The History of The Quran by Theodor Noldeke

E
=
m
c
2
1

A

Uploaded By
$am$exy98
theBooks

This eBook is downloaded from
www.PlentyofeBooks.net
PlentyofeBooks.net is a blog with an aim
of helping people, especially students,
who cannot afford to buy some costly
books from the market.
For more Free eBooks and educational
material visit
The History of the Qurn
Texts and Studies
on the Qurn
Editorial Board
Gerhard Bwering
Yale University
Jane Dammen McAuliffe
Bryn Mawr College
Bilal Orfali
American University of Beirut
Devin Stewart
Emory University
VOLUME 8
The titles published in this series are listed at brill.com/tsq
The History of the Qurn
By
Theodor Nldeke
Friedrich Schwally
Gotthelf Bergstrer
Otto Pretzl
Edited and translated by
Wolfgang H. Behn
LEIDEN BOSTON
2013
Library of Congress Cataloging-in-Publication Data
Nldeke, Theodor, 1836-1930.
[Geschichte des Qorns. English]
The history of the Qur'an / by Theodor Noldeke, Friedrich Schwally, Gotthelf Bergstrasser, Otto
Pretzl ; edited and translated by Wolfgang H. Behn.
p. cm. (Texts and studies on the Qur'an ; v. 8)
Includes bibliographical references and index.
ISBN 978-90-04-21234-3 (hardback : alk. paper) ISBN 978-90-04-22879-5 (e-book)
1. Koran. I. Schwally, Friedrich, 1863-1919. II. Bergstrsser, Gotthelf, 1886-1933. III. Pretzl, O. (Otto).
IV. Behn, Wolfgang. V. Title.
BP130.N613 2013
297.1'221dc23
2012041328
This publication has been typeset in the multilingual Brill typeface. With over 5,100
characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for
use in the humanities. For more information, please see www.brill.com/brill-typeface.
ISSN 1567-2808
ISBN 978-90-04-21234-3 (hardback)
ISBN 978-90-04-22879-5 (e-book)
Copyright 2013 by Koninklijke Brill NV, Leiden, The Netherlands.
Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing,
IDC Publishers and Martinus Nijhoff Publishers.
All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV
provided that the appropriate fees are paid directly to The Copyright Clearance Center,
222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA.
Fees are subject to change.
This book is printed on acid-free paper.
CONTENTS
Preface to This Edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Nldeke [Preface to the German Edition, Part 1, pp. viiviii] . . . . . . . . . . xvii
Schwally [Preface to the German Edition, Part 1, pp. ixx] . . . . . . . . . . . . . xix
Zimmern [Preface to the German Edition, Part 2, pp. iiiiv] . . . . . . . . . . . xxi
Fischer [Preface to the German Edition, Part 3, pp. 220224] . . . . . . . . . . xxiii
Pretzl [Preface to the German Edition, Part 3, pp. viiix] . . . . . . . . . . . . . . xxv
The Origin of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Notion of Prophecy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Type of Muammads Prophetic Endowments . . . . . . . . . . . . . . . . 2
Jewish and Christian Influences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Literacy in Pre-Islamic Arabia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Muammads Relation to Zayd b. Amr and Umayya b. Ab l-alt . . . 13
Pagan Influence and Muammads Personal Contribution to the
Establishment of His Religion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Muammads Revelations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
The Various Types of Muammads Revelations . . . . . . . . . . . . . . . . . . . . 17
Psychological Stages of Excitement. The Alleged Mentor of the
Prophet. Daya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
The Length of the Revelations, Their Names, Style, Refrain and
Wordplay. The Construction of the Verses. . . . . . . . . . . . . . . . . . . . . . 24
Written Notes of Koranic Passages. Additions and Changes
Arising from Muammad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
The Seven aruf (Sets of Readings.) Abrogation of Revelations . . . . 40
The Originality of the Koran and Its Connection with the
Revelations of the Prophet Maslama . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
The Origin of Individual Parts of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Aids in Establishing the Chronology of the Sras . . . . . . . . . . . . . . . . . . . 47
The Transmitted Lists of the Chronology of the Koran . . . . . . . . . . . . . 48
The Individual Parts of the Current Koran: The Meccan Sras. . . . . . . . . . 55
General Chronology of the Meccan Sras . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Content and Characteristics of the Meccan Sras . . . . . . . . . . . . . . . . . . 58
vi contents
Classification According to William Muir, Hubert Grimme, and
Hartwig Hirschfeld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
The Sras of the First Meccan Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
General Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Formulas of Invocation at the Opening of Many Sras . . . . . . . . . . . . . 63
Comments on Sras 96, 74, 111, 106, 108, 104, 107, 102, 105, 92, 90. . . . . 64
Comments on Sras 94, 93, 97, 86, 91, 80, 68, 87, 95, 103, 85, 73, 101,
99, 82, 81, 53, 84, 100, 79, 77, 78, 88, 89, 75, 83, 69, 51, 52, 56, 70,
55, 112, 109, 113, 114 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
The Sras of the Second Meccan Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
The Divine Name Ramn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Comments on Sras 54, 37, 71, 76, 44, 50, 20, 26, 15, 19, 38 . . . . . . . . . . . 100
Comments on Sras 38, 36, 43, 72, 67, 23, 21, 25, 17, 27, 18 . . . . . . . . . . . 107
The Sras of the Third Meccan Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Comments on Sras: 32, 41, 45, 16, 30, 11, 14, 12, 40, 28, 39 . . . . . . . . . . . 118
Comments on Sras 29, 31, 42, 10, 34, 35, 7, 46, 6, 13 . . . . . . . . . . . . . . . . . 126
The Medinan Sras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Political and Religious Conditions at Yathrib before the Hijra . . . . . . 135
The Unprecedented Success of Islamic Propaganda at Yathrib . . . . . 136
The Waverers (munfiqn) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
The Pagan Population and the Jewish Tribes . . . . . . . . . . . . . . . . . . . . . . . 138
Content and Style of the Medinan Sras . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Comments on Sras 2, 98, 64, 62, 8, 47, 3, 61, 57, 4, 65, 59 . . . . . . . . . . . 141
Comments on Sras 33, 63, 24, 58, 22, 48, 66, 60, 110, 49, 9, 5 . . . . . . . . 167
Muammads Uncanonical Promulgations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Non-Extant Revelations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
The Uncanonical Sayings of the Prophet . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Miscellaneous Revelations Preserved in the Traditions . . . . . . . . . . . . 207
contents vii
Written Collections in Muammads Lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . 209
The Preservation of the Manuscripts of the Revelation during
Muammads Lifetime, Based on References in the Koran
and on the Literary Form of the Sras . . . . . . . . . . . . . . . . . . . . . . . . . . 209
The Indirect Collectors of the Koran or the Custodians of the Oral
Revelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Popular Acquaintance with the Koran under the First Caliphs . . . . . . . . . 217
Collecting and Editing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Al as a Collector of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
The Collection of Slim b. Maqil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
The (First) Collection of Zayd b. Thbit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
The Prevailing Tradition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Divergent Traditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
Criticism of the Traditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Form and Content of the First Collection . . . . . . . . . . . . . . . . . . . . . . . . . . 232
The Other Pre-Uthmnic Collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
The Personalities of the Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Dissemination and Preservation of the Editions . . . . . . . . . . . . . . . . . . . 237
The Koran of Ubayy b. Kab According to the Transmission of
al-Fihrist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
The Koran of Ubayy b. Kab According to the Transmission of
al-Suys al-Itqn, and Its Relation to al-Fihrist . . . . . . . . . . . . . . . 239
The Sras Peculiar to the Koran of Ubayy b. Kab . . . . . . . . . . . . . . . . . . . 240
The Question of Authenticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
The Relation of the Transmitted Lists of Ubayy b. Kabs Sras
with One Another and with the Canonical Edition . . . . . . . . . . . . 243
The Koran of Ibn Masd According to the Transmission of
al-Fihrist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
The Koran of Ibn Masd According to the Transmission of
al-Suys al-Itqn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
The Relation of the Two Lists to One Another and to the
Uthmanic Recension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
The Relation of the Korans of Ubayy, Ibn Masd, and Ab Ms
to One Another and to the Canonical Version. . . . . . . . . . . . . . . . . . 247
Obscure and Dubious Codices of the Koran . . . . . . . . . . . . . . . . . . . . . . . . 249
viii contents
The Genesis of the Authorized Redaction of the Koran under the
Caliph Uthmn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
The Established Tradition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Deviating Traditions and Their Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Criticism of the Established Tradition: The Biographies of the
Members of the Commission, and the Qualification of the
Members for Their Task . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
The Procedure of the Establishment of the Text of the Koran, and
the Qualification of the Members of the Commission for
Their Task . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
The Arrangement of the Sras in the Uthmnic Koran . . . . . . . . . . . . 263
The Mysterious Letters Preceding Certain Sras . . . . . . . . . . . . . . . . . . . 267
The Basmala . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Sectarian Reproach against the Uthmnic Text . . . . . . . . . . . . . . . . . . . . . . . . 279
The Alleged Corruption of the Text of the Koran by Ab Bakr and
Uthmn: Reproach of Christian Scholars of the West . . . . . . . . . . 279
Reproach of Muslim Sectarians, Particularly the Shites, against
Uthmn. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
The Shite Sra of the Two Lights ( ) . . . . . . . . . . . . . . . . . . . . . 293
Administrative Measures for the Realization of the Uthmnic Text . . . . 305
The Islamic Canon and Its Relation to Its Christian and Jewish
Counterparts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
The Founders of Judaism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
The Position of Jesus in Christianity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
The Development of Canonical Islam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
The Islamic Sources on the Origin of the Revelations and the Genesis
of the Book of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
The Muslim Sources: The Foundations of the System of
Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
The Foundations of the System of Transmission . . . . . . . . . . . . . . . . . . . 316
The Biography of the Prophet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
The Canonical adth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
The Literature of adth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
The Biographies of the Companions of Muammad . . . . . . . . . . . . . . . 338
contents ix
The Peculiarity of the Islamic Exegesis of the Koran: The Exegetic
adth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
The Exegetic adth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
The Creators of the Exegesis: Ibn Abbs and His Pupils . . . . . . . . . . . . 346
The Extant Commentaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Important Writings Containing Commentaries on History and
Exegesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Shite Commentaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
Writings on the Occasion of Revelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
The Introductions to the Koran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
Poetry as Source of History: The Poetical Examples in the
Biographic and Exegetic Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
Nineteenth-Century Christian Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
Criticism of the System of Tradition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
The Christian Biographers of the Prophet . . . . . . . . . . . . . . . . . . . . . . . . . . 372
Separate Studies in History and Interpretation . . . . . . . . . . . . . . . . . . . . . 380
The Exegesis of the Koran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
The History of the Text of the Koran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
The Consonantal Errors of the Uthmnic Text . . . . . . . . . . . . . . . . . . . . . 389
Variants of the Uthmnic Copies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
The Orthography and Its Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
The Cardinal Orthographic Peculiarities of the Uthmnic Text . . . . 408
The Orthography of Agnes S. Lewis Sinai Palimpsests . . . . . . . . . . . . . 426
The Non-Uthmnic Orthographic Variants and Readings: The
Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
The Text of Ibn Masd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
The Text of Ubayy b. Kab. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
The Text of the Lewis Palimpsests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
The Alleged Syriac Translation of a Non-Uthmnic Text of the
Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
The Victory of the Uthmnic Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461
The Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
Basic Problems: The Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
The Connection with the Consonantal Text . . . . . . . . . . . . . . . . . . . . . . . . 472
Linguistic Accuracy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
The Principle of Tradition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
The Principle of Majority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
x contents
Standardization of Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488
Koranic Teaching and the Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . 493
Criticism of the Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498
Orthodox Teaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
The Readers and the Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
Survey of the Older Readers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507
The Historical Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
The Systems of the Seven, the Ten, and the Fourteen Readings . . . . 529
The Characteristics of the Canonical Variant Readings and Their
Distinctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
Literature of the Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
The Earliest Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
The Rise of the Canonical Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . 550
The Rise of the Classic System of the Seven . . . . . . . . . . . . . . . . . . . . . . . . 552
The Extension of the System of the Seven . . . . . . . . . . . . . . . . . . . . . . . . . . 561
The Literature of the Uncanonical Readings (shawdhdh) . . . . . . . . . 565
Writings on the Individual Variant Readings (mufradt) . . . . . . . . . . . 566
The Writings on Tajwd. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Writings on Pause in Koranic Recitation (waqf) . . . . . . . . . . . . . . . . . . . . 570
Writings on the Enumeration of Verses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573
Writings on Koranic Orthography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574
The Commentaries on the Koran as a Source of the Science of
Variant Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575
Manuscripts of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585
The State of Manuscript Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585
The Script of the Older Korans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 586
The Provision of the Koran with Reading Aids, Verse Dividers,
and Names of the Sras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
Verse Dividers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
Names of the Sras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592
The Dating and Provenance of Manuscripts . . . . . . . . . . . . . . . . . . . . . . . . 601
New Editions of the Koran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603
Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 605
contents xi
Bibliography and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613
Printed Books and Articles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613
Select List of Manuscripts Arranged by the Numbers of the
Respective Catalogues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645
PREFACE TO THIS EDITION
The Geschichte des Qorns was originally Nldekes doctoral thesis, submit-
ted in 1860 to the Universitt Gttingen. It is burdened by over-documenta-
tion in the footnotes, which amount to well over 3,110. He became the most
successful scholar to work out a chronology of the Koran. This attempt at
a chronology was followed in general until the present day by three schol-
ars: Alfred Guillaumewho in 1955 still considered Nldeke indispensable
to critical studyRgis Blachre, and Montgomery Watt, the translators
sometime teacher at Toronto.
The present History of the Qurn is a translation of the second edition
of Geschichte des Qorns, and which was dedicated to Ignaz Goldziher and
Christiaan Snouck Hurgronje as well as Frau Marga Bergstrer. The fourth
reprint of the Leipzig edition, 19091938, appeared in 2000.
In 1909, Nldekes eyesight had deteriorated to the point that he could no
longer consider large scale academic work. Insuch circumstances, Friedrich
Schwally, Nldekes former student and friend, felt morally obliged and aca-
demically honouredtostart the updating andcompletionof Nldekes study
in accordance with the wishes of his publisher. It then took three scholars to
complete the second edition. Two died, one after the other, before the work
was completed: Schwally was one of the casualties of the Anglo-American
starvation blockade and expired on 5 February 1919, and Bergstrer died in
a mountaineering accident on Mount Watzmann on 16 August 1933.
Pretzl was largely destined to abandon his own plans and finish Berg-
strers work, with which, of course, he had been largely familiar, since he
had witnessed its genesis and growth. He became the heir to the project
but was not its originator. He considered it a duty to continue and pre-
serve this heritage with which he came to identify himself. Although he
came to realize that his pet project, the science of qirt, was of secondary
importance when viewed in proper perspective, he became so interested in
this dry subject that early in his research he took lessons in Koranic read-
ing from a Turkish muqri. He became unbelievably competent in the most
varied minutae of the practical aspects of the qirt, so much so that he
once even astonished an old Damascene muqri with this particular compe-
tence. Anton Spitaler, one of Pretzls students, was a witness to the surprise
and admiration which Pretzls proficiency caused at this muqris reception.
This competence in a field which is the innate domain of Muslims was
xiv preface to this edition
undoubtedly one reason why Pretzl was a favourite in Muslim circles. In no
time was he able to winthe hearts of all, fromthe peevishand sceptical fi
of a mosque library to the most inaccessible director of a museum. When-
ever he returned to Istanbul he was a welcome guest of dignified sheikhs
and scholars. But Otto Pretzl became the only author to see the complete
History of the Qurn in print. He died in an aircraft crash in 1941 while on
military duty in Germany.
1
Germanwritings aroundthe eve of the twentiethcentury are a nightmare
for a translator. For example, Bergstrers need to accentuate the point he
is making by spaced type is a reflection on the kind of style the reader has
to wade through. In the original, some of Schwallys paragraphs run up to a
solid five pages. An attempt has been made to break up paragraphs of more
than one page in length, but this has not always been possible.
It also makes rough reading when, for example, several lines of refer-
ences are wedged in (p. 150 footnote 188) between subject and predicate
in the footnotes. The over-use of spaced type, exclamation marks (on one
occasion there are 17 on 14 pages), and the superlative are signs of stylistic
weakness.
Wherever possible the references to Germanwritings have beenreplaced
by English translations that have appeared over the years. This applies par-
ticularly to Ignaz Goldzihers monographs, but also to the writings of Adam
Mez, Fuat Sezgin, Aloys Sprenger, and Jan A. Wensinck. Conversely, the
English originals have been usedindicating volume and pagewhere
Nldekes original edition included German translations, namely the writ-
ings of J.L. Burckhardt, Richard Burton, E.W. Lane, Adam Mez, Wm. Muir,
E.H. Palmer, and George Sale.
Some of the Arabic texts are fully vocalized in the German edition. Most
of such texts appear here either without vowel marks at all or with the
vocalization reduced to the decisive vowel.
The translations from the Koran are normally those of Arthur Arberry,
but occasionally, for the sake of harmonizing with the context, they have
been modified or replaced by some other rendering. The translator is much
obliged to Mrs. Anna Evans, Professor Arberrys daughter, for permitting the
quotations from her fathers work.
Muslim personal names are listed under their first element. If, however,
a Muslim writer is not generally known by the first part of his name, that
1
Anton Spitaler, Otto Pretzl, 18931941; ein Nachruf. Zeitschrift der Deutschen Morgen-
lndischen Gesellschaft, 99 (1942), 161170.
preface to this edition xv
partor those partsof the name by which he is generally known in
Anglo-American writings appears in small CAPITALS.
To facilitate the location of a passage in the German original, in the
left margin reference is made to the first new paragraph of a page of the
separately paginated three parts of the German text, but as some German
paragraphs run over several pages, there are sometimes long gaps between
these references. However, as an aid, newdivisions have been introduced at
the discretion of the translator, wherever this was possible.
Thanks are also due to Jeremy Kurzyniec and Stewart Moore for their
meticulous copy-editing and for Ali Rida Rizek for checking the Arabic text.
W.H. Behn
NLDEKE
[PREFACE TO THE GERMAN EDITION, PART 1, PP. VIIVIII]
Inthe year 1898 the honourable publisher surprised me withaninquiry as to
whether I would be prepared to produce a second edition of my Geschichte
des Qorns, or, in case of a negative reply, I could suggest a suitable scholar
for the task. For a number of reasons I was unable to oblige to produce
such a revision in the formwhich would somehowsatisfy myself. After brief
considerations I suggestedmy oldstudent andfriend, Professor Schwally, for
the enterprise; and he obligingly agreed. The book which I had completed
half a century earlier in a rush, he brought up to current requirements as far
as this was possible. I purposely say as far as possible because the traces of
youthful boldness could not be entirely obliterated without resulting in an
entirely different work. Many a thing that I had presented with more or less
certainty later turned out to be rather dubious.
My personal copy contained haphazard unimportant notes which
Schwally was free to use. The result, which is nowpresented in printed form,
I proof-read once. In so doing I made all sorts of marginal notes but left it to
him whether or not to incorporate them. I did not check every detail, and
by no means did I make researches as if it had been my own rewritten text.
Thus, the second edition has the advantage of being the result of two schol-
ars researches, but also the disadvantage that the responsibility is divided
between the two of them.
It seems doubtful that I shall be able to proof-read the second part since
my failing eyesight is making reading increasingly difficult.
Herrenalb (Wrttemberg), August, 1909.
Th. Nldeke
SCHWALLY
[PREFACE TO THE GERMAN EDITION, PART 1, PP. IXX]
When I was entrusted with the honourable task of preparing a second edi-
tion of Th. Nldekes Geschichte des Qorns I did not doubt for a moment
that the new edition of this book, which in the world of learning was con-
sidered a standard work, must proceed with great care. Although it would
have been far easier to produce a new book by making use of the first edi-
tion, I did not consider myself justified to do so. Rather, I attempted to bring
the text up to the current state of research by making as few changes as
possible. Only when such means failed did I decide on radical changes or
extensive additions. In spite of this conservative procedure the volume of
the present first part has grown by five sheets of paper. In viewof this work-
ing procedure it turned out to be impossible to indicate changes from the
first edition.
Nearly all the discussions regarding Muir, Sprenger, and Weil I retained.
Even if the view of these scholars is now largely outdated, their research is
of lasting importance. Relatively fewworks on the genesis of the Koran have
appeared during the last four decades. The number of valuable publications
is evenfewer. If anything substantial has beenomittedthis is purely acciden-
tal.
In general, the Arabic works of tradition are quoted according to books,
chapters and paragraphs respectively. Whenever passages were too volumi-
nous, references to volume and page number of a certain edition had to be
added. In the literature of adth there is regrettably no counterpart to the
established pagination of the Talmud.
For advice and corrections I am grateful to Th. Nldeke, my dear teacher
and author of the first edition. I am much obliged to the two scholars to
whom this work could be dedicated, my highly esteemed friend Professor
Dr. Ignaz Goldziher of Budapest, and Staatsraad Professor Dr. C. Snouck
Hurgronje of Leiden. Only after my manuscript had been completed, and
uponmy request, did Th. Nldeke and I. Goldziher make their private copies
of the book available to me for a few days.
The Preuische Akademie der Wissenschaften, and the Ministry of State
of the Grand Duchy of Hesse enabled me with their financial support to
conduct research in Cairo, the very centre of Muslim learning, for which I
here take the opportunity to express my most respectful thanks.
xx schwally [preface to the german edition, part 1, pp. ixx]
The work on this project dragged on for a long time because under the
pressure of other literary commitments as well as anextensive teaching load
I was able to pursue Koranic studies only with lengthy interruptions. The
supervision of the printing, which had commenced in the Spring of 1908, I
had to interrupt for half a year because of most urgent research in Turkey.
The second section of the work, including the literary introduction, is
scheduled to appear next year. The preliminaries for the third section came
to a halt at an important moment since it has not been possible for me to
study the old manuscripts of the Koran at the libraries of Paris, London, and
Petersburg. During last years visit to Constantinople none of these codices
was accessible to me. Still, I am optimistic that also those treasures, which
have been too well-guarded for too long, will be made accessible to me.
Giessen, August 27, 1909
Fr. Schwally
ZIMMERN
[PREFACE TO THE GERMAN EDITION, PART 2, PP. IIIIV]
On 5 February of this year the author of the second section of the book,
my dear brother-in-law, Friedrich Schwally, expired in his fifty-sixth year
as one of the many casualties of the Anglo-American starvation blockade
to which his frail health finally succumbed. Until the last weeks before his
death he was strenuously busy with the completion of the manuscript of
the Geschichte des Qorns. Thus, at the time of his death, the manuscript
of the second part of this work was nearly ready to be sent to the printers.
In such circumstances it could easily be seen through the press also by a
Semitic scholar who, like the present writer, was not anArabist. This taskwas
supported by my local colleague, August Fischer, who kindly agreed to share
inthe professional proof-reading andrevision, thus guaranteeing the proper
and consistent choice and romanization of Arabic names and book titles
which had not been entirely completed in the manuscript. August Fischer
also supplied some additions which led to the correction of some actual
errors as well as some references to important works which have appeared
recently.
The present second part of the History of the Koran together with its
literary-historical supplement is not only more voluminous than the cor-
responding second part and the literary introduction of the first edition
of Nldekes edition because Schwally went much further and considered
a wealth of new source material and the important advances in the field
during the last sixty years. In comparison to the first part, this second
part changed to such an extent that, as he repeatedly emphasized, little of
Nldekes original work remained. This second volume constitutes largely
Schwallys own contribution. As pointed out above, Schwallys manuscript
could essentially be sent straight to the printers without textual changes.
The final part of the literary-historical supplement, whichtreats more recent
Christian investigations, has been taken over without changes, although,
unlike the rest of the work, there existed no final draft, and it is likely
that Schwally would have made minor changes and possibly additions. This
applies in particular to additional translations of the Koran as well as the
corrections at the end, which have been incorporated in the present text.
There are indications in the manuscript to supplementary information but,
unfortunately, this could no longer be discovered.
xxii zimmern [preface to the german edition, part 2, pp. iiiiv]
For Schwallys third part, reading variants of the Koran only prelimi-
nary notes were found among his papers, nothing ready for the printers.
Upon my request, Gotthelf Bergstrer, Schwallys successor to the chair
at the Universitt Knigsberg, kindly agreed to look after this third part
by utilizing Schwallys remaining materialfor which, incidentally, he had
already completed all sorts of preliminary investigation during his stay at
Constantinopleas soon as his other literary commitments would permit.
There is, thus, a good chance that a new edition of Theodor Nldekes
excellent first publication does not remain unfinished so that it can be
presented to the public within a short time. May our respected old master
live long enough to see this.
Leipzig, September, 1919
Heinrich Zimmern
FISCHER
[PREFACE TO THE GERMAN EDITION, PART 3, PP. 220224]
Obviously this constitutes nothing but additions and corrections, which
would certainly have met with Schwallys approval. There are, first of all,
those that he would have most likely made himself if he had been granted to
put the final touches on his manuscript, and then see it through the printers
himself. Then there are such that he would have readily accepted when
suggested by competent outsiders. With the consent of the publisher, the
type of additions and corrections which required only minor changes in the
type I tacitly made directly on the proofs.
[The four pages of additions that follow here have been incorporated
directly in the English text.]
August Fischer
PRETZL
[PREFACE TO THE GERMAN EDITION, PART 3, PP. VIIIX]
Sixty-seven years have passed since the first edition of this work was pub-
lished. Death swept away two scholars who had been entrusted with the
production of the second edition. Friedrich Schwally died on 5 February,
1919, having worked on the first two volumes until he breathed his last. Got-
thelf Bergstrer thencontinuedwiththe thirdvolume. He was still working
on the third and final instalment of the History of the Text of the Koran
when he died on 16 August, 1933. The first two instalments had been pub-
lished already in 1926 and 1929 respectively. The reason for the long delay
of the remaining part was the accumulated bulk of unknown manuscript
source material which had been unearthed on his initiative. It thus hap-
pened that Bergstrer spent the last years of his busy life doing preliminary
work without being able to make use of it himself. As his colleague it fell
to me to complete the work. In 1929, he himself saw the text through the
printers down to page 173. For the continuation of the part on the historical
development I discovered an outline among his papersas main source
for this served a nearly complete edition of Ibn al-Jazar, abaqt al-qurr,
authored by him. It was pure accident that we had agreed ona mutual work-
ing scheme, where I became responsible for the literature of the reading
variants according to manuscript sources. This was intended to constitute
the main part of the third instalment. Precisely on this subject I had just
then published a lengthy paper entitled Die Wissenschaft der Koranlesung
(ilm al-qirah), ihre literarischen Quellen und ihre Aussprachegrundlagen
(ul.) [the science of qirah, its literary sources and principles of pronun-
ciation.] Ever since, additional visits to libraries enabled me to discover and
photograph newmaterial. The results have been used mainly in the chapter
on manuscripts of the Koran.
Bergstrer left valuable material on variant readings of the Koran nearly
ready togotothe printers, viz. a complete collectionof canonical variants, to
which he intended to add a complete list of the uncanonical readings. For
his two editions of the shawdhdh works of Ibn Jinn and Ibn Khlawayh
he had completed the preliminary research, without leaving written notes.
I could not get myself to include Bergstrers completed collection of the
canonical variants without the far more important uncanonical ones. I think
that I can justify this change of his plan since the arbitrarily selected reading
xxvi pretzl [preface to the german edition, part 3, pp. viiix]
of the Sevenconsiderably limited by traditionis of too little interest. An
edition of the shawdhdh would have unduly postponed the publication of
the work, and this particularly since the most comprehensive and impor-
tant source material became known only after the death of Bergstrer. I
hope that my presentation of the main differences of the pronunciation
of the readers of the Koran (qurr)with special reference to the general
tajwd-rules of pronunciationsatisfies the most urgent need of scholar-
ship. The realization of Bergstrers plan is intended to fit the framework
of his projected apparatus criticus to the Koran.
I am much obliged to Dr. Anneliese Gottschalk-Baur for her detailed
indexes to the three parts which she competently compiled.
I acknowledge with thanks the generous support from the Bayerische
Akademie der Wissenschaften which enabled me to complete my prede-
cessors work. This assistance enabled me to study the much neglected field
of the variant readings of the Koran at libraries in Europe and the Orient,
and accumulate a large photographic archive of manuscripts of the Koran
as well as Koran-related works. For this purpose I received considerable
funds for the acquisition of photographs from the Einjahrhundert-Stiftung
[centenary foundation] of the Universitt Mnchen as well as fromthe Uni-
versittsgesellschaft Mnchen.
I was much encouraged to carry on the difficult task of my great prede-
cessors, in particular the kindly promised and readily granted help of Herrn
Geheimen Hofrat Dr. A. Fischer, Leipzig. He read the proofs and corrected
many a mistake and obscurity. Prof. A. Jeffery, Cairo, obligingly shared with
me and my predecessor the treasure of his own Koranic researches. My sin-
cere thanks are due to these gentlemen as well as to my assistant, Dr. Anton
Spitaler, who was most helpful and displayed great understanding and inde-
fatigable diligence when it came to reading the proofs.
Since I can no longer thank the scholar to whomI amindebted most, my
respected and unforgettable teacher, Gotthelf Bergstrer, I am asking his
wife to accept the dedication of the book, in the completion of which she
participated like no one else as the partner of her husband.
Mnchen, 26 January 1937
Dr. Otto Pretzl
THE ORIGIN OF THE KORAN
The Notion of Prophecy
Although it is beyond question that something resembling prophecy ap-
peared at various times among many different peoples, it was only among
the Israelites
1
that prophethood developed from very primitive origins into
a force influencing the entire realm of religion and state. The essence of a
prophet is that his mind becomes so filled and taken by a religious idea
that he ultimately feels compelled, as though driven by a divine force,
to announce that idea to his peers as a God-given truth.
2
Why prophecy
appeared particularly among this people, and what influence it exerted in
turn upon their history, we cannot discuss in detail at this point.
3
While the
prophetic movement receded in Judaism, it never vanished altogether, as is
evident from the various so-called false messiahs and prophets of Roman
times. Jesus of Nazareth wanted to be more than a prophet. He felt that he
was the promised messiah of Israels prophets and the founder of a new
religion of the heart and sentiment. Indeed he knew how to instil in his
community the belief that he, as the Son of God and Lord of the faithful,
would enter into the glory of the Father in spite of his martyrdomand death.
Among the original Christian communities, too, the prophetic spirit beat its
wings, although after the decline of Montanism it was forced to retreat to
the most distant corners of obscure sects.
The most powerful prophetic movement recordedsubsequently by eccle- [i/2]
siastic history arose suddenly and unexpectedly on the outermost fringe
of Christian missionary activity, in the immediate vicinity of the Kaba of
1
The ancient Arabian khin, pl. kuhhn (soothsayers) are likely to have been familiar
with a similar phenomenon, but we know too little about them. Let us state here that all the
other Semitic languages derive their expression for prophet from the Hebrew .
2
Prophecy in its broadest sense is thus a divine art. Yet as soon as an attempt is made
to teach it in schools or pass it on, and organize prophets in guilds, it quickly becomes
reduced to a profession. Typical of the essence of the genuine prophet is Amos, 7:14: I am
no prophet (by class), neither a prophets son, but an herdsman, and a gatherer of sycamore
fruit: And Yahweh took me as I followed the flock, and said unto me, Go, prophesy to my
people Israel.
3
Cf. thereon Heinrich Ewalds introduction to Propheten des Alten Bundes.
2 the origin of the koran
Mecca, the central sanctuary of the pagan Arabs. That Muammad was a
true prophet
4
must be conceded if one considers his character carefully and
without prejudice, and properly interprets the notion of prophethood. One
could perhaps object that the main tenets of his teaching are not the prod-
uct of his own mind but rather originate from Jews and Christians. While
the best parts of Islam certainly do have this origin, the way Muammad
utilized these precursors spiritually, how he considered them a revelation
descended from God, destined to be preached to all mankind, shows him
to be a true prophet. Indeed, if possessing entirely new and unprecedented
ideas were the only prerequisite of prophethood, would not then every last
man of God and founder of a religion be denied the title of prophet? On the
contrary, we must recognize the fervour of prophethood, frequently border-
ing on fanaticism, in Muammads receiving those external ideas, carrying
themwith himduring his long solitude, and allowing themto influence and
shape his own thinking until, at long last, his decisive inner voice obliged
him to face his countrymen and attempt to convert them, despite danger
and ridicule.
The Type of Muammads Prophetic Endowments
The more one becomes acquainted with the best biographies of Muam- [i/3]
mad, as well as with the uncorrupted source for our knowledge of his mind,
the Koran, the more one becomes convinced that Muammad sincerely
believed in the truth of his mission to replace the false idolatry of the
Arabs
5
with a higher, soul-saving religion. How else could he have preached
so fervently in the Koran against the deniers, whom he threatened with
the most horrible torments of hell, confessing that he himself would have
suffered divine chastisement if he had not proclaimed the complete reve-
lation?
6
How could so many noble and sensible Muslims, particularly his
4
This is the view of writers of the recent past like Henri de Boulainvilliers, Vie de
Mahomed (1730); J. von Hammer-Purgstall, Gemldesaal der Lebensbeschreibungen, vol. 1;
Thomas Carlyle, On heroes, hero-worship and heroic in history (1840); Aloys Sprenger, Life of
Mohammad (1851); Ernest Renan, Mahomet et les origines de l islamisme (1851); cf. now
Chr. Snouck-Hurgronje, Une nouvelle biographie de Mohammed (1894).
5
The Meccans were insulted not so much by the new message per se but by the accom-
panying attack on their ancestors. They venerated their ancient gods, but without true faith,
their cult was sacred only in so far as it was taken over fromtheir ancestors and, like all other
traditions, mere superstitio.
6
Sras 5:71, 6:15, 10:16, 39:15.
the origin of the koran 3
close friends Ab Bakr al-iddq
7
and Umar b. al-Khab,
8
have stood by
him in good days and bad if he had been nothing but an impostor? Added
to the testament of such a numerous following is especially the fact that
menfromnotedfamilies, raisedinthe pedigree-arrogance of the thoroughly
aristocratic Arab, joined a sect consisting largely of slaves, freedmen, and
individuals fromthe lowest strata of society, even though their countrymen
considered this to be the greatest shame, solely because of their enthusiasm
for the Prophet and his teaching. Furthermore, there is the fact, which the
Muslims naturally tried to hide, that Muammad was by nature a soft, even
fearful person who initially did not dare to make public appearances. His
inner voice, however, allowed him no peace. He was compelled to preach
and, whenever he felt discouraged, to rally his spirits in the face of the
ridicule and insults of his early friends.
9
Muammads mind, however, suffered from two serious impediments [i/4]
that affected his authority. If prophecy in general originates from excited
fantasy anddirect impulse of feeling rather thanfromspeculative reasoning,
it is the latter that Muammad was lacking. Although endowed with great
practical acumen, without which he would have never been triumphant
over his enemies, he was almost totally incapable of logical abstraction.
As a consequence, he regarded whatever moved his inner self as coming
from external, heavenly sources. He never questioned his belief, relying
on his instinct as it led here and there, for it was precisely this instinct
that he considered to be the voice of God, destined uniquely for him. The
superficial, literal interpretation of the revelation, which forms the basis of
Islam, follows from this.
Connected with this is the fact Muammad presented those sras that he
clearly produced with conscious effort, using foreign stories, as the first fruit
of his excited mind and a real, divine message. This reproach, however, can
equally be heaped upon the Israelite prophets who presented their literary
products as the words of Yahweh Sabaoth. Yet in general such claims, here
or elsewhere, are not made purposely to deceive but rather follow from a
nave belief. Prophets, after all, are the mediumof the deity not only in their
7
EI
2
; EQ; G. Juynboll, Encyclopedia of canonical adth, p. 460, col. 1.
8
EI
2
; EQ; Juynboll, Encyclopedia, p. xxsqq.; Sezgin, GAS, vol. 8, p. 171, v. 9, p. 16 & 31.
9
Of course we cannot trust all reports of persecutions he suffered before his emigration.
There is little likelihood that his enemies at any time could resort to bodily maltreatment
because the honour of his protectors and all the Ban Hshim, believing and disbelieving
alike, wouldhave demandedrevenge. Alsothe reports concerning Muammads unprotected
followers are certainly an exaggeration.
state of ecstasy; all of their thoughts andactions canappear tothemtobe the
direct emanation from the Divine Essence. In spite of thisas we shall see
later
10
Muammaddidnot intendevery revelationto be proclaimedinthe
Koran, nor indeed did he present all of his communications as revelations.
Since Muammad was unable to distinguish precisely between religious [i/5]
andmundane matters, he frequently usedthe authority of the Korantoissue
ordinances that are not at all related to religion. When reviewing these facts
it must not be overlooked that at that time religionand the social order were
closely connected, andthat by involving Godinthe most humanaffairs daily
life thus became elevated to a higher, divine sphere.
The nave thinker that he was, Muammad was forced to consider per-
missible everything that didnot blatantly conflict withthe voice of his heart.
Since he was not endowed with an acute and robust perception of good and
evilwhich alone can save a person walking in the heights of humanity
from the most suspicious lapseshe did not hesitate to use reprehensible
means, even pious fraud,
11
to spread his belief. While Muslimwriters tend to
hide these traits, European biographers of the Prophet are easily shocked at
one moral indignation after another. Both conceptions are equally unhis-
torical. It would be a miracle if prophets were without blemish and sin,
particularly in the case of Muammad, who was at the same time a mili-
tary leader and statesman. If we knew as much of the private lives of other
prophets as we know of Muammad, some of them would be less exalted
than they now appear to be on the basis of the fragmentary surviving liter-
ature, endlessly sifted through throughout the centuries. Muammad was
no saint, and did not aspire to be one (47:21; 48:2, etc.) We will hardly ever
be able to tell for sure how much of our criticism owes to the semi-barbaric
conditions of the time, to his good faith, or to the weakness of his charac-
ter. The central point is that until he breathed his last he was struggling for
his God, for the salvation of his peopleeven all of humanityand that he
never lost faith in his divine mission.
10
In the chapter on Muammads uncanonical promulgations.
11
A. Sprenger rightfully says in his Life of Mohammad, p. 124sqq. enthusiasm, in its
progress, remains as rarely free fromfraud, as fire fromsmoke; andmenwiththe most sincere
conviction of the sacredness of their cause are most prone to commit pious frauds. This
applies not only to the field of religion, but equally to political and other affairs.
Jewish and Christian Influences
The principal source of the revelations was undoubtedly Jewish scripture, a [i/6]
source, according to the rude faith of the Muslims, as well as to the entire
Middle Ages and even a fewof our contemporaries, literally infused into the
prophets. Muammads entire doctrine carries already in its first sras the
obvious traces of this origin. It would be superfluous to explain here that
not only most of the histories of the prophets in the Koran but also many
of the dogmas and laws are of Jewish origin.
12
In comparison, the influence
of the Gospels on the Koran is much slighter.
13
A closer investigation of
the apparent Jewish and Christian elements in the Koran will lead to the
conclusion that the primary elements shared by Christianity and Islam are
of Jewish colouring. For example, the familiar Muslim creed,
is
derived from a Jewish formula; verse II Samuel 22:33 = Psalms 18:32
and appears in the Targum as and also in the
Syriac Peshitta .
This is not to say that all Jewish elements can be traced back to Jewish [i/7]
authorities. Jews were, of course, numerous in several regions of Arabia,
particularly in the vicinity of Yathrib, which had multiple connections with
Muammads place of birth; they must have visited Mecca frequently as
well. By the same token, Oriental Christianity ingeneral was strongly infused
with Jewish ideas. In education and edification throughout the ancient
Church the New Testament always took the second place to the Old Tes-
tament. Christianity in particular had experienced a notable proliferation
in the Arabian Peninsula
14
among the tribes of the Byzantine-Persian bor-
der (Kalb, Tayyi, Tankh, Taghlib, Bakr), in the interior among the Tamm,
and in the Yemen, which for a long time had been politically subservient
to Christian Abyssinia. Wherever Christianity had not become established
at least some knowledge of it prevailed. Some of the best-known pre-Islamic
poets, though remaining pagan, display familiarity with Christianity in their
attitude and thinking. We must therefore recognize that apart from Jewish
influence on the Prophet there was also a Christian counterpart. In view of
12
It would be desirable that a profound scholar of ancient Arabia, Islam, and Jewish lit-
erature would continue Abraham Geigers penetrating study, Was hat Mohamed aus dem
Judentum aufgenommen (1833). The Arabic as well as the Jewish sources (Midrashim), are
now of such abundance that the mere reprint of the work (Leipzig, 1902) was totally super-
fluous.
13
Cf. thereonNldeke, Hatte Muammad christliche Lehrer? ZDMG, 12 (1858), 699708.
14
Cf. Julius Wellhausen, Reste arabischen Heiligtums (Berlin, 1897), pp. 234242.
much evidence it must remain to be seen fromwhich source it reached him.
In some instances the Christian origin is beyond doubt. For me this includes
the institution of vigils, several forms of the prayer rite, the designation of
revelationas , whichcanhave beenderivedonly fromChristianAramaic
(furqn in the meaning of redemption, cf. below, the detailed explanation
on p. 25, foot-note 59), then the central importance and notion of the Final
Judgement, and, above all, Jesus superiority over all the prophets.
These facts might lead to the conclusion that Islam is basically a reli-
gion following in the footsteps of Christianity, or, even further, that it is the
manifestation in which Christianity entered Greater Arabia. This combi-
nation would find a welcome support in the verdict of Muammads con-
temporaries. The Arabian disbelievers frequently called his followers Sabi-
ans, who were closely related to certain Christian sects (Mandaeans, Elke-
saites, Hemerobaptists). On the other hand, Muslims consider themselves
to be descendants of the anfs, people who, disillusioned with paganism,
sought fulfilment in Christian and Jewish teachings. Since this name was
also applied to Christian ascetics, this would strongly suggest that the Mus-
lims were particularly close to Christians. The emigration of some followers
of the Prophet to the Christian King of Abyssinia would also fit into that pic-
ture.
There can thus be no doubt that Muammads prime source of informa- [i/8]
tion was not the Bible but uncanonical liturgical and dogmatic literature.
For this reason the Old Testament stories in the Koran are much closer to
Haggadic embellishments than their originals;
15
the New Testament stories
are totally legendary and display some common features with the reports
of the apocryphal Gospels, e.g., sras 3:41 and 43 as well as 19:17 with Evan-
geliumInfanti, cap. 1, EvangeliumThomasi, cap. 2, and Nativity of the Virgin,
chapter 9. The only brief passage of the Koran copied verbatimfromthe Old
Testament is sra 21:105: For We have written in Psalms, The earth shall be
the inheritance of My righteous servants; cf. Psalms, 37:29.
Sra 61:6, however, where Jesus predicts that after him God shall send [i/9]
a messenger whose name shall be Amad,
16
does not refer to any precise
passage in the New Testament.
15
For details cf. A. Geiger, Was hat Mohamed aus demJudentumaufgenommen.
16
According to the most likely guess, Muammad took this to refer to himself, and with
allusion to his name
, the promised messenger, . Cf. Ibn Sad (abaqt) Biographie

