The Qur'ân's Self-Image: Writing and Authority in Islam's Scripture

THE QUR'AN'S SELF-IMAGE Writing and Authority in Islam's Scripture

DANIEL A. MADIGAN Islam is frequently characterized as a "religion of the book," and yet Muslims take an almost entirely oral approach to their scripture. Qur 'an mean s "recitation" and refers to the actual words Muslims believe were revealed to Muhammad by God. Many recite the entire sacred text from memory, and it was some years after the Prophet's death that it was first put in book form. Physical books play no part in Islamic ritual. What does the Qur' an mean , then , when it so often calls itself kitab , a term usually taken both by Muslims and by Western scholars to mean ''book"? More than any other canon of scripture the Qur 'iin is self-aware. It observes and discusses the process of its own revelation and reception; it asserts its own authority and claims its place within the history of revelation. Here Daniel Madigan reevaluates this key term ,

kitab, in close readings of the Qur 'iin's own declarations about itself. He presents a compelling semantic analysis, arguing that the Qur'an understands itself not so much as a completed book but as an ongoing process of divine " writing" and "re-writing ," as God 's (cominued

011

back flap)

THE QUR'AN'S SELF-IMAGE WRITING AND AUTHORITY IN ISLAM'S SCRIPTURE

DANIEL

PRINCETON UNIVERSITY PRESS

A.

MADIGAN

PRINCETON AND OXFORD

Copyright Q 2001 by Princeton University Press Published by Princeton University Press 41 William Street, Princeton, New Jersey 08540

CONTENTS

In the United Kingdom: Princeton University Press, 3 Market Place Woodstock, Oxfordshire OX20 1SY LIST OF FIGURES

ix

LIST OF TABLES

ix

PREFACE

xi

All Rights Reserved Figures from God and Man in the Koran, by Toshihiko lzutsu, are reprinted by permission of the publisher, Keio Institute of Cultural and Linguistic Studies. Copyright 1964 by Keio

xiii

A NoTE FOR THE NoN-ARABisT

University, Tokyo. TABLE OF TRANSLITERATIONS AND ABBREVIATIONS

Madigan, Daniel A.

XV

3

INTRODUCTION

The Qur'!n's self-image: writing and authority in Islam's scripture I Daniel A. Madigan p.

em.

ISBN 0-691-05950-0

1.

THE QuR'AN AS A BooK

13

2.

THE QuR'AN's REJECTION oF SoME CoMMON CoNCEPTIONS OF KITA.B

53

3.

SEMANTIC ANALYSIS AND THE UNDERSTANDING OF KITAB

79

4.

THE SEMANTic FIELD oFKITAB I: VERBAL UsES OF THE RooT K-T-B

107

5.

THE SEMANTIC FIELD OF KITAB II: TITLES AND PROCESSES

125

6.

THE SEMANTIC FIELD OF KITAB III: SYNONYMS AND ATTRIBUTES

145

7.

THE ELUSIVENESS OF THE KlTAB: PLURALS, P ARTITIVES, AND INDEFINITES

167

8.

THE CoNTINUING LIFE oF THEKITABIN MusuM TRADmoN

181

Includes bibliographical references and index. Library of Congress Control Number: 2001088949

The publisher would like to acknowledge the author of this volume for providing the camera-ready copy from which this book was produced.

This book has been composed in Adobe Minion. Princeton University Press books are printed on acid-free paper, and meet the guidelines for permanence and durability of the Committee on Production Guidelines

APPENDIX: THE PEOPLE OF THEKITAB

193

BIBLIOGRAPHY

21~

INDEX OF QuR'ANIC QuoTATIONS

22~

GENERAL INDEX

23!

for Book Longevity of the Council on Library Resources.

