8, 12–13, 15, 17, 34, 44, 49 (ns. 47, 50), 51, 54–7, 65–6 (n. 5), 66 (n. 12), 70–1, 76, 78, 80, 82–3, 89–90, 92, 94–7, 100–1, 109–11, 130–1, 133 (n. 25), 196, 204–5, 214 (ns. 7, 8), 227–8, 230, 249–54, 255, 257–9; Arabic-Aramaic (or Syriac) “mixedlanguage” 15, 38–40, 78–9, 115–16, 227; Judaeo-Arabic 256; poetic koinè 13, 36–8, 95; script 10, 16, 45, 57–65, 66 (n. 7), 68 (n. 42), 73–5, 100, 254 Aramaic language (cf. Syriac language) 10, 12, 15, 23 (n. 48), 24 (ns. 75, 76), 36–41, 51–3, 59, 70, 78, 85 (n. 39), 90, 92–3, 94, 96, 99–100, 102 (n. 17), 104 (n. 33), 110, 115, 128, 136 (n. 69), 194, 196, 204–5, 212–13, 214 (n. 7), 215 (n. 19), 227, 250, 253–8; script 53, 57, 60, 114, 125 Armenia 162 (n. 33), 184, 187, 199 (n. 14) Armenian language 59, 190, 196, 226 asatir al-awwalin 89, 102 (n. 15) Astrology 52 Augustine 162 (n. 32) Bab al-Lan (Darial Pass) 186 Babylon 51, 180 Badr, battle 72, 93 Baghdad 45 Bahira (Sergius), monk 82, 87 (n. 56), 91, 102 (n. 14), 103 (n. 26) al-Baladhuri 92 al-Balkhi, Abu l-Qasim 93 Banu Kinda 53, 63 Bar Hebraeus (Abu l-Faraj) 52, 121, 122 Basilides 154 Basra 42 289 INDEX OF PEOPLE, PLACES AND SUBJECTS al-Bayhaqi 91 Beck, E. 30, 31, 50 (n. 52), 113, 132 (n. 21), 244 Bergsträsser, G. 3–5, 6, 9, 19, 20 (ns. 10, 11, 15, 17), 21 (ns. 18, 22, 23, 27), 30, 31, 45, 50 (n. 52), 89, 101 Bethlehem 170, 171, 219 Bible 59, 78, 101, 142, 144, 145–6, 225, 227, 250, 255; and Aramaic 59, 136 (n. 82); Diatessaron 81, 99, 112, 166, 171, 172, 173 (n. 15), 226; Ethiopic Bible 149, 204–13, 215 (n. 10), 216 (n. 23); New Testament 23 (n. 48), 78, 87 (n. 60), 143, 145, 150–1, 156, 158, 160, 227; Old Testament/Hebrew Bible xiii, 88, 143, 257; Peshitta 78; translation into Arabic 15, 70 Blachère, R. 30, 36, 100, 161 (n. 9) Bostra (Busra) 55, 60, 87 (n. 56) Böwering, G. 18, 205 Butrus al-Bayt Ra’si 143–4, 155 Byzantine Empire 10, 13–14, 60, 63, 90, 102 (n. 17), 160, 184–5, 188–9, 190, 191, 193, 218–19 Cain 236 Cairo 3, 4 Callimachus 168 Caspian Sea 186, 192–3, 201 (n. 40) Caucasus 179, 184, 186, 187, 188, 192, 201 (n. 40) Christ, v. Jesus Christians 10, 12, 14, 15, 33, 36, 39, 52, 59–60, 63, 77, 80, 81, 86 (n. 51), 102 (n. 17), 110, 112–13, 115, 117, 118, 120, 121–2, 127, 128, 131, 143, 145–6, 149, 150–1, 152, 155, 156, 160, 165–6, 169, 171–2, 192, 210, 217–20, 235, 254, 259; East Syrian (“Nestorian”) Christians 13, 32, 59, 110, 112, 121, 204, 217–18; Jacobites (“monophysites”) 13, 32, 110, 112, 121–2, 130, 137 (n. 93), 184, 185, 188, 190–1, 202 (n. 61), 204, 217–18, 