Muhammads bis zur Flucht, p. 64sq. It is knownthat sra 61:6 has beenreferredto as evidence
that Muammadallegedly hadreadthe Bible. Marraccis idea (Pronomi adrefutationem, Alco-
rani, vol. 1, p. 27, and the note on sra 61:6) to take to mean and
It is very difficult to determine the form and extent of the religious liter- [i/10]
ature prevalent among the Jews and Christians of the period. The Arabian
Christians, who, by the way, were far less numerous than Sprenger thinks,
were to a large extent rather superficially converted. Caliph Al is supposed
to have remarked regarding one of the tribes among whomChristianity had
translate this by , which would prove the unthinkable, i.e., Muammad knew Greek, is
modified by both Sprenger, Life of Mohammed, p. 97, note 1, Das Leben, vol. 1, p. 158, and Muir,
Life of Mahomet, vol. 1 (1858), p. 17, because ina contemporary Arabic translationof the Gospel
of St. John is translated by . But also this is wrong. Such a corruption of the
text cannot be explained by no matter what; this cannot be documented either in a Syriac or
Arabic transmission. Rather, the different forms of the name Paraclete encountered among
Muslims pretty much all correspond to with or without the of the Aramaic stat.
emphat. (Marracci, loc. cit.; al-Shahrastn, vol. 1, p. 167; Ibn Hishm [EI
2
; EQ; F. Sezgin, GAS,
vol. 1, pp. 297299,] p. 150). If Ibn Hishm, loc. cit., on the basis of the Gospel of St. John 15:26
lists also as a name of Muammad, this is nothing but the usual translation of -
mnahmn inthe dialect of ChristianPalestinianAramaic (cf. Schwally, Idioticon) which,
misled by the outward sound, he erroneously associates with
. In Talmud and Midrash the

name of the Jewish Messiah is not uncommon, cf. J. Levy, Neuhebrisches Wrterbuch,
vol. 3, p. 153; G. Rsch, Die Namen, p. 439. Mani, the founder of Manichaeism, also consid-
ered himself a Paraclete; cf. G. Flgel, Mn, pp. 51, 64, and 162sq.; Eusebii Histori, libri vii,
p. 31; Efrem, ed. Rom. II, 487. Incidentally, there are still other Aramaic names for the Prophet,
e.g. , i.e =
, like i.e. = (cf. Diyrbakr, Tarkh

al-khams, vol. 1, 206, and Goldziher, ber muhammedanische Polemik, p. 374). Sprenger
(Leben und die Lehre, vol. 1, pp. 155162) carried the above hypothesis regarding Amad even
further, claiming that also Muammad was not really the name of the Prophet but a by-
name, which he adopted only at Medina, conforming to Jewish diction and belief to appear
as the promised and anticipated Messiah. But all the arguments supporting this view and
advanced by Sprenger himselfand afterwards by H. Hirschfeld (New researches, pp. 23sq.
and 139), Fr. H.C. Bethge (Ramn et Amad, p. 53sq.), and L. Caetani (Annali dellislam,
vol. 1, p. 151)are invalid because: (1) Muammad always appears as the proper name of
the Prophet in the entire ancient historical tradition, and in indubitable genuine documents
like the Constitution of Medina (Ibn Hishm., pp. 341 sqq.), the Pact of al-udaybiyya (Ibn
Hishm, 747), the diplomatic correspondence with the Arabian tribes (Wellhausen, Skizzen
und Vorarbeiten, Heft 4), and lastly the Koran. (2) If the name had been originally an epithet,
andthe fact that it does not evenonce occur withthe article, wouldbe difficult tounderstand,
despite Sprenger, vol. 3, p. 31, n. 2. (3) The Jewish Messiah never had a name derived fromthe
verb to long for; the Messianic interpretation of passages like Haggai 2:7 and Canticles,
2:3 is pure invention. (4) Muammad had been a common masculine name in Arabia even
before the rise of Islam. Ibn Sad (al-abaqt, vol. 1, part 1:) Biographie Muhammads bis zur
Flucht, p. 111 sqq., IbnQutayba, Wstenfelded., p. 276, andIbnRustah, Kitbal-Alk an-nafsa,
p. 194, quote three men with this name, Ibn Durayd lists five, adding that he found fifteen in
another source entitled . There is not the least reason to distrust these references. What
on earth ought to have been the reason for the forgery? By the way, the name on
a Greek inscription from Palmyra, anno 425 Seleuc. = 114/115ad (Corpus inscriptiorum Grae-
carum, vol. 3, no. 4500) is to be equated with as the Aramaic text reads in de Vog,
p. 124, l 4.
been best established: The Taghlib are no Christians, and adopted only
their wine-drinking habits.
17
Literacy in Pre-Islamic Arabia
Wherever the missionaries of the two religions of the Book went, it is incon-
ceivable that they should not have carried with themsome kind of religious
literature, be it in Hebrew, Aramaic, Ethiopic, and, now and then, probably
also in Greek. The rabbis and clergy were thus obliged to translate foreign
prayers, liturgies, hymns and homilies into Arabic. Although it is unlikely
that they ever expressed their own theological writings in Arabicas can
be seen from existing Syriac works of ancient Arab clericsit is neverthe-
less conceivable that people beganinpre-Islamic times tofix inwrittenform
those oral Targums. As the art of writing among the Meccans and Medinans
in the age of Muammad was anything but common (cf. below, p. 11 sqq.)
and, as it seems, it was usual to put in writing important correspondence
(e.g., Muammad with the Bedouins) and treaties (al-udaybiyya, Consti-
tution of Medina), it can reasonably be assumed that the Arabic script was
used also to record the productions of poets, singers, and story-tellers. Liter-
ature presupposes ad hoc writing. Sheets of note-paper ( ), panegyrical
songs or satirical poems (cf. Goldziher, introduction to Dwn al-ueja,
p. 18) etc. are likely to have been widely circulated (Ab l-Faraj al-Ifahn,
al-Aghn, vol. 20, p. 24, and vol. 2, p. 16; Hudhayl (al-Sukkar, The Poems of
the Huzailis,
18
) p. 3 n. 4; Jarr al-Mutalammis,
19
p. 2 n. 2; Labd b. Raba,
20
Der
Dwndes Lebd, edited by Chalidi, p 47, l 1; Aws b. ajar,
21
Gedichte, p. 23, l. 9,
etc.). Collections of pre-Islamic authors work, however, are not known.
As far as Muammads relation to Jewish and Christian literatures is con-
cerned, it is undoubtedly safe to say that he had no access to material in
languages other than Arabic, even if only because of his ignorance of for-
eignlanguages. The superstitious fear withwhichthe Jewslong before the
Muslimswere watching the
would not be an insurmount-

able obstacle per se, not to mention that this inviolability by followers of
other faiths applied to canonical books only. But the matter of whether or
17
al-abar, al-Zamakhshar [EI
2
; EQ,] and Bayw, Ansb al-tanzl, on sra 5:7.
18
Ed. in the Arabic from an original manuscript and translated by J.G.L. Kosegarten.
19
Edited by Karl Vollers; EI
2
; EQ; Sezgin, GAS, vol. 2, pp. 173175.
20
EI
2
; Sezgin, GAS, vol. 2, pp. 126127, sqq.
21
Sezgin, GAS, vol. 2, pp. 171172.
not the Prophet was able to readandunderstandwrittenArabic translations
cannot easily be determined from either the Koran or the traditions.
On these points the statements of Muslims differ. Worst of all, those who [i/12]
agree as well as those who disagree are less concerned with the truth of
the matter than with the pursuit of dogmatic or political interests. Both
parties resort to the weapon of fabricated or twisted traditions, a method
that was widespread in early Islam. The Sunnites, generally, tend to think
that Muammad could not read and write; the Shiites hold the opposite
view.
22
The latter consider it beneath the dignity of the Prophetwhom
they consider that he could have been lacking the foundation
of learning. Added to this is the attempt to excuse the pact between Al
and Muwiya (Ibn Ab Sufyn)
23
which they considered offensiveby
the example of the Prophet. At al-udaybiyya he allegedly signed a similar
agreement where, in his own handwriting, he replaced the words (Rasl
Allh) to which the disbelievers objectedand substituted them
with . A different version of the same account merely states that
after Als objection, the Prophet himself crossed out those words and that
Al then inserted the new words. There are still other reports stating that
Al wrote both the new and the previous words.
24
In these circumstances
we cannot hope to settle the matter, particularly when we remember that
the word can refer not only to a persons actually writing but also
to someone elses writing his words, namely dictation. In the letters of
Muammad that have come down to us in Ibn Sad we frequently read

where only dictating canbe meant, whichis made quite clear by the
addition at the end of . Also in Ibn Hishm, loc. cit., particularly in
the account of the peace pact, we read
, where also only the indirect formwriting can be meant. The insertion of
, intentionally or mistakenly, could thus easily be explained as another
distortion of this tradition.
22
A. Sprenger, Life of Mohammad, p. 101, n. 2; and Leben und die Lehre, vol. 2, p. 398,
where he mentions that Muammad b. Muammad IBNNUMN[al-Mufd b. al-Muallim]
d. 413/1022 [EI
2
; Sezgin, GAS, vols, 1, 4, and 8] wrote a monograph to prove that Mohammad
could write.
23
EI
2
; G. Juynboll, Encyclopedia, p. 11 sqq.; Sezgin, GAS, vol. 8, p. 172.
24
Cf. Ibn Hishm, p. 747; abar, Akhbr, vol. 1, 1546; al-Mubarrad, al-Kmil, vol. 1, p. 540;
al-Bukhr,
( ); K. al-Shur, 15; Muslim, a vol. 2, p. 170sq.; (al-

Qasalln, vol. 7, p. 415sqq., K. al-Jihd, 29); Fakhr al-Dn al-Rz on sra 48:25; al-Khab
al-Tibrz, Mishct-ul-mabh; or, Acollection of the most authentic traditions, pp. 345 and 347
= 353 and 355, Bb al-ul); al-Bajr [Sezgin, GAS, vol 1, p. 158,] al-Mawhib al-laduniyya, Bb
al-bayat al-[ri]wn, where this matter is treated in detail.
Another tradition is equally uncertain: On his death bed Muammad [i/13]
allegedly requested a reed and tablet in order to write down something
that was to protect the Muslims from error.
25
But this tradition going back
to Abd Allh IBN ABBS (d. 68/687)
26
is suspicious when compared with
another tradition, openly betraying its bias, in which isha tells the story
that Muammad thus intended to fix in writing the appointment of Ab
Bakr as his successor.
27
It is consequently certain that this entire tradition,
which is not found in Ibn Hishm, was forged to defend Ab Bakrs claimto
succession. But even if this were not the case, also in this instance the words
so that I write can be interpreted as so that I dictate; once again we would
lack a sound argument.
Even the Koran itself does not afford any more certainty, regardless of
howone interprets the frequently occurring verb , particularly the passage
in sra 96:1 and 3. If it simply means to lecture, to preach, it is a priori
irrelevant. If, however, this means to read or to lecture on what has been
read, even this interpretation does not contribute anything towards solving
the problem since it is in the nature of heavenly textswhich are beyond
human speech or writingand is therefore comprehensible only by divine
inspiration.
We can see that the evidence for Muammads ability to read and write [i/14]
is very weak. But what about the evidence generally marshalled to prove
the opposite? The main argument is that in sra 7:156 and 158 Muammad
is called
, words that nearly all commentators take to mean the

Prophet who could neither read nor write. However, when we make a
thorough investigation of all the Koranic passages that contain
we find
that it is used everywhere to mean the opposite of , namely, not a
person capable of writing but the owner (or expert) of the Holy Scripture;
sra 2:73 even says that there were
who have a poor understanding of

the Scriptures. In relation to Muammad, this word must mean that he was
not familiar with ancient divine texts and knew the truth only from divine
inspirationcharacteristics frequently mentioned on other occasions as
25
al-Bukhr, a, Bb mawt al-nab, appendix to , K. al-Ilm, 40; Muslim,
vol. 2, p. 78sq. (al-Qasalln. vol. 7, p. 95sq.,
4); al-Khab al-Tibrz, Mishct-ul-

mabh; or, A collection, p. 540 ( 548); cf. G. Weil, Das Leben Mohammeds, p. 329sq.;
Caussin de Perceval, Essai sur l histoire des Arabes, vol. 3, p. 321.
26
EI
2
; G. Juynboll, Encyclopedia, pp. 12; Sezgin, GAS, vol. 8, pp. 2122.
27
Muslim, vol. 2, p. 457 (al-Qasalln, vol. 9, p. 257, ), and later al-Khab
al-Tibrz, Mishct-ul-mabh; or, A collection, , fal 1, 3. But Ibn Sad, abaqt,
ed. I, IV, [sic] p. , 7, (i.e. .) [This reference cannot be verified.]
well that do not mean that he could not read or write at all.
28
In addition,
Muammad says insra 29:47 that before the revelationof the Koranhe had
not read a book;
29
yet even these words, taken by themselves, are nothing
other than argument in ones own favour. Finally, it is claimed that to the
first revelation, the Angels command to
, he replied with the words

.
30
But even this is of little importance since this entire tradition is
extremely embellished,
31
others insteadreporting his reply as or or
, what am I supposed to read?
32
Both parties, thus, offer nothing but pretences. Worthless, too, are state- [i/15]
ments that Muammad could write, but only little and not well. In a tradi-
tion regarding the first revelation he claims to be unable to read well,
33
and
in the above-mentioned account of the Pact of al-udaybiyya some say
, he did not write well but he nevertheless did write.
34
It is
only too obvious that both variants are the weak attempts of an uncritical
mind to mediate between two contradicting traditions.
Neverthless, biased traditions may also contain a grain of truth. To some
degree it seems quite likely that a man in whose neighbourhood some fifty
people could read and compose notes in writing
35
I know of forty-four,
merely fromIbn Sad, Biographien der medinischen Kmpfer Muhammeds in
der Schlacht bei Bedr, p. 2; Wellhausen, Medina vor dem Islam, p. 105sqq.;
and al-Baldhur,
36
p. 471 sqq.not only understood enough of the craft
28
Umm is derived from ummah, i.e. = = Aramaic lmy. The Jews call people
ignorant of the Scriptures and the Law am h-re. The etymologies which Muslims supply
for umm we can pass in silence. Cf. H.L. Fleischer, Kleinere Schriften, vol. 2, pp. 115sqq.
29
Arberry translates Not before this didst thou recite any Book, or inscribe it with your
right hand
30
Cf. foot-note 32.
31
In comparison, Sprengers interpretation is unsuitable; it purposes to neutralize the
evidence of these words by saying that I am not reading merely means I do not read, but
in no case I cannot read (Life of Mohammad, p. 95, footnote; Das Leben und die Lehre, vol. 1,
p. 332 n. 2.) As it reads in Ibn Hishm, p. 226 l. 14,
, and frequently in traditions,

where it can only refer to the ability to write, in the same way that these words must
be interpreted in the context of the story. And they thus translate the Turkish al-Mawhib
al-laduniyya quite accurately with , nonsumlector (p. 27).
32
Cf. Ibn Hishm, p. 152, and al-abars original, Tafsr, vol. 1, 1150 (cf. A. Sprenger, Notice
of a copy of the fourth volume of the original text of Tabary, p. 115). Others combine both,
like the Persian al-abar ( ), and al-Itqn, p. 53.
33
G. Weil, Das Leben Mohammeds, p. 46, n. 50.
34
al-Bukhr; al-Khab al-Tibrz, Mishct-ul-mabh; p. 347 (p. 355, ); Fakhr
al-Dn al-Rz on sra 48:25, somewhat past the middle.
35
Cf. also Goldziher, Muslimstudies, vol. 1, pp. 105sqq.
36
EI
2
to function as a merchant, dealing with notes, prices, and names, but also
might have attempted to progress further, given his interest in the holy
scriptures of the Jews andChristians. But since we are deprivedof all reliable
information we must be content with the fewimportant certainties, namely
that (1) Muammad himself did not want to be considered literate and
therefore hadothers readthe Koranas well as letters tohim;
37
andthat (2) on
no account had he read the Bible or other important works. Still, Sprenger
at all cost wants to make him a learned man. He considers it a fact
38
that
Muammad read the
, a book on dogmas and legends.

39
Asir
al-awwaln
40
(roughly translated The Legends of Old) was the name applied
by the Quraysh to Muammads edifying but boring stories, exactly as in
the Koran the dites call the sayings of the Prophet Hd
. It is
correct that Sprenger should be looking also for the title of a book. Yet
it would be totally against the habit of the Prophet, who only referred to
his own revelations, to make use of a generally known
41
book and then
pursue a useless defence against the accusation. But even if he did indeed
mean a book, he would have probably not said this is only asr but this
is
. It makes even less sense when Sprenger also declares

as books used by Muammad the (sras 53:37sqq.; 87:19), i.e.,
according to Muammad, the revelation made to Abraham and Moses.
42
37
Cf. al-Wqid, K. al-Maghz, edited by A. von Kremer, p. 202, l. 12sqq.
38
Life of Mohammed, p. 99sq., Leben und die Lehre (Berlin, 1869), vol. 2, p. 390, he only
thinks so. Weil turns against him, Mahomet savait-il lire et crire?
39
For details regarding its origin etc., see Sprenger, Life of Mohammed, p. 99, foot-note 3;
Das Leben und die Lehre, vol. 2, pp. 390397.
40
is considered the plural of a singular () or () (cf. .) The origin of
the word is not clear. One could think of the Syriac , chirographum (Mishnaic
or Sabaic inscription) which must undoubtedly go back to Babylonian sharu,
to write. The same applies to Arabic
, line,

, ruler,
, to write (sras 17:60, 52:2,

68:1, etc. also Sabaic),
(compare the Hebrew designation of a civil servant,

which also originates from Babylonia, , as well as Job 38:33, , which too cannot
be explained with certainty). In a tradition traced back to Ibn Abbs, and transmitted in
al-Suys al-Itqn f ulm al-Qurn, p. 311,
sra 17:60 and are considered

outright Himyarite. I now consider it more likely to be a derivation of but not
the other wordsfrom (Georg W. Freytag in Lexicon Arabico-Latinum; A. Sprenger,
Observations on the physiology, p. 119, and his Das Leben und die Lehre, vol. 2, p. 395;
H.L. Fleischer, loc. cit., vol. 2, p. 119), now more likely; also Th. Nldeke himself has been
suspecting this for quite some time.
41
This must have been this book since the asr al-awwaln are mentioned nine times in
the Koran, and on totally different occasions.
42
Sprenger, Das Leben und die Lehre, vol. 2, p. 367. It is already written in the Talmud
(Abd Zr, 14b) that Abraham knew and followed the Halakah. He is later credited with
the authorship of the Kabbalist book Yezr or at least the lost Liber de idolatria (Johann
He is supposed to have laid open his own sources so carelessly for all the
world to see!
We must therefore continue to resist the belief that Muammad [i/17]
utilized written sources; in contrast, he undoubtedly received the great-
er part of his his dogma by way of oral transmission from Jews and Chris-
tians. The Koran seems to allude to this fact in the following words