Printed in the United States of America

2 4 6 8 10 9 7 5 3 1

vii

D

LIST OF FIGURES 3.1 The semantic field ofimdn 3.2 The semantic field of kufr 3.3 Qa'Ml in the semantic field of kufr and ~ird.t 6.1 The semantic field of ~ukm 6.2 The semantic field of 'ilm 6.3 The semantic field of kitdb

LIST OF TABLES 6.1 Qur'anic vocabulary related to ~ukm: authority 6.2 Qur'anic vocabulary related to 'ilm: knowledge

ix

PREFACE

A study such as this rarely originates in a pure and disinterested desire for knowledge, but more often in the pressing questions that arise out of experiences and commitments·. The origin of this work lies in an experience in Pakistan of the interaction between Muslim and Christian understandings of scripture. The emerging local Christian theology, even though often showing little appreciation of Islam, seemed to take for granted a distinctly Islamic approach to scripture and revelatiori: Furthermore, many Muslims took it for granted that Christians would be their allies against a secular world that was extremely skeptical of any claim to be in possession of literal divine revelation. This raised for me two questions: what was the understanding of scripture in the Muslim community's earliest days, and why was the Qur'An so adamant that God's revelation is common to all? My questions were further focused and sharpened by contact with William A. Graham's work on the Qur'An and on the oral use of scripture in Islam as well u in other religious traditions; His investigations of the importance of orality made the prominence of the word kitab-'book' or 'writing'-in the Qur'~n's aelf-description all the more perplexing. Professor Graham has been a valued suide and mentor from the time this project began to take shape during a year at dle Center for the Study of World Religions, Harvard University. This book began as a dissertation for the religion department at Columbia University, and I owe a substantial debt of gratitude to Professor Peter Awn and to the late Professor Jeanette Wakin, who from the start treated me more as a colleague than a student. They were unfailing in their support and concern through occasionally difficult times. My thanks to colleagues who have been generous in reading chapters and offering encouragement and suggestions, particularly to Lance Laird, Paul Heck, Greta Austin, and Clark Lombardi. Carolyn Bond did a flne job of editing the manuscript and teaching me some economy of style. Columbia University's Department of Religion was a wonderfully congenial atmosphere in which to work; it is a place where religion is valued and considered

-

'P"""'"'

~II

1'111 QtJR'AN'S SELF-IMAGE

wmthy of the most serious and careful study. The Graduate School of Arts and Scil:nccs was very generous in its support over my years in New York and (:ambridge. I see no way of repaying that debt except by a further commitment to the study and teaching of religion. One advantage of having moved to Rome is that I have been alerted to the work of my colleague, Arij A. Roest Crollius, S.J. He made me aware of his significant but unfortunately too little-known thesis on the ~ur'an and Hindu scriptures just as this book was going to press. I regret not having discovered it earlier so as to profit from his insights and engage with them further in the course of this investigation. We take rather different starting points, and differ in methodology, but our findings bear each other out, and our conclusions as to the understanding of divine writing and the symbolic nature of the kitab converge substantially. Fortunately the way is now open to collaboration in this area, and the issues on which we vary will provide avenues for further exploration. No undertaking like this can ever be accomplished alone, and I am much indebted to those who gave me the encouragement to see it through to its completion, especially my dear friend Dr. Nikolaos George, and my Jesuit companions in Lahore, New York, Boston, Cambridge, Rome, Berlin, and of course Australia.

A NOTE FOR THE NON-ARABIST

At the first sight of a book so laden with quotations in the original language and script, the reader unfamiliar with Arabic may feel an apology is in order. However, let me offer encouragement rather than apologies. This book has a dual purpose: to reexamine a consensus long held by both Muslim and Western scholars about the way the Qur'an understands itself, and to outline an alternative view for the specialist as also for those whose expertise does not lie principally in the study of Islam, but whose interest is in the burgeoning field of comparative study of scripture and hermeneutics. The extensive use of Arabic is necessary to the main argument, which engages closely with the text of the Qur'an. However, it is not intended to exclude the non-Arabist reader. All Arabic, Syriac and Hebrew have been translated and, in places where it might be useful, have also been supplied in transliteration. For those unfamiliar with the structure of Semitic languages, I offer the following short explanation. Though it falls far short of an introduction to Arabic, it should suffice to enable the non-Arabist to follow the linguistic argument of the book. If we examine the sets of English words sing, sang, sung, song, and ring, rang, rung, we recognize that the basic meaning is given us by the consonants, while the changing vowel indicates what part of speech the word is and its tense. This phenomenon is rather limited in English and not very regular-compare bit, bat and but, or hit, hat, hut and hot-but in Arabic this kind of pattern is of the essence of the language. One way of conceptualizing Arabic and other Semitic languages is that almost all words are based on roots consisting of usually three consonants. These roots are often thought of as the bearers of one or more basic meanings. For example, the triliteral root k-t-b (also represented in this book as ...fk-t-b) usually carries the idea of writing. Particular grammatical forms are derived from the root by affixes, infixes, doubling of the root letters and also by the arrangement of vowels linking the letters of the root. So, for example:

xiii

xi v

THE QUR'AN'S SELF-IMAGE KAT ABA

he wrote

KUTIBA

it >Vas written

YAKTuBu

he writes

TAKTuBu

she writes

KATIB

writer, scribe

MAKTOB

MAKTAB

desk, office

wri~ten

The root KH-1. -Q is connected with creation and gives rise, for exnple, to these words: KHALQ

c.:reation

KHALAQA

he created

YAKI:iLu Q u

he creates

TAKHLuQu

she creates

.KtlALIQ

creator

MAKHLOQ

created

,KH ALIQA

nature

TABLE OF ARABIC TRANSLITERATIONS

t

'

Note that the root letters always remain in the same order. Having observed how numerous words share a common root and hence a related meaning, now see how the pattern of non-root letters gives a common grammatical form. Words that use the pattern cAclc (where C stands for any root consonant) are all active participles of the basic verb:

KATln

one who writes

i.e., a scribe

'ALIM

one who knows

i.e., a scholar

KHALIQ

one who creates

i.e., a creator

QATIL

one who kills

i.e., a murc\;;rer

?-A LIM

one who misuses

i.e., a wrongd'i>er, a::.·oppu'>sor

FATI}1

one who breaches

i.e., a conqueror

known

MAKHL0Q

created

MAQT0L

killed

M.A4L0M

wronged, oppressed

MAFT0}1

breached, defeated

This phenomenon of the combination of root letters with affix/vowel patterns is repeated thwughout the language, and is easi;y recognized afll'f a relatively short acquaintance with it.

r

._;

f

a

.)

z

J

q

~

b

J"

s

.!l

k

.;:.,

t

J"

ill

J

.!,..

1h

J"

~

r

m

../

9

.:;

n

.

c c.

I]

.1

t

kh

j;,

.)

d

t

~

.dh

t

h ?

J

w/u y/i

gh

ABBREVIATIONS

Ef

The Encyclopaedia of Islam. New ed., prepared by a number of leading orientalists. Edited by an editorial committee consisting of H. A. R. Gibb and others. Leiden: E.}. Brill, 1954-.

csco

Corpus Scriptorum Christianorum Orientalium, Louvain.

GdQ

Noldeke, Theodor. Geschichte des Qorans. Zweite Auflage, bearbeitet von Friedrich Schwally. 3 vols.l909-38. Dritte teil von G. BergstraBer und 0. Pretzl. Reprint, Hildesheim and New York: Georg Olms Verlag, 1970

lAOS

Journal of the American Oriental Society

!SAl

Jerusalem Studies in Arabic and Islam

/SS

Journal of Semitic Studies

MID EO

Melanges de l'Institut Dominicain des Etudes Orientales

SI::I

The Shorter Encyclopedia of Islam. Edited by H. A. R. Gibb and J. H.

Similarly the passive participles share a common pattern (MAce 0 c): MAKTOB written MACLOM

.)

~

Kramers. Leiden: E.

J. Brill, 1953 . v

Introduction

Islam is commonly characterized as a religion of the Book, not only in popular parlance but also in scholarly circles, and is even considered by many the most fully developed example of this type of religion, which had long been emerging in the Near East. 1 This view of Islam has seemed to many scholars almost selfevident, and of course is not entirely without foundation. It is true that the words of scripture occupy a position and play a role in the faith and practice of Muslims that is much more exalted and central than perhaps in any other religion. However, this way of approaching Islam fails to acknowledge that Islam is also characterized by an almost entirely oral approach to its scripture. One finds no physical book at the center of Muslim worship; nothing at all reminiscent of the crowned T6dh scroll or the embellished lectionary. On the contrary, the simple ritual and the recitation of the Qur' an that forms part of it are carried out from memory. Even the prodigious effort of memory required to have the entire sacred text by heart is not considered at all out of the ordinary for a Muslim. To have to consult a written copy to quote the Qur'an is thought a failure of piety. Yet Muslims themselves would surely not dispute the claim that the Book is at the heart of their religion. The Qur'an uses the term kitdb (pl. kutub, usually translated as 'book' or 'scripture') hundreds of times, and for commentators it is axiomatic that al-kitdb means the Qur'an. However, full weight must also be given to the fact that at the foundational level of Islam-in the Qur'an itself-the precise meaning of the term kitdb is not so easily ascertained. This word kitdb is pivotal to the Qur'an's perceptions both of itself and of God's dealings with humanity over the centuries. However, the term's continuing importance in the self-understanding oflslam makes it all too easy for us to read later developments and usages back into the verses of the Qur'an itself. That is to say, we too easily