219; Jewish-Christians 33, 48 (n. 19), 77, 79–81; Melkites 112, 217, 220 (n. 2) Circumcision 143 Companions of the Cave 1, 109–37, 241 Constantinople 184 Cook, M. 25 (ns. 90, 91), 30, 32, 88 Coptic language 59, 75, 77, 114, 151, 249, 257–8 Crone, P. 25 (n. 91), 30, 32, 88 Crucifixion, v. Jesus, Crucifixion Damascus 37, 50 (n. 51), 52, 53, 55, 60, 68 (n. 35), 170, 171 Darband 186–7, 192–3, 201 (ns. 40, 42) David, king, prophet 164–5 Decius, emperor 123–4, 126, 129, 130–1, 135 (n. 63), 240–1 Devil 71, 78, 80, 84 (n. 8), 98, 155, 164, 199 (n. 20), 207 Dhu l-Qarnayn, v. Alexander the Great Diatessaron (Harmony of the Gospels), v. Bible, Diatessaron Doctrina Jacobi 14 Dome of the Rock 14, 43, 188, 220 Donner, F. 18, 50 (n. 60), 203 (n. 64) Dumat al-Jandal 60 East Syrian Christians, v. Christians, East Syrian Christians Ebionites 33, 79, 80, 151 Edessa 51, 52, 59, 121, 184–5, 199 (n. 13) Egypt 2, 4, 20 (n. 2), 168–72, 179, 180, 184 Elijah of Nisibis 147 Elkesaites 99 Elusa 54, 57, 63 Enoch 162 (n. 27), 200 (n. 28); Book of 200 (n. 28) Ephraem 109, 112–13, 132 (ns. 18, 20, 21), 188, 220 Epiphanius 53, 57, 63, 162 (n. 32) Eschatology, v. Apocalyptic, Heaven, Hell Ethiopia 32, 77, 94, 149, 168, 204–6, 212–13, 214 (n. 3), 216 (n. 23) Ethiopic language 70, 75, 106 (n. 62), 151, 206–12, 256 Eucharist 23 (n. 48), 78, 80, 148–51, 211 Euphrates River 170, 179 Eusebius of Caesarea 162 (ns. 32, 33) Eutychius of Alexandria 130, 143 al-Farra’ 241 Fischer, A. 3, 4, 100 Fredegarius 186, 187, 191, 193–4 Gabriel, angel 9, 10, 91 Gärima, saint 211 Geiger, A. 16, 30, 32 Ghassanids 58, 60, 67 (n. 22), 68 (n. 30), 111, 112, 121, 190, 202 (n. 61) Gilgamesh 176, 197 (n. 6), 198 (n. 12) Gilliot, C. 18, 80, 82, 227 Gnosticism 154–5, 156, 160 290 INDEX OF PEOPLE, PLACES AND SUBJECTS Gog and Magog 179, 181, 186–8, 193–4, 197 (n. 4), 199 (n. 20), 201–2 (n. 47), 236 Goliath (Jalut) 236 Greek language 49 (n. 50), 52, 55–6, 57, 58, 59, 66 (n. 14), 67 (n. 18), 83 (n. 2), 90, 91, 102 (n. 17), 104 (n. 33), 145, 176, 186, 190, 196, 200 (n. 28), 201 (n. 42), 206, 207, 208–9, 212, 215 (n. 15), 226–7, 228, 235, 249, 258 Gregory of Nyssa 143 Gregory of Tours 120 Griffith, S. 15, 18, 83 (n. 2), 241 Hadith 2, 20 (n. 7), 42, 81, 166, 170, 171, 251, 259 Hagar 218, 220 hajj 35 al-Hajjaj b. Yusuf 3, 35, 48 (n. 29), 100 Haloras River 180, 199 (ns. 13, 14) hanif 76, 81–2, 86 (n. 54), 238–40 Harran 49 (n. 50), 52, 55, 56, 57, 60 Harut and Marut, angels 236 al-Hasan al-Basri 90, 103 (ns. 21, 22) Heaven 16, 78, 113, 123, 128, 