(sra 25:5), and

(sra 16:103). The Commenta-

tors onthe latter passage list several contemporaries of the Prophet towhom
the reference applies ( , , [EQ], , and ). These and other
such reports from tradition do not mean very much. Even if there is a ker-
nel of truth in the legend that associates Muammad with a Syrian monk
Bar or Nestorios, such encounters can hardly have been of importance
for his prophetic mission. And no matter howoften Muammad might have
gone to Syriahundreds of his fellow countrymen made this journey year
after yearit was hardly necessary for a pagan Meccan to go to Syria or
Abyssinia, or a Syrian or Abyssinian Christian to come to Mecca, in order
to gain acquaintance with revealed religions. As has been pointed out on
pp. 45, numerous Jews and Christians were living not far away. There must
have been abundant and multifaceted channels through which religious
knowledge reached Muammad. Yet givenhis enthusiastic convictionof his
divine mission, there was only one real source of truthfor him: AllhandHis
Divine Book.
Muammads Relation to Zayd b. Amr and Umayya b. Ab l-alt
Sprenger adds Zayd b. Amr b. Nufayl
43
to the oral sources. According to [i/18]
some reports
44
unfortunately modelled entirely on the Islamic point of
viewZayd b. Amr had denounced the idolatry of the Meccans for some
Albert Fabricius, Codex pseudepigraphus Veteris Testamenti, Hamburg, 1722, vol. 1, p. 400).
In contrast, Epiphanius, Haer. 1, cap. 8, does not speak of books (thus [B.] Hamburger,
Real-encyklopdie, s.v.) but of eight children (liberi) of Abraham.
43
EI
2
; EQ; Sezgin, GAS, vol. 2 (1975), p. 271.
44
See about him Ibn Hishm, p. 145sqq.; al-Bukhr ( ); al-Aghn,
vol. 3, pp. 1517; Ibn Qutayba, al-Marif, p. 29; al-Masd, Les prairies d or, vol. 1, p. 136; and its
English translation, Historical encyclopaedia, by A. Sprenger. Cf. Sprenger, Life of Mohammed,
p. 41 sqq., and his Leben und die Lehre, vol. 1, pp. 8289, pp. 119124; Caussin de Perceval, Essai
sur l histoire des Arabes, vol. 1, p. 323. We must always remember that all this information is
strongly influenced by the endeavour to present Islam as an old divine teaching that existed
before the time of Muammad.
time before Muammads appearance. It is conceivable that Muammad
was stimulated by this man to think about religionpossibly for the first
time. But details are lacking; Sprenger
45
in any case goes too far when he
concludes from Zayds sermon, which in its transmitted form has much in
common with the Koran, that Muammad borrowed from him not only
the dogma but also his expressions. This sermon,
46
however, so obviously
carries the mark of a fabrication of Koranic quotations by a Muslim that we
may ignore it the same way we doZayds false poems, whichare containedin
IbnHishmandinal-Aghn (vol. 3, pp. 1517). It is highly farfetchedthat not
only did Muammad memorize by heart Zayds speeches and insert them
later into the Koran, but also that, concurrently, someone else passed on the
same speeches to posterity.
Clment Huart
47
claims for himself the honour of having discovereda new [i/19]
source of the Koran in certain poems of Umayya b. Ab l-alt.
48
Yet all the
passages he quotes in support of his hypothesis are under strong suspicion
of being forgeries of the Koran. Other similarities can be traced to the fact
that Umayya b. Ab l-alt, like Muammad, drewfromJewish and Christian
sources.
49
Pagan Influence and Muammads Personal
Contribution to the Establishment of His Religion
The ancient belief of Muammads people was not unimportant as a source
of his dogma. No reformer can completely escape the prevailing concepts
with which he grew up. Additionally, the founder of Islam retained many
ancient superstitions (e.g. jinns) and opinions from the
[Jhiliyya]
period. Certain other influences he deliberately retained more or less com-
pletely. He adapted to his dogma the customs at the Kaba and during the
ajj,
50
believing them to be of Abrahamic originwhich, incidentally, was
45
Life of Mohammed, pp. 95 and 98.
46
Life of Mohammed, p. 41; Das Leben und die Lehre, vol. 1, p. 121 sqq.
47
C. Huart, Une nouvelle source du Qorn. The poems are from a fifth century work by
al-Maqdis, edited and translated under the title Livre de la cration et de l histoire [Sezgin,
GAS, vol. 7, pp. 277288].
48
EQ; F. Sezgin, GAS, vol. 2, pp. 298300, vol. 9, p. 277.
49
See now Friedrich Schulthe, Umajja ibn Abi-s-Salt.
50
R.P.A. Dozy inhis Die IsraelitenzuMekka(1864) intendedtoshowthat the Meccansanc-
tuary and the festival were established in the time of David by the Israelites, and particularly
by the tribe of Simeon (after him= Ismaelites = Gorhum). This attempt, however, was a total
failure; cf. C. Snouck Hurgronje, Het Mekkaansche feest. Further, belowonsra 16:24. [Inorder
completely unknown to the ancient Arabs. Some ancient Arab legends,
which we frequently come across in geographic names and old poetry, refer
briefly or in abbreviated forms to d and Thamd, to the
[sayl
al-Arim,
51
] etc. Muammad adopted these but changed themso completely
in accord with his Jewish histories of the prophets that little remained of the
originals.
52
It was out of varied elements that the new religion, destined to shake [i/20]
the world, developed in Muammads inner consciousness.
53
His own posi-
tive contribution was insignificant in comparison with foreign borrowings,
except for the second principle of Islam,
(sra 48:29.) Although

many religious figures of the past (Noah, Israel, Lot, Jethro, Moses, Aaron,
Jesus, Hd, li) are identified as such in the Koran, it was Muammad
who placed himself far above them by claiming the finality of his prophetic
importance (sra 33:40 ) [sic, Seal of the Prophets].
to follow up several of the references in the History of the Koran, considerable parts of Het
Mekkaansche feest by Nldekes student had to be read. The English notes will later serve as
the basis for a total English translation of this study by the translator.]
51
This is an event that fixed itself ineffaceably in the memory of the Arabs, and which is
known in their traditions as the Flood of the Dyke.
52
One of Muammads creations seems to have been the Prophet li, of whom we
find no trace anywhere. Traditions about li in Aloys Sprenger, Leben und die Lehre, vol. 1,
pp. 518525.
53
While the general terms for religion in the Koran, Persian , and Aramaic
, are of
foreign influence, the specific expression , Islm (sras 3:17, 79; 6:125; 39:33; 49:17; 61:7)
is truly Arabic, and probably coined by Muammad himself for his religion. Apart from the
absolute usage of the corresponding verb (fifteen times), we also find the connection
(four times) or with
(four times). D.S. Margoliouths idea (On the origin

and import of the names Muslim and Hanif , p. 467sqq.) that the name Muslim originally
indicated a follower of the prophet Musaylima [EQ; G. Juynboll, Encyclopedia of canonical
adth, pp. 72 and 73] was soon refuted by Ch.J. Lyall (The words Hanif and Muslim,
p. 771 sqq.) Aslama surrender, however, might be an old borrowing from the Aramaic. Cf.
also I. Goldzihers article Islm in the Jewish encyclopedia, vol. 6, p. 651, col. 2.
MUAMMADS REVELATIONS
The Various Types of Muammads Revelations
Muammad claimed to have received his revelations
1
fromthe divine spirit, [i/20]
, (Hebr.) andconsideredit to be anangel
2
who, inthe Medinan
sras, is also called , Gabriel.
3
The revelations, however, did not always
come to the Prophet in identical circumstances. Before enumerating them
in detail, however, we should point out that for Muslims the word way
( ),
4
revelation, does not refer to the Koran alone but to every single
1
O. Pautz in his Muhammeds Lehre von der Offenbarung says a great deal, but does not
even explore all possibilities, nor is his thesis in any way conducive to the matter.
2
Sras 16:104; 26:193sq.; Ibn Sad (al-abaqt al-kabr, vol. 1, part 1): Biographie Muham-
mads bis zur Flucht, p. 125. In the poemby Kab b. Mlik [EI
2
; Sezgin, GAS, vol. 2, pp. 293294]
in Ibn Hishm, Sra, vol. 1, p. 528, l 13, both are found side by side, and .
3
Only sras 2:9192; and 66:4. In adth, however, the Angel is quite important. Muam-
mad probably pronounced Jabrl (identical with the reading of Abd Allh IBN KATHR AL-
DR [EI
2
; EQ]) or more according to Arabic vocalization Jibrl, since this is the form (- -)
most frequently encountered in the poetry of his contemporaries. In a poem on his death in
the notes to Ibn Hishm, p. 219, l 5, however, we find the form more closely corresponding
to the Hebrew rhyme
( .) The identical form occurs outside the rhyme in a poem

from the time of Muwiya [Ibn Ab Sufyn; EI
2
; Juynboll, Encyclopedia of canonical adth;
Sezgin, GAS, vol. 8, p. 172,] and in al-Aghn, Bulaq ed., vol. 13, p. 167, l 27. Al-abar (Tafsr,
vol. 1, p. 328sq.), and al-Bayw (on sra 2:91) supply detailed discussions of the different
vocalizations of the word. Also ulaya had his Gabriel: al-abar, Tafsr, vol. 1, p. 1890, l 13,
and al-Baldhur, Fut al-buldn, p. 96. The earliest references to Gabriels divine mission
are St. Luke 1:19 and Daniel 8:16 as well as 9:21. Unless I am wrong, this is the figure of Nabu,
the Babylonian god.There is no support in the Koran for the opinion that during the first
three years of his prophethood Muammad had anything to do with Seraphel (Isrfl) [EI
2
;
EQ], al-abar, Tafsr, vol. 1, p. 1249, l 4sqq., p. 1255, l 10sqq.; al-Suy, al-Itqn, p. 104.
4
only in sras 11:39; 20:113; 21:46; 23:27; 42:50; and 53:4; the corresponding verb,
, however, is very frequent. The meaning of divine inspiration, revelation can most
readily be traced back to early Arabicwhere it is not infrequently foundto hint or to
give intimations or indications ( [the way of the eyes is their speech, Watt,
Muhammads Mecca, p. 57]; Yaqt, Mujam vol. 3, p. 520, l 7, which also explains sra 11:39 =
23:27; amsa, p. 606 bottom), to encourage someone ( with Pers. [Divans of the six
ancient Arabic poets, ed. by W. Ahlwardt, p. 211]; Alqama, no. 13, l 26; Muslim b. al-Wald,
Diwan, ed. by Michael Jan de Goeje, no. 15, 2; with Acc. Pers. encourage to fight; Yqt,
vol. 4, p. 102, l 14). Fromthis basic meaning even in the pre-Islamic era was applied to the
mysterious and puzzling (al-Maydn, ed. Freytag, Arabumproverbia, cap. 26, no. 90) traces of
the inscriptions [Divans of the sixancient Arabic poets, ed. by W. Ahlwardt]; (Antara, no. 27, l 2;
Muallaqat Labd [Ibn Raba], p. 2; Zuhayr (Ibn Ab Sulm), 15, l 5; 17, l 3, appendix, p. 4, 1); cf.
18 mu
hammads revelations
inspiration of the Prophet, as well as any divine commandment that he
received, evenif these words were never claimedtobe fromthe Koran.
5
Most
of the categories of revelation that Muslims enumerate do not refer at
all to a Koranic revelation.
6
There are conflicting old traditions regarding
the classification of the revelations. Only later were they combined in an
artificial system according to dogmatic considerations. When Muammad
was asked howhe had received the revelations he allegedly replied to isha
that they were accompanied at times by a noise, like that of a bell, which
particularly grabbed him; at other times, he conversed with the angel as
though with a human being so that he easily understood the words.
7
Later
writers, who refer to still other traditions, differentiate between even more
categories.
Inal-Suys al-Itqn, p. 103, the following types of revelationare enumer- [i/22]
ated: (1) Revelations with bells ringing; (2) Inspiration of the divine spirit in
Muammads heart; (3) Gabriel appearing in human form; (4) Revelations
directly from God, either when wide awake, like during the Ascension to
Heaven (mirj), or in a dream. One writer (Imm diq) concurs with this
enumeration, and Sprenger (Life of Mohammad, p. 154) follows him in this
regard.
Yet in al-Qasallns al-Mawhib al-laduniyya
8
these categories ( ) are
listed as follows: (1) In dreams; (2) Gabriels inspiration in the Prophets
heart; (3) Gabriels appearing to him in human form, mostly resembling
a man by the name of Daya (or Diya)
9
b. Khalfa al-Kalb; (4) those
below, pp. 3738; Goldziher, Muslim studies, vol. 2, p. 20; Siegm. Fraenkel, Die aramischen
Fremdwrter, p. 245. This is the origin of the later use as the technical term for scriptio,
al-abar, Tafsr, vol. 3, p. 2524, l 8, see below, Michael J. de Goeje inthe glossary, Lisnal-Arab,
vol. 20, p. 257, l 20sq.
5
Cf. al-Suy, al-Itqn, p. 102. Also the inspirations of Musaylima and ulaya are
considered way, al-abar, Tafsr, vol. 1, p. 1917sq.; al-Bayhaq, ed. Fr. Schwally, p. 33.
6
al-Suy, al-Itqn, p. 104.
7
Mlik b. Anas, al-Muwaa, vol. 1, p. 70; al-Bukhr at the beginning; Kitb Bad al-khalq
5; Muslim, al-Jmi al-a, vol. 2, p. 430 = al-Qasalln, Irshd al-sr; vol. 9, p. 182
; al-Nas, al-Sunan, p. 106 = vol. 1, p. 147sq., Kitb al-Iftit 37; Ibn Sad, vol. 1, part 1:
Biographie Muhammads, p. 131 sq.; al-Khab al-Tibrz, Mishct, vol. 1, p. 514 (522
); al-Tirmidh, al-Jmi al-a, vol. 2, p. 204 ( , bb, 5). Cf. G. Weil, Das Leben
Mohammeds, p. 44; Wm. Muir, The Life of Mahomet, vol. 2, p. 88; A. Sprenger, Das Leben und
die Lehre, vol. 1, p. 272, and generally, pp. 269275.
8
Maqad, 1.
9
Vowelled Diya or Daya. Cf. al-Dhahab (Cod. Lugd., 325). Ibn Durayd, K. al-Ishtiqq,
ed. F. Wstenfeld, p. 316, and al-Nawaw, Tahdhb al-asm, ed. F. Wstenfeld, p. 239; Wilh.
Ahlwardt, Verzeichnis, no. 10169 (Codex Sprenger, 282). Also good manuscripts and Indian
printed books frequently have both vocalizations (e.g., al-Tirmidh, al-Shamil, bb 1).
mu
hammads revelations 19
accompanied by a ringing of bells, or an indistinct sound, etc.; (5) Gabriel in
his true appearance, which he revealed only twice; (6) revelation in heaven,
such as, for example, the commandment of the five daily prayers; (7) God in
person, but veiled ( ); (8) God revealing directly, without veiling.
Other writers are supposed to have added two additional categories: First,
Gabriel in the shape of another human being and, secondly,
10
God in person,
appearing in a dream.
It is quite obvious that many of these categories originate from the inac- [i/23]
curate interpretation of either the traditions or Koranic passages. This
becomes quite evident from the early controversy among Muslims as to
whether Muammad did or did not see God and receive revelations from
Him.
11
isha, showing extreme indignation, allegedly declared those who
maintained this point of view to be godless.
12
This latter opinion prevailed,
even though it was against Muammads view and originated only from
the inaccurate interpretation of some passages in sra 81, and particularly
in sra 53. Other writers tried to soften the rigidity of that view, and con-
cluded fromsra 53:11 that the Prophet had seen God with his heart ( or
).
13
By the same token the other category should also be rejected, in which
Gabriel appears before Muammad in the shape of Daya.
14
Although some
writers say this has happened frequently, or most of the time
15
(
)
this entire view did not develop until after the events of the year 5/626,
when the army mistook Daya (Ibn Khalfa), who was running ahead, for
10
Alsoas a woman( al-abar, Tafsr, vol. 1, p. 1262, l 6sq., al-Tirmidh, Manqib),
and even as a biting camel stallion, Ibn Hishm, p. 191, l 1, cf. p. 258, l 8.
11
Regarding related dogmatic matters cf. al-Qasalln, al-Mawhib al-laduniyya; al-
Khab al-Tibrz, Misct-ul-mabh; or, A collection of the most authentic, p. 493 (501
). Goldziher, Schools of Koranic commentators, pp. 6970.
12
al-Bukhr and al-Tirmidh in on sra 53; al-Bukhr in (35,
and 52); Muslim, a, vol. 1, pp. 127sqq. = al-Qasalln, vol. 2, pp. 96sqq.; Ab l-Layth
al-Samarqand, Tafsr on sra 6:103; cf. A. Sprenger, Life, 122, note 5: Thou hast said a
blasphemy, at which my hair stands.
13
al-Tirmidh, Tafsr; al-Khab al-Tibrz, al-Mishct, 493 (501); al-Mawhib al-laduniya;
al-Bayw on sra 53:11.
14
Cf. thereon al-Wqid, p. 72 (by Wellhausen, Muhammed in Medina, p. 211); Ibn Sad,
Biographiender Muhgirn, p. 184sq.; al-Bukhr in , s.v.
at
the end (ii, 182),
, 1; Muslim in al-Qasalln, vol. 9, p. 333; al-abar, Ab

l-Layth al-Samarqand, and al-Zamakhshar on sra 6:8sq.; Ibn ajar al-Asqaln [EI
2
; EQ],
al-Iba f tamyz al-aba, vol. 1, no. 2378; Ibn al-Athr, Usd al-ghba f marifat al-aba,
vol. 2, p. 130. Several of them mention as an honour to Gabriel that Daya was very beautiful
(cf. sra 19:17, and above, p. 6 n. 16).
15
al-Zamakhshar on sra 6:9.
20 mu
hammads revelations
Gabriel.
16
Additionally, the sixth category developed from the story of the
Ascension to Heaven, while the fifth category stems from a different expla-
nation of sras 81 and 53.
Psychological Stages of Excitement.
The Alleged Mentor of the Prophet. Daya
In contrast, much material regarding the fourth category has survived. It [i/24]
is related that when Muammad received a revelation he had a serious
attack, foaming at the mouth, his head drooping and his face turning pale or
glowing red; he screamed like a young camel; perspiration broke out, even
though it was wintry.
17
This attack, to which we could add still other indi-
cations, al-Bukhr
18
and al-Wqid called a paroxysm of a fever (
); but
Weil (Das Leben Mohammeds, p. 42sqq.) was the first to show that Muam-
mad suffered from a kind of epilepsy, which already the Byzantines had
suspected,
19
although more recent scholars deny this.
20
But since lack of
16
Cf. Ibn Hishm, p. 685; G. Weil, Das Leben Mohammeds, p. 251, footnote; and cf. above,
footnote 14.
17
al-Muwa, 70; Ibn Hishm, Sra, p. 736; al-Wqid, History, p. 322; Ibn Sad, vol. 1, part 1,
Biographie Muhammeds bis zur Flucht, p. 131 sq.; al-Bukhr, al-a, beginning,
passim, kitb al-tafsr on sra 74; Muslim, a, vol. 1, p. 672sq., vol. 2, pp. 116, 631
(= al-Qasalln, Irshd al-sr, vol. 5, p. 185, K. al-ajj, vol. 7, p. 211, vol. 10, p. 229); al-Nas,
al-Sunan, 106 = I, 147sq.; al-Khab al-Tibrz, Mishct, p. 211, 514 = 219, bb , fal 2,
end, 522; al-Mabn li-namal-man, capt. 4, [W. Ahlwardt, Verzeichnis, no. 910 (= Wetzstein,
no. 103), p. 94, etc.]. Cf. G. Weil, Das LebenMohammeds, p. 48, foot-note, and G. Weil,
Sur un fait relatif Mahomet, pp. 108112; Sprenger, Life of Mohammed, p. 112, Sprenger,
Das Leben und die Lehre des Mohammad, vol. 1, p. 208sq., 269286, considers Muammad
a hysteric.
18
In adth al-ifk (K. al-Shahdt, 15, K. al-Maghz, 36).
19
: Theophanes, the Confessor, [title not supplied] vol. 3, p. 512 (Cor-
pus scriptorum historiae byzantinae, no. 28); Leo Grammaticus, [title not supplied] p. 153
(Corpus scriptorumhistoriae byzantinae, no. 31); Constantinos VII, Porphyrogenitus, Emperor
of the East, [title not supplied], vol. 3, p. 91 (Corpus scriptorum historiae byzantinae, no. 5);
Georgius Monachus Hamartolus, Chroniconbreve, quodex variis chronographis , ed. Eduard
von Muralt (1863), p. 592; : Ioannis Zonaras, [title not supplied] vol. 3, p. 214
(Corpus scriptorum historiae byzantinae, no. 30); : Michael Glycas, .
Annales, a mundi exordio usque ad obitum (Paris, 1660), p. 514 (Corpus scriptorumbyzantinae
historiae, no. 16); epileptica: Henricus Canisius, Thesaurus monumentorum ecclesiasticorum
et historicorum, ed. by J. Basnage (Amsterdam, 1725), vol. 4, p. 440. Cf. also Verlegung des
Alcoran Bruder Richardi, Prediger Ordens, cap. 11; Johann H. Hottinger, Promtuariumsive Bib-
liotheca orientalis exhibens catalogum(German by Dr. M. Luther, Heidelberg, 1658), p. 14sqq.;
Marracci on sra 74:1, etc. This opinion, which was advanced against the dignity of Muam-
mad, seems to have been widespread among Oriental Christians.
20
S. Ockley, History of the Saracens, vol. 1, pp. 300301: As for the Byzantines, their
mu
memory is but one of the symptons of epilepsy proper, it must rather be con-
sidereda matter of psychological fits of excitement (Rob. Sommer). Muam-
mad is supposed to have suffered from it since his early youth.
21
Arabs as
well as other ancient peoples considered such persons to be bewitched
( ).
22
Muammad, who at first shared this belief, later seems to have con-
sidered these attacks a manifestation of the One and True God. He is likely
to have been repeatedly afflicted by such attacks ever since he appeared as
a prophet, particularly during the initial phase when his mind was highly
incited; but they happenedoccasionally also after the flight.
23
Thus, whenhe
suddenly fainted while enwrapped in deep thought, he believed that he had
been guided by a divine power; but, as we observed above, he did not real-
ize it was a revelation until the Angel had departed,
24
i.e., when, after great
excitement, he regained full consciousness. According to Muslim accounts,
he encountered these attacks, which were particularly aided by his frequent
mental excitement, during the revelationof Koranic pasages
25
as well as dur-
ing divine commandments about other matters.
26
authority in this matter is of no great weight at all, especially considering they always made
it their business to represent Mahomet as full of all manner of imperfections, both of body
and mind as possible. As to being wrapped up in blankets, there might be many occasions of
that besides the falling-sickness, and his being troubled with disease having no foundation
in any Arab historian, is to be rejected among the rest of those idle stories which have been
told of Mahomet by the Christians. George Sale on sra 73; Jean Gagnier, La Vie de Mahomet
(1732), p. 91; Caussin de Perceval, Le combat de Bedr; pisode de la vie de Mahomet, Journal
asiatique, 3e srie, t. 7 (1839), pp. 97145. Incidentally, the matter is by far less important than
generally claimed.
21
Cf. the passages on sra 94, referred to below. Such an attack seems to be referred to
in the affair as told by Ibn Hishm, p. 117, l 1317; (cf. its foot-note); al-Bukhr, K. al-alt,
8; Ibn Sad, ed. Sachau (al-abaqt al-kabr), vol. 1, part 1: Biographie Muhammads bis zur
Flucht, p. 93; al-Azraq, p. 105 or 107 bottom; Muslim, a, vol. 1, p. 217 = al-Qasalln, vol. 2,
p. 407sq. (K. al-ay). Muslims, however, interpret the matter differently. But this tradition
is not reliable. There is some indication that the Prophet was afflicted by these attacks only
after his religious calling. Cf. also M.J. de Goeje, Die Berufung Muhammeds in Orientalische
Studien Nldeke gewidmet (1906), vol. 1, p. 5.
22
Ancient views of epilepsy as holy illness in Oeuvres compltes d Hyppocrates, ed. mile
Littr, vol. 6, p. 5.
23
This includes his fainting in the Battle of Badr: Ibn Hishm, p. 444; al-abar, Tafsr,
vol. 1, p. 1321; al-Wqid, p. 65; al-Aghn, vol. 4, p. 27; cf. Gustav Weil, Das Leben Mohammeds,
p. 157.
24

or . The sources, p. 16 n. 17, above.

25
Cf. e.g. Umars tradition: Fakhr al-Dn al-Rz and al-Tirmidh (tafsr) on sra 23 (begin-
ning); al-Zamakhshar at the end of sra 23.
26
Cf. e.g. Yals [GAS, vol. 2, p. 414]; tradition in al-Bukhrs
(vol. 3, p. 45),
, 2 end (vol. 3, p. 145) = , 10 (vol. 1, p. 202); al-Khab

al-Tibrz, Mishct-ul-mubh; or, A collection of the most authentic traditions, p. 522 (530);
22 mu
hammads revelations
Only suchmorbidphysical andmental conditions canexplainthe visions [i/26]
and dreams which raised Muammad above human existence. The best
known incident of this type is the Night Journey ( ) or Ascension to
Heaven (
) which, as will be demonstrated below, was a vision. That

the reports on his mental behaviour are generally correct can best be seen
from the fantastic, wild passages of the Koran, which Muammad skillfully
proclaimed during the first years of his prophethood.
In this context we must remember that a great many of these revelations
seemto have occurred at night,
27
a time when the mind is far more receptive
to fantasies and moods than during daytime. We know for certain that in
the early period Muammad quite often stayed awake for part of the night
in nocturnal devotions (
, sra 17:81), and that he fasted a great deal. Such

exercises, however, considerably sharpen the senses (Matthew 4:2; Apoc.
Esdra, the beginning), as has been recognized by more recent physiology
(Johannes Mller, [18011858]).
Evena casual observer will readily agree that the whole of the Korancould
not have come about in the highest degree of ecstasy. Its spirit ranges widely
from ecstasy to simple, serious reflection. In any case, Muammad, in his
state of excitement, cannot be imagined to have grasped entire parts of the
Koran but rather mere words and ideas. For historical research, therefore,
an independent passage of the Koran taken by itself is no revelation; rather
it is the literary formin which the Prophet expressed the message as he had
conceived it. Naturally, the strength of prophetic frenzy exerted a decisive
influence on his literary style. Since this enormous excitement diminished
as time went on, the sras became more temperate, initially still driven by
some poetical force but soon afterwards turning merely into the words of
a teacher and lawgiver. If Muammad did retain the form of Gods direct
speech, as he always maintained, this claim is not just an idle description
but rather a genuine expression of his own conviction. The most common
species of revelationinthe Koranis that inwhichthe unseenAngel inspired
his heart, although Muslims prefer to apply this classification to other
revelations more than to the Koran.
Gustav Weil
28
now surmises that Muammad received some of the rev- [i/27]
elations straight from a human being who was playing a game with him,
also al-Mabn li-nam al-man, cap. 4 [Wilh. Ahlwardt, Verzeichnis der arabischen Hand-
schriften, no. 910 (= cod. Wetzstein, no. 103)].
27
This certainly applies, e.g., to sra 73:1 sqq., and likely to sra 74:1 sqq., etc.; al-Suy,
al-Itqn, claims that most of the Koran was revealed during daytime.
28
Das Leben Mohammeds, foot-note to p. 598, and Historisch-kritische Einleitung in den
Koran, p. 57sqq., and 2nd ed., p. 66sqq.
mu
arguing that the verses in which Muammad is addressed can hardly be
interpreted in any other way, particularly in the early period. In addition,
Weil thinks that Gabriel is said to have resembled Daya (Ibn Khalfa al-
Kalb). This viewcannot be accepted. The fact that Muammadis addressed
by God in no way clashes with his overall manner of prophetic diction,
least of all in the earliest period, when he likely believed that he saw angels
frequently. In the later sras he retained this and other modes of commu-
nication, if only out of habit. But most of the passages quoted by Weil in
support of his view refer to rather late sras. Thus, until shortly before the
hijra, Muammad would not even have been aware of the fraud! How can
we believe that a reformerwhich the true author of those verses must
certainly be consideredinstead of appearing himself, would have chosen
someone else, easily trickable, to present his teachings, employing a ruse
that deprives the truth of all its value. If Weil thinks that those verses cannot
be reconciled with the sincerity that permeated Muammad at the begin-
ning, we are faced with the following dilemma: either the unknown author
produced only those verses, which in any case are unimportant, or he also
produced other verses which, though they are his, had to appear as though
they came from Muammad himself. In either case there is the same con-
flict. Finally, the the reference to Daya is totally out of place, as this man,
who played no important role at all, was honoured only accidentally as an
imposter of Gabriel.
29
Evenafter the hijra he remaineda disbeliever for some
time, wandering about as a trader,
30
and could not possibly have had any
close relation with Muammad.
Sprenger, too (Leben und die Lehre, vol. 2, 2nd. ed., pp. 348390), takes [i/28]
great pains
31
to showthat there was certainly one other person busy behind
the scene (p. 366) or plotting with him (p. 362), and is most inclined to
consider him to be identical with Bar, the Prophets mentor, author of
the uuf. But Sprengers arguments are not convincing.
32
In any case, it is unlikely that a superior and self-confident mind like
that of Muammad could have become dependent on a contemporary to
such an extent. Least of all could we imagine a conspiracy of fraud between
29
See above, p. 19sq.
30
See below, on sra 62.
31
Cf. A. Sprenger, Mohammed und der Koran, p. 58, and earlier his Mohammads Zusam-
menkunft mit dem Einsiedler Bahyr, and against Th. Nldeke, Hatte Muhammad christli-
che Lehrer, p. 699sqq.
32
Cf. also Hartwig Hirschfeld, New researches into the composition and exegesis of the
Qoran, p. 22.
24 mu
hammads revelations
the two. Despite all his faults, Muammads life and work unconditionally
presuppose the sincerity of his commission (see above p. 4).
The Length of the Revelations,
Their Names, Style, Refrain and Wordplay.
The Construction of the Verses
The length of the individual revelations varies considerably. Tradition dis- [i/29]
agrees on this point as well as on many others. Some claim that the Koran
was revealed in single letters and verses (
), with the excep-

tion of sras 9 and 10, both of which he received complete, one at a time.
33
According to others, the revelations were received in one or two verses at
a time.
34
In contrast, some claim they came down in units of one to five or
more,
35
while others say five toapproximately ten,
36
or, according toone final
opinion, always inunits of five.
37
Inaddition, it is saidof some sras that they
descended fromheaven all at once, e.g., sra 6
38
and others.
39
In this respect
most ambiguous is al-Kalb:
40
]

By way of concluding this record of contradictionwhich could easily be
continued, and only shows how little credence is to be given to tradition
let me add Ab l-Layth al-Samarqands words on sra 6:

.
Thus, all the verses were revealedat the same time, apart fromtwothat came
down elsewhere!
If we read the Koran without prejudice we recognize that multiple verses
always belong together, that often the number of verses undoubtedly re-
vealed at the same time is rather large, and even that many srasnot only
33
al-Zamakhshar and al-Bukhr on sra 9, at the end.
34
Ab l-Layth al-Samarqand on sra 2:181.
35
36
Ibid.
37
al-Itqn, p. 99.
38
Ab l-Layth al-Samarqand and al-Bayw on sra 6 at the end; Cod. Lugd., no. 674
(a Masoretic book from the end of the fifth century); al-Diyrbakr, Tarkh al-khams, p. 12;
al-Suy, al-Itqn, p. 19. But in al-Itqn, p. 84sq., this tradition is classed weak ( ).
39
al-Itqn, p. 84sq.
40
In W. Ahlwardt, Verzeichnis, no. 732 (= Cod. Sprenger, 404), an anonymous fragmentary
commentary on the Koran.
mu
the very brief ones, which probably no one would want to divide, but also
rather lengthy ones such as sra 12must have been created all at once.
Some sras are well arranged, having not only a proper beginning but
also an appropriate ending. Generally, however, the diction of the Koran
is rather inconsistent, so that the context is not always clear; there is the
danger that related parts can easily become separated. Of course, we cannot
deny that some revelations were very brief. Independent investigation must
try to discern the original form through the most minute consideration of
the context. The false view of Muslims regarding the original brevity can be
tracedbacktoa variety of causes. It was knownthat certainlaws, particularly
those enacted in Medina, had been very brief, and hence it was deduced
that this principle applied to others as well. Frequently there were differing
traditions regarding the occasion of closely related verses, leading to the
conclusion that they must originally have been separate; we also hear of
longer passages named after single verses (e.g. the Ftia), which produced
the erroneous interpretation that it was a reference to those individual
verses. Finally, this view may have arisen out of the idea that Muammad
received all of the verses of the Koran during his epileptic attacks, which
were not considered to have had a long duration.
41
Incidentally, Muammad often put together, or joined, Koranic passages [i/30]
that originated fromdifferent periods. Insome instances this is very obvious
while in others we may suspect it; in still other instances these connections
may remainhiddenfromus. After all, whowoulddare toseparate verses that
differ little in time and language when their author had combined them?
Muammad called a single, self-contained promulgation sra or qurn.
The former expression is met with nine times in Meccan and Medinan pas-
sages: 2:21; 9:65, 87, 125 &128; 10:39; 11:16; 24:1; and 47:22. Muslims made many
futile attempts at explanation,
42
but its origin has still not been sufficiently
41
Cf. Sprenger, The Life of Mohammad, p. 152, n. 4: According to Ab Hmid it is incon-
sistent with the office of a prophet that he should be subject to fits of madness (jonn), or to
swoons (aghm) of long duration. Balqyny, in his glosses to the Rawdhah, differs from him.
Sobky thinks that the swoons of the prophets differ as much from those of other persons as
their sleep. Soyty, Anmzaj allabyb fy Khayi alhabyb.
42
It is derived either from the root and interpreted as grandeur, rank (a meaning
which, by the way, is ascertained by the word
; cf. the more frequent
), where always
one part is taken to be more elevated than the other; or, it is considered to be derived from
by taking hamza to be weak while others are said really to pronounce sura. Then it is
said to mean
, al-abar in the introduction to the Tafsr, Cairo ed.,

vol. 1, p. 34sq. Cf. Ab l-Layth al-Samarqand on sra 24:1; al-Zamakhshar and al-Bayw on
sra 2:21; Ibn Aiyya, al-Jmi al-muarrar al-a, al-Qurub, al-Jmi al-muarrar; 25 recto;
i and Qms, s.v.; al-Itqn, p. 121. However, it may well be pointed out that words of this
root do not indicate part in any Semitic language at all but only the leftover.
26 mu
hammads revelations
determined. It was considered to be a borrowing from Hebrew
43
row
(of people [e.g., Mishna Sanhedrin 4:4] and of objects), on the basis of
which sra could easily be interpreted as a line from the heavenly book;
the meaning line in books and letters can be documented only in recent
New Hebrew. It can hardly remind one of the expression guide
line,
44
or even consider sra a corruption
45
of Hebrewsidrh. But the mean-
ing of section, as a synonym of ,
46
is not a bad conjecture.