1

See, for example, Wilfred C. Smith, "Scripture as Form and Concept: Their Emergence for the Weatern World," In Rethinking Scripture: Essays from a Comparative Perspective, ed. Miriam Levering (Albany: State Unlvenlty of New York Preu, 1989 ): 29-57.

I NTROIJUC'IION

1111 QuR'AN' s SELF-IMAGE

that an understanding of scripture which only gradually emerged among Muslims during the centuries of their community's development was actually pn:~cnt and fully enunciated in the text of the Qur'&'l', , :';,v'cr!4r J}f),

1'152); (;eo Widengren, The Ascension of the Apostle and the Heavenly Book (Ki,'>;:; .,,..; llppsula Univcr~itcts Arsskrift 1950:7 (Uppsala: A.-B. Lundequistska Bokh'l-.a~. ~. ~~5( ): m) >:: Mulwrrmuu/, tlw Apostle of God, and His Ascension (King and Saviour V), UppsIJ II[ Islet Ill (Nt•w York: Kluv, IW>7); Tor Andr:u·, Mul111mrruul: '11tr Mw. wJt/J,;., htitlt

( Nrw Vutk :
One scarcely expects to find at the end of such an investigation that a singll' word is adequate to translate kitdb. Nor is one likely to oppose translation altogether. Rather, I hope that some of the complex connotations of the Qur'i\n's "book" and "writing" language will come to light as it is allowed to "interpret its 16 own concepts and speak for itself. " If the claim to have discerned the Qur'an's particular conception of kitiJI• is to be convincing, however, there should also be evidence that it has survived in the Islamic tradition. It would be merely presumptuous to claim that the Qur'an's intent had been entirely misunderstood from the beginning by the very community that was constituted by it. For this reason, the concluding chapter attempts to show how the rich conception of kitab present in the Qur'an's discourse and the understanding of revelation and canon that it signifies have continued to exert their influence on the tradition in various ways. In speaking for itself, the Qur'an acknowledges that it is speaking a language already used by other groups-the ahl al-kitab. It was from these people-according to tradition, the Jews of Madina, the scattered Christian ascetics of the desert, and perhaps others as well-that the people of the l:fijaz region learned the language ofkitab that made the Qur'an's claims and exhortations comprehensi~le to them. So in the appendix, I turn to the revelations that the Qur'an sees as 1ts kin and to the communities that cherished them, to see whether the notion of kitab emerging from our semantic analysis would have made sense to the other peoples who were defined by the phenomenon of the kitab. It remains to say something about the overall approach adopted in this study. A

considerable weight of scholarship in recent years has focused on the question of the origins of Islam and its scripture, and the figure of John Wansbrough has hovered over all this work, not always named but ever present either as mentor or as adversary. Wansbrough rejects the idea that the Qur'an was an already complete collection of the revelations of Mu~mmad before the time the Arabs had expanded beyond Arabia to other parts of the Middle East. According to Wansbrough, the Qur'an did not originate in Arabia, nor indeed did Islam. What is customarily considered to be the expansion of Islam was rather the expansion of a recently united and militarily powerful Arab confederation, one that had as yet no distinct religion of its own. In the process of military and economic expansion they came upon a new sectarian development within the Abrahamic and monotheist religious environment of Mesopotamia. This they adopted as their own, rewriting its history and giving it an Arab imprint in tlw

15

( .'orw ptJ i11 tlw C)rdt111, Md ;ill Ishunk Studies I (Montreal: McGill University Press, 1966). 3-41. 1 '' l:t.IIIMI, ( ,'mfc'r'fH.'I, \.

II II ( JillfAN'~, Sill - IMA( .1

prtKl'.~s.