136 (n. 79), 148, 154, 161 (n. 13), 179, 180, 182, 198 (n. 12), 199 (n. 25), 200 (n. 28), 210, 212, 243; and virgins, v. Hur Hebrew language 12, 24 (n. 75), 37, 70, 76, 78, 80, 92–4, 143, 145, 149, 200 (n. 28), 209, 215 (n. 15), 226, 235–6, 249–50, 255–6, 257; script 93, 241 Hegra (Mada’in Salih) 53, 54, 57 Hell 10, 234–5 Heraclius 13, 160, 184–5, 186, 187–8, 189, 190–1, 193, 195, 202 (n. 61) Heresy 13, 153, 160, 162 (n. 32), 163, 172, 218, 219, 220 hijra 35, 71, 72, 213, 214 (n. 3), 217 Himyarites 67 (n. 27), 121 Hippolytus 162 (n. 32) Hira’, Mt. 9 al-Hira 58, 60, 77, 100, 111, 170 Hisham b. ‘Abd al-Malik, caliph 14 Hopkins, S. 97, 227 Hoyland, R. 18 Hur (houris, virgins of paradise): xii, 16, 78, 100, 112, 242–3 Ibn Abi l-Zinad 93 Ibn ‘Asakir 165, 170–1 Ibn Hisham 14, 91, 159 Ibn Ishaq 14, 87 (n. 60), 91, 104 (ns. 31, 33, 34), 130, 164–5, 169 Ibn Kathir 170 Ibn Mas‘ud 93 Ibn Mujahid 2, 20 (n. 7) Idris 153, 162 (n. 27) ijma‘ 14 ‘Imran 142–3, 145, 163–4, 166 Imru’ al-Qays, king 53, 58 Imru’ al-Qays, poet 145 Injil (Revelation of Jesus; Gospel) 81–2, 99 Iran (Persia) 114, 179–80, 184–8 Irenaeus of Lyons 154–5 Ishmael, v. Isma‘il Ishmaelites 218–19 Isma‘il 76, 85 (n. 24), 220, 244, 254 Israelites (banu Isra’il) 81, 142, 144, 145, 165 Jacob, son of Isaac 164, 165 Jacobites, v. Christians, Jacobites Jacob of Serugh 59, 109, 120–4, 126, 127–30, 134 (n. 42), 135 (ns. 51, 53, 57, 62), 136 (n. 72), 137 (n. 92), 176, 188, 198 (n. 8) Ja‘far al-Sadiq 171 al-Jahiz 171 Jalut, v. Goliath Jeffery, A. 3–6, 7, 15, 16, 19, 21 (ns. 22, 23), 24 (n. 76), 30, 89, 110, 145, 204, 241, 257, 259 Jerome, saint 53, 57, 150, 151 Jerusalem 76, 135 (n. 51), 160, 170, 171, 179, 180, 184–5, 186, 188, 190, 192, 219 Jesus 10, 59, 80, 81, 91, 99, 112, 118, 141–3, 145–9, 152–9, 162 (n. 28), 163–72, 188, 192–4; birth/Christmas 17, 78, 85 (n. 31); creating bird from clay 146, 161 (n. 9); crucifixion 99, 152–5; form of the word ‘isa 235–6; and table from heaven (ma’ida) 148–51, 204, 206–7, 210–11 Jews 12, 32, 33, 36, 39, 51, 81, 92, 102 (n. 17), 118, 142, 145, 152–4, 160, 218, 220, 221 (n. 21) jihad 16, 76 Joachim (Yuwakim), father of Mary 142, 165 291 INDEX OF PEOPLE, PLACES AND SUBJECTS John Chrysostom 143 John of Aksum 211 John of Ephesus 121, 122, 129, 136 (ns. 72, 90), 137 (n. 92) John of Sedreh 219 John the Baptist (Yahya) 141, 142, 149, 159 Joseph, husband of Mary 164–5, 167 Joseph, son of Jacob 164 Josephus, Flavius 181, 187 Julian, emperor 99 Ka‘ba 76, 220 al-Ka‘bi, v. al-Balkhi, Abu l-Qasim Karbala’ 