or, with a weak hamza,
,
47
indicates not only a single section [i/31]
of the revelation
48
but also, like the Jewish mir, several or all of them
together.
49
This meaning later became so generally accepted that it was
applied to the recension commissioned by Muammads successors.
50
In
form it is identical to a common infinitive of
51
according to the not infre-
quent pattern of fuln. This, however, does not answer either the question
of the sense of its original meaningsince the usage of is somewhat
obscureor the question of the real origin of the word, since yet another
possibility should be considered.
43
The etymology is obscure. It has nothing to do with wall. Cf. Siegmund Frnkel,
Die aramischen Fremdwrter imArabischen, p. 237sq.
44
Paul de Lagarde, Sra, p. 296.
45
H. Hirschfeld, Newresearches into the composition and exegesis of the Qoran, p. 2, n. 6.
46
Already found in G. Sale, The Koran, commonly called the Alkoran of Mohammed. Pre-
liminary discourse, section iii, p. 44, bottom.
47
This is likely to have been Muammads own pronunciation since there was a prefer-
ence for a soft hamza throughout the ijz (see below). In assn b. Thbit it reads (Ibn
Hishm, p. 526):
( ) and (Ibn Hishm, p. 713, l 1 = Dwn, p. 45, l 9)

. This is the way Ibn Kathr read the Koran, and this is why you find in old
Kufic manuscripts (i.e. Qurn, not Qurn). By the way, Kab b. Zuhayr [EI
2
; Sezgin, GAS,
vol. 2, pp. 229235] has
: Ibn Hishm, p. 891, l 13. Cf. also Karl Vollers, Volkssprache und
Schriftsprache imalten Arabien, p. 91, and generally, pp. 8397.
48
E.g., sras 72:1, and 10:62.
49
E.g., sras 15:87, 17:84, and 25:34, equivalent to the heavenly book.
50
It may be mentioned in passing that some Muslims derive
not fromto read, rather

it is likely to be influenced by sra 75:17,
, with the meaning, unite, collect

(cf. ) so that it is taken to indicate what binds the individual
. This was the opinion of

(Ab l-Khab) QATDAH [EI
2
; EQ; G. Juynboll, Encyclopedia, pp. 438449; F. Sezgin, GAS,
vol. 1, pp. 3132] and AB UBAYDAH al-Naw Mamar b. Muthann [Sezgin, GAS, vol. 3,
pp. 363364, vol. 7, 343, vol. 8, 6771, and vol. 9, 6566] (al-Jawhar, i, s.v.) Cf. al-abar in
the introductionto the Tafsr, Cairo edition, vol. 1, p. 31 sq.; Lisnal-Arab, vol. 1, p. 124; al-Itqn,
p. 118sq. Cf. also Ibn Qutayba, Liber posis et potarum, ed. J.M. de Goeje, p. 26, l 4 and 5.
51
Sra 17:80, and 75:17sq. Such a poet in Ibn Qutayba, Handbuch der Geschichte, edited
by F. Wstenfeld, p. 99; (according to Ibn Abd al-Rabbih, al-Iqd al-fard, cap. and
Ibn al-Athr, al-Kmil f l-tarkh, vol. 3, p. 151, it is assn b. Thbit)
(by
glorification and recitation of the Koran), al-abar, vol. 1, p. 2196, l 17 (vol. 1, p. 3063 and in the
dwn the verse is missing); additional examples in a and Qms, Ibn Aiyya, al-Qurub,
loc. cit., al-Mabn li-namal-man, vol. 3.
mu
In the Koran means to lecture, to recite (sras 16:100, 17:95, 69:19, [i/32]
73:20, and 87:6) from a text or by heart,
52
and in other cases also to dictate
to a writer.
53
On several occasions it is said in the traditions that Muam-
mad had said something and , where it can mean nothing other than
reciting Koranic passage by heart. The usage in Muslim, a, vol. 1, p. 80 (=
al-Qasalln, vol. 1, p. 449)
, where it refers to an
ordinary saying, is seldomfound; the applicationof the expressionfor mem-
orizedrecitationof Koranic passages toadthcaneasily be explained. Since
a cultural termlike to read cannot be Protosemitic, we may assume that it
came to Arabia fromthe outside, most likely fromthe north, where the orig-
inal meaning of to call is still quite common in Hebrewand Aramaic. This
meaning is unknowninArabic. Althoughit survives unchangedinthe famil-
iar phrases
54
and
55
the close relation that here
exists of to the Aramaic greeting (Hebrew )
56
raises doubts that
the entire phrase is after all not a borrowing, despite the fact that it cannot
be documented in early Aramaic. If sra 96:1 really were to be translated as
preach then this case ought to be viewed in a similar way.
57
The fact that
Syriac, in addition to the verb , has the noun eryn with the double
meaning of and tends to support the conjecturein
relation to what has just been arguedthat the term qurn is not just an
exclusively Arabic development froman infinitive of the same meaning but
is, rather, a borrowing from the Syriac word which simultaneously assimi-
lated to the pattern fuln.
in fact does not mean a book; it is rather a neutral expression for [i/34]
revelation and is used for Muammads inspirations (sras 3:2, 25:1, 2:181)
as well as for those of other prophets like Aaron and Moses (sras 2:50 and
21:49.)
58
52
A. Sprenger, ber das Traditionswesen bei den Arabern, p. 4; al-Itqn, p. 254sq.;
Sprenger, Life of Mohammed, p. 96, n. 2; Sprenger, Leben und die Lehre des Mohammed, vol. 1,
pp. 298463, and vol. 3, p. xxii.
53
E.g., Ibn Sad, [al-abaqt al-kabr], vol. 3, part 2, p. 59, l 15, p. , l 20 ( ).
54
al-Bukhr, Kitb al-mn, 19; al-Maqdis, Bad al-khalq, 5, i; al-Aghn, vol. 1, p. 15, l
18; tim al-, ed. Schulthe, p. 83, l 15; amsa, p. 604, verse 2.
55
al-Muwaa p. 175, l 3 frombottom; al-Wqid, p. 189, l 2 frombottom; al-Tirmidh, Tafsr
on sra 3:163, and often. The Turkish ms interprets the idiomas

he delivered someone verbal regards.
56
Goldziher, Der Diwan des Garwal b. Aus, p. 22sqq., has shown that the salmgreeting
was known long before Islam. But it cannot be Protosemitic. The foreign vocabulary in the
Koran is also partly old borrowing. Muammad himself is unlikely to have added much.
57
More about this see below, s.v. Cf. also Snouck Hurgronje, Une nouvelle biographie de
Muhammed; also his Mekka, vol. 2, p. 225, foot-note.
58
This word, like the Ethiopic ferqn, is derived from the Aramaic cf. A. Geiger,
28 mu
hammads revelations
Because the style of the Koran differs considerably according to the date
of composition it will be better for us to discuss the issue later when deal-
ing with the different periods. Whereas some of its older parts are wild and
excited, others are serene; we find still other parts in which the language
is quite ordinary, almost prosaic. Their common characteristicswith few
exceptionsconsist only in Gods speaking directly and in certain rhetor-
ical touches prevailing throughout. There is one aspect that we must keep
in mind, namely that the Koran is more rhetoric than poetry. Even though
little importance can be attached to Muslimstories purporting to showthat
their Prophet was unfamiliar with pre-Islamic poetry
59
since they are but
an outgrowth of the statement in sra 69:41, its entire
intellectual purpose is far more concerned with didacticism and rhetoric
than with pure poetry. This would explain that at a time when the greatest
of the Arab poetsShanfar,
60
al-Nbigha al-Dhubyn,
61
Maymn b. Qays,
calledAL-ASH,
62
etc.were at their prime, or haddiedonly shortly before,
Muammad preferred a poet like assn b. Thbit, d. 40/659
63
over all the
others, and admired the poetry of Umayya b. Ab l-alt,
64
even though bor-
loc. cit., p. 55sqq.; Siegmund Fraenkel in his doctoral thesis, De vocabulis in antiquis Arabum,
p. 23; and Fr. Schwally, Lexikalische Studien, p. 134sqq. The same word appears in the
HebrewTargumas , , , and in NewTestament Greek as , ,
(e.g. Luke 21:28; Romans 3:24; Ephesians 1:7, and 14; 1; Colossians 1:14; Hebrews
9:15), (e.g. Luke 1:69; Apocrypha7:10; 12:10). Inthe latter meaning Muammaduses the
word twice in sra 8 (verses 29 and 42). The meaning revelation does not exist in Aramaic.
It is thus possible that it came into use in the Arabic-speaking world only. If one does not
merely want to suppose a misunderstanding on the part of Muammad, it might be worth
considering whether this change of meaning did not occur in a community where its entire
religious thinking was dominated by the hope for liberation and redemption, i.e., primarily
and most likely among Christians, or otherwise in Messianic oriented Jewish circles.Al b.
Ab lib (Ibn Hishm, p. 518, l 7):

Regarding the derivation of this word, too, there is much inaccuracy to be found among
Muslims. Cf. al-abar, Tafsr, vol. 1, p. 32sq.; al-Bukhr on sra 24:1; the encyclopaedias, etc.
59
Cf. Ibn Hishm, p. 882; al-Aghn, vol. 20, p. 2; Ibn Sad, (al-abaqt al-kabr): Biogra-
phien der Muhgirn und Ansr, die nicht bei Bedr mitgefochten, p. , l 25sq.
60
EQ; EI
2
; Sezgin, GAS, vol. 2, pp. 133137.
61
EI
2
62
EQ; EI
2
63
EI
2
; EQ; Juynboll, Encyclopedia; Sezgin, GAS, vol. 2, pp. 289292.
64
Cf. Muslim, a, vol. 2, p. 399sq. (= al-Qasalln, vol. 9, p. 100sq. ( ));
al-Khab al-Tibrz, Mishct-ul-mabh, p. 401 (409, ); al-Tirmidh, Shaml,
bb, 37; al-Aghn, vol. 3, p. 190sq.; al-Bukhr, K. al-Adab, 90.
mu
rowed ideas
65
and the rhetorical pomp
66
replaced true poetry in his work.
Only once in his life does Muammad seem to have composed a most sim-
ple distich,
67
and only rarely did he resort to the poetry of others.
68
Still, Muammads adversaries considered him a poet. This shows that [i/36]
the form in which he made his promulgations, the so-called saj ( ), was
still considered poetical, although for a long time poets had been using a
diction defined by strict rhyme and metre.
69
Saj is produced when speech
is made up of short parts, in which two or more lines always rhyme with
one another, although in such a way that the final syllable of the individual
parts is pronounced not according to the minute rules for the end of a
verse but according to normal pause (waqf); the parts also have a much
more liberal rhyme (qfiya).
70
This style of diction, which dominated the
65
Cf. his verse regarding those who carry the Throne of God:
[so for ] (Ibn ajar al-Asqaln, vol. 1, p. 261; al-Aghn, vol. 3, p. 190,
l 19; al-Damr, K. ayt al-ayawn, vol. 2, p. 154 (s.v. )): . Ibn Abd Rabbih, al-Iqd,
vol. 3, p. 96, andal-Qazwn, Kosmographie, ed. Wstenfeld, vol. 1, p. 56 ; al-Qazwn [Ajib
al-makhlqt] is the only one to read , which we must undoubtedly take as an
allusion to Ezekiel 1:10, but particularly to Apocr. 4:7.
66
Cf., e.g., the elegies in Ibn Hishm, p. 531 sqq. and the other fragments of his poetry
in al-Aghn, vol. 3, pp. 186192, vol. 16, p. 71 sq.; Abd al-Qdir (ibn hir) al-Baghdd,
Khiznat al-adab, vol. 1, p. 118sqq.; Jamharat ashr al-Arab, p. 106sq.; Ibn Qutayba, Liber
posis, pp. 279282; al-Masd, Prairies d or, vol. 1, pp. 136142. The other references can now
be found in the above-mentioned article by Fr. Schulthe, p. 14 n. 19, who also discussed
the content of the fragments, particularly the theological and historical connections. What
attracted the Prophet to the man was his almost Islamic philosophy.
67
The frequently quoted rajaz (e.g. al-Bukhr, K. al-Maghz 55; al-abar, vol. 1, p. 1662,
and on the tafsr on sra 9:15, Cairo edition, vol. 10, p. 64; al-Wqid, p. 273, l 19; al-Khab
al-Tibrz, Mishct, p. 417, bb al-mufkhara, fal, 1; al-Diyrbakr, Tarkh, vol. 2, p. 103, section
on the battle of unayn, etc.):
*

*
.
68
Cf. al-Bukhr, Muslim, al-Tirmidh, Shaml, in the passages of the previous foot-
note 69.
69
Cf. Goldziher, Abhandlungen zur arabischen Philologie, vol. 1, pp. 5783.
70
The mainpurpose of pause andrhymedprose is the dropping of the final short vowels as
well as tanwn and the pronunciation of as . The artificial pronunciation with half vowels,
which are called rawm(not Rmas Sylvestre de Sacy vocalizes since it is the simple infinitive

); cf. the i and the orthography of good manuscripts like W. Ahlwardt, Verzeichnis,
657 (= cod. Petermann i, p. 159; Ms. Leiden, cod. Golius, 46) and al-ishmm have of course
been discussed by Sibawayh (al-Kitb, vol. 2, p. 282) [GAS, vol. 9, 5163] but it is doubtful
whether they emanate from real life or merely from the schools. Regarding Masoretic works
like the Jazariyya with its commentaries (e.g. cod. Vindob. A.F. 377c. = Flgel, 1636, A.F., 309
b. = Flgel, 1630); Ibn al-Jazars great work, W. Ahlwardt, Verzeichnis (= cod. Petermann i,
no. 159); W. Ahlwardt, Verzeichnis, 591 (= cod. Sprenger, 382); Gotha, cod. Mller, 65; Itqn,
p. 210, etc. The laws of pause are explained more precisely in Sibawayh, ed. Blq, vol. 2,
pp. 277291; al-Zamakhshar, al-Mufaal, ed. Broch
2
, p. 160sqq.; Ibn Mlik, Alfiyya, cap. 69;
cod. Gotha, fol. 25 r.; al-Suy, al-Itqn, p. 209sqq. Cf. also the notes in Ewalds Grammatica
30 mu
hammads revelations
special form adopted by the old soothsayers (kuhhn), Muammad also
made use of, although with several changes. He disregarded the rule of
making the individual parts of roughly equal length, gradually lengthened
the verses in the later sras, and ended up making such a free use of rhyme
that the Muslims came to differentiate, not unjustly, between the Koranic
rhyme (
) and rhymed prose ( ).

71
Since this rhyme, when
viewed closely, but cautiously, is of prime importance for the improvement
of certainpassages, for the proper arrangement of the verses, the recognition
of the connection of longer passages, and the rearrangement of misplaced
verses, it is warranted here to delve into it in greater detail. In the case of
rhymed prose Muammad takes all available poetic licences, even going
beyond in some instances. At times he totally silences
, that ought to
be pronounced at the end of a verse,
72
omits the final or of verbs
ending in or ,
73
lengthens the of nab in nouns and verbs to as in
rhymed verse,
74
and completely omits the of the first person
75
or changes
it to
which is also frequently found in rhymed verse.

76
But Muammad
goes one step further and makes similar consonants, particularly and ,
critica linguae arabicae, vol. 1, p. 373sq., vol. 2, 335sqq.; Wm. Wright, AGrammar of the Arabic
language, 3rd ed., vol. 2, pp. 368373.
71
Ibn Khaldn, Muqaddima, cap. 6, 45; al-Itqn, p. 693sqq. It is generally forbidden to
call the rhyme of the Koran because it is no (loc. cit., p. 695); on the other hand, it is
debatable whether because of its form can be applied to a greater degree.
72
Sra 58:2 (end of the verse?); 90:6; 74:33, etc. At times (Wm. Wright, A Grammar of the
Arabic language, 3rd ed., p. 369 B) this occurs also in a verse, e.g. in Labd.

(Dwn, ed. by Brockelmann, no. xxi, 4) for
.
73
Sra 55 verses 26, 44 and 54; 75:27, etc. Incidentally, this is also not rare in ordinary
pause, and generally common in some dialects. Audacious is the dropping of in sra
75:26, but not unprecedented. Cf. onthis Sibawayh, al-Kitb, vol. 2, p. 289sq.; al-Zamakhshar,
al-Mufaal, p. 161 sq.
74
Sra 33 verses 4, 10, 49, 66; 84:14; 74:15. Al-Zamakhshar on sra 33:10 puts it this way:
here, an in fila is added as in other cases in qfiya.
75
Sra 13:32, etc. The Mufaal (p. 163) permits this also in regular pause; occasionally it
is found in poetry, e.g. Ab Tammm, Hamsa, p. 362,
for

; in Labd:

(for
, Labd, vol. 2, no. xxxix, l 1); in al-Ash:

(in the Mufaal cited for ). Cf. R. Geyers review of Karl Vollers, Volkssprache und
Schriftsprache imalten Arabien.
76
Sra 69:19sq., 25sq., 28sq. So also
, sra 101:7. This, too, is possible in rhymed

prose or poetry.
mu
somewhat less frequently and , rhyme indiscriminately, extending this
rhyme in later sras even to quite different consonants, with the result
that the rhyme is reduced to mere assonance.
77
Conversely, the change of
vowels, unless permitted in versefor example, from to , and the change
of the short vowels
78
before a consonantis extremely rare. As far as the
arrangement of the verses is concerned, occasionally an unrhymed verse is
foundbetween
79
or after
80
rhymed ones. Freely rhymedsras, however, show
that in longer passages the rhyme is observed more consistently.
81
In better prose,
82
in contrast to proper poetry, the Arabs used to change
the rhyme after some of the short parts of speech. This also happens quite
frequently in the Koran, particularly in the earliest sras.
83
In most of them,
however, the rhyme continues throughout most or all of the verses, particu-
larly in the case of the longer ones. The majority of the Koranic rhymes end
with n, n, m, d, r, etc., mainly in a closed syllable with long vowel. Sus-
tained rhyme with ( and respectively) is found far less frequently
and is primarily limited to the Meccan sras (17; 18; 19; 20; 25; 53; 71:5sqq.; 72;
73; 76; 78; 79; 80; 87; 91; 92; 93; and 99); among the Medinan sras it is limited
tosras 33; 48; and65. Innearly anequal number (16) of srasnamely with
the exception of 47, all Meccan sras (37:411; 54; 74 passim; 75:713; 81:118;
82:15; 84 passim; 86; 90:15; 93:911; 94; 96:15; 108; 111; 112; and 113)the
rhyme consists of a closed syllable with short vowel, e.g., ib, kum, hum, ar, ir,
ur, etc. Less frequent is the rhyme with (, ) 69:124; 75:1 sqq. and 14sqq.;
79:614; 80:11 sqq. and 38sqq.; 88:15 and 816; 101; and 104 (all early Mec-
can sras) and 98 (Medinan). At the end of a verse, there is sporadically a
closing syllable with a double consonant (97; 103, Meccan), and a closed syl-
lable with a diphthong (106, Meccan), which might also be assigned to the
preceding category. This case is worth a special investigation.
77
Isolated instances already in the earlier sras like 106:1, 2, 3 (which actually closes with
). At least in rajaz you find isolated instances of rhyme without completely identical
consonants (see talqb al-qawf in Wrights Opuscula arabica, Talqb al-qawf, p. 57). It is
also not rare to find in the well-composed qada now and then there is an instead of ,
which usually predominates in rhymed syllables, or for .
78
Cf., for example, sra 54; Talqb al-qawf, p. 55sq.
79
So sra 70:10 (where the previous rhyme is repeated); sra 82:6 (where the later rhyme
appeared once before); sra 80:32.
80
So in sras 53, 82, 93, and 96.
81
For example, insra 18 the rhyme is ( ), but fromverses 66 to82
(except inverses 78
and 80) with a preceding unvowelled consonant.
82
Similarly in the short rajaz which was not quite recognized as
.
83
Occasionally a previous rhyme reappears later; e.g., in sra 80 the rhyme
. Cf. nowKarl
Vollers (above, pp. 26f.), pp. 5580.
32 mu
hammads revelations
Yet even these brief statistics, which do not take into account all the [i/39]
sporadic rhymes within sras, demonstrate well the gradually increasing
monotony of the style of the sras. From among the enumerated types of
rhyme in Medina only no. 2 can be supported by documentary evidence,
and this three times only, while nos. 3 and 4 once each, and nos. 5 and 6
not at all. In the late Meccan and Medinan sras only two rhymes prevail,
which can easily be formed by grammatical endings and frequently used
words,
84
namely the one ending with n, n, m, m, and the other ending
in with following consonants. It is rare to find a sra with uniform rhyme
interruptedby verses withanother rhyme.
85
The casual applicationof rhyme
becomes more apparent the less fitting it is for the prosaic tenor of the later
passages. Particularly in ordinances and similar passages we must consider
the rhyme a bothersome fetter that does not even adorn the speech.
It might be mentioned in passing that the influence of the rhyme on [i/40]
the diction of the Koran is by no means without importance. In order to
maintain the rhyme, the form,
86
and even the sense, of words are occasion-
ally changed. For example, when in sra 55 two gardens are mentioned
(verse 46), therein two fountains of running water (verse 50), therein of
every fruit two (verse 52), and besides shall be two gardens (verse 62), it
is obvious that the duals are used to support the rhyme. The same applies
to sra 69:17, whose eight (angels) shall carry above them the Throne of
thy Lord would be puzzling if did not fit the rhyme. Finally, there is
the peculiar influence of every poetical form (metre, rhyme, stanza, etc.)
upon the order of the construction
87
and the flow of ideas.
88
Of no small
84
Like , , etc.; , , ,
, etc.
85
As is the case in sra 55:16sq., and 55:43.
86
In sra 37:130 there is instead of ; sra 95:2 instead of (or as others
read
to avoid the un-Arabic form
). These forms caused Muslims a great deal of

headache.
87
For example, in sra 2:81 (
) we find that the last word takes the place

of , which is required for the parallelism of the rhyme. For the same reason a verbum
finitum is frequently paraphrased by with participle or with the genitive. Also some
Muslims recognized this influence, and Shams al-Dn b. al-igh in his book

made minute observations (listed in al-Itqn, p. 699sqq.) although at times going

too far.
88
Apart fromthe Koran, Muammadrarely seems to have made use of but this applies
in particular to prayers as in the frequently quoted
* *

al-Bukhr, K. al-Jihd, 96; al-Tirmidh, ibid., 28. Other prayers of this kindsee al-Muwa,
164; Ibn Hishm, p. 756sq.; al-Khab al-Tibrz, Mishct, K. al-Masjid, fal 3, 7, and 8, K.
al-Witr, fal 2, 8; Ibn Sad (al-abaqt al-kabr), I, IV [sic], p. 14sqq., etc. In his sermons he is
mu
importance is the impact that rhyme had on the composition of the Koran.
This will become apparent when we later analyse the sras. At the present
moment we seek only to highlight the main points of view. Uniformity of
rhyme can never serve as proof of the unity of a sra; rather, it should be
considered only as a product of internal circumstances. One must always
anticipate the possibility that disparate passages of identical rhyme were
later inserted by either Muammad himself or in a later recension. At times
the Prophet might have purposely composedanadditiontoanalready exist-
ing revelation in the rhyme of the original.
Muslims hold very different views regarding the rhyme of the Koran (al- [i/41]
Itqn, p. 697sqq.): Some will admit that rhyme is needed throughout the
Koran. There are others, however, who deny this outright because they con-
sider such inaccuracies of its components, even of the common sajnot
to mention the Divine Bookto be unseemly. A third faction attempts to
compromise by suggesting that inthe Koran, as inthe rhetorical prose of the
Arabs, rhymed and blank verse alternate. Some people therefore introduce
a pause after every verse, claiming that the Prophet also followed this cus-
tom.
89
When determining the pauses, however, most people pay attention
only to the syntactic construction and, wherever the rhetorical formation
does not coincide with the former, they accordingly pronounce the final
words of the verses as though in the middle of speech ( ) so that the
rhyme is hidden.
There are three sras with a refrain: sra 54 (verses 15, 17, 22, 32, 40, [i/42]
51verses 16, 18, 21, 30, 37, 39); sra 55, where it is repeated ad nauseam,
namely starting with verse 12, thirty-one times the words (
), and
sra 57, verses 21, 29verses 11, 17verses 15, 26, 27. But, like a refrain, single
verses are repeated several times in some sras, particularly in the histories
of the prophets, which in certain parts are quite similar one to another.
90
said to have totally avoided this formof speech, al-Khab al-Tibrz, Mishct, 28 (36 K. al-Ilm,
3, 4). al-Bukhr, K. al-Adab, 6
(cf. al-Qasalln, vol. 9, p. 6sq.), in a slightly different arrangement,

al-Bukhr, K. al-Riqq, 22. Cf. alsoGoldziher, Abhandlungenzur arabischenPhilologie, vol. 1,
p. 68.
89
al-Tirmidh, al-Shamil, 44, ; al-Tirmidh, Sunan,
17; al-
Khab al-Tibrz, Mishct,
, fal 2, 8; [AB AL-QSIM] Umar b. Muammad

[IBN ABD AL-KF]; Leiden Ms. Warner, no. 653, on sra 79. There is no doubt that this
was Muammads pronunciation, but such a tradition is without much credibility since it is
known that later interpreters attempted to present their opinions as coming from Muam-
mad al-Tirmidh, too, does not trust this tradition ( ).
90
For example, in sra 19 (verses 15, 3475, 98), sra 37 (verses 110, 121), sra 26 (verses 7
34 mu
hammads revelations
Wordplay, which was far less common (though not entirely absent
91
)
among ancient Arabic poets than among later ones, who made it their
primary poetic goal, is also found occasionally in the Koran.
92
This did not
elude Muslims.
93
Such a play on words indeed divides a verse into several
small parts, e.g., sra 10:63,
; and sra 71:5,
, etc.
94
Some years ago, David H. Mller,
95
in his monograph on the prophets in [i/43]
their original form, attempted to identify the structure of the Koranic verses,
based on the sras 7, 11, 15, 19, 26, 28, 36, 44, 51, 54, 56, 69, 75, 80, 82, 90,
and 92. Most suitable for his hypothesis are sras 56 and 26. In order to
judge for ourselves it is sufficient to look at the construction of these two
sq., 67sq., 103sq., 121 sq., 139sq., 158sq., 174sq., 190sq.), sra 7 (verses 64, 76sq., etc.); and sra
56 (verses 73, and 96).
91
Cf. al-Shanfars Lmiyya, verse 4:
; in Labd (Ibn Hishm, p. 941, l 10:)

, and l 13, ( = Dwn, ed. al-Khlid [GAS, v. 1], pp. 17 and 19); in
al-Khans [GAS, v. 2, p. 311] (Dwn, Beirut, 1888), p. 24, l 4; similarly p. 32, l 8, and p. 37,
l 15: opposite of and ; in Bashma [Ibn al-Ghadr (Sezgin, GAS, v. 2, p. 118)], the
uncle of Zuhayr [Ibn Ab Sulm, EI
2
; Sezgin, GAS, vol. 2, pp. 118120]:
(in al-Buturs [EI

2
; EQ; Sezgin, GAS, vol. 2, pp. 560564] amsa, cap. 9, andwithout
identifying the poetal-Itqn, p. 302, against al-Mufaaliyyt, ed. Thorbecke, p. 11, l 2,

; Imru al-Qays, The divans of the sixancient Arabic poets, ed. Ahlwardt, no. 52,
verse 58 (p. 154); Ibn Hishm, p. 519, l 2, ); very much of the wordplay in ancient
proverbs, etc. Some of the ones quoted here occur also in the Koran and seem to have been
in common use. See sra 21:90 (
); sra 2:181 ( and ; cf. sras 65:7 and 92:7 and

10).
92
Apart fromthe passages mentionedcf.
(sra 104:1, cf. AbZayd, al-Nawdir, p. 76,

l 14); (sra 27:45); (sras 12:84; 12:19; 30:42; 24:37; 56:88; 55:54.)
Here belongs also the compilation of similarly sounding names, which have been changed
only for this particular purpose, for example, and (sra 2:96); and
(sras 18:93, and 21:96; cf. Imru al-Qays in The divans of the six ancient Arabic poets, ed. by
Ahlwardt, p. 204, no. 25, l 4); and (sra 2:250sqq).
93
R.E. Brnnow, Das Kitabu-l-Itbi wa-l-Muzwaati des Ab-l-Husain Ahmed ibn Fris
(d. 395/1004) [al-Qazwn]; al-Suy, Muzhir f ulm al-lugha wa-anwhih (Bulaq, 282/
1865), vol. 1, pp. 199201, cap. ; Abd al-Malik b. Muammad AL-THALIB, Fiqh
al-lugha wa-sirr al-Arabiyya (Cairo, 1317/1899) p. 303, ; and p. 314sq., . The
latter one correctly says that the figure of speech was rarely used in pre-Islamic poetry, but
became extraordinarily frequent later on. The matter is worth a monographic investigation.
Max. T. Grnert in Die Alliteration supplies a mass of 224 examples, but they are almost
exclusively from literary works not from primary sources.
94
Still more artificial we find this reflected in poetry, e.g., al-Sukkar, Lieder der Hud-
hailiten, p. 15, verse 2sqq.
95
Die Propheten in ihrer ursprnglichen Form; die Grundgesetze der ursemitischen
Poesie, erschlossen in Bible, Keilinschriften und Koran und in ihren Wirkungen erkannt in
den Chren der griechischen Tragdien, [The prophets in their original form; the basic laws
governing Proto-Semitic poetry, from the Bible, cuneiform and Koran and in their effects as
reflected in the choir of the Greek tragedy], vol. 1, pp. 2060, 211 sqq.
mu
sras. After a brief introduction (verses 19), sra 56 describes the three
categories into which mankind shall be divided at the Final Judgement: The
sbiqn(fourteenverses, vv. 10 to23), the Companions of the Right (sixteen
verses, vv. 24 to 39), and the Companions of the Left (seventeen verses,
vv. 40 to 56). The introduction, verse 57, leads to three questions addressed
to mankind regarding their attitude toward the seed you spill (five verses,
58 to 62), the toil you till (four verses, 63 to 66), the water you drink
(verses 67 to 69), and the fire you kindle (verses 70 to 72). The first and
second verses both have the same beginning ( and respectively). In
sra 26, the introduction (vv. 1 to 6) and all of the seven following sections
regarding the past prophetsvv. 9 to 66 (fifty-eight verses), vv. 69 to 102
(thirty-four verses), vv. 105 to 120 (sixteen verses), vv. 123 to 138 (sixteen
verses), vv. 141 to 157 (seventeen verses), vv. 160 to 173 (fourteen verses), and
vv. 176 to 189 (fourteen verses)conclude with this two-verse long refrain,
[Surely in that is a sign, yet most of them are not believers. Surely thy
Lord, He is the All-mighty, the All-compassionate:] In addition, starting
with verse 105, with the exception of the names, each of the first verses of
the last five sections have the same phrase, [ cried lies to the Envoy.] It
cannot be denied that both sras represent an artistic, literary work, with
proper disposition, with skilful application of rhetorical forms of style, and
with purposeful proportions of the individual length of the sections. On
the other hand, there is so much irregularity in the composition, so much
licence and arbitrariness, that it cannot be called a strophic structure in the
full meaning of the word.
Written Notes of Koranic Passages.
Additions and Changes Arising fromMuammad
The revelations allegedly have been recorded as follows:
96
[i/44]
] [

or . Yet at the same time it is claimed that the division of
the sras was introduced only after the revelation of the words
96
al-Tirmidh, Sunan, p. 502 (vol. 2, p. 134, tafsr); Fakhr al-Dn al-Rz; al-Zamakhshar; al-
Bayw on sra 9; al-Khab al-Tibrz, Mishct, p. 186 (194 fail al-Qurn, end); al-Qurub,
Jmi al-akm (Ahlwardt, Verzeichnis, no. 810, vol. 1, 23
r
); al-Mabn li-nam , part 3; al-
Khab al-Tibrz, Mishct, p. 141; in the text there are several unimportant variants.
36 mu
hammads revelations
,
97
which, of course, are consideredby some tobe the earliest of the reve-
lations.
98
But I cannot remember having readinany reputable ancient writer
the statement that every individual part of the Koran was put between two
boards or covers (
) as soon as it was written down and frequently

taken out to be copied.
99
Below, we shall demonstrate the probability that
this must be considereda Shiite fabrication. Alsothe traditionthat Muam-
mad assigned a definite place
100
to every single verse immediately after its
promulgation cannot be historical, even if he occasionally made additions
to certain sras. This tradition evolved first from the superstition that the
existing order of both the verses and the sras was certainly of divine origin
and must have been copied exactly by Muammad himself, and secondly
fromthe erroneous opinion that the individual revelations were quite short
and put together only at a later time. G. Weil has already pointed out the
absurdity of this entire matter.
101
It is doubtful that Muammad put down
in writing all the revelations of the divine book from the start.
102
During the
first years of his divine commission, when he hardly had any followers, he
might have forgotten some of the revelations before outsiders learned of
them. Some other revelations might have been retained only in the mem-
ory of Companions, as attested by several traditions that say that he recited
Koranic passages to his followers until they knew them by heart. Still, it is
likely that already many years before the flight he dictated entire verses
to a scribe,
103
not merely single verses, as Muslims claim. After all, when
Umar embraced Islam, passages of the Koran had already been recorded,
if indeed
104
reports of this event can be trusted. To prove that there had been
written sras certainly in 2/623 one might refer to assn b. Thbit, who
97
al-Khab al-Tibrz, Mishct, p. 185 (193 fail al-Qurn, fal 3 2); al-Wd, Asbb
al-nuzl, in the introduction, p. 5; al-Mabn li-nam , part 3; al-Suy, al-Itqn, p. 184sq.
(Several traditions of Sad b. Jubayr [EQ; EI
2
; Juynboll, Encyclopedia, p. 44sqq., Sezgin, GAS,
vol. 1, p. 28, no. 2], are traced back to Ibn Abbs and Ibn Masd.)
98
al-Wid, loc. cit. Both are wrong; see below.
99
Kazem-Beg, Observations sur le Chapitre inconnu du Coran, publi et traduit par
M. Garcin de Tassy in Journal asiatique, 4e srie, t. 2 (1843), pp. 375sqq.; cf. George Sale, The
Koran; Preliminary Discourse, section 3, pp. 4454. Al-Bukhr, fail al-Qurn, 16. Kazem
Beg is following in his article uncritically almost exclusively contemporary Shiite authors.
100
101
Das Leben Mohammeds, p. 361, and foot-note on p. 569.
102
Details thereon below in the chapter, Muammads uncanonical promulgations.
103
Against this we have the tradition that Muammad gave the scribe precise calligraphic
instructions (A.I. Silvestre de Sacy, Recueil de diffrens traits, p. 357). This must be consid-
ered the fabrication of a scribe anxious about the external appearance of the Koran.
104
See below regarding sra 20.
mu
says in a poem
105
about the Battle of Badr that the former resting place of
Zaynab was like kha al-way, on smooth parchment. It is unfortunately
not quite clear what the meaning of revealed writing is, or, generally,
mysterious, effaced writing,
106
with which pre-Islamic poets are inclined
to compare the traces of deserted settlements.
When Muslims put down in writing the sras they had memorized, they [i/46]
probably often put passages together that had originated in the same period
and had the same rhyme. This would plausibly explain how the individual
parts of the long Medinan sras, which could not have been created at the
same time, still belonged largely to the same period.
When Muammad recited Koranic passages to be memorized or written
down, it might have been that he only then decided on their final version.
This is quite evident from the following account, which is supplied by most
of the Commentators
107
on sra 6:93.
When Muammad once dictated the beginning of sra 23 to Abd Allh
[Ibn Sad] b. Ab Sar (died 57/676677),
108
whom he frequently employed
as a scribe,
109
he was so enraptured by the description of Gods creation that
he exclaimed: . Then the Prophet explained that the
exclamationwas intotal agreement withthe words of the Koranand that they
belong here.
Abd Allhs words evidently appeared to Muammad to be so fitting as to
introduce them on the spot.
Muammad, who did not hesitate either to repeat verses or to change [i/47]
or to abrogate passages, and whose work concentrated to a great extent on
the immediate circumstances, was not at all inclined to arrange the sras
according to chronology or subject. But this is no reason that we should
105
Dwn, ed. Tunis, p. 10, l 12; Ibn Hishm, p. 454.
106
Cf. the passages inTh. Nldeke regarding Labds Muallaqa, Fnf Moallaqt, p. 65; Ibn
Hishm, p. 702, l 11; Yqt, Geographisches Wrterbuch, vol. 4, p. 422, l 18; Lisn al-Arab, vol. 2,
p. 19, verse 1; vol. 5, p. 229, verse 1; vol. 9, p. 46, verse 1; above, p. 18.
107
E.g., al-Zamakhshar, al-Bukhr, al-Baghw; al-Zamakhshar also on sra 23:14.
108
EI
2
.
109
This man is to be added to Uthmn, Muwiya, Ubayy b. Kab, and Zayd b. Thbit,
known as
(Sprenger, Leben und die Lehre,