IN 1KOl >UC II! lN

Thl' t)ur':ln Cllll'rged out of a diversity of sourn·s as part of this process,

with its canonical form gradually separating itself from a body of prophetic sayings that had originated in this sectarian environment. The process was so gradual that Wansbrough would maintain that one cannot speak of a fixed version until about 800 C.E! 7 Partly because of the radical nature of his thesis and partly because of the opaqueness of his original works, it took some years before Wansbrough's thought was fully engaged with by his critics. 18 More and more, however, Wansbrough's critique is being seriously evaluated, though still 'to a large extent rejected. The rejections notwithstanding, most will recognize that Wansbrough has done a valuable service to the field of Qur'anic studies by challenging the naivety with which western scholars have approached the sources and accepted traditional accounts of the emergence of Islam and of the Qur'an's history. Wansbrough and those who have adopted his approach have stimulated a great deal of critical and creative thought about the early history of the Qur'an. By calling for a more sophisticated approach to the traditional sources, they have opened the Qur'an text to new readings-£eadings less controlled by the rather too self-assured interpretations of one particular moment in Muslim tradition. This study owes a debt to Wansbrough's work even though it does not accept his more radical conclusions. It approaches the Qur'an text as a unity, while recognizing that this does not necessarily entail accepting that it was codified and that it achieved a position of authority in precisely the way the traditional accounts claim. The processes of collection and canonization were obviously more complex and time-consuming than they are often presented as being, and 17

John Wansbrough, Quranic Studies: Sources and Methods of Scriptural Interpretatron, London Oriental Series, vol. 31 (Oxford: Oxford University Press, 1977) and his The Sectarian Milieu: Content and Composrtion oflslamrc Salvation History, London Oriental Series, vol. 34 (Oxford: Oxford Umversity Press, 1978). Wansbrough's scenario makes a good deal of the fact that we have no complete copy of the Qur'an datable earlier than the ninth century. However, there are certainly fragments which can be attributed to a much earlier period. In spite of the lateness of tue extant complete trancripts, one needs to separate the question of the canonization of the text from the matter of its being widely written and copied. In the case of a sacred text that functioned and, indeed, continues to function in a predominantly oral way, the lack of a complete transcript dating from an early period is not as significant as might at tfisrt be thought. Estelle Wheelan has argued that the custom of writing Qur'ans long pre-dates the first datable texts (E. Wheelan, "Forgotten Witness: Evidence for the Early Codification of the Qur'an," lAOS 118 [ 1998]: 1-14). 18 A small group of scholars who have followed Wansbrough's lead (among them G. R. Hawting, Herbert Berg and Andrew Rippin) have succeeded not only in explaining his position more clearly than he originally did, but also in keeping the discussion of his approach open. Their panel at the 1996 meeting of the American Oriental Society at the University of Pennsylvania, and the special issue of Method and Theory in the Study of Religio11, 9 (1997) that followed from it, played no small role in spurring a new engagement with Wansbrough's thought.

I

II

we can find evidence of that fact even in the traditional accounts. Nonetheless, I approach the text with the presumption that, however it may have come by its historical forms and its present content, it existed as a sacred text from a much earlier time than Wansbrough would posit. The extent to which the entire text was known, transcribed, memorized and used in law or liturgy in the early period is still not clear and much research remains to be done on the matter. However, without foreclosing on the legitimate questions about the Qur'an's provenance and early history that still remain, I have chosen to treat the text as a coherent whole. I do so because that is the way it functions within the community that canonized it and that looks to it for guidance. The notion of kitab is the overarching theme that proclaims and maintains that coherence.

CHAPTER ONE

The Qur'an as a Book

WESTERN SCHOLARSHIP AND THE WRITTEN NATURE OF THE QUR'AN One very striking feature of the Qur'an is its insistent claim to kinship with the revelations to the Christians and Jews. This insistence that seems to be what has led most non-Muslim scholars to stress that the production of a canonical volume parallel to those of the other faith communities lies at the heart of the Qur'an's identity and Mu}:lammad's mission. The Swedish scholar Geo Widengren went so far as to claim: "That M~ammad himself, by committing his revelations to paper, purposely aimed at creating a Holy Book in competition with the T6rah and 'Evangel', is perfectly clear."1 The Book had been sent down to Mul}ammad and now, Widengren claims, "he was going to set it down on paper, just as 'the Book which Moses brought, a light and a guidance to man, which ye set down on paper, publishing a part, but concealing a part,' as he reproached the Jews in Sll.ra 6:91." 2 The relevant part of the verse is:

43.,J.:5 ~') ;,;):.;.5 vW! ~~) ~~; ~; ~ ~G- ($:D' :,..~, J}(:;. Ji ;J'j4T~) ~~ (;J:7 ~~ ~J I~ ;:,_,l.Wj Say, "Who was it who sent down the kitab that Moses brought to be light and guidance for humanity and that you put on (lit., make into) parchments [or papyri] that you display, but you also conceal a great deal? You have been taught what neither you nor your forebears knew." Sllrat al-Arfdm 6:91 1 Wldensren,Muhammad,150. 2 Wlden1ren, H1brew Proph111, 54,

2

THE QuR'AN's SELF-IMAGE

14

Widengren's principal conccrn is with the motifs of th~ \cavenly book and the apostle in Near Eastern religious history, and he more t!1an adequately demonstrates their significatKc in the development of Islamic thought. The two archetypes of the messenger--one who ascends to heaven and returns with the message and the other upon whom the message descends while he remains in the human sphere-both figure prominently in the development of Muslim belief about the Prophet, even though the Qur'an presents him as being only of the second type. 3 Widengren points out that much speculation about Mu}:tammad's night journey and ascent to heaven ignores and even contradicts the explicit doctrine of the Qur'iln. However, Widengren has allowed both his own overall vision of the development of the idea of scripture in the religious history of the Near East and also his reading of the later Islamic ~ral~ition to prejudice his reading of the Qur\1n itself. He sums this up at one poir.t in his discussion: "The endeavour of Mul,lammad to create, as it were, an Arabic Yersion of the Heavenly Book, as we have seen, is dictated by the pattern of the Ancient Near East, directing the Apostle to exhibit to his adherents, in a visible form, the Book he had been given by God.'>4There is no evidence at all ofMul}ammad's ever having "exhibited" a visible book-in fact quite the contrary. But this fact seems not to faze Widengren in the least, so intent is he on his thesis. The verse that Widengren quotes in support of his contention (Q 6:91) is more logically read as an indication that Mu}:tammad m;ght not have intended to produce a written document. We learn from it that he was cer~ainly familiar with the Jewish custom of writing down the kitab on qaratf,; ('papyrus' or 'parchment' rather than 'paper'). 5 Therefore, if he had indeed taken such a practice as his model, one would expect to find indications of that in the traditions about the written Qur'anic materials that remained after his death. But we find no reference to papyrus or indeed to anything that would suggest an intention to produce a codified document. Al-SuyiiP lists several types of material on which, according to the traditions, fragments of the Qur'fm were written: <usub 'the bark of palm3

Cf. Widengren, Muhammad, IOOff., 204-8. He summarize5 his findings in "Holy Book and Holy Tradition in Islam,» in Holy Book and Holy Tradition, edited by F. F. Bruce and E. G. Rupp, (Manchester: Manchester University Press, 1968), 21(}-236; here 218-220. 4 Widengren,Muhammad, 151. 5 The term qir,tds is almost certainly from the Greek XUP't1lS. probably through the Syriac q"r,tisa. Cf. Jeffery, Foreign Vocabulary, 235-6. Roberts and Skeat (Codex, St. n. I) stress that in Greek usage in the early Christian era it meant a roll or scroll, usually undrnplccl/1 / tlcc· 1\ur·,,, Mccllll\1 tif•l• 111 I Icc· I lllc'lllccl 1"'1111111' (I :hi';'ll": \lncvc·r~ily of( :hiutv,o l'c'c'"• 1'1:1'1), ~••1.