171 Khadija 80, 90–1, 101, 103 (ns. 26, 28) Khazars 187, 193, 202 (n. 48) Khosro II, Persian king 160, 184, 192 Korah (Qarun) 236 Kropp, M. 19, 20 (n. 1), 22 (n. 38), 89 Lakhmids 58, 59, 111, 121 Latin language 12, 39, 92, 99, 120, 150, 169, 186, 193, 207, 226–7, 234, 258 Law, v. shari‘a Lüling, G. xiii, 5, 6, 8, 9–11, 14–15, 18, 21 (ns. 19, 26), 23 (ns. 50, 52), 30–4, 48 (n. 19), 74–7, 80, 81, 96–7, 99, 101 Luxenberg, C. xi–xii, 8, 9, 15–17, 18, 23 (n. 48), 24 (n. 74), 25 (n. 87), 30–1, 33–4, 37–40, 49 (ns. 41, 46), 50 (n. 51), 51–3, 63, 74, 77–81, 85 (ns. 28, 31), 88–9, 97–101, 110, 113–14, 125, 136 (n. 69), 225–8, 237–45, 249, 254, 259–60 ma’ida, v. Jesus, and table from heaven Madaba 57, 63 Mada’in Salih, v. Hegra Madigan, D. 34 al-Mahdi, caliph 159 mahdi (messiah) 158 Mandaeans 81, 255 Mani 99 Manicheans 81, 99, 250 Marcion 99 Marqus b. al-Qunbar 161 (n. 7) Mary, mother of Jesus 19, 91, 99, 112, 141–3, 145, 146, 148, 152, 156, 159, 161 (n. 2), 163–72, 173 (ns. 2, 15), 210 Mecca 10–11, 12, 18, 33, 36, 37, 49 (n. 35), 64, 65, 67 (n. 27), 70, 71, 72, 75–82, 89, 90, 93, 95, 100, 113, 118, 141, 175, 184, 189, 190–1, 227, 259 Medina 70, 71, 72, 73, 75, 80–1, 82, 92, 93, 141, 190–1, 220, 255 Mediterranean Sea 205, 207, 212 Meletius 162 (n. 33) Melkites, v. Christians, Melkites Mesopotamia (al-Jazira) 23 (n. 49), 32, 52, 114, 184, 186–7, 190, 198 (n. 12), 199 (n. 14), 254–5 mhaggrayê 217–20 Michael the Syrian 88, 121, 221 (n. 21) Midrash 70, 83, 86 (n. 52), 260 Mingana, A. 227, 249–50, 253–4 Miriam, sister of Moses 143, 145, 164, 166, 173 (n. 2) Monophysite Christians, v. Christians, Jacobites Moses 80, 82, 91, 142–5, 147, 164, 166, 176, 199 (n. 25), 235–7, 244 Mourad, S. 19 Movses Dasxuranc‘i 187–8, 192, 194, 202 (n. 51) Muhammad, prophet 9–11, 14, 20 (n. 4), 31–3, 40–3, 45, 63, 70, 71–83, 83 (n. 6), 84 (ns. 8, 13), 87 (ns. 56, 60), 88–95, 97–101, 103 (ns. 27, 28), 110–12, 114, 117–20, 125–6, 129–31, 132 (n. 24), 135 (n. 50), 141–2, 144, 152–3, 156, 161, 161 (n. 7), 163–4, 166, 170, 172, 175–6, 182, 184, 189–91, 194–6, 204–6, 210, 213, 213 (n. 3), 214 (n. 3), 218, 220, 221 (n. 21), 226, 252; the sira 9, 11, 14, 17, 24 (n. 58), 84 (n. 13), 87 (n. 60), 91, 93, 130, 159 Muhammad b. Salih 93 Muqatil b. Sulayman 90, 170 Muslims xii, 2, 10, 12, 19, 33, 70, 71, 77, 86 (n. 54), 110, 114, 130, 149, 150, 192, 195, 210, 217, 220, 229, 234, 249, 252, 254, 257–8, 259 Mu‘tazila 19, 93, 98, 229 Nabataean language 52–3, 57–61, 114, 160, 254–5 Nabataeans 51–2, 53, 63 Najaf 171 Najran 