2
vol. 3, p. xxxi). Some others mentioned
as Muammads scribes will have looked after Muammads correspondence (see al-abar,
vol. 1, 1782; Ibnal-Athr, Usdal-ghba, vol. 1, s.v. ; al-Nawaw, [Tahdhbal-asm] Biographical
dictionary, ed. F. Wstenfeld, p. 37; A.I. Silvestre de Sacy, Mmoire sur l origine et les anciens
monuments de la littrature parmi les Arabes, p. 332, foot-note; A.I. Silvestre de Sacy, Recueil
de diffrens traits relatifs l orthographie et la lecture de l Alcoran, p. 357; G. Weil, Das
Leben, p. 552, foot-note). Cf. the letters in Ibn Sad, and above, p. 11 sq.; EI
2
; Goldziher, Schools
of Koranic commentators, p. 24, n. 11.
38 mu
hammads revelations
seriously reproach him as Gustav Weil
110
did. Could the Prophet really have
foreseenas Weil thinksthat a dispute would arise after his death over
the very letter of the revelation, especially given that he was an unlettered
man with no idea of the veneration of letters? His spirit, which naturally
was aiming at the most immediate goal, could by no means anticipate the
strange development that Islamwas to followafter his death. Leaving worry
about the future to his God, he likely never pondered over the fate of the
Koran, or least of all the choice of a successor. The complete collection of
the entire Koran was beyond the feasibility of even its author. Not only
according to Muslimtradition,
111
but even the evidence of the Koran
112
shows
that he himself had already forgotten some passages and had deliberately
changed others. The following example makes it unequivocally clear that
Muammad occasionally made expedient additions to the established text.
When those who did not participate in military campaigns were severely [i/48]
reproached in the Koran, two blind men came and anxiously asked whether
the reproach also applied to them; the Prophet ordered Zayd b. Thbit [ibn
al-ak]
113
to add a fewwords excepting handicapped persons.
114
We shall
see below in more detailed discussion that entire passages were clearly
interchanged after short or long intervals. But some parts Muammad
recited in different versions to different people, partly because he wanted
to improve them oras seems to apply in most casesbecause his mem-
ory failed to retainthemunchanged. Regarding this subject there are several
traditions, the best of which can be traced back to Umar (Ibn al-Khab)
and Hishm b. al-akm [Ibn izm
115
], who were quarreling over their
110
Historisch-kritische Einleitung, p. 42sq., 2nded. p. 53. Inthe final analysis all the founders
of the great religions might be reproached with just as much or rather, with just as little jus-
tification.
111
Cf. ishas tradition frequently found in al-Bukhr (e.g., K. al-Shahdt 11) and
Muslim, vol. 1, p. 443sq. (= al-Qasalln, vol. 4, p. 72sqq., Fail al-Qurn, bb 2):

or merely
. In this case the verses had already been communicated to others.

112
Sra 2:100 (where admittedly others read = ) and sra 87:6sq.
113
EI
2
; EQ; Juynboll, Encyclopedia, pp. 232sqq.; Sezgin, GAS, vol. 1, pp. 401402.
114
The story is supported by the evidence of several persons, among them also Zayd
himself. See al-Bukhr in 31; Fail al-Qurn 7; al-Tirmidh, K. al-Jihd, 22;
al-Nas, ibid. 3; Muslim, vol. 2, p. 231 (al-Qasalln, vol. 8, p. 114sq., jihd); Ibn Sad
(al-abaqt al-kabr): Biographien der Muhgirn, p. 154sq.; al-abar, Tafsr, vol. 6, p. 134;
Ab l-Layth al-Samarqand, al-Wid, al-Zamakhshar, al-Bayw on sra 4:97. Cf. Silvestre
de Sacy, Mmoire sur l origine et les anciens monuments de la littrature parmi les Arabes,
p. 424.
115
Juynboll, Encyclopedia, p. 343, col. 1.
mu
respective readings of sra 25. When they consulted the Prophet he decided
that they were both correct about the revelations, since the Koran was
revealed , all of which were correct.
116
It is reported also by
Ubayy b. Kab that he once heard a man recite the Koran in a mosque
in a reading unknown to him. He ignored it but soon thereafter another
person did it again. He thereupon went to the Prophet, and he approved
of the reading. When Ubayy b. Kab was frightened by this, fearing that he
be considered a liar, the Prophet calmed him with an answer similar to the
one which he had given to Umar and Hishm[Ibn al-akmb. al-izm].
117
In the same vein we must consider the undeniable differences among
the variant readings of the Companions of the Prophet, which are reflected
in the following passage (further details below in Otto Pretzls section The
history of the text of the Koran):
118

119

120

All such variations, which we can easily explain, caused great troubles for
Muslims. Particularly the explanation of the words,

or, as one variant
121
has it, , meant a lot of hard work for
Muslims. Several traditions are subservient to writers ends;
122
and already
116
al-Muwaa, p. 70; al-Bukhr, K. al-Bad al-khalq, bb 5, 10; K. Fail al-Qurn 4;
khumt 3 (al-Qasalln, loc. cit, vol. 4, p. 237, lists parallel cases; Muslim, vol. 1, p. 457
(al-Qasalln, vol. 4, p. 97sqq., fail); al-Tirmidh, al-Qirt, bb 2, 1; al-Nas, al-Sunan,
p. 107sq. (I, 149, K. al-Iftit, 37
); al-Khab al-Tibrz, Mishct, fail al-Qurn,

bb 3, fal 2; al-abar, Tafsr, introduction, vol. 1, pp. 924). Frequently repeated by later
writers like IbnAiyya, al-Qurub, vol. 1, p. 18
v
, Ibnajar, andIbnal-Athr, Usdal-ghba, vol. 1,
s.v. ; Silvestre de Sacy, Mmoires, 50, p. 425; Goldziher, Schools of Koranic commentators,
pp. 2527; etc.
117
Muslim, Fail al-Qurn, 13; al-Khab al-Tibrz, Mishct, ibid., bb 3, fal 2; al-Nas,
Iftit, 37; al-Qurub, vol. 1, p. 18
v
f.; al-abar, Tafsr, vol. 1, p. 9sqq.
118
In Ibn Sad, (al-abaqt al-kabr): Biographien der mekkan. Kmpfer, p. 270, l 8sqq.
119
This is Ibn Masd. The authenticity is of course not established.
120
According to Yqt, vol. 3, p. 218, etc., this al-Saylan is located in Iraq not far from
ra, and is possibly identical with the Hebrew (A. Neubauer, La gographie du Talmud,
p. 262). The proverb more famous and direct than the road to al-Saylan I have found
nowhere else.
121
al-Mabn li-namal-man, vol. 4; al-abar, Tafsr, vol. 1, p. 12, l 2: 6 or 7 [sic].
122
E.g. Gabriel told the Prophet that the Koran ought to be read ; but he
objected because the Muslims were too weak; God then agreed to two sets of readings, then
upon a renewed request, to five, and finally to seven (Muslim, al-Qasalln, vol. 4,
p. 102sqq.; al-Azraq, p. 436; al-Khab al-Tibrz, Mishct, p. 184 (192); al-Qurub, i, p. 16
r
).
40 mu
hammads revelations
Ab tim Muammad IBN IBBN al-Bust (d. 354/965)
123
was able to
collect thirty-five to forty different types of explanation, most of whichor
at least the most important of whichwe find in a variety of books.
124
(Abd
al-Ramn b. Isml) AB SHMAH (d. 665/1266
125
) composed a book on
this particular subject.
126
But since all of them are largely worthless, even
ridiculous, and contrary to the text of the traditions, we shall limit ourselves
to a few examples.
The Seven aruf (Sets of Readings). Abrogation of Revelations
The seven aruf are supposed to indicate the seven subjects of the Koran, [i/50]
namely stories, commandments, interdictions, etc., or sevendifferent senses
(one outwardandsix inwardones), or the sets of readings of the sevensubse-
quent readers (see below; this point of viewis considered a sign of ignomin-
ious ignorance in al-Itqn, p. 115), or the seven languages, fromwhich words
are allegedly borrowed for the Koran,
127
etc. Some Shiites take the easy way
out and reject this entire tradition. Even some Muslims
128
have recognized
that the number seven is of little consequence and that, instead, it serves
here, as elsewhere, to represent an unknown quantity, regardless of whether
Muammad himself may have fixed it or it was inserted later. is letter,
reading. The words thus express no more than the permission to read the
Koran in different sets. This differencealso admitted by some Muslims,
thinking that it might have been permissible to exchange single words with
others of identical meaning
129
might have been rather extensive, compris-
ing the omission and addition of entire verses.
The frequently mentioned explanation that the seven indicate [i/51]
seven Arabic dialects must be dismissed. Muammad certainly left it to
Similar traditions abound.; cf. al-Tirmidh, al-Nas, and al-Khab al-Tibrz, Mishct, loc.
cit. (p. 33 n. 116); Mabn ix; Ibn Aiyya; al-Qurub, p. 16sqq.; al-Itqn, p. 105sqq, etc.
123
EI
2
124
Ibn Aiyya; al-Mabn li-nam al-man, ix; al-Qurub, loc. cit.; Cod. Lugd. 653 Warn;
Abd al-Ramn IBNAL-JAWZ (cod. Gotha 1671 = Pertsch, Arabische Handschriften, no. 544);
al-Itqn, loc. cit.; the Shiite Tafsr al-Qurnof Muammadb. Murtad, Ahlwardt, Verzeichnis,
no. 899 (= cod. I Petermann, no. 553).
125
Encyclopaedia of Islam, new edition.
126
Cf. the great work of Ibn al-Jazar, al-Nashr f l-qirt al-ashar, Ahlwardt, Verzeichnis,
no. 657 (= cod. Petermann i, no. 159), folio 9
r
.
127
Arabic, Greek, a Coptic dialect (
), Persian, Syriac, Nabataean, Ethiopic!

128
Ibn al-Jazars great work, folio 11
r
; Itqn, p. 107.
129
As, for example,
, , , , ,
. Al-Qurub, vol. 1, Jmi akm al-Qurn,

folio 16
v
; al-Itqn, p. 108sqq., etc.
mu
each person to pronounce the Koran according to his native dialect,
130
but
this type of difference was in no way so great that his Companions might
have started to quarrel. Even in Ibn Aiyyas introduction to his al-Jmi al-
muarrar al-a al-wajz f tafsr al-Qurn, and al-Suys al-Itqn, p. 111,
we find the appropriate remark that this does not fit with the story of Umar
and Hishm, as both were Quraishites.
Incidentally, in the historical account itself I found instead of
only in Kazem-Beg, loc. cit. The list of the seven dialects is completely
arbitrary, with the most unfortunate result that a dialect was assigned to
every tribe dwelling inthe sacredterritory of Mecca or its immediate vicinity
(Quraysh, Kinna, Khuza, Thaqf, etc.), or even to tribes to whom this did
not apply at all.
131
These sets of readings are frequently mentioned in connection with the [i/52]
tradition that Gabriel habitually recited the Koran to the Prophet once
a yearor every Raman(i.e. if the tradition makes any sense at all,
those parts of the Koran that had been revealed to this date); if he then
had omitted or added anything, the Companions would have memorized
this,
132
thus accounting for the variants.
130
Ibn Masd is said to have permitted a person who could not say
(sra 44:44) to say

, and to read or (al-Mabn li-nam al-man, vol. 9; al-Itqn, p. 109), but this
example is likely to have been fabricated to serve some kind of theory: Either that person was
able to pronounce every initial only as then it would have been absurd to expect him
every time to search for another wordor this peculiarity applied only to a few words; in
this case he could have easily conformed to the prescribed pronunciation instead of having
to look for an entirely different word. It must be added that the words and interfere
with the rhyme. Moreover, it is unthinkable that Ibn Masd, who is reported to have said
instead of , should have been unable to tolerate such a minor divergence and rather
have taken a completely different word. Incidentally, this change from to is dialectically
documented in ancient and modern times.
131
Mentioned are, e.g., Quraysh, Kinna, Asad, Hudhayl, Tamm, abba, Qays; or Quraysh,
Sad ibn Bakr, Kinna, Hudhayl, Thaqf, Khuza, Asad, abba, or five tribes of the back-side
( ) of Hawzin and from the lowest ( ) of the Tamm. Most of the writers select the
dialects from among the Muar ( ), with preference for the Quraysh (who, however, are
missing from the list mentioned last!), and Hawzin, among whom, according to legend,
Muammad was educated; still others enumerate Quraysh, Yaman (a collective name which
compriseddifferent tribes), Tamm, Gurhum[alsoGorhum] (anoldsemi-legendary people!),
Hawzin, Qua (belonging to the Yaman!), ayyi (the same). But the names mentioned by
Kazem Beg, loc. cit., p. 379, among whom there is even imyar, I found nowhere else.
132
al-Bukhr, Fail al-Qurn, 7, al-awm, 7, al-way; Muslim (al-Qasalln, vol. 9,
p. 162, , p. 337, Fail, Fima); al-Mabn li-nam al-man, vol. 3; al-Qurub,
fol. 22
r
, and frequently; al-Khab al-Tibrz, Mishct, 175 (183, bb al-itikf); al-Shshw, cap.
1 [his al-Fawid al-jamla al l-yt al-jalla, edited by Idrs Azzz (1989)]; al-Itqn, p. 116;
Silvestre de Sacy, Recueil de diffrens, p. 357. Occasionally it is added that this happened to
42 mu
hammads revelations
Yet when it is claimed that Muammad forbade his Companions to quar-
rel about the advantages of variant readings
133
this tradition appears obvi-
ously to be the fabrication of a man who saw in this controversy a threat
to the faith. It is a consistent feature of largely fictitious adths to put later
teachings in the mouth of the Prophet.
There is a difference between what Muammad changed and what is
abrogated ( , sra 2:100). That one revelation can possibly abrogate
another is suchanunprecedentedconcept that it cannot have beenMuam-
mads brain child; rather it would seem to be related to the Christian idea
of the abolishment of the ordinances in the Gospels (e.g. Ephesians 2:15;
Colossians 2:14). It is in this context that the word for the foreign con-
cept is likely to have been borrowed, even if that particular meaning of
is not contained in the Aramaic that we know. Hibat Allh b. Salma
al-Baghdd
134
(d. 410/1019), whose
135
attained tremendous
authority and became the source and model for many later studies of this
subject, classified the abrogated passages as the following:
136
first, passages
abrogated by the sense but retained by the letter of the Koran; second, those
abrogated by the letter but valid according the sense; and third, those abro-
gated by the sense and the letter.
This classificationquite obviously relates tothe contemporary formof the [i/53]
Koran as decreed by Muammad at the divine behest of God. Consequently,
all that has been lost without the will of the Prophet or inadvertently not
included in the collection of the Koran by his successors is regarded as
being abrogated. In addition to Hibat Allh b. Salma, Muslims consider to
be abrogated many verses that are of no practical relevance because their
cause has disappeared. For example, all those verses requiring Muammad
patiently to bear insults and persecution are considered to have been abro-
gated after his situation drastically changed and the matter was no longer
relevant. Al-Suy quite clearly saw
137
that the category of the abrogated
Muammad twice during the last year of his life, or the final reading is the one which Gabriel
used at his last encounter with the Prophet.
133
al-Bukhr,
, 37; al-Khab al-Tibrz, Mishct, ibid., bb 3, fal 1, 2;

al-abar, Tafsr, vol. 1, p. 10 bottom; Ibn al-Jazar, folio 16
v
; al-Suy, al-Itqn, p. 195.
134
Sezgin, GAS, vol. 1, p. 47, vol. 9, p. 183.
135
Frequently found in our libraries (cf. Brockelmann, GAL, vol. 1, p. 192); printed in the
margin of al-Wids Asbb al-nuzl (Cairo, 1316/1898). Other works on the subject are listed
in Flgels edition of Ibn al-Nadm, [EI
2
; EQ; Sezgin, GAS, vol. 8, p. 17sqq.] al-Fihrist, p. 37.
136
Cairoedition, p. 9sqq.; cf. al-Diyrbakr, Tarkhal-khams, vol. 1, p. 14; al-Suy, al-Itqn,
p. 516sqq.
137
al-Itqn, p. 516sqq. A close investigation of the vacillations of tradition is relevant only
mu
matter had grown to become ridiculous. When we consider only the current
composition of the Koran we find in it either the abrogated and the abro-
gating side by side, or only the abrogated,
138
or only the abrogating ones. But
in reality we have to differentiate between two different types of abrogated
passages: first, the validity of a verse is abrogated by an explicit revelation
that applies particularly to law, where it has to be adapted to the require-
ments of contemporary conditions and, second, by a simple prohibition
issued by Muammad to forbid his Companions from reading and copy-
ing any particular passage. The volume of both types cannot be considered
to have been large. Still, we hear of one tradition, which might contain a
kernel of truth, that Muammad personally crossed out a passage of the
Koran
139
that he hadonly recently dictatedtohis followers. Whoever is famil-
iar withthe strange viewMuslims holdabout the Koranwill not be surprised
that there are some who dismiss the whole doctrine of the abrogation, even
thoughthis is clearly statedinthe Koran.
140
This view, however, is considered
heresy.
141
Individual revelations which are missing from the current version of the [i/55]
Koran but have survived in some other way and which, according to the
aforenamed Muslim categories, are reckoned among the abrogated pas-
sages, shall be treated below.
for the history of dogma and fiqh. Interesting is, for example, howin al-Bukhr, Way, 18,
the relevance of the manskht to sra 4:9 is challenged

. . Adetailedinvestigation
of the matter in al-abars Tafsr on sra 2:100.
138
In this case Muslims assume that the Koranic passage was abrogated by the sunna. But
on this point Muslims are greatly divided. Cf. al-Bayw and al-abar on sra 2:100; cod.
Petermann I, p. 555 (a book independent of Hibat Allh b. Salma [Sezgin, GAS, vol. 1, p. 47])
on by Abd Allh IBN hir al-Baghdd (d. 429/1037); al-Suy, al-Itqn, p. 515;
I. Goldziher, Muslimstudies, vol. 2, pp. 2930.
139
The most simple version of this tradition is found in Hibat Allh b. Salma, Cairo
edition, p. 12 (cf. L. Marracci, Prodromus, part 1, p. 42; Ibn Hishm, translation by Gustav
Weil, Das Leben Mohammeds, p. 597, note), where Muammad replied to Ibn Masd, who
was surprised about the disappearance of the writing:
(here ,
abrogate, tollere is used with the identical meaning as ). Asomewhat different version
is found in al-Qurub on sra 2:10, and wonderfully embellished in al-Itqn, p. 526, where
two men forget a sra at the same time.
140
Sra 2:100; cf. sra 16:103. From the Koran this concept was then applied to adth.
141
Ab l-Layth al-Samarqand and al-Qurub on sra 2:100; Hibat Allh b. Salma, Cairo
edition, p. 26; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 14.
44 mu
hammads revelations
The Originality of the Koran and Its Connection
with the Revelations of the Prophet Maslama
At the end of this general discussion of the Koranic revelations it would not
be inexpedient to answer the question of why Muammad dared to chal-
lenge all opponents to bring ten sras (sra 10:16) to question his exclusive
prophetic claim, and, when they were unable to do so, to bring even only a
single one.
142
The fact that no one could meet this challenge, even at a time
when Arabia abounded in master rhetoricians, Muslims to this very day rec-
ognize as irrefutable proof of the divine origin of the Koran, which by its
nature discredits all human art. This point of view, which entails many a
controversy, is expounded in several works entitled .
143
But when we take a close look at Muammads challenge we find that
he was not asking for a poetic or rhetorical equivalent to the Koran but
rather for something essentially equal to the Koran. It was in the nature
of the request that his opponents could not comply. Should they defend
their ancient belief in their gods in the same way as Muammad espoused
the unity of God and the related dogmas? Should they let the gods speak
for themselves? This would have been nothing but satire and absurdity. Or
should they equally become enthusiastic about the unity of God and restrict
their opposition to Muammads prophethood? In this case they could only
copy the Koran, which they intended to rival, but an image can never rival
the original. Muammads faith was a novelty for his people, and therefore
produced an inimitably original expression. The difficulty was substantially
enhanced by his clumsy style.
In spite of everything, Muammads challenge was not entirely without [i/56]
response. Still during his lifetime, and shortly thereafter, men appeared at
various places of the Arabian Peninsula and claimed to be prophets of their
people and to be receiving divine inspiration: Laq b. Mlik in Oman (al-
abar, vol. 1, p. 1977, l 7sq.), Abhala b. Kab al-Aswad in Yemen, the Asadite
ulaya, the Tamimite Musaylima, and finally the prophetess Saj.
144
They
142
Cf. thereon Martin Schreiner, Zur Geschichte der Polemik , pp. 663675. [Arberrys
translation: Bring a Koran other than this, or alter it.]
143
Sra 10:39; and 2:21. The founder of the Bbs, Mrz Al Muhammad of Shrz [Goldzi-
her, Schools of Koranic commentators. p. 33 n. 70] claimed such a mission, inproof of which he
had produced verses and a book like the Kurn, but surpassing it in wisdomand eloquence.
(E.G. Browne, The Babis of Persia, p. 916sq.). Regarding imitations of the Koran in later peri-
ods see Goldziher, Muslimstudies, vol. 2, p. 363sqq.
144
Cf. J. Wellhausens reflections in his Prolegomena zur ltesten Geschichte, pp. 737.
mu
all spread revelations, yet only Musaylimas fragmentary sayings
145
have sur-
vived to give us a vague idea of the mans religious ideas. As a religion aware
of its strength, and wanting to appear as the best of the world, the young
Islam, fighting for survival, unhesitatingly declared all these movements to
be nonsense and the work of Satan. Success proved it right, yet in every
other respect this verdict is unfair and false. Musaylimas and Muammads
doctrines are of course closely related. Both of them have in common the
important, fundamental components of Islam, such as eternal life ( , al-
abar, vol. 1, p. 1917, l 29), the divine name Ramn (al-abar, Tafsr, vol. 1,
p. 1933, l 12; p. 1937, l 3, cf. p. 1935, l 14; and al-Baldhur, Fut al-buldn,
p. 105, l 6), fasting (al-abar, p. 1916, l 14; 1917, l 1), proscription of wine (1916,
bottom), and the three
146
appointed times of prayer (al-abar, vol. 1, p. 1919, l
2sqq.). Still, this similarity most likely does not constitute a borrowing from
Islam but rather a mutual dependence on Christianity. Musaylimas teach-
ing, however, contains peculiar elements that originate from Christianity
but are foreign to the Koran, e.g., the commandment of sexual abstinence
as soon as a male child has been born (al-abar, vol. 1, pp. 1916 and 1917, l
47), and the eschatological concept of the kingdom of heaven
147
(

, al-abar, vol. 1, p. 1917, l 2). Rhymed prose is even less likely to
have been borrowed, since it had been a favourite form of religious diction
among the Arabs since long before Muammad. Moreover, Musaylima dis-
plays so much originality in his expressions, particularly in his similes, that
his allegedimitationof the Koran(
, al-abar, vol. 1, p. 1738, l 17; Ibn

145
al-abar, vol. 1, p. 1738, l 14 and 17sq.; p. 1916, l 10; p. 1917, l 4; p. 1933, l 2; p. 1934, l 6;
p. 1957, l 4, and5. The reliability of the transmissiongenerally cannot be doubted. The obscene
dialogue (al-abar, vol. 1, p. 1917, l 12 to1918, l 10) is of course a malicious fabrication. The mans
real name was Maslama as e.g. al-Mubarrad, al-Kmil, p. 443, l 5, ina verse, but the diminutive
of his name became commonamong Muslims withthe meaning of satirical belittlement (Ibn
Khab al-Dahsha, Tufat dhaw l-Arab, ed. by Traugott Mann, s.v.), the same as the name of
the prophet Tala was changed to Tulaya (al-Bayhaq, ed. by Fr. Schwally, p. 33, l 5).
146
According to M.Th. Houtsma (Iets over den dagelijkschen alat [Some remarks about
the ritual prayer]) it is quite likely that Muammad instituted only two daily canonical
prayers, to which he later added a third one, the middle (al-wus) alt. Goldziher, Carra
de Vaux, Le Mahomtism, p. 385; and Goldziher, Islam in Jewish encyclopedia, vol. 6 (1904),
p. 653, col. 1, expanded this thesis, saying that after the Jewish pattern soon were added the
other two in imitation of the five gh ( ) of the Parsees. Cf. also Leone Caetani, Annali
dellislam, vol. 1, 219; vol. 2, tom. 1, p. 354sqq., and p. 635sqq.
147
The expression ( ) found nineteen times in the Koran is not,
or at least not primarily, to be interpreted eschatologically, rather it is merely a statement
that Allh is the Lord of the World (also 38:9). It is obviously easy to associate this with
eschatological ideas (e.g. 45:26).
46 mu
hammads revelations
Hishm, p. 946, l 14) becomes less probable. This originality also stands as a
notable argument for the essential authenticity of the revelations attributed
to him. If this were founded purely on the invention of Muslimtheologians,
one would expect to find a greater similarity with the Koran.
THE ORIGIN OF
INDIVIDUAL PARTS OF THE KORAN
Aids in Establishing the Chronology of the Sras
Whenstudying the individual parts of the Koranwe must keep inmind both [i/58]
the period and the occasion of a revelation. In order to give the reader an
idea of the limitations of such an investigation from the outset, we must
first explain the research aids available to help solve the problem as well as
outline the difficulties that will be encountered.
Our prime source is the historical and exegetical tradition. This is most
reliable whenit is relatedtothe great historical events inthe history of Islam.
For example, no one can doubt that sra 8 refers to the Battle of Badr, sra
33 to the Battle of the Trench, and sra 48 to the Pact of al-udaybiyya.
However, the number of these most reliable facts is not very large and
applies only to the Medinan sras, as Muammad remained much in the
background at Mecca, where he did not initiate great historical events.
Substantial doubt hangs over the very numerous traditions regarding the
myriadminor events mentionedby historians andexegetes tothrowlight on
single verses of the Koran. Since we shall discuss the origin of this exegetic
tradition in the literary survey, let us supply here only some examples of
its unreliability, namely that the origin of verses universally regarded as
Meccan frequently turns out to lie in events after the emigration, and that
frequently two closely related verses
1
are attributed to entirely different
occasions, though these explanations often do not fit the context of the
passage. Still, among the mass of doubtful and fraudulent information one
also finds more reliable data that, supported by historical events, can be of
great benefit to anyone using it with discretion. Such criticism is not easy,
since the bias underlying an individual tradition does not become apparent
until all potential traditions from the same source have been collected. As
long as there is no systematic investigation of the exegetic traditions, we
have no choice but to check the reliability of each and every tradition. From
1
See above.
48 the origin of individual parts of the koran
these innumerable, fraudulent, often mutually contradictory explanations
of Muslim tradition only a limited selection can be offered here.
In order to obtain a representative example, we shall in most cases con- [i/59]
sider the traditions on the original locality of revelations of both complete
sras or single verses, as found from time immemorial not only in his-
torical and exegetic works but also in Masoretic texts and in most of the
manuscripts of the Koran.
The Transmitted Lists of the Chronology of the Koran
A chronological list of sras has been transmitted to us. The list, however, [i/59]
considers only the beginning of the sras and not the verses that might
have been added later.
2
As the texts of this catalogue frequently differ con-
siderably, it might not be superfluous to compile an exact inventory of the
transmissions.
3
In the fifth-century book (also in Miguel Casiri, Biblioteca
Arabico-Hispana Escurialensis, v. 1, p. 509, without title) of (ABAL-QSIM)
Umar b. Muammad IBNABDAL-KAF
4
(Leiden Ms. 674, Warner) 13 verso,
sq., we find the following enumeration.
(1) Meccan sras: 96, 68, 73, 74, 111, 81, 87, 92, 89, 93, 94, 103, 100, 108, 102,
107, 109, 105, 113, 114, 112, 53, 80, 97, 91, 85, 95, 106, 101, 75, 104, 77, 50, 90,
86, 54, 38, 7, 72, 36, 25, 35, 19, 20, 56, 26, 27, 28, 17, 10, 11, 12, 15, 6, 37, 31,
34, 39, 40, 41, 42, 43, 44, 45, 46, 51, 88, 18, 16, 71, 14, 21, 23, 32, 52, 67, 69,
70, 78, 79, 82, 84, 30, 29, 83.
(2) Medinan sras: 2, 8, 3, 33, 60, 4, 99, 57, 47, 13, 55, 76, 65, 98, 59, 110, 24,
22, 63, 58, 49, 66, 62, 64, 61, 48, 5, 9.
Missing is the first sra, which is attributed equally to Mecca and Medina
(see below). As for all the others omitted fromthe list, the explanation is, of
course, only an error in the text.
This version of transmission is identical to that of al-Mabn li-nam al- [i/60]
man, vol. 1, and al-Suy, al-Itqn, p. 21 sq., except sras 58sqq., which are
omitted in the former.
2
This is how it reads in at least in K. al-Mabn li-nam al-man, cap. 1. This is also
the only sensible way to present those sras that have been brought together from different
periods in a chronological order.
3
Three scholars have previously drawn attention to these lists: J. von Hammer-Purgstall
(Der Islam und Mohammed, p. 82sqq.), G. Weil (Mohammed der Prophet, p. 364sqq.), and
G. Flgel (ber Muhammad bin Ishks Fihrist, p. 568).
4
Sezgin, GAS, vol. 1, p. 16 n. 24.
the origin of individual parts of the koran 49
Another version (al-Mabn li-namal-man, no. 3) differs only in so far
as it is left undecided whether sra 98 is of Meccan or Medinan provenance.
This version can be traced back to Ibn Abbs through A b. Ab Rab
[Aslam al-Qurash, d. 114/732.]
5
Another version found in al-Diyrbakr, Tarkh al-khams (Cairo edition,
p. 10), inadvertently omits sras 68 and 73, and places sras 50 and 90 before
95, 61 before 62, and 9 before 5.
The version in al-Suys al-Itqn, p. 20, which al-usayn b. Wqid
6
and
others trace back to Ikrima b. Abd Allh al-Barbar
7
and al-asan b. Ab
l-asan, omits some sras but places sra 44 after 40, sra 3 after 2, and
makes sra 83 the first of the Medinan sras.
The fourth version, in al-Mabn li-nam al-man, can be traced back
through [Ab Muammad] SAD IBN AL-MUSAYYAB [Ibn azn al-Makh-
zm, 13/63494/713]
8
to Al and Muammad. It considers the first sra to
be the oldest, puts sra 53 among the last of the Medinan sras (sic), places
84 after 83, and omits sras 111 and 61.
The first one in the same book with an isnd including (Muammad b.
al-Sib) AL-KALB,
9
AbliBdhm[al-Kf, d. 120/738
10
], andIbnAbbs
places sra 93 before 73, 55 after 94, 109 after 105, 22 before 91, 63 before 24,
and considers 13 to be the first of the Medinan sras, ending with sras 56,
100, 113, and 114.
Although al-Yaqb
11
(Historiae, vol. 1, 32sq., 43sq.) mentions these au-
thorities, with respect to the aforementioned differences he has only the
first two and the last one identical with al-Mabn li-nam al-man, vol-
ume 1. His arrangement of the above list is as follows: sra 1 after sra 74;
sra 100 as Medinan; sra 109 missing; sras 113 and 114 as Medinan; sra 112
missing; sra 56 as Medinan; sras 34 and 39 behind 43; sra 32 as Medinan
and confused with sra 13; sras 69 and 84 are missing; sra 83 is the first
Medinan sra; sra 59 before 33; sra 24 before 60; 48 before 4; 99 missing.
Starting with sra 47, the differences are considerable: 47, 76, 65, 98, 62, 32,
40, 63, 58, 66, 49, 64, 61, 5, 9, 110, 56, 100, 113, 114.
5
EQ; EI
2
; Juynboll, Encyclopedia, pp. 139140; Sezgin, GAS, vol. 1, p. 31 sqq., vol. 5, p. 24,
vol. 8, p. 22.
6
He is also mentioned in al-Diyrbakr, Tarkh al-khams.
7
EI
2
; EQ; Juynboll, Encyclopedia, p. 241; Sezgin, GAS, p. 23sqq.
8
EI
2
; Juynboll, Encyclopedia, p. 4sqq.; Sezgin, GAS, vol. 1, pp. 276.
9
EI
2
; EQ; Sezgin, GAS, vol. 1, pp. 21 and 34.
10
EQ.
11
EI
2
; EQ.
Flgels al-Fihrist, p. 25sq., according to the transmission of al-Wqid
12
[i/61]
from Mamar b. Rashd (d. 154/770)
13
from Muammad b. Muslim AL-
ZUHR, d. 124/742,
14
from Muammad b. Numn b. Bashr, has the follow-
ing list: 96:15, 68, 73 (
), 74, 111, 81, 87, 94, 103, 89, 93, 92, 100,
108, 102, 107, 105, 112, 113, 114 (according to others, Medinan,
15
) 53, 80, 97, 91,
85, 95, 106, 101, 75, 104, 77, 50, 90, 55, 72, 36, 7
16
( ), 25, 35,
17
19, 20, 56, 26, 27,
28, 17, 11, 12, 10, 15, 37, 31 (
), 23, 34, 21, 39 to 45, 46 (
), 51, 88,
18 (
), 6 (
), 16
18
(
), 71, 14, 32, 52, 67, 90, 70, 78, 79,