THE QuR'AN AS A BOOK

15

branches'; likhaf'thin stones'; riqa' 'scraps' (al-Suy'll~ i understands these to have been of jild 'leather'; raqq 'parchment'; or kagl!ad 'paper'); qita' al-adim 'pieces cut from a skin'; aktdf'shoulder-blades'; aqtab 'the wood of camel saddles'; and adlac 'ribs'.6 It is difficult to see in such a motley collection of materials any indication that Mul:tammad had a book in mind. It is quite likely, of course, that this list has little historical reliability. Nevertheless, it indicates that the community of al-Suyilti's time ·(d. 1505) was unaware of anything that might correspond to what Widengren envisages as the Prophet's intended format. Furthermore, if this verse was actually an expression of Mul)ammad's own intentions-as Widengren takes it to be-one might expect its wording to have exerted some influence on the tradition, at least to the extent that qara.tis (or the 7 singular qirtas) would appear in the traditional list of materials. But such has not been the case. In a somewhat idiosyncratic reading of an important verse, Widengren finds further support for his conviction that "Mu}:tammad himself has both read and 8 written the Divine Revelations":

0~~ yG~~I~!~~~) y8' .:r ...L,i .:r 1:,t; d

~)

Before it you used not recite a-ny hltab nor do' yo~ tr~~;crib~ it with your right hand, for then those who follow vanities might have doubted. Surat al-' Ankabut 29:48 It is difficult to see how this verse which is traditionally taken as proof of the Prophet's inability to write, if not also to read, 9 could be taken by Widengren as an indication of quite the opposite. The text neither confirms nor denies Mu}:lammad's literacy, but Widengren uses it to prove that the Prophet was literate. 10 On the other hand, traditional Muslim commentators, perhaps with 6 Al-Suy\l,ti, AI-Itqan fi r..S

day.; of Kis l';\.' V. V. l'olrL,in,S/ol'll t'

l'ol'l L..S' as "as if their r GdQ, 1: 47. At the time ofMu~ammad's death, no discussion had been give• , tCI the succession, so it was up to the community to decide the matter-and even to decide by what metLo~. and what criteria the decision should be arrived at. The question of succession in the earl)' period of Islam remained an extremely controversial issue, and gave rise ultimately to the divisi on into Sunni and Shill, IK I'I. k· l\l,llhi'J ~ ...;IJ ..>LA.. 1 ,._:.

-r o.r' ;,__ jS' ..>T_r<JI r'>L.Jt ...,.U j,,..~ ~ 4-J.,;J t'\.,. ~ ~ JT_,.-IJt""Y ~t.r-"'"" .r-i'PJ ~ ..:U1 J;-; ¥ <E'_ ...LS" .___:S .;T_..i.JI _,lS' .li, On the prohibition of writing, seeM. J. Kister," . .. Lil taqra'lll-qur'tlna •alt11-mu_.,llll/iyyln wa-111 ta~milu 1-•i/ma 'alii 1-~a~afiyyin . . . :Some notes on tht· Transmission of 1./tlllitll," /SAl 12 (I'JYH ): 127-t.2 . 5 '' AI-Suytip, al ltlllalr• on the lnh·wity uf thr l)u r' ~n." Studi11 /il•"""" l'l ll'l'J Ill ''• 1•1,

THE QuR'AN AS A BOOK

II

I:Iamida bint Abi Yunus said, "My father recited to me when he was eighty years old from the mu?haf of cA'isha «Surely God and his angels ask blessings [yu~alluna] on the Prophet. 0 you who believe ask blessings and invoke peace upon him .... »" She said, "That was before 'U1hman 71 changed [yughayyir] the codices." Zirr b. I:Iubaysh said: " 'Ubayy b. Kacb asked me, 'How do you count [the verses in] Surat al-Ahzab?' I said, 'Seventy-two [or seventy-three] verses.' He said, 'It used to be the same length as Surat al-Baqara and as part of it we used to recite the stoning verse.' I said, 'What is the stoning verse?' He said, '«.... ~>:>- }JI...,;; t:;~>:- J ...,_.:S , -L-1'

~·• l~uot:J :_;b

For /o! it is a respected scripture.\\ Futility-;anno; app;oach Jt fr~m -the fro~t or fro~ /Jehirul. I low, they ask, could it be 7 IIC, .