59–60, 65, 77, 111, 112, 121, 169, 174 (n. 30), 206 292 INDEX OF PEOPLE, PLACES AND SUBJECTS Narsai of Edessa 109, 121 Nestorians, v. Christians, East Syrian Christians Neuwirth, A. 8, 12, 25 (n. 92), 31, 34, 35, 41, 81, 97 Nile River 3, 168 Noah 142–4, 166, 200 (n. 28), 244 Nöldeke, T. 3, 9, 12, 16, 21 (n. 19), 23 (n. 45), 31, 60, 72, 73, 95, 100, 120, 133 (n. 30), 175–7, 183, 195, 197 (ns. 4, 8), 198 (n. 8), 204, 226, 229, 230, 233, 238, 240, 245 (n. 3) 38, 45, 49 (n. 43); script and orthography 16, 20 (n. 11), 33, 35–8, 41, 44–5, 46, 49 (ns. 40, 46), 51, 74, 75, 76, 79, 92, 111, 125, 194, 199 (n. 21), 233, 235, 240, 242; ‘Uthmanic recension 4, 7, 12, 40, 41–2, 49 (n. 40), 50 (n. 59), 73, 77, 80, 89, 93, 100–1; variae lectiones (qira’at; variant readings) 1, 2–4, 5, 7, 12, 20 (n. 11), 21 (n. 25), 30, 41, 42, 46, 66 (n. 11), 74, 84 (n. 18), 93, 96, 100–1, 182, 228–9, 232 Quraysh 36, 76, 90, 93, 94 Origen 120, 134 (n. 37), 143 Ramadan 35 al-raqim 1–2, 125–7, 133 (n. 29), 136 (n. 76), 240–1 Red Sea 77, 205, 212, 213 Reynolds, G.S. 46 (n. 1) Rippin, A. 19, 20 (n. 1), 31, 88 Rome 55, 57, 60 Rusafa (Sergiopolis) 59, 134 (n. 42) Pagans 10, 12, 14, 18, 59, 76, 81, 86 (n. 51), 113, 127, 207, 238, 240, 242 Palmyra 51, 55, 67–8 (n. 29) Paret, R. 72, 86 (n. 49) Partaw (Bardha‘a) 188, 202 (n. 51) Paul, apostle 80, 135 (n. 57), 147–50, 151, 157 Pelagius 160 Pentecost 211 People of the Book, v. ahl al-kitab Persia, v. Iran Persian language 24 (n. 75), 39, 58, 91, 109, 149, 186, 234, 250, 251, 252–3, 256 Peshitta, v. Bible, Peshitta Peter, apostle 149, 207, 210 Petra 54, 57, 63, 255 Pharaoh 199 (n. 25) Porphyry 99 Pretzl, O. 2, 5, 9, 19, 20 (n. 11), 21 (n. 20), 30, 89, 101 Proto-Evangelium of James, v. Apocrypha (Biblical) Psalms xiii, 78, 82, 83 (n. 2), 87 (n. 60), 209 Qarun, v. Korah Qaryat al-Faw 53, 64 qira’at, v. Qur’an, variae lectiones Qur’an, v. Table of Contents; epigraphy and Qur’an i, 13, 18, 22 (n. 38), 24 (n. 70), 25 (n. 87), 36–8, 39, 43, 45, 49 (n. 39), 50, (n. 50), 51–65, 67 (n. 27), 68 (n. 42), 213, 254; form of word qur’an 35, 78, 240, 258; numismatics and Qur’an 14, 208; San‘a’ Qur’an manuscripts 7–8, 22 (ns. 35, 39), 37, Saadi, A. 19 Sa‘adya 255 Sabaean language (Sabaic) 53, 204, 205 Sabbath 211 Samaritans 220, 250 Samir, S.K. 19 San‘a’ Qur’an manuscripts, v. Qur’an, San‘a’ Qur’an manuscripts Satan, v. Devil; form of word shaytan 7, 22 (n. 38), 207 Saul (talut) 236 Sawirus (Severus) b. al-Muqaffa‘ 155 Sebeos 186, 218 Septuagint, v. Bible, Septuagint Sergius, saint 