82, 84, 30, 29, 83, 54, 86.
Medinan
19
2, 8, 7
20
(
), 3, 60, 4, 99, 57, 47, 13, 76, 65, 98, 59, 110, 24, 22,
63, 58, 49, 66, 62, 64, 61, 48, 5, 9.
We see that the arrangement from96 to 87, from108 to 105, from53 to 90,
from 25 to 17, from 39 to 18, from 52 to 83, and from 76 to 9 is identical with
al-Itqn, p. 20; everywhere else great diversity prevails.
The sequence ascribed to Ab l-Shatha JBIR IBN ZAYD al-Azd
21
(d. 93/
711)
22
and Al in al-Itqn, p. 56sq., differs still more. It places sra 42 after
sra 18, and, starting with sra 42, reckons as follows: 32, 21, 16:140, 71, 52,
23, 67, 69, 70, 79, 82, 84, 30, 29, 83 (Medinan) 2, 3, 8, 33, 5, 60, 110, 24, 22, 63,
58, 49, 66, 62, 64, 61, 48, 9. Al-Suy himself calls this .
When we now select even the very best from among these versions of [i/62]
transmissiontheir diversity, as canbe seen, being rather considerable, and
their origin impossible to trace back to a single archetypewe still arrive
at no useful result. In all of these cases, sras revealed to be very old by
various reliable indicators are, nevertheless, placed after much later ones,
turning Meccansras unequivocally intoMedinanones. This tradition, even
12
EI
2
13
EI
2
; Juynboll, Encyclopedia, pp. 401, 587, 626, 630; Sezgin, GAS, vol. 1, pp. 290291.
14
EI
2
; EQ; Juynboll, Encyclopedia, pp. 690730; Sezgin, GAS, vol. 1, pp. 280283.
15
According to a note at the end of the list also sra 114 is Medinan.
16
This sra is again listed later, but under its common name (
) as the third Medi-

nan.
17
The words can be interpreted differently. Either sra 35
is listed under two of its common names and was introduced only inadvertently, or this
designation, Sra of the Angels, indicates sra 33, which would otherwise be missing from
the list.
18
Fihrist, p. 25, l 32sq. at the end of the Meccan sequence has the following addition:

.
19
al-Fihrist, p. 26, l 2sqq.: 85 sras originate from Mecca, 28 from Medina (according to
Ibn Abbs). This makes 113 sras. Thus the Ftia does not seem to be reckoned as sra.
20
But see above, note 16.
21
EI
2
; EQ; Juynboll, Encyclopedia, pp. 128, 442; Sezgin, GAS, vol. 1, p. 586.
22
Juynboll, Encyclopedia; Sezgin, GAS, vol. 1, p. 586.
if very old and possibly stretching back to Ibn Abbs, can be nothing other
than a rough attempt at drawing up a chronological order on the basis of
an extremely uncritical foundation and pure fantasy, with mere reference
to some good traditions. An accurate transmission of the chronology of
the early sras, or of the Meccan sras in general, is nearly unthinkable.
Or are we to assume that Muammad kept a record of the chronological
order of the sras? This would be a nice counterpart to Gustav Weils ironic
pigeonholes for the individual sras, where later revelations could be added
as received.
Moreover, there are plenty of traditions that differ considerably fromthis
one. In al-Itqn, p. 23sq., for example, the Medinan sras are listed in two
different ways. They agree, however, on the chronological sequence of the
Meccan sras. It is said that, excepting individual verses of other sras, the
controversy over a pre or post-hijra origin is limited to sras 18, 55, 61, 64, 83,
97, 98, 99, 112, 113, and 114. This is wrong, however, because the controversy
applies to many more sras. Moreover, the enumeration of the Medinan
sras as found inal-Qurub, folio 23 verso, and, with only minor differences,
in al-Shshw, cap. 20, is again different from the two versions referred to
previously.
Thus, if we, like later Muslims, were to depend solely, or almost solely, on [i/63]
transmissions fromolder teachers, we would rarely arrive at a solid, or even
less frequently, at an accurate result. Yet there still remains one reliable aid
that leads to a profitable use of traditions, namely a precise appreciation
of the sense and diction of the Koran itself. By careful observation even
the casual reader of the Koran will become increasingly convinced that the
passages with passionate diction and ideas must have been promulgated
earlier than those with serene, broad content. We realize that Muammad
moved from the first style to the second gradually rather instantaneously,
and that he displays individual gradations in both.
An important element is the length of the verses. The moving, rhythmic
diction of the earlier period, more closely related to the true saj, requires
far more pauses than the later style, which gradually moved closer to pure
prose. A comparison of two passages with identical subjectseven if they
do not originate from entirely different periodscan occasionally suggest
the likelihood that one originated earlier than the other. Since Muammad
often repeats himself explicitly, it is sometimes possible to distinguish the
original from the later version. Like all writers, Muammads diction in
different periods displays preferences of word and phrases that facilitate
the establishment of a chronological order. By observing the rhyme, the
language in the widest sense, and especially the context of his ideas, we
can identify the individual parts of which sras are often made up. Of
course, when considering the context we must not hastily presume an
interpolation whenever a logical connection seems to be lacking. It is a
consistent characteristic of the Koranic style that ideas seldom develop
calmly, instead jumping from here to there. Yet careful observation easily
shows that there is at least an inkling of connection.
Muslims, too, tried to go beyond the rudimentary tradition and follow [i/64]
a more critical methodology examining linguistic usage. For example, they
readily concede that passages containing
are Medinan, but

that the address, , though mostly occurring in Meccan verses, is
occasionally also found in Medinan verses,
23
or that the Meccan verses are
shorter than those from Medina.
24
Occasionally Muslims even attempt to
dismiss traditions with arguments taken from the passage itself. A case in
point is al-abar in his Tafsr, as well as al-Farr al-BAGHAW,
25
who refute
the tradition that sra 13:43 refers to Abd Allh b. Salm [Ibn al-rith]
(d. 43/663)
26
on the grounds that it is supposed to be a Meccan sra. Such
criticism we find in al-Itqn, p. 25sqq., and p. 37sq., where it is stated
directly (p. 31) that
there
are some people who in the case of exception (i.e. regarding individual
verses that have been promulgated at different places from that of the
sra in which they occur) rely on individual judgement without regard for
tradition. But this analysis, particularly if it goes beyond matters that are
generally obvious, has no solid, critical foundation, even amongst Muslims.
Such attempts at interpretation are nearly useless for our purpose.
Careful consideration of the auxiliary means, which tradition and the [i/65]
Koran itself offer, enables us to obtain much more accurate knowledge of
the originof single Koranic passages. Yet our knowledge of this matter leaves
muchto be desired; some of it remains totally uncertain, while other aspects
are at least doubtful. This is even more the case because we have very few
European predecessors in the field of critical investigation of the Koran.
27
23
Ab l-Layth al-Samarqand on sra 4:1, and 5:1; al-Zamakhshar on 2:19; [AB AL-
QSIM] Umar b. Muammad [IBNABDAL-KF]; [Sezgin, GAS, vol. 1, p. 16, no. 24] (Leiden,
cod. 674, Warner) on sra 22; less explicitly Ab l-Layth al-Samarqand, Tafsr. Less accurately
al-Bayw on sra 2:19.
24
Ibn Khaldn, al-Muqaddima (1886) fal 1, 6, p. 87.
25
EI
2
; EQ; Sezgin, GAS, vol. 1, p. 155, no. 2.
26
EI
2
; Juynboll, Encyclopedia, pp. 303, 304, 324; Sezgin, GAS, vol. 1, pp. 304305.
27
Cf. the Literarische Einleitung [literary introduction]; which lists important contem-
porary works, particularly those of Goldziher, SnouckHurgronje, andWellhausen; inaddition
Leone Caetani, Annali dellislam, volumes 1 and 2 as well as Hirschfeld.
The revelations of the Koran consist of two classes: those of Mecca and
Medina. This divisionis quite logical, as Muammads emigrationtoMedina
gave an entirely new direction to his prophetic activity. From the earliest
period Muslims rightfully recognized this, and we must accordingly respect
the distinction. It is worthnoting, however, that, following the customof the
majority of Muslims (al-Suy, al-Itqn, p. 17sq, etc.), we call all passages
promulgated before the hijra Meccan and all later ones Medinan, even if
they were not exactly promulgated at Mecca and Medina proper.
We attempt to adhere to the chronology as much as possible, although
individual passages belonging a different period are best discussed in the
context of the relevant sra, so as not to separate them unduly. An exact
chronological order of the individual parts wouldbe unworkable andimpos-
sible. Moreover, we shall allowourselves a fewexceptions to the chronolog-
ical arrangement for the sake of convenience.
THE INDIVIDUAL
PARTS OF THE CURRENT KORAN:
THE MECCAN SRAS
General Chronology of the Meccan Sras
The historical traditions offer little reliable help when it comes to study- [i/66]
ing the Meccan sras. Even the very first subject of investigation, fixing of
the span of time to which these promulgations belong, is uncertain. Mus-
lims transmit many figures regarding the various periods of Muammads
life but these differ greatly. Far too often in this regard, unfortunately, Mus-
lims fail to admit their ignorance of certain matters, instead conjecturing
according to untenable principles. It is worthwhile to demonstrate this with
an example.
It is certain that Muammad died on Monday, the 12th of First Rab in
11/632.
1
Since it is said that he had been active for a number of years in Med-
ina and Mecca, these years were simply calculated as complete years, so that
the most important periods of his life came to be dated to the 12th of First
Rab, a Monday, or, in any case, to the same month. He is thus assumed to
have arrived on Monday the 12th of First Rab at Qub
2
or Medina,
3
and also
1
That he died on a Monday we know best from a contemporary witness, namely from
a verse of an elegy on his death by assn b. Thbit (Ibn Hishm, p. 1024, l 16; Ibn Sad,
Ahlwardt, Verzeichnis, no. 9648 (= cod. Sprenger, no. 103, folio 166 recto = Dwn, ed. Tunis,
p. 24, l 7)). All traditions are agreed on this point: Mlik b. Anas, p. 80; Ibn Hishm, p. 1009sq.;
al-Tirmidh, Shamil, bb waft rasl Allh; al-Nas, p. 216 (I, 259, K. al-Janiz, 8), al-
abar, Tafsr, vol. 1, pp. 1256 and 1815; al-Yaqb, vol. 2, p. 126, etc. Cf. the evidence col-
lected by Sprenger in his ber den Kalender p. 135sqq. Since among the days of First
Rab which are mentioned as those of his death, only the 12th or 13th fall on a Monday (al-
abar, loc. cit.; al-Yaqb, loc. cit.; Ibn Qutayba, Handbuch, p. 82; al-Masd, Prairies d or,
vol. 4, p. 141 sq.), the second date, which is also mentioned as the day of death (al-abar
Sprenger, loc. cit.), cannot be considered. Sprenger, too, settles definitely (loc. cit.) for the
12th, but he has the most important proof of al-assn b. Thbit only from a secondary
source.
2
Juynboll, Encyclopedia, 5, 6, 289, 321, 595.
3
Ibn Hishm, pp. 333 and 415; al-Wqid, p. 2; Ibn Sad (al-abaqt al-kabr): Biographie
Muhammads bis zur Flucht, p. 157; Ibn Qutayba, Handbuch, ed. Wstenfeld, p. 75; al-abar,
Tafsr, vol. 1, p. 1255sq. Other writers here mentioned only the day of the month, not the day
of the week. It is quite possible that he arrived in this month at his new residence. Other
writers mention the 2nd of First Rab (al-Wqid, loc. cit.; Ibn Sad, loc. cit.), and it remains
to be seen how the afore-mentioned erroneous date of death was established.
56 the individual parts of the current koran: the meccan sras
to have been born
4
and called to his mission
5
both on Mondays. Other writ-
ers add still other events in the life of the Prophet that supposedly occurred
on a Monday.
6
We generally know very little of the chronological order of
the events before the emigration; not even the years of the main periods
are known. The majority of writers fix the period of his prophetic activity
at Mecca to either thirteen
7
or approximately fifteen,
8
while still others to
ten
9
years or somewhat more (Muslim al-Qasalln, vol. 9, p. 197; al-abar,
vol. 1, p. 1248) or, indeed, to only eight years (al-abar, vol. 1, p. 1250, l 4;
Ibn Sad, loc. cit., p. 151). A compromise between the first and the third
points of view seems to be the hint that Muammad received his mission-
ary call at the age of forty-three and subsequently spent another ten years
at Mecca.
10
This tradition does not seem to take into account the three
years
11
during which his public preaching is said to have been interrupted,
particularly as nearly all agree that he receivedhis call toprophethoodat the
age of forty. Little reliability can be attributed to this figure, as the impor-
tance Orientals attach to the number forty is well known.
12
Nevertheless,
4
Ibn Hishm, p. 102; Spr[enger], loc. cit., p. 138sq.; contemporary Muslims celebrate this
day as the Prophets birthday. Other writers mention different dates (Ibn Sad, al-abaqt
al-kabr: Biographie Muhammads bis zur Flucht, 62; Spr[enger], loc. cit., p. 137sqq.) but they
are all agreed on the month; some mention Monday only.
5
Ibn Sad (al-abaqt al-kabr): Biographie Muhammads bis zur Flucht, p. 129; al-abar,
Tafsr, vol. 1, p. 1141 sq., 1255; al-Khab al-Tibrz, Mishct, p. 171 (p. 179,
, fal 1, 10);
al-Wid in the introduction to the Cairo edition, p. 10. Al-Masd, Prairies d or, vol. 4, p. 154,
mentions in addition the First Rab. That this is an error we shall see below.
6
al-abar, Tafsr, vol. 1, p. 1141 sq., p. 1255sq.
7
Various traditions in Ibn Hishm, note on p. 155, l 9; Ibn Sad, ed. Sachau, loc. cit.,
p. 151 sq.; al-Bukhr, al-a, vol. 2, p. 205, bb ( ), p. 211 (bb ); Muslim
= al-Qasalln, vol. 9, p. 196, p. 198, fail, bb 26); al-Tirmidh, al-Shamil (bb al-sinn);
al-abar, Tafsr, vol. 1, p. 1246sq.; al-Yaqb, vol. 2, p. 40; al-Khab al-Tibrz, Mishkt, p. 513
(521, bb al-mabath, beginning); al-Masd, Prairies d or, vol. 4, p. 132, 138sq., vol. 9, p. 50.
8
Muslim, vol. 2, p. 346 (al-Qasalln, vol. 9, p. 199); Ibn Sad, loc. cit.; al-abar, vol. 1,
p. 1248; al-Khab al-Tibrz, Mishkt, loc. cit.; Ibn Sad (al-abaqt al-kabr): Biographie
Muhammads bis zur Flucht, p. 151, l 20: fifteen or more years.
9
Muslim, vol. 2, p. 434 (al-Qasalln, vol. 9, p. 195sqq.); al-Bukhr, vol. 2, p. 173,
, and other passages; al-Tirmidh, al-Shamil, loc. cit.; al-abar, vol. 1, p. 1255; al-Masd,
Prairies d or, vol. 4, p. 148sq.; Ibn Sad, loc. cit., p. 127, p. 151; al-Wid on sra 24:54. Regarding
the last foot-notes see the collection of traditions in Spr[enger], loc. cit., p. 170sq.
10
Ibn Sad, loc. cit., p. 151; al-abar, Tafsr, vol. 1, p. 1245sq.; al-Masd, Prairies d or, vol. 4,
p. 148sq.
11
A similar attempt at solving this dilemma is the foot-note in Ibn Hishm, p. 155, l 9.
12
In Jewish writings the number forty as a rounded off numeral is frequently found:
Genesis 7:12 and 17; Exodus 34:28, Numbers 14:33, Ezekiel 29:13, 1 Kings 19:8, Jonah 3:4, Acts of
the Apostles 1:3, Apoc. Baruch Syr. 76:3, Mishnah, Pirk bth, vol. 5, p. 21; Talmud, Abdah
Zrah, folio 5 b, top. From the Islamic world cf.
and (Goldziher, Al
the individual parts of the current koran: the meccan sras 57
that Muammadwas publicly active as a prophet for more thantenyears we
see fromthe words of a song, quoted several times by historians and usually
ascribed to [Ab Qays] irmah b. Ab Anas
13
or, less frequently and reliably,
to assn b. Thbit as well.
14

15

16

17

18

19

He lived among the Quraysh for ten and some years, warning them,
expecting possibly to find a friend who would meet him,
and presenting himself to the visitors of the markets.
Such a verse says more than twenty traditions do, although Muslims (al- [i/69]
Qasalln, vol. 9, p. 197) who prefer this verseand undoubtedly this is the
verse in questionare generally reprimanded. It is also this verse that ruins
the entire scenario concocted by Sprenger in his article, which has been
referred to again and again. The conjecture that Muammad was active
for ten years at Mecca is, it seems, tendentious and can be traced back
to a man who wanted to present the Prophets entire public life in two
equal parts, divided by the emigration, so as to give it an outwardly uniform
appearance. The claim that for seven years he heard only the divine voice,
and for eight subsequent years received revelations, is even more difficult
B Mubrak, p. 351, as well the favourite collection of forty traditions on certain objects
(Ahlwardt, Verzeichnis, nos. 14561550), etc.). Sprenger, loc. cit., p. 172, makes the pertinent
reference to sra 46:14. It was for me a great pleasure that also Sprenger became convinced
that Muammad was ignorant of the day of his birth, loc. cit., p. 172. But I add that he also
did not knowthe year. All data are based mostly on rough calculations backwards, including
the synchronisms with the Persian kings. Cf. Th. Nldeke, Geschichte der Perser und Araber,
pp. 168, 172, etc.; Leone Caetani, Annali dellislam, vol. 1, 23. Mahmd Efendi made the futile
attempt to produce an exact astrological calculation regarding the unreliable data (Journal
asiatique). It is quite a different matter if you want to consider only the conventional date as
Sprenger has done.
13
Sezgin, GAS, vol. 2, p. 294.
14
Ibn Hishm, p. 350; al-abar, vol. 1, pp. 1247 and 1248; al-Azraq, p. 377; Ibn Qutayba,
pp. 30, 75; al-Masd, vol. 1, p. 145, vol. 4, p. 141; al-Nawaw in Muslim, al-Qasalln, vol. 9,
p. 197; Ibn al-Athr, Usd al-ghba, vol. 3, p. 18; Ibn ajar al-Asqaln, vol. 2, p. 486; Ibn al-Athr,
al-Kmil f l-tarkh, vol. 2, p. 83.
15
al-abar, vol. 1, p. 1248, . This is a fabrication based on the traditions mentioned on
page 49.
16
al-Masd, vol. 1 and 4
. It is doubtful that this variant can be verified.

17
Ibn Qutayba, p. 75,
Nawaw
.
18
al-Azraq .
19
al-Masd, vol. 1, p. 145; Ibn Qutayba, p. 30
.
to reconcile with this verse.
20
In this case the Prophets actual activity would
have lasted only eight years. I would not dare to come down firmly on the
side of fifteen or thirteen years as the period of Muammads first prophetic
activity. For the time being let us leave it at the latter number of years, as is
generally done.
Content and Characteristics of the Meccan Sras
This example demonstrates the uncertainty of the chronology of events
in Muammads life before the hijra. Only in a very few cases is it pos-
sible to give an approximate idea of how many years before the hijra (as
the only definite date) something happened. Even the best available biog-
rapher, Abd Allh Muammad IBN ISQ b. Yasr,
21
supplies almost no
chronology for the entire Meccan period.
22
In the case of the Meccan sras,
where reference to precise historical events is extremely rare, it is hardly
possible to establish any kind of chronology or establish individual peri-
ods.
The fewchronological cluesof whichnot evena single one is absolutely
certainare as follows: first, sra 53 refers to the flight to Abyssinia,
23
which
is said to have occurred in the fifth year of Muammads mission; second,
sra 20, according to the common story, was revealed before Umars con-
version, which is dated to the sixth year before the hijra; and third, sra
30:1 sqq. is likely an allusion to the war between the Persians and the Byzan-
tines,
24
events that occurred probably in the seventh and eighth year after
Muammads call.
If we use this vague classification as a basis, we can then attribute the
sras of the second category to approximately the years ah5 and 6; the
longer periods before and after would be for the first and third categories.
This division is quite appropriate for the internal character of the individ-
ual periods, yet it poses the problem that the seventy-second sra, which
certainly belongs to the second period, is usually interpreted to refer to
the journey to al-if undertaken by the Prophet after the death of Ab
20
Ibn Sad, Biographie Muhammads bis zur Flucht, p. 151; Muslim, vol. 2, p. 437 (al-Qasal-
ln, vol. 9, p. 499), and al-Khab al-Tibrz, Mishct, p. 513 (521) add .
21
EI
2
; Juynboll, Encyclopedia, pp. 419423; Sezgin, GAS, vol. 1, pp. 288290.
22
Only Ibn Sad does this somewhat more often.
23
See below on sra 53:19.
24
See below on the subject.
lib [Ibn Abd al-Mualib
25
] and Khadja,
26
not many years before the emi-
gration in the tenth year of his prophetic call. However, we could possibly
avoid this difficulty and, by following certain traditions, completely sepa-
rate the journey to al-if from the appearance of the jinn which is men-
tioned here.
27
We cannot pay any attention to the details concerning the
Ascension to Heaven mentioned in sra 17, since its dating is totally vague.
When looking at the sras of the individual periods we will consider only
the internal development without regard for the utterly vague chronol-
ogy.
The single, higher goal of Muammad in the Meccan sras is convert- [i/70]
ing humanity to the only true God andwhat remains inseparable for
himto belief in the resurrection of the dead and the Final Judgement.
Muammad did not attempt to convince his listeners with logical argu-
ments, however, instead appealing to their emotions with rhetorical presen-
tations. Of particular importance are both the description of eternal bliss
for the pious and the torments of Hell for the sinners. The impression that
such descriptionsparticularly the latterleft upon the fantasy of simple
minds, untouchedby, or unfamiliar with, any similar theological imagery, we
must consider to be one of the most powerful means
28
of spreading Islam.
During this period the Prophet frequently resorts to almost personal attacks
on his pagan adversaries, threatening them with eternal punishment. On
the other hand, however, while living in a total pagan community, he sel-
dom quarrels with the Jews, who are much closer to him, and hardly ever
with the Christians.
29
25
EI
2
; Juynboll, Encyclopedia, p. 19, col. 1, 25.
26
EI
2
; Juynboll, Encyclopedia, s.v.
27
Further, see below.
28
Cf. Snouck Hurgronje, De Islam, p. 256sq.; Une nouvelle biographie, p. 150. According
to Hubert Grimme (Mohammed, vol. 1 (1892), p. 14; Mohammed (1904), p. 50) Islam did not
enter by any means as a religious system, rather it was a kind of socialist attempt to counter
certain growing mundane abuses. This assertion, which defies the entire tradition, Snouck
Hurgronje subjected to a thorough review(Une nouvelle biographie, particularly p. 158sq.).
Cf. also Fr. Buhl, Muhammeds Liv, p. 154sq.
29
Not all the passages where Muammad declaims the doctrine that God has progeny
( ) must be interpretedas polemics against the teaching of Christ, the Sonof God. The pagan
Arabs called their goddesses, al-Lt, Mant, and al-Uzz [EI
2
; EQ, Juynboll, Encyclopedia,
p. 143,] daughters of Allh. Most likely, the name means no more than divine female beings
(cf. Wellhausen, Reste arabischen Heidentums, 2nd ed., p. 24sq.). It was not far-fetched to
assume that idolators responded to Muammads overwhelming evidence of the unity of
God by saying that they, too, recognized this since their goddesses were but daughters of
God; cf. sras 37:149sqq., 6:100sq., etc. This sentence in the form in which it has come down
In the different styles of the sras we recognize different types which [i/72]
must each be closely related chronologically. Two large groups particu-
larly stand out: the emotionally moving earlier sras and, secondly, another
group froma later period that closely resemble the Medinan style. Between
both of these groups there is yet a third intermediate group, which leads
gradually from the former to the latter. We must thus differentiate between
the sras of three periods.
30
Classification According to WilliamMuir,
Hubert Grimme, and Hartwig Hirschfeld
In the second part of his Life of Mahomet and history of Islam
31
WilliamMuir
establishes a different arrangement of the sras, which, though it differs in
some parts with ours, is identical in its main points. He divides the Meccan
sras into five stages that he categorizes chronologicallyalbeit without
any support whatsoeveras follows: (1) Sras preceding sra 96, which are
thus before his actual call to prophethood; (2) from the earliest sras up to
his first public appearance; (3) to the fifth year of his call; (4) to the tenth
year; (5) to the hijra. The first three of these stages, however, comprise nearly
all of the sras we combine in the first period, with the result that Muirs
second stage corresponds to the group of sras we consider the oldest and
his first and third stages to all the rest. Muirs fifth stage is, for the most part,
equivalent to our third period. Most numerous in his fourth stage are those
sras which we reckon to be of the second period, although there are many
more added fromother periods. But this difference is considerably reduced
when we realize that Muir attributes seven sras from our first period to
his fourth stage and, conversely, eight sras of the final years of our second
period to his last stage. Thus, the main difference is that Muir puts an earlier
start and end to our second period, his fourth stage. However, there still
remainsix sras that Muir puts inhis fourthstage that we, incontrast, assign
to our last period.
to us in many a Muslim tradition (the idolaters considered the angels to be daughters of
God) cannot be consideredanoldMeccandoctrine. Muslims are incapable of discussing the
nature of other religions, and tinge them all Islamic. They thus have the Qurayshites discuss
resurrection, prophets, etc.
30
G. Weil was the first scholar to establish these three classes in his Historisch-kritische
Einleitung in den Koran.
31
P. 132sqq., but particularly pp. 318320.
The main error of Muirs classification consists in his attempt to arrange
the sras in a strict, chronological order in every respect. Although he is
sufficiently modest to admit that he has not quite reached his goal, in fact
his goal is itself unattainable. In addition, he fails to divide those sras that
are assembled from various pieces, and places entirely too much emphasis
on the length of sras, which is far less important than the length of the
individual verses.
Hubert Grimme
32
basically follows us in his estimation of the Medinan [i/73]
period and the grouping of the Meccan sras. From the first period he does
not include sras 51, 52, 53, 55, 56, 1, 97, 109, and 112, of which he places the
first five inhis secondperiodandthe final four inhis thirdperiod. Otherwise,
he includes in his second period only sras 14 (with the exception of the
Medinan verses 38 to 42), 15, 50, and 54, whereas he assigns sra 76 to the
first period and all the rest to the third period.
Hartwig Hirschfeld,
33
however, dismisses the criteria established by Gus-
tav Weil, WilliamMuir, and the present writer regarding the arrangement of
the Meccan sras, although his own categories (first proclamation, confir-
mation, declamatory, narrative, descriptive and legislative revelations) are
nothing but a different transcription of our principles. Except for one sra
(98), there is total agreement regarding the classification of the Medinan
period. With the exception of sras 51, 1, 55, 113, and 114, Hirschfelds three
first categories consist of the sras of our first Meccan period plus sras 26,
76, and 72 from our second period, and sra 98 from our Medinan period.
Apart from the these qualifications, his three final categories are a mixture
of our second and third Meccan periods.
Over the course of the many years that I have studied the Koran I became [i/74]
increasingly convinced that certain individual groups among the Meccan
sras can indeed be identified, although I also realized that it is impossible
to establish any kind of exact chronological order. Many an indicium that I
had earmarked turned out to be unreliable, while some of my claims, which
at the time seemed quite certain, upon newand careful scrutiny turned out
to be uncertain.
32
Grimme, Mohammed; vol. 2 (1895), pp. 2527.
33
Newresearches into the composition and exegesis of the Qoran (1902), pp. 143145.
THE SRAS OF THE FIRST MECCAN PERIOD
General Comments
The sras of this period can, I think, be identified with some certainty [i/74]
by their style. The power of the Prophets enthusiasm, which moved him
during the first years and allowed him see the Godsent angels face-to-face,
is necessarily reflected in the Koran. The God who inspires him is the
speaker; man remains completely in the background, as was the case with
the great old prophets of Israel.
1
The diction is grandiose, lofty, and full of
daring images, while the rhetorical energy still retains a poetic coloring. The
passionate flow of language, quite frequently interspersed with simple yet
forceful, rather serene admonitions and colourful descriptions, is reflected
in the brief verses; the entire diction is rhythmically moving and often of
great, yet still natural, harmony. The Prophets emotions and premonitions
are reflected occasionally in a certain obscurity of meaning, which generally
is alluded to rather than expressed.
Formulas of Invocation at the Opening of Many Sras
A peculiar but characteristic phenomenon of the sras of this period is the
abundance of conjurationsthirty times against only once (64:7) in the
Medinan srasby which Muammad purports to confirmthe truth of his
address, particularly at the beginning of the sras. As was the case with saj,
he borrowed this custom from the pagan soothsayers (khin, kuhhn), who
used to introduce their predications with solemn oaths appealing less to the
gods than invoking the most diverse natural objects,
2
such as landscapes,
road marks, animals and birds, day and night, light and darkness, sun, moon
and the stars, the heaven and the earth.
3
In his capacity as the Messenger
1
Cf. Ewald, Die Prophetendes AltenBundes, 2nded., vol. 1, p. 31 sq. Inthe earlier periodthis
mode of speech of Muammad is not merely an outward form, rather it has a deep meaning,
although becoming different later on.
2
The question to what extent these formulas are originally based on animistic concepts
cannot here be discussed.
3
Sah in: Ibn Hishm, p. 10, l 14, p. 11, l 5, 11 sq. al-Mustaraf f kull fann mustaraf of
al-Ibshh, bb 60; al-Masd, Les prairies d or, vol. 3, p. 394; Shiqq: in Ibn Hishm, p. 12, l
64 the sras of the first meccan period
of Allh he swears by the Revelation (36, 38, 43, 44, 50, 52, and 68), by the
Day of Resurrection (75), by the Promised Day (85), and by his Lord.
4
Most
difficult of all has always beenfor Muslimexegetes
5
as well as for usthe
interpretation of a third category of formula, in which, the majority of
cases, the oath is sworn by female
6
objects or beings. This type also has its
extra-Koranic parallel.
7
Most of the sras of this period are shortof the
forty-eight sras, twenty-three consist of less than twenty, and fourteen of
less than fifty versessince the extreme mental excitement that produced
them could not have lasted long.
When Muammad now presented such revelations to his countrymen [i/76]
he was bound to be considered by most of them a lunatic or a liar. He was
called a crazy poet, a soothsayer
8
associated with jinn, or a possessed person
(majnn). For some time, it seems, he must have shared the latter opinion
to some extent,
9
but after having been convinced of his divine commission
he naturally had to fight such views with all the rhetorical power at his
disposal. The vehement attacks against his opponents, which culminated
in damnation, singling out some of them personallyonce even by name
(cf. below, sra 111)played a great role in these sras.
Comments on Sras 96, 74, 111, 106, 108, 104, 107, 102, 105, 92, 90
[Sir William] Muir holds the peculiar view that eighteen sras had been
revealed prior to Muammads prophetic call in sra 96, and that they were
1.
, Ibn al-Athr, al-Kmil, vol. 2, p. 11; al-Maqrz, Die Kmpfe und Streitigkeiten, ed.
G. Vos, p. 10; Mustaraf, loc. cit.; in al-Masd, vol. 3, p. 381; Musaylima, al-abar,
Tafsr, vol. 1, p. 1933, l 3sq., l 12sq.; ulaya, al-abar, p. 1897, l 9sq.; cf. sras 52, 85, 86, 75, 68,
89, 92, 93, 103, and 95.
4
From the mouth of Muammad only in 34:3, 64:7, and 51:23; in other passages of the
Koran where it is sworn by God the speakers are introduced as other men (37:54, 26:97, 21:58,
12:73, 85, 91, and 95), or God (19:69, 4:68, 70:40), or Satan (38:83). Except sras 4:65, and 4:68,
all these passages are from the Meccan period.
5
For this reason Ibn Qayyim al-Jawziyya (d. 751/1350, cf. C. Brockelmann, GAL, v. 2,
p. 105sqq.; EI
2
) composed a book entitled
, the explanations of the oaths