QuR'AN AS A BOOK

Drawing on observations of the actual shapes used for the various letters in t hl' oldest inscriptions and papyri, Bellamy argues that these early forms of letters or combinations of letters could easily have been mistaken by later copyists ( o1· those reading the original text aloud to them) once the script had been furthl' l" developed and standardized. Given the variations of script found in early inscriptions and papyri/ 05 it is conceivable that, for example, the combination ba' -alif could be taken as the letter fb.'. Thus the combinations bd' -alif-sin or bii' -alif-sin-mfm-the first letters of the basmalah-could conceivably have been read as _ta-sfn or as _td-sin-mim-two of the combinations of 'mysterious let· ters'-by someone not recognizing an abbreviation. An abbreviation here might not have been recognized because it might well not have been expected, given that the basmalah held a controverted status within the text of the Qur'an.""' Since there was no oral tradition of reciting these letters and therefore they wert' arguably not part of the Qur'an,107 they first proved mysterious to the community only when early transcripts for some reason later came to prominence. No standard abbreviation had yet been adopted for this invocation of uncertain status and so each early scribe had devised his own (though several are repeated). Thus the letter combinations vary. The importance of Bellamy's work lies not so much in his uncovering of thl' "original" text, for these formulas have since taken on a mystical or dogmatic lik of their own quite apart from their original significance. Rather, if he is correct in his surmise, his work is a salient reminder of the further limitations of any early transcript of the Qur'an. Given these limitations, it should be no surprise to discover that, both in the understanding of the Prophet and in the life of th l'

105

These have been tabulated by Abbott in North Arabic Script, table 5, and by Adolf Grohmann in Arab1sche Papyruskunde ( Handbuch der Orientalistik, Erste Abteilung, Erganzungsband II, l' rsl c Halbband [Leiden: E. ). Brill, 1966], 49-118), table 10. Abbott is of the opmion (p. 48) that Ar;ltll r writing was equal to the task of transcribing the revelations in the time of the Prophet. She bast•s this conviction upon the clearly developed script evident in the papyrus PERF n" 558-from Egypt, c. 2! A.H. (plate 4 but more clearly decipherable in Grohmann, plate 2, 1]. arguing plausibly eno ugh that, if such were the development in Egypt, then writing in the h eartland of the language must havr I>/;,,,M'i/ t11tl'lr't111 ). AI 1\0qillani is quoted as .say in.: tlr.tt I ;ahr·id IIS-1 'II,,.>-_;...~, .;r r~ r-' ,1,_JI.;T ..r' ..J_,... 1 '-# YJJ •JW JJI....J.; L. ~ ..:...,U

>~L ~

:._.'\ft JI.!J.I) j~' .li .) 1( 4 ~·I) ,, 1 ~~ .._,.:;I

J.L..-. ...,~ ..:nt , .ll

.:. .•l~ .JI

yo Ll,.;. IJ.u., rJ ,, l. l,l_.. , 1 ..•ll , 1,1 ... 1. 10 ,,, ).L..o.il ~ ..,T_,..i.ll [ ..L,.l.-; ;ll..Y..,: y , & ._,.11 th< J

•J..)~I ~ jo.Y.,:) .....:S i f ..Jl;lJ ~1

,[ •..w,;.:;

.;.)> .;i ..).&- ~L&;I_, $1»1 j_,...J .._,t.....,..l J)l.!..:; .u.., '\II ~I~ ~J ,~_,.... o.r"-1 'If_,,~

..:...>r_, .~~,.., ;1_,..... JS' .;T ~; $ .,lj

$

J~~;t ..J~ ..JI ~

. ...,.. ,;1 ,,.utt.l•, : Jt>

._,.11 y ...:..J._;. ;_.'1ft

..,lJ , ii'>l.:JI ..:.,.\))

'-'11

,:_.1,1,..0)1

.... ..Ac_, ~u, J ,,, rJ,

• . ....,

48

THE

QuR'AN' s SELF-IMAGE

THE

and arranged by God, however, becomes much more problematic. The tradition implicitly recognizes this difficulty when it pays so little attention to structure and context in its approach to the text. The commentary tradition treats the Qur'an as a set of discrete verses rather than as coherent suras or pericopes. Concern for the interrelationship of verses is principally tied to the question of abrogation; in those cases the clues to the chronology of the Qur'an come not from within the text itself but from reconstructions of the context of each verse from the biography of the Prophet and the accounts of his Companions. The difficulty is compounded when one projects the claim that the presen: structure is ideal back onto the heavenly archetype, as though the mu?haf were an accurate transcript of the lawh mahfu+ 'preserved tablet' of Q 85:22. Can God's Book really be as fragmentary, haphazard, specific, and, one might even say, parochial as the text of the mu~haf? By asserting that it is, the community argues itself into the unenviable position of having to claim that from all eternity God has been concerned about such minutiae as the domestic arrangements