55, 59, 134 (n. 42) al-Shafi‘i 251 shari‘a 14 Shi‘a (Shiites) 2, 170–1, 174 (n. 45) Simon of Cyrene 154 Sinai, Mt. 143, 226, 236 Sinai, peninsula 67 (n. 19), 114, 202 (n. 51) sira, v. Muhammad, the sira South Arabian language 36, 53, 58, 66 (n. 7), 204 Sozomenus 162 (n. 32) Spirit (of God; Holy Spirit) 10, 123, 146, 156, 159–60, 164 Sprenger, A. 88, 103 (n. 21) Stephen of Ephesus 120 293 INDEX OF PEOPLE, PLACES AND SUBJECTS Stewart, D. 19 al-Suyuti 71, 170, 230, 232, 235, 237, 252, 253, 256, 257, 260 (n. 12) Syria 13, 18, 34, 36, 42, 49 (n. 50), 51, 52, 53, 56, 59, 77, 80, 102 (n. 14), 114, 170, 171, 172, 183, 184, 187, 206, 207, 212, 214 (n. 7), 254 Syriac language (cf. Aramaic language) 15–19, 24 (ns. 75, 76), 33–41, 49 (n. 50), 51–2, 54, 59–60, 63, 67 (n. 28), 70, 76–9, 87 (ns. 56, 60), 89, 92–3, 96–101, 102 (n. 15), 109–31, 131 (n. 7), 132 (n. 20), 134 (n. 37), 136 (n. 72), 136 (n. 82), 137 (n. 97), 143, 145, 149–50, 160, 175–86, 188–96, 197 (ns. 4, 6), 198 (n. 8), 199 (ns. 20, 21), 200–201 (n. 30), 204–5, 215 (n. 18), 217–20, 220 (n. 2), 221 (n. 4), 225–7, 235–8, 240–2, 249–59 Türks (Kok Türks) 184, 186, 187–8, 192–4, 202 (n. 48) Tyre 206 ‘Ubayd b. ‘Umayr al-Laythi 91 ‘Umar b. al-Khattab, caliph 68 (n. 42) ‘Urwa b. al-Zubayr 91 al-‘Utaridi 91 ‘Uthman, caliph 3, 40, 77, 84 (n. 17), 95, 189; ‘Uthmanic recension of Qur’an, v. Qur’an, ‘Uthmanic recension van Bladel, K. 19 Van Reeth, J.M.F. 81, 99–101, 108 (n. 104) Vollers, K. 13, 30, 36, 95 Virgins of Paradise, v. Hur al-Tabari, Abu Ja‘far 16, 80, 91, 97, 130, 164, 166, 169–71, 232, 251–2 Talmud 54, 197 (ns. 6, 8) Talut, v. Saul Targums 254, 255 Tatian 166, 226 Tawrat (Revelation of Moses; Torah) 78, 82, 91, 99, 240 Tayyayê 67 (n. 28), 218–20 Temple (of Jerusalem) 165–6, 220 Theodoret of Cyrrhus 160, 162 (n. 32) Theodosius II, emperor 124, 126, 129–30, 135 (ns. 64, 65), 136 (n. 90) Thot, god 208 Tigris River 179, 183, 199 (n. 14) Timothy, patriarch 159 Trinity 10, 23 (n. 52), 33, 75–6, 80, 86 (n. 51), 112, 156 Wahb b. Munabbih 137 (n. 97) Wansbrough, J. 8–9, 11–14, 18, 24 (n. 58), 25 (n. 91), 30–1, 34–5, 41–2, 48 (n. 25), 50 (n. 58), 73–4, 84 (n. 20), 88 Waraqa b. Nawfal 80, 82, 85 (n. 39), 90–2, 101, 102 (n. 14) Watt, W.M. 31, 144, 214 (n. 3) Yahya, v. John the Baptist Yemen 7, 32, 37, 65, 77, 112, 206 Yunus b. Bukayr 91 Zayd b. ‘Amr 82 Zayd b. Thabit 82, 92–3 Zebed 49 (n. 50), 55, 60, 66 (n. 13) Zechariah (Zakariyya) 141, 166, 199 (n. 23) Zoroastrians 81, 218, 250 al-Zuhri 91, 104 (n. 36), 105 (n. 39) 294 REVELATION