in the Koran (Ktib elebi, no. 2401).
6
Sras 37, 51, 77, 79, and 100.
7
al-abar, vol. 1, 1934, ll 35 (Musaylima).
8
Although the ancient Arabs believed in a special relation between the khin and jinns,
this belief did not correspond to the Muslim notion according to which jinns and Satan
ascend to heaven, there spy on the angels, and communicate the content to the soothsayers.
Cf. J. Wellhausen, Reste arabischen Heidentums, p. 137.
9
Ibn Hishm, p. 154; al-abar, vol. 1, p. 1152; al-Bukhr in several passages, particularly
in bb al-way; Ibn Sad (al-abaqt): Biographie Muhammads bis zur Flucht, p. 130, l 10.
the sras of the first meccan period 65
insertedinthe Koranonly later; inthese Muammadis speaking for himself;
it is not God, who enters as speaker only with sra 96. The English scholar
[and missionary] was evidently somewhat infused with sympathy for the
Prophet throughhis acquaintance withMuslimsources andtries, at least for
a little while, to defendhimfromthe highblasphemy of forging the name of
God.
10
This opinion, however, does not have any positive arguments on its
side, contradicts tradition, and, in the case of certain sras, can readily be
disproved. Given that Muammad argues against the enemies of religion
in many of these sras, against antagonists who refute the faith (

), and preaches and conversely praises the believers, these sras cannot
possibly originate from a time when he had not yet come to terms with
himself, whenhe had not yet realized that he was destined to be the Prophet
of Allh, and had not yet proclaimed the faith ( ).
Already sra 103, which Muir considers to be the first to have been
revealedprobably because in its current version it is the shortestdeals
withthe enemies of Muammad(103:2), andwithhis followers whobelieve,
and counsel each other to be steadfast in the face of persecution (103:3).
Thus, it can only have been revealed at a time after his public preaching
stirred obvious antagonism. There is an abundance of similar passages in
the sras to which Muir refers, e.g., sra 82:9, 92:16, etc. Here also belong
those passages in which Muammad recalls the decline of the enemies of
God in former times (89:6sqq., 91:11 sqq., and sra 105) as a warning exam-
ple for his adversaries. Finally, it is not at all true that God Himself never
appears as the speaker, for even if we assume that all the passages in which
Muammad is being addressed are soliloquies (Wm. Muir, p. 60) and disre-
gard those verbal forms that may easily be turned from a grammatical first
person into another person by the mere change of diacritical points
11
(e.g.,
for , etc.), there still remain the following passages: 90:10, 94:2, 108:1,
95:45. Muir (p. 62) thinks that some of these verses are represented as pro-
nounced directly by the Deity, but probably as yet only by poetical fiction.
Why does he not do this elsewhere? One could even possibly add that those
passages were intentionally changed later on. On the basis of such an unrea-
sonable, unsupported hypothesis, however, one should not venture equally
untenable assumptions.
10
The Life of Mahomet, p. 75.
11
For a discussion of variant readings related to the peculiarity of the Arabic script see
Goldziher, Schools of Koranic commentators, pp. 23.
We can thus see no reasonat least not in Muirs argumentto depart [i/78]
from the generally accepted Islamic tradition
12
that sra 96:15 is the old-
est part of the Koran and contains Muammads first call to prophethood.
Since the revelationof these verses was accompaniedby a visionor dream, it
is conceivable that even shortly afterwards the precise circumstances of the
apparition eluded him. Even less so can we rely on Muslim accounts of the
matter. The best of them is the tradition which Urwa b. al-Zubayr
13
reports
from isha,
14
although she is very unreliable. Moreover, Muammad can-
not have told her of the event until many years later, as she had not even
been born at the time. According to this tradition, the revelation began with
unmistakable visions ( ) which illuminated the Prophet like the
radiance of dawn. He then retreated to the solitude of Mount ir.
15
After
spending quite some time there in devotional exercises, the Angel (
)
16
visited him and brought the behest: Recite, to which he replied: I can-
not recite ( ). The Angel put much pressure on him (
)
17
and
12
Ibn Hishm, p. 152sq.; Ibn Sad (al-abaqt al-kabr): Biographie Muhammads bis zur
Flucht, p. 130sq.; al-Bukhr, Tafsr; Muslim, vol. 1, p. 113 = al-Qasalln, vol. 2, p. 38sqq. (bb
bad al-way), al-Azraq, p. 426sq.; al-abar, vol. 1, p. 1147sq.; al-Masd, vol. 4, p. 133; Fakhr
al-Dn al-Rz, and other Commentators on sra 96; al-Wid in the introduction; al-Khab
al-Tibrz, Mishkt, p. 513sq. (p. 521 sq., Bb al-mabth wa-bad al-way, beginning); Itqn,
p. 52sq., etc.; cf. Caussinde Perceval, Essai, vol. 1, p. 354; Weil, Das LebenMohammeds, p. 45sq.;
Muir, Life of Mahomet, vol. 2, p. 85; Sprenger, Life of Mohammad, p. 95sq., Das Leben und die
Lehre, vol. 1, p. 297sq.; but particularly Sprenger, Notice of a copy of Tabary, p. 113sqq.;
Leone Caetani, loc. cit., vol. 1, pp. 220227. When reference is sometimes simply to sra 96 as
the earliest sra, this is merely a general reference. Many scholars emphasize that only the
first five verses are that old, and that the rest was revealed later. Al-Bukhr, Bb bad al-way
refers only to the first three verses.
13
EI
2
; EQ; Juynboll, Encyclopedia, p. 17sqq; Sezgin, GAS, vol. 1, pp. 278279.
14
The text of this traditionsometimes short sometimes long, with many variantscan
be found in al-Bukhr; Muslim, loc. cit.; al-Wid, loc. cit.; al-abari, vol. 1, p. 1147sqq.; al-
Azraq, loc. cit.; Fakhr al-Dn al-Rz, loc. cit.; al-Khab al-Tibrz, Mishct, loc. cit.; al-Suy,
al-Itqn, p. 52; abbreviated in Ibn Sad, loc. cit.; Ibn Hishm, p. 151. Cf. Sprenger, Notice of
a copy Tabary, p. 113sq.; on the following page (114sq.) there is a different version which
Sprenger rightfully considers embellished and confused. A comprehensive survey of the rel-
evant traditions in Sprenger, Leben, vol. 1, pp. 330349.
15
So, , the best manuscripts read; this vocalization is established to be the only
permissible forminYqt, vol. 2, p. 228; al-Bakr, p. 273; al-arr, Durrat al-ghawws f awhm
al-khaw, ed. H. Thorbecke, p. 140; and Ahlwardt, Verzeichnis, no. 10169 (= Cod. Sprenger
282), anonymous Ms, .
16
The question whether these and other apparitions were hallucinations or nebulous
ghosts (de Goeje, Die Berufung Mohammeds) cannot be answered. The fact remains that
Muammad believed in the corporal apparitions of the Angel. For a historian of religion they
thus are of identical reality as in comparative cases of the Bible.
17
There are the variants
(Majd al-Dn IBN AL-ATHR, Nihya).

repeated the command. After this happened three times the Angel finally
revealed those five verses. Muammad was deeply shaken and hurried to
his wife Khadja to take comfort.
Another tradition, most likely originating from the same source and [i/79]
found in Ibn Hishm, p. 151 sqq., and al-abar, vol. 1, p. 1149sq., is transmit-
ted from Ubayd b. Umayr b. Qatda.
18
It is interesting, as it mentions in
particular that this event occurred in a dream. When Muammad awoke,
the words of the revelation had already been impressed in the Prophets
heart. The tradition continues with the remark that Gabriel brought a silken
cloth (

) with the words he had to read (al-Itqn, p. 53). The
Koran nowhere mentions such writing material, knowing only
(parch-
ment) and (paper). But that this Koranic revelation was a commu-
nication from a divine document is beyond doubt.
19
An indication of this
is not only the linguistic usage of as explained above on p. 17sqq., but
also the numerous passages that mention the sending down ( ) of kitb,
i.e., writtenrevelation; further, sra 85:21 sq., where the aforementioned pas-
sage is called a Koran preserved on a well-guarded table ( ), and finally,
96:4, because the words your Lord Who taught man the use of the writing
reed
20
pertain most readily to a document in Heaven, which is the source
of all true revelation, including the Jewish and the Christian as well as the
Islamica reminder of the standard phrase . The tradition that
Allh had the complete Koran first sent down to the lowest heaven and that
the Angel thencommunicatedindividual parts to the Prophet as required
compare the Commentators on sra 97thus presupposes a thoroughly
correct point of view. These conceptions of the mechanism of revelation
are of course no arbitrary invention; on the contrary, they are based on
the Judeo-Christian tradition in which books written either by the Hand
of God, or fallen from Heaven, or delivered by an angel, play an important
role.
21
18
(Ab im al-Layth) Ubayd b. Umayr b. Qatda; EQ; Juynboll, Encyclopedia, s.v.
19
According to sras 20:113, 25:34 [sic], 53:5 and 10, 75:18, and 81:19 a revelation was
delivered as follows: Muammad did not read himself from the divine book but an angel
recited the words and the Prophet repeated them until they were embossed in his memory.
20
Arberry translates Thy Lord is the Most Generous, who taught by the Pen, taught man
what he knew not. Regarding this translation cf. Th. Nldekes review of Reste arabischen
Heidenthums, by Julius Wellhausen, ZDMG, 41 (1887), p. 723.
21
Exodus, 31:18, 32:16, 34:1; Deuteronomy, 9:10, 4:13; Ezekiel, 3:13; Apoc. St. John, 10:10; the
Apostolic Father, Hermas, 2nd vision; Eusebii Historiae ecclesiasticae, vol. 6, p. 38; Hippolyte,
Haeres, refut., 9:13.
More recent interpreters of the 96thsra have more or less repudiatedthe [i/80]
Islamic exegetic tradition. Weil
22
believes that in this instance Muammad
is receiving the command to present a revelation that had previously been
made. This interpretation is not only contrary to tradition but also against
its internal probability. For what reason would Allh have commanded the
Prophet to present or recite a particular relevation when it had already been
revealed?
Sprengers statement in his Life of Mohammed, p. 95sq. that
means [i/81]
here to seek for truth in the books of the Jews and Christians clearly goes
against the meaning and is sufficiently refuted by Muammads previously-
mentioned lack of familiarity with the Bible. Sprengers later interpreta-
tion (Das Leben, vol. 1, pp. 298 and 462, and vol. 3, p. xxii), that
means
enounce, must also be rejected on the grounds that it is not supported by
usage.
AbUbayda al-Naw saysaccording toFakhr al-Dnal-Rz, loc. cit.
that the phrase is equivalent to
, where is , i.e., added to

indicate clearly the object, and that is equivalent to to mention. ,
however, does not have this meaning anywhere.
23
Hartwig Hirschfeld
24
translates the phrase proclaim the name of thy
Lord. Since this meaning is foreign to Arabic, he quotes as an authority
the frequently occurring Old Testament phrase . Of course
means to proclaim, to reveal but is likely not its object and
instead means by the use of the name of Yahweh.
25
Only in this sense
(proclaim in the name of your
26
Lord) can the possibility of a borrow-
ing from Hebrew usage be accepted. In support of this fact it could be
asserted that several traditions, according to which Muammad replied to
the Angels command, , with , display a very suspicious relationship
with Isaiah 40:6 ( ). In this case, however, sra
96 would represent a totally isolated usage never imitated in the least in
the Koran, in adth, or in liturgy.
27
On the contrary, is used everywhere
22
Historisch-kritische Einleitung in den Koran, 2nd ed., p. 65; in the first edition he trans-
lated as to read.
23
means he read somethingin a book or something similaror he followed
this or that vocalization like he expressed this or that opinion; cf. M.J. de Goeje in
the glosses in al-abar.
24
Beitrge zur Erklrung des Korn, p. 6; New researches, p. 18sq. Likewise even earlier
Gustav Weil, Mahomet savait-il lire et crire?, p. 357.
25
Cf. also B. Jacob, Im Namen Gottes, p. 171 sqq.
26
This is how Th. Nldeke is now inclined to interpret these words.
27
Regarding the phrase cf. above, p. 27sq.
in the Koran to denote the mere reciting or chanting of holy texts, while
the meaning of reading texts evolved only gradually. It is advisable also in
our passage not to abandon the usual meaning of the verb to present, to
recite.
28
If we follow this and reduce the prevailing tradition (see p. 57sq.) to its [i/82]
essentials we may understandthe development of that revelationas follows.
After living for a long time in solitude as an ascetic and, through an inner
struggle, becoming unbelievably excited, Muammad
29
is finally destined
through a dream or vision to assume prophethood and proclaim the truth
as he had come to understand it. This commission becomes firmly rooted in
his mind as a revelation in which Allh commands him in the name of his
Lord, the Creator of mankind, to present to his countrymen those parts of
the divine book with which he had become acquainted. The time of the first
revelation seems to be indicated in the Koran itself as the so-called Night
of Power (laylat al-qadr), which undoubtedly took place in the month of
Raman.
30
It must remain to be seen whether sra 96:15 is indeed the earliest [i/83]
part of the entire Koran. Because of its compelling invitation to recite,
even if it seems obvious to associate this part with the history of revelation,
the chronological determination of the text does not follow at all. On the
basis of its content the words can be attributed to any time at which a
new part of the divine book was communicated to the Prophet. But its
concise style as well as the short rhythm would suggest a composition
in the first Meccan period. A somewhat more precise definition emerges
from the relation of verses 1 to 5 with the rest of the sra. This latter part
28
Cf. above, p. 27sq.
29
Regardless of how much scholars are divided regarding the meaning of verse 1, they
are all agreed that it can only have been addressed to Muammad. As far as I can make out,
Dozy in his Essai sur l histoire de l islamisme, pp. 2729, is the only person who thinks that
the verses 1 to 5 are a later admonition meant for a disbeliever or half-convert.
30
Compare 97:1, and 44:2 with 2:181. This is also the general opinion; Muammad himself,
however, probably never paid attention to the date. For this reason the particulars are at
variance even in the earliest period (cf. al-Muwaa, p. 98sq.; Ibn Hishm, p. 151 sq., 155;
al-Bukhr at the beginning; Fakhr al-Dn al-Rz on sra 96; al-Diyrbakr, vol. 2, p. 280sqq.,
etc.). As we have been able to see above on p. 46sq., other writers reckon Rab 1 as the
month of the commission. Related to this is the tradition that Gabriel once a year showed
the Heavenly Book to the Prophet, but only in the year of his death this was done twice; and
that in the month of Raman he always devoted himself for ten days to special religious
observances ( ) but in the year of his death this was extended to twenty days (Ibn Sad
(al-abaqt al-kabr) I, iv [sic] ed. Sachau, 4, p. 3, ll 58, vol. 8, p. 17, l 14sq.).
cannot possibly originate fromthe time when Muammad first received his
prophetic commission, since it is already aiming at an enemy of the faith
who was keeping a godfearing slave
31
from the ritual prayer of the young
Muslim congregation (verses 910). Therefore this part of the sra can have
originated only after Muammads prophetic commission. The same dating
would apply to verses 15 if sra 96 had originally been created as a single
unit. Unfortunately, howerver, there is no evidence that it was. All that can
be said is that there seems to be a certain relation between verses 5 and
6; it is worth noting in verses 5 and 6, and the particle introducing
verse 6,
(kall), which does customarily stand at the beginning of direct

speech in the Koran.
32
Nevertheless, if verses 6sqq. should turn out to have
been added on, we would have to consider the opening of the sra to be
older.
After Muammad accepted his call to prophethood it seems that he [i/84]
was not quite sure of himself. Under such circumstances he could not
dare to preach publicly. Unfortunately, details are completely wanting. In
al-Bukhr
33
we have a tradition regarding his enormous mental suffering
during this period, which is appendixed to the above-mentioned tradition
(p. 58, n. 14) of isha on sra 96:

34

35

31
It is known that at first many slaves accepted the new belief (cf. Ibn Sad (al-abaqt):
Biographie Muhammads bis zur Flucht, ed. Sachau, p. 132sq.; Sprenger, Life, pp. 159163;
Sprenger, Leben, vol. 1, p. 356sq.; L. Caetani, Annali, vol. 1, pp. 237 and 240sq.). These people
might have been quite rude to the ancient gods and the penalty was not undeserving. In
al-Wid, p. 336 on sra 92:5 it says about Bill:
. I hardly need
to add that the interpretation of (verse 7) by man as opposed to = God (Sprenger,
Leben, vol. 2, p. 115, Diener Gottes [a servant of God]) is totally unsuitable.
32
When the Koran begins a direct speech with the rebuttal of a fictious sentence, without
actually having been uttered, not
is used but only (cf. sras 75:1, 90:1, 56:74, etc.) It may
be mentioned that there are variant views on this subject; see Wright, Grammar of the Arabic
language, part third, p. 305, CD.
33
In Kitb al-iyal, 16, Bb al-tabr. Excerpts from it in al-Khab al-Tibrz, Mishct,
p. 514 (522), and Fakhr al-Dn al-Rz on sra 96. All the others omit this addition or have
only the first two words. In one passage ( on sra 96) al-Bukhr has the first seven
words.
34
Fakhr al-Dn al-Rz, .
35
So far al-Khab al-Tibrz, Mishct, loc. cit. (bb al-mabath wa-bad al-way).
We cannot quite tell, however, if this mental state, which nearly drove
the Prophet to suicide, really belonged to an earlier time,
36
namely before
his prophetic commission, when he was still leading a lonesome life in the
mountains (
) before being confronted with the revelation ( or

).
Because of the connection of this tradition with the fact that, at first, [i/85]
Muammad attempted, not quite openly, and probably for a longer period,
to convert relatives and friends,
37
but particularly in order to balance the
chronological gap, Muslims construeda periodlasting fromtwoanda half to
three years, called the fatra. This unbelievably long intermission in the rev-
elation Sprenger first repudiated as insupportable in his frequently-quoted
article.
38
In his early studies Sprenger considered this fatra to be an important
period during which Muammadwho could only have been prompted to
prophethood by his unshakeable belief in Allh and the Final Judgement
begantodevelopa systemof faithof his own, andevenstudiedthe Bible.
39
To
this period Sprenger also attributed several sras that urge the Prophet to
defend himself against the suspicion of his friends that he was possessed.
But all the sras in which Muammad refutes such allegations are undeni-
ably directed against the enemies of the religion he was proclaiming.
The usual view regarding the end of this fearful situation is reflected in
the following, well-known tradition of Ab Salma [Nubay b. Shar
40
] from
JBIR IBN ABD ALLH [Ibn Amr al-Khazraj, d. 78/697:]
41
After an intermission of revelation
42
Muammad suddenly recognized the
Angel who had appeared to himin divine glory at al-ir. Frightened, he ran
36
On the other hand, Muammads doubts regarding the final success of his prophetic
mission as well as the struggle with his conscience, which contrary to his innate timidity
compelled himto public proclamation, did not cease entirely until his emigrationto Medina.
All the single stages, which Muslims here ostensibly achieve by means of contrivances of
angels, are of little value.
37
Ibn Hishm, p. 166; Ibn Sad, [al-abaqt al-kabr] vol. 1, part 1: Biographie Muhammads
bis zur Flucht, edited by Sachau and Mittwoch, p. 132sq.
38
ber den Kalender der Araber, p. 173sq., where the individual citations are listed.
The original account (e.g., in Ibn Sad, ed. Sachau, loc. cit., p. 131, top) merely says that the
revelation had been interrupted for some time (
).
39
Sprenger, The Life of Moammad, p. 104sq.
40
41
Juynboll, Encyclopedia; Sezgin, GAS, vol. 1, p. 85, no. 3, etc.
42
The tradition begins as follows: . This evidently relates to an
earlier tradition regarding the first revelation, unless we assume that these words were not
introduced until the tradition of isha.
to Khadja and cried:
43
or

44
wrap me in garments. After this was
done
45
the Angel deliveredthe beginning of sra 74. Thereafter the revelations
followed in quick succession.
46
Since the first words referring tothe intermission[fatra] are missing insome [i/86]
versions of this tradition, some people
47
soon claimed that sra 74 was the
oldest of all. It is always added, however, that this view causes surprise, as
sra 96 is considered the oldest according to the previously noted tradition.
Generally, sra 74:1 sqq. is regarded as being the first call to public preach-
ing.
48
Yet not even this can be deduced with any degree of certainty from
the words of the tradition, which was influenced considerably by the form
of sra 96. The assocation of the sra that begins with with this
tradition is probably only due to the word .
49
But we know that when
Muammad was afflicted with fits he was often wrapped in garments.
50
This
43
is without doubt correctly explained by . All interpretations of the root
are ultimately related to (to age, etc.) or to denominatives of .
44
This meaning can frequently be documented also in writings other than the Koran:
Muallaqt Imru al-Qays, verse 77 = Ibn Hishm, 905, vol. 1; [Muammad b. Yazd] AL-
MUBARRAD, [d. 285/898, EI
2
; Sezgin, GAS, vol. 7, p. 350, vol. 8, p. 98, vol. 9, pp. 7880];
al-Kmil, ed. William Wright, p. 483; al-abar, vol. 1, p. 1822, l 10; Ibn Sad, [al-abaqt, vol. 3,
part 2]: Biographien der medinischen Kmpfer, ed. Sachau, p. 105, l 26, etc.
45
Some add that water had to be poured on him.
46
al-Bukhr, Bad al-way, tafsr; Muslim, Bad al-way (al-Qasalln, v. 2, p. 49); al-
Tirmidh, al-Shaml, kitb al-tafsr on sra 74; al-Wd in the introduction and on sra 74;
Fakhr al-Dn al-Rz on sra 74; K. al-Mabn li-namal-man iii; Itqn, p. 53sq. Shorter quo-
tation, al-Zamakhshar, and al-Bukhr on sra 74; cf. A. Sprenger, Life of Moammad, p. 110,
n. 3. According to a different tradition, on the day of death a particularly significant increase
of revelation occurred; Ibn Sad ed. I, IV p. 2, 7.
47
al-abar, vol. 1, p. 1153.
48

The prophethood of the blessed messenger

of God preceded his mission ; the revelation of sra 96 was accompanied by prophethood,
but only with the revelation of sra 74 began his mission. al-Diyrbakr, Tarkh al-khams,
juz 1, p. 282.
49
Since
, which is nearly an equivalent, frequently takes the place of in tradi-

tion (e.g., Leiden Ms 653, Warner), sra 73, beginning with
, is often confused with

sra 74.
50
Cf. Ibn Hishm, p. 735, l 17sq.

; al-Wid on sra 93
Compare this withsra 73:1, andpossibly the account

referred to above on p. 19 n. 21, Ibn Hishm, p. 117, including parallels; also Muslim, Kitb
al-ajj, 1 (al-Qasalln, vol. 5, p. 189). The same custom is to be found with two prophetic
contemporaries of Muammadwho, according totradition, veiledthemselves whenever they
expected a revelation. In al-abar, vol. 1, p. 1890, it says about the prophet ulaya

, and al-Bayhaq, Masin, ed. Schwally, p. 33, l 15sq.,

. This is probably the explanation for the nickname, Dh

practice, according to the most likely guess, was not based on medical con-
siderations but rather on a superstitious fear.
The words of the sra at least showus that they were revealed in the ear- [i/88 ]
liest period of prophethood.
51
Considering all the circumstances, we must
emphasize that this refers only to the verses 1 to 7 and 1 to 10 respectively;
the following verses, in which a single outstanding adversary is attacked, are
later, although still very old.
52
Inserted in this section is a paragraph from a
muchlater period, namely the verses 31
53
to34, whereas the end,

, might belong to the older portionandconstitute the original continua-
tion of verse 30. The interpolation, which might go back to the Prophet, is in
any case of Medinan origin,
54
because it differentiates between four classes.
(1) the Jews (those who received the Scripture); (2) the Muslims (those who
believe); (3) the Hypocrites
55
(those with sickness in their hearts); and (4)
the idolaters. These might still be from the first Medinan period, since the
Prophet is still friendly with the Jews and considers them in line with the
believers, whereas he soon regarded them as his most bitter enemies. The
verses 41 sqq. are of later provenance but still fromthe first period. The rarely
used term for hell, saqar, in verse 43, which occurs twice in the first part of
this sra but only once more in all of the Koran, may serve as evidence that
l-Khimr (the Veiled One) of the Yemenite prophet Abhala b. Kab as well as the pagan seer
Awf b. Raba (Tj al-ars, ed. 1395, vol. 3, p. 188 bottom; Izz al-Dn IBN AL-ATHR, al-Kmil,
vol. 1, p. 377, l 1 sqq.; al-Aghn, vol. 8, p. 66, l 2sqq.; Wellhausen, Reste, 2nd ed., p. 135 n. 2).
This habit is probably mainly rooted in the common (cf. e.g. Exodus 34:33sqq.) opinion that
looking at the divine is harmful for man ( , Herodotus, vol. 1, p. 32).
Here, I cannot dwell on the interesting problem of religious disguise.
51
There is a different version of the revelation of sra 74 in Ibn Hishm, p. 184, l 8sq. (not
according to Ibn Isq) without listing the authority.
52
Cf. the words verse 8; later this becomes always .
53
Here, Flgel has a totally inaccurate arrangement in his edition of the Koran. Muslim
tradition considers verses 31 to 34 to be one verse.
54
This was the feeling of Weil (Das Leben Mohammeds, p. 365), even though he did not
want to say so.
55
Munfiq is derived from the Ethiopic menfeq; its verb nfaqa, with the meaning of
to doubt, to be inconstant quite common in Geez. It is probably derived from the noun
menfeq, and would be in accordance with the fact that the participial forms are found
in thirty-two passages in the Koran, while the corresponding verbal forms only four times.
Arab tradition rightfully identifies munfiq as an Islamic word however falsely deriving it
from nfiq mouse-hole (e.g. al-Mubarrad, al-Kmil, Cairo ed., v. 1, p. 158). The common
rendition of Hypocrites is in so far not quite pertinent as the majority of the men called
munfiqn in the Koran and tradition are by no means feigned in the sense of the word,
rather they demonstrated at every possible occasionthat their hearts had not yet beentotally
won over since they had accepted Islam less for reasons of conviction but were obliged by
circumstances.
they were originally connected with the preceding verses. Still, this term is
likely to have been inserted into verse 43 only accidentally from those two
passages, here to take the place of the older term jam,
56
since the context
requires a rhyme with m.
Sra 111 is considered by all to be one of the earliest revelations. There [i/89]
is general agreement on the broad outline of the circumstances of its rev-
elation: after much hesitation, Muammad finally summoned his men,
or, according to a more likely account, his clan, the Ban Hshim,
57
and
demanded their acceptance of God. But his uncle, Abd al-Uzz IBN ABD
AL-MUALIB, called Ab Lahab, said:
, Go to the hang-
man! Is this why you called us?
58
Upon hearing these words from a highly
respected man of the familywhich were likely not meant as maliciously
as they sounded
59
the assembly dispersed, as Muammads address did
not make sense to them. The Prophet then cursed Ab Lahab and his entire
clan with the words of sra 111, making himself their most dangerous antag-
onist.
At the same time, one should not be overly impressed by the extensive [i/90]
consensus of tradition. The reference to hands in the first verse might
56
Next to the very frequent nr (fire) as well as jahannam (hell), jam is the most
frequently used word (26 times) for hell in the Koran. Other equivalents are sar (16 times)
and la ( once).
57
We have many different accounts regarding the meeting itself and the other details.
Some of them are wonderfully embellished, others fabricated in favour of Al, who, at that
time, was still very young. Cf. Ibn Sad, [al-abaqt al-kabr] ed. Sachau: Biographie Muham-
mads bis zur Flucht, vol. 1, part 1, pp. 42sq. and 132sqq.; al-abar, vol. 1, p. 1170; al-abar,
transl. Zotenberg, vol. 2, p. 405; al-Bukhr in Kitb al-Tafsr; al-Bayw, al-Zamakhshar
on sra 111; Muslim al-Qasalln, vol. 2, p. 185 (Kitb al-mn 77); al-Tirmidh, Kitb al-
Tafsr; al-Khab al-Tibrz, Mishct, Bb al-Indhr, fasl 1, 2, Bb al-Mabath, fal 1, 9; al-
Wid on sra 111; Fakhr al-Dn al-Rz on sra 26:214; al-abar, Tafsr, vol. 19, p. 67, v. 30,
p. 190sq. These accounts are confusing in al-Zamakhshar on sra 26:214, etc. See also Weil,
Das Leben Mohammeds, p. 53; Caussin de Perceval, Essai, vol. 1, p. 316sq.; Sprenger, Life,
p. 177sq., Sprenger, Leben, 2nd ed., vol. 1, p. 526. Both Muir, Life of Mahomet, vol. 2, p. 113sq.,
and Leone Caetani, Annali, vol. 1, p. 239sq., have well-founded doubts about the reliability of
these accounts.
58
Some sources add
.
59
The general view which holds that Ab Lahabs words constitute a curse is not true.
We here have the exclamation of a man who is angry because he had been summoned to a
great andimportant meeting andwhohears nothing but nonsense. There is basically nothing
bad about this and can be compared to the easily uttered words , to hell with you,
Goddamn, etc. For example, in Ab l-Faraj al-Ibahn, al-Aghn, vol. 16, p. 159, it says the
poet Aba b. Quray [Sezgin, GAS vol. 2, p. 191] once called his countrymen together to tell
them a bad joke,
. In this case it is only annoying jesting.

It is quite another story when, after having been deeply injured by his uncle, the Prophet
shouts:
.
suggest an assault upon the Prophet. The description that Ab Lahab had
thrown stones at his nephew during the afore-mentioned meeting is men-
tioned only in late writings (al-Bayw, al-Nasaf). Other traditions, accord-
ing to which Ab Lahab threwdung or carrion in front of the Prophets door
(Ibn Hishm, p. 276sq., Ibn Sad, Biographie Muhammads bis zur Flucht, p.
sq.), do not refer to our sra, whereas Ibn Hishm, p. 233, and all Com-
mentators construe a connection with a similar action by his wife, referring
to verses four and five. Alesser number of traditions
60
connect the sra with
other events, whichare no less reliable. The impressionremains that evenat
a very early period there exsisted no reliable tradition and that all we have is
speculation on the part of the exegetes. Another noteworthy aspect of this
sra is that only one other sra, 33:37, similarly mentions a contemporary
by name.
61
Sra 106 admonishes the Quraysh to thank the God of the Kaba,
[i/91]
,
62
for being able to send out two caravans annuallythe source of
prosperity for the trading community.
63
The sympathetic mood expressed
here indicates that the sra originates from before the conflict with the
Quraysh.
64
There is no mention of the Kaba in any other later Meccan sra.
For lack of a historical guide to the remaining sras of the first period, the
chronological order must be abandoned outright. We therefore will aim for
a topical arrangement, although when arranging the individual classes we
intend to use the gradual development of style and thought as a guidleline
as far as possible.
60
Ibn Hishm, p. 231 connects the sra with an event of the late Meccan period. Al-Azraq
(p. 81 sq.) and al-Wqid (ed. Wellhausen, p. 351) even date Ab Lahabs curse to year 8ah,
when after the conquest of Mecca this uncle of the Prophet, after the destruction of the
idols of al-Uzz and al-Lt respectively, promised to take care of the goddesses. Ab Lahab,
however, had beenlong dead by that time. Al-abar inhis Tafsr (juz/vol. 30, p. 191) mentions
yet another reason but without supplying a date.
61
A more detailed study in our chapter The collecting and editing of the Koran.
62
This contradicts Muirs view (Life of Mahomet, vol. 2, p. 140 and 154sq.) that before the
revelation of sra 53 Muammad rejected the existing Qurayshite worship as a whole by
reason of its idolatry and corruption.
63
That the Quraysh put together the winter and the summer caravans, and may they
continue to do so. This is howSprenger finishes the explanationas presentedinhis Muam-
mads Zusammenkunft mit dem Einsiedler Bahyr [Muammads encounter with the her-
mit] at the same time completely misunderstanding the Hebrew term . Like so many
stories in honour of Muammads own clan of Hshim it is certainly not true that these
two caravans were first organized not until Hshim. Even Ibn Hishm, p. 87, l 12, adds to
the account his critical words . The verses quoted in its support are false.
64
Cf. Leone Caetani, Annali dellislam, vol. 1, 234, n. 2.
Of the various sras that primarily serve to attack enemies, sra 108 is [i/92]
likely to be among the earliest. In this sra, God comforts the Prophet after
suffering abuse. The target is in most cases al- b. al-Wil,
65
less frequently
it is Uqba b. Muay
66
or Kab b. al-Ashraf [d. 3/625].
67
They were supposed to
have reproachedhimfor being a tailless man, i.e. a manwithout sons.
68
But
God says that He has given him plenty
69
of goods. The view held by a few
writers, who consider this sra to be Medinan,
70
and who think this refers
to the death of his son Ibrhm (Ibn Muammad),
71
does not warrant a
serious refutation. As a matter of fact, the general expression, he that hates
you [ ], might not even refer to any particular person but rather to
an entire group of adversaries, an interpretation which, following older
exegetes, already al-abar (Tafsr, vol. 30, p. 186) is inclined to suspect. Like
the other sras (48, 71, 97, and 108) beginning with inn ( = verily, we), this
one, too, might have lost its original beginning.
Sra 104, which, according to Hibat Allh b. Salma,
72
some writers con- [i/93]
sider to be of Medinan origin, attacks rich and arrogant men.
Sra 107, verse 4 (cf. 104:1), also hurls woe unto those that pray and
refuse charity. As these words somehow seem to fit the Hypocrites in Med-
65
Cf. Ibn Hishm, p. 261; Ibn Qutayba, ed. Wstenfeld, p. 145; al-Masd, part 5, p. 61;
[Izz al-Dn] Ibn al-Athr, al-Kmil f l-tarkh, vol. 2, p. 54; al-Wid, and the Commentators;
Sprenger, Das Leben und die Lehre des Mohammad, vol. 2, p. 4; Juynboll, Encyclopedia, p. 89,
col. 1.
66
al-abar, Tafsr, loc. cit., vol. 30, p. 186; Juynboll, Encyclopedia, p. 531, col. 1.
67
al-abar, ibid.; Juynboll, Encyclopedia, p. 577, col. 1.
68
It is known that for times immemorial the greatest blessing for Semites is many sons,
resulting in power, honour, and wealth, cf. Schwally, Das Leben nach dem Tode (1892), p. 29
sqq.; G. Freytag, Einleitung in das Studiumder arabischen Sprache (1861), p. 210.
69
is actually an adjective meaning much, plentiful, abundant cf. the examples in
Ibn Hishm, p. 261; from there it is the enormous dust (Dwn der Huailiten, 92, verse 44);
thus in this case the abundance, the mass. The corresponding verb is to be of large
quantity, e.g., dust (cf Ab Tammm, amsa, 106 verse 5). Already in Ibn Hishm, p. 261 sq.,
there is an old, yet inaccurate, explanation that Kawthar is the name of a river in Par-
adise.
70
Like Al al-Dn (Al b. Muammad AL-KHZIN AL-BAGHDD, Tafsr al-Khzin al-
musamm Lubb al-tawl f man l-tanzl) says according to the tradition of asan (al-Bar
[EI
2
; Juynboll, Encyclopedia; Sezgin, GAS, vol 1, pp. 591594]); Ikrima, and Qatda; al-Suy,
al-Itqn, p. 30; Brockelmann, GAL, suppl. vol. 2, p. 135.
71
Cf. al-Suy, Asbb al-nuzl, where also another tradition is mentioned, according to
which the sra was revealed on the day of al-udaybiyya; so also al-Itqn, p. 45.
72
Apart from this sra there are many others which, according to [Ab l-Qsim] Umar
b. Muammad [IBN ABD AL-KF] (Leiden, Cod. 674, Warner) all (old exegetes, like the
students of Ibn Abbs, etc.) consider to be of Meccan origin, while some others consider
them to be from Medina, e.g. sras 25, 53, 57, 67, 80, 87, 89, 90, 92, 102, and 110. [Sezgin, GAS,
vol. 1, p. 47; vol. 9, p. 183.]
ina, some exegetes hold that all of this sra,
73
or at least verses 4 to 7,
74
are of
Medinan origin.
According to a single proponent, sra 102 refers to the Jews of Medina.
75
Sra 105 is probably to demonstrate with an example from historyand
particularly from Meccas own historyhow God punishes their like.
76
Sra 92, like so many others, was, according to some authors, revealed
totally or partially only after the hijra.
77
Already sra 90 seems to be somewhat later. The isolated viewthat it is of [i/94]
Medinan origin was recognized as false already by al-Suy, al-Itqn, p. 29.
No less erroneous is the viewof writers who accept as Meccan only the first
four or the first two verses, in which they identify a reference to Mecca.
78
Comments on Sras 94, 93, 97, 86, 91, 80, 68,
87, 95, 103, 85, 73, 101, 99, 82, 81, 53, 84, 100, 79, 77,
78, 88, 89, 75, 83, 69, 51, 52, 56, 70, 55, 112, 109, 113, 114
The following sras are of miscellaneous content and do not, according
to general consesus, have as their primary purpose either fighting against
adversaries or describing eschatology.
73
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; (according to Ibn Abbs,
al-asan al-Bar, and Qatda). al-Zamakhshar, al-Bayw, al-Itqn.
74
Hibt Allh b. Salma, al-Itqn, p. 37 (not exact), and [Ab l-Qsim] Umar b. Muam-
mad [IBNABDAL-KF]. These two sras, too, are allegedly addressed to particular persons:
First.Sra 104, to Akhnas b. Sharq [EI
2
] (Hibat Allh [IBN SALMA], al-Zamakhshar, al-
abar, Tafsr), Umayya b. Khalaf [EI
2
; Juynboll, Encyclopedia, p. 460sqq.]; (al-Zamakhshar);
Wald b. al-Mughra [EI
2
]; (al-Zamakhshar, Naysbr al-Qumm in the margin of al-abar,
juz 30 (Gharib al-Qurn wa-raghib al-furqn), p. 161), to Jaml b. mir (al-abar.) Sec-
ond.Sra 107 to b. Wil (Hibat Allh [IBN SALMA], al-Wid, al-Naysbr al-
Qumm), Ab Sufyn b. arb (Ibn Umayya), [EI
2
; G. Juynboll, Encyclopedia, p. 108sqq.; Sez-
gin, GAS, vol. 2, p. 283] (al-Wid, al-Naysbr al-Qumm), Wald b. al-Mughra and Ab Jahl
(al-Naysbr al-Qumm [EI
2
; EQ; Juynboll, Encyclopedia, p. 73, col. 1]). All this is, of course,
pure fabrication.
75
Cf. al-Bayw, al-Wid, and al-Itqn, p. 30, which is in accordance with this view.
76
Reference is to the familiar expedition against the sacred territory of Mecca, in the
course of which the Abyssinian army was destroyed, apparently by plague. It is likely that the
lore of the Meccans had already embellished this event. Cf. Ibn Hishm, p. 29sqq., al-Azraq,
p. 86sq., Dwn der Huailiten, p. 112sqq., al-abar, vol. 1, p. 935sqq., al-Masd, Prairies
d or, vol. 3, 158sqq., and the Commentators. Weil, Das Leben Mohammeds, p. 10, Caussin de
Perceval, vol. 1, 279, Sprenger, Life of Moammad, p. 35, Sprenger, Leben und die Lehre, 2nd
ed., vol. 1, 461, F. Buhl, loc. cit., p. 21, Caetani, Annali, vol. 1, p. 143sqq. A discussion of all the
relevant problems to be found in Th. Nldeke, Geschichte der Perser, pp. 204208.
77
For example, al-Itqn, p. 29, etc.
78
al-Itqn, p. 37. The phrase alone guarantees verse 12 and its context its
Meccan origin.
In sra 94,
79
as well as the apparently somewhat later sra 93, God pur-
poses to console Muammad for his current situation by reminding him
that He even earlier saved him from calamity. In those days when there
were but few people who believed in himnearly all of the lower strata
of societyand there was little hope of succeeding with his message, there
must have been ample opportunities for such a consolation by Allh. Con-
sequently, these sras do not need to be the result of any particular incident.
Even if this were the case, however, it would still be highly unlikely that any
such report, alone among so many similar reports,
80
was properly transmit-
ted to posterity.
Sra 97 concerns the Night of Power,
81
inwhichthe angels andthe Spirit
descend upon the earth
82
with the revelation. Because of reference to a
tradition retold in al-Suy, al-Itqn, p. 29, it is falsely held to be of Medinan
origin.
83
The wording of the first verse makes it likely that the real opening
of this sra has been lost.
84
In sra 86 the first three verses seem to indicate that it was revealed at [i/95]
night under the impression of a glowing star.
85
In sra 91, which opens with an disproportionately large number of
solemn conjurations (verses 1 to 8), the Prophet demonstrated to his con-
temporaries the sin of the ancient Thamdites who had accused a messen-
79
From an inaccurate literal rendering of sra 94:1 combined with the tradition of
Muammads epileptic fits in his childhood developed the miserable myth which we find in
Ibn Hishm, p. 105sq.; Ibn Sad, ed. Sachau, [al-abaqt] vol. 1, part 1: Biographie Muhammads
bus zur Flucht, p. 74sq.; al-Bukhr in bb al-mirj, and other sources; Muslim, kitb al-mn,
72 (al-Qasalln, vol. 2, p. 60sqq.); al-abar, transl. Zotenberg, vol. 2, p. 241 sq.; al-Masd,
vol. 4, p. 131; al-Khab al-Tibrz, Mishct, p. 516 (524 bb alamt al-nubuwwa beginning), etc.
Cf. Weil, Das Leben Mohammeds, note 11 [sic]; Sprenger, Life, p. 78, Leben, 2nd ed., vol. 1, p. 168;
Muir, Life of Mahomet, vol. 1, p. 21, etc. Other writers establish a relation between the story of
the opening of his chest and his midnight journey to the seven heavens (see the quotations
on sra 17.)
80
Cf. the Commentators; al-Bukhr, Kitb al-kusf, abwb al-taqsr, 24, fail al-Qurn
1; Muslim, Kitb al-jihd, 24 (al-Qasalln, vol. 7, p. 439sq.); al-Waid.
81
See above, p. 47.
82
This is the first sra in which the root is used with reference to the revelation of the
Koran.
83
al-Bayw. [Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF] (Leiden Cod.
674, Warner); Al al-Dn (Al b. Muammad al-Baghdd AL-KHZIN), vol. 4, p. 464; and
al-Suy, al-Itqn, p. 56, with reference to the commentary of al-Nasaf from al-Wqid, it is
held to be the earliest of the Medinan sras. Hibat Allh b. Salma not even once mentions
that some writers hold it for a Meccan sra.
84
See above, p. 76sq., on sra 108.
85
According to al-Wid these words were revealed while Ab lib IBN ABD AL-
MUALIB was having a meal and was frightened by a shooting star. The three verses, how-
ever, only fit a planet or at least a large fixed star.
ger of Allh of deceit and killed him but were in turn punished by annihila-
tion. Muammadoftenreferredtothis story
86
later on(altogether twenty-six
times in the Koran).
In sra 80, God reprimands the Prophet for turning away fromIbn Umm
Maktm, a poor blind man
87
who was anxious for instruction, whilst he
was engaged in conversation with al-Wald b. al-Mughra, one of the chiefs
of Makkah [Sprenger].
88
Muammad reproaches himself for his failure in
having unduly favoured the mighty of his town. It is a surprise and also char-
acteristic of the most humane of all the revealed religions that these words
foundtheir way intothe Koran. Hibat Allhb. Salma al-Baghdd is the only
writer tomentionthat there is noagreement as tothe place of this revelation
( ). August Mller
89
sees in verse eleven the beginning of a newfrag-
ment, probably of some later years, whereas David H. Mller
90
holds that
the second part, which apparently is not at all related to the other part,
begins only with verse sixteen.
Sra 68 is held by some to be the earliest sra
91
or, in any case, the second [i/96]
earliest, following immediately upon sra 96.
92
People most probably linked
the initial words of the sra, , tothe opening of sra 96, andconsequently
also construed a chronological relation. Of course, verses that polemicize
directly against the enemies of the faith can by no means be that old. Yet
verse 17sqq., of which vv. 17 to 33 and vv. 48 to 60 are sometimes considered
to be of Medinan origin,
93
were apparently added to the older sra only in
the second period.
94
86
al-Bukhr and al-Tirmidh (in Kitb al-Tafsr) relate a funny episode regarding this
matter.
87
Juynboll, Encyclopedia, pp. 320, 470, 493, 494.
88
Usually(al-Muwa, p. 70sq.; Ibn Hishm, p. 240; Ibn Sad, ed. Sachau, Biographien
der Muhgirn(vol. 4, part 1), p. 153; al-Tirmidh, Kitbal-Tafsr; al-Wid; Ibnajar al-Asqa-
ln, vol. 2, p. 1245; the Commentators; Sprenger, Life, p. 186; Sprenger, Leben und die Lehre,
vol. 2, p. 317; Muir, vol. 2, p. 128; Caetani, Annali dellislam, vol. 1, p. 297)the man is called
[Abd Allh] Ibn Umm Maktm [EI
2
; Juynboll, Encyclopeadia, p. 320sqq.] but this person
must serve everywhere as the representative of the blind. In this case, you expect a man from
the lower strata of society while the other one belonged to the Quraysh clan of mir b. Luay,
and his mother even originating from the Makhzm clan, of equal importance as the Abd
Shams. Cf. about him Ibn ajar al-Asqaln, loc. cit., and Izz al-Dn IBN AL-ATHR, Usd al-
ghba f marifat al-aba, vol. 4, p. 127.
89
Der Koran, translated by Friedrich Rckert, ed. by August Mller (1888), p. 545.
90
Die Propheten in ihrer ursprnglichen Form, p. 57.
91
Hibat Allh [Ibn al-Salma al-Baghdd.]
92
See above, the chronological arrangement on p. 43sqq.
93
Leiden, Cod. 674; al-Itqn, p. 36.
94
One ought to note, for example, the greater length of most of the verses as well as some
Sra 87 is yet another example of the careless interpretations and con- [i/97]
clusions of some of the ancient exegetes. The invitation to praise God (87:1
and cf. 87:15) was taken to refer to the five daily prayers instituted not long
before the hijra, and the sra was taken to be Medinan without hesitation.
95
They reached the same conclusion about both sra 95, in the third verse
of which there is a clear reference to the sacred territory of Mecca, and sra
103, which might possibly be a mere fragment.
96
These two sras are likely
to have come to us in a revised version. It seems to me that 95:6 was added
later, since it is disproportionately long, its content excessively weakens the
impression of the context, and the phrase
was not
common until the late Meccan period. The first and the third argument
apply also to the current form of 103:3.
Sra 85 warns the believers of the example of the pious people who
long ago were tortured and killed by accursed
97
men.
98
Verses 8 to 11 were
probably added later, possibly by Muammad himself. They differ from the
other related verses in terms of their greater length, long-winded diction,
and slightly different rhyme.
99
unusual expressions not used in the first period like
verse 28,
verse 48;
cf. generally, below on sra 52. H. Hirschfeld, Newresearches, p. 60, Somewhat later but still
of a very early period are vv. 34 to 52 with an allusion to Jonah, the man of the fish.
95
Leiden Cod.Warner 674; al-Bayw.
96
Leiden Cod. Warner 674; Hibat Allh b. al-Salma.
97
Naturally, this alone is meant by
as some of the exegetes realized.

98
Generally, these are held to be the Christian martyrs of Dh Nuws [EQ; EI
2
,] king of
the Jews in Najrn. Cf. Ibn Hishm, p. 20 and 24; al-abar, vol. 1, p. 925; the Commentators;
Sprenger, Life, p. 36sq., Sprenger, Leben und Lehre, vol. 1, p. 464sqq.; Muir, Life of Mahomet,
vol. 2, p. 146. There is something to this; the event of October 523 created a great uproar
wherever it became known. The most reliable source, the letter of Simeonof BthArshm(cf.
Nldeke, Geschichte der Perser und Araber, p. 185sq.) relates that at that time the church was
put on fire, killing everybody who had there taken refuge, priests as well as ordinary church
members (Anecdota Syriaca, vol. 3, p. 236, l 12sqq.). All who did not renounce their faith
died by the sword. The letter knows nothing of an actual stake or ditches. Later writers only
(cf. W. Fell, Die Christenverfolgungen in Sdarabien, pp. 8 and 32) report of fiery ditches,
which, according to Fr. Praetorius ZDMG, 23 (1869), p. 625 [wrong; this reference is not to
Praetorius but to M. Grnberg, Nachtrge zu den Bemerkungen ber die Samaritaner])
can only have been inferred from our sra. In such circumstances see A. Geigers conjecture
(loc. cit., p. 192) that those verses might refer to the three men in the burning fiery furnace
(Daniel, 4), cf. O. Loth, Tabaris Korankommentar, p. 621. This explanation is also to be
found in Muslim commentaries to the Koran, among others by al-abar in his Tafsr, s.v.
(cf. O. Loth, loc. cit., p. 610), and by al-Baghw (

). Of course, Geigers reasoning is without foundation, claiming that Muammadwho

at that time can hardly have had even an inkling of the discrepancy between his doctrine and
Christianitycould not have called Christians believers.
99
In other cases all verses rhyme with d or d, only verses 20 (t) and 22 (z) differing
Sra 73, as mentioned above,
100
is occasionally held to be one of the earli- [i/98]
est sras because of the similarity of its beginning tothat of sra 74.
101
Equally
off the mark are those who, when explaining its origin, mention isha,
whomMuammad did not marry until muchlater.
102
Verse 20 is so obviously
of Medinan origin that this could not possibly have escaped Muslims.
103
Of
course Weil also realized this.
104
The verse must belong to a time when fight-
ing with the disbelievers had already started. As its content is similar to that
of the first versecf. particularly verses two and threewe must assume
that either Muammad himself or one of the Companions must have pur-
posely attached it to the other sras. For unknown reasons verse 20 is also
held by some
105
to be Medinan.
The remainder, and greater portion, of the revelations of this period deal
with the enormous upheavals of nature that will accompany the advent
of the Final Judgement, or describe the joys of heaven and the horrors
of hell with dazzling colours. There are no more grandiose sras in all of
the Koran, and none in which the emotional excitement of the Prophet
appears in fuller force. It is as if ones very eyes see how the earth splits
asunder, mountains scatter, and stars are thrown about. Another group of
sras, whose descriptions are more serene and prosaic, must on the whole
be attributed to a later period.
The emotional turmoil appears immediately in the short verses of sra [i/99]
101. The argument that August Fischer
106
advances for an interpolation of
the verses 7 and 8 is inconclusive as far as I am concerned. More easily
acceptedalthoughhe does not evenconsider thiswouldbe the possibil-
ity of a lacuna between the verses 6 and 7. But this, too, is both unnecessary
and unlikely.
slightly; but verse 10 rhymes with q, and verse 11 with r, a very frequent change of rhyme
later on.
100
The original reference is to Seite i/81, where no sra is mentioned at all.
101
This must be the reference of al-Baghws statement regarding verse 1, God addressed
the Prophet before his public preaching (
) with the words
;
he seems to place even this sra before sra 111 as well as other passages that refer to the
public sermon.
102
al-Bayw.
103
al-Suy, al-Itqn, pp. 20and36. But one traditionfromisha has this verse be revealed
a whole year after the others. Cf. Fakhr al-Dn al-Rz on verse 4; [Ab l-Qsim] Umar b.
Muammad [IBN ABD AL-KF].
104
Historisch-kritische Einleitung, 1st ed., p. 56; 2nd ed. p. 65.
105
al-Suy, al-Itqn, pp. 20 and 36. Al al-Dn (Al b. Muammad al-Baghdd AL-
KHZIN), vol. 4, p. 338.
106
Inhis article Eine Qoran-Interpolation. I concur withFischer insofar as the possibility
of interpolations in the Koran must be unconditionally admitted.
Sra 99, which impresses with its gradiose, rhetorical and rhythmical
opening, is considered by many
107
to be of Medinanorigin, probably because
verse seven was interpreted as dealing with mundane affairs, i.e., the victory
of the Muslims over the pagans.
108
Similar to this sra but embellished with more colourful imagery are
sras 82 and 81. One would like to connect the latter to sra 53, which
belongs to the later sras of the first period, although not to this third
section. Both sras are related in content, and both deal with the apparition
of the Angel. Sra 81 mentions only one apparition, whereas sra 53 is
concerned with two, for the one mentioned at the beginning of this sra is
identical with the one in sra 81,
109
cf. particularly 53:1 and 81:23. In sra 53,
however, there is an allusion to yet another apparition, when the Prophet
imagines himself to be in Heaven. Sprengers assertion
110
that verse 15 is a
later additionwe cannot accept.
111
Giventhe connectionof these apparitions
withthe later dreamof the Night Journey toJerusalem(sra 17), as well as the
influence of Jewish and Christian examples,
112
the legend of Muammads
ascension was not born until some time after his death. In the description
of this matter Muslims like to resort to the words of sra fifty-three.
When Muammad publicly recited sra 53 and reached the verses 19 to [i/100]
22, where the pagans are asked whether they had ever seen their goddesses,
al-Lt, al-Uzz, and Mant, in the same way that he had seen the Angel, it
is reported that either he or Satan imitating the Prophets voice said at that
moment: These are the exalted gharnq, whose intercession (from God)
can be counted on.
113
The episode is explicable by Muammads uneasy
107
[Ab l-Qsim] Umar b. Muammad [IBNABDAL-KF]; al-Bayw, al-Zamakhshar,
andal-Itqn, pp. 20and30. Hibat Allh[IbnSalma] does not evenmentionthat some writers
hold it to be from the Meccan period. Cf. also the classification of the sras on p. 39sqq.
108
Cf. al-Suy, al-Itqn, p. 30.
109
This apparitionis probably tobe considereda dream; according tosra 81:1518 it seems
to me likely that the revelation originated by the night swarming, by the dawn sighing.
110
Sprenger, Life, p. 133, note; Leben, vol. 1, p. 307, note; maw (verse 15), Sprenger considers
to be a house outside Mecca near which Muammad had the apparition. This idea, which
surprisingly August Mller thought plausible, misled Caetani in his Annali, vol. 1, p. 231, also
to suspect in sidrat al-muntah (verse 14) a place name near Mecca.
111
Sprenger, Life, p. 123sqq.; Leben, vol. 1, p. 306sqq.
112
I point out the familiar celestrial journey of the Ecstatics, for example, Isaiah (ascensio
Jesaiae). Further, see below on sra 17.
113
A quite common version of these words runs as follows:
*

*

*
Ibn Sad (al-abaqt): Biographie Muhammads bis zur Flucht, p. 137, l 11; Majd al-Dn IBN
AL-ATHR, Nihya, vol. 2, p. 58; al-Wid, Ab l-Layth al-Samarqand, Fakhr al-Dn al-Rz,
mood when looking for a compromise with the old belief by recognizing
those goddesses as Allhs subservient, benevolent beings.
Both Muir
114
and Sprenger
115
affirmthe reality of this event, recognizing in [i/101]
it a welcome motive to denounce the Prophet once more as an impostor.
On the other hand, Leone Caetani,
116
the most recent of Muammads biog-
raphers, tries to prove that this is a later fabrication. His main arguments are
as follows: (1) The isnd of this tradition is unreliable. (2) It is doubtful that
the Qurayshwho only a short time previously had obliged the Muslims
to emigrate to Abyssinia, and who persecuted everyone who recited only a
few verses of the Koranquietly listened to an entire sra and then prayed
together with Muammad. (3) Other compromises with the pagan cult, e.g.,
the incorporation of the Kaba into Islam, demonstrate an entirely different
tactic. (4) Such a gross error as the admittance of the three pagan goddesses
into the Islamic cult would have destroyed the Prophets entire early lifes
work.
and al-Bayw on sra 22:51; Hibat Allh [Ibn Salma] on sra 20:113. Al-Jurjn in the recen-
sion of al-Tirmidh, preface, p. 3, and al-Damr, s.v., . The most common variants are
as follows for : , al-abar, Annales, vol. 1, p. 1193, l 6, p. 1194, l 1; al-abar, Tafsr on sra
22:51 (Cairo edition, vol. 17, p. 119sqq.) no. 3;
, Yqt, Geographisches Wrterbuch, vol. 3,

p. 665, l 20; , al-abar, Annales, vol. 1, p. 1195, l 8; codices al-Bayw, and M. [not listed
among Nldeke-Schwallys abbreviations], al-abar, Tafsr, loc. cit., no. 1; al-Zamakhshar on
sra 22:51.For
: , al-abar, Tafsr, no. 4; Hibat Allh (Ibn Salma) on sra 22:51; Ab l-

Layth al-Samarqand on sra 17:75; , al-abar, Tafsr, no. 6; al-Diyrbakr, Tarkhal-khams,
vol. 1, p. 289;
, Ab l-Layth al-Samarqand on sra 22:51. , al-abar, Tafsr, no. 6; al-

Diyrbakr, Tarkhal-khams, vol. 1, p. 289; , Abl-Laythal-Samarqand onsra 22:51;
, Ab l-Layth al-Samarqand on sra 22:51.For : , al-abar, Annales, vol. 1,
p. 1195, l 8; al-alab, al-alabiyya, vol. 2, p. 4; , Hibat Allh (Ibn Salma) on sra 22:51,
and Ab l-Layth al-Samarqand, loc. cit.; , al-abar, Annales, vol. 1, p. 1192, l 14 codices al-
Bayw, and M. [abbreviation not identified]; al-abar, Tafsr, nos. 1 and 4; , al-abar,
Annales, vol. 1, p. 1192, l 14, p. 1193, l 6, p. 1194, l 1; al-abar, Tafsr, nos. 2 and 3.Regarding
the word which is used for a variety of birds, and in the Dwn der Huailiten, no. 157, l
2, edited by Wellhausen, for water-birds, cf. the dictionaries of Majd al-Dn IBN AL-ATHR
(Nihya), and al-Damr (al-ayawn), s.v. Regarding and with the meaning of
tender young men and refined people cf. Ab Tammm, amsa, pp. 608 and 607; Ab
Zayd, al-Nawdir [f l-lugha], p. 44, l 18, p. 45, l 7; other passages in J. Wellhausen, Reste ara-
bischen Heidentums, 1st ed., p. 30, 2nd ed., p. 34. It remains to be seen how these meanings
can be reconciled with each other, and how the word in that remark of the Prophet is to be
understoodin this case its common translation is swan. Incidentally, there are frequent
allusions to this event without specifically mentioning these words; e.g. in al-Bukhr. Cf.
also Weil, n. 64 [sicincomplete]; Sprenger, Life, p. 184sq., and Muir, Life of Mahomet, vol. 2,
pp. 150151.
114
Life of Mahomet, vol. 2, p. 149sqq.
115
Das Leben und und die Lehre, vol. 2, p. 16sqq.; also H. Grimme, Mohammed, vol. 2,
p. 16sqq., and Frants Buhl, Muhammeds Liv, p. 180sq., but without any arrire pense.
116
Annali dellislam, vol. 1, p. 278sqq.
There are several objections to these contentions. The evidence of the [i/102]
above-mentioned isnd was suspicious even to scholars of the fifth to the
seventh centuries. Apart from the fact that in the final analysis they were
guided by dogmatic considerations, their criticism of adth cannot really
be taken seriously. In particular Goldziher
117
has shown that the literature is
full of fictitious traditions despite the fact that their isnds are formally quite
sound. Caetanis contention in the second argument is correct but this does
not mean that the claim is improper per se. No matter how many fictitious
details a tradition might contain, it might still be based on a historical core.
Caetanis last two arguments are also not convincing. The remark about the
gharnq does not attempt to equalize the pagan goddesses with the One
Allh but rather considers them subordinate beings with only the privilege
of intercession. Moreover, the central theme of Muammads sermon was
resurrection and Final Judgement rather than rigid monotheism.
118
In the
same way he initially also did not assail the Christians despite the resem-
blance of their doctrine of the Trinity to polytheism. It is even more serious
that Caetani cannot explain the origin of the ostensibly false tradition. It
goes without saying that Muslims did not fabricate a story that depicts their
Prophet in such an unfavourable light.
119
But ifas some Muslim theolo-
gians have itheretics concocted this, it simply would not have penetrated
orthodox tradition. There is consequently no way out of the dilemma but to
accept the event in its basic outline as a historical fact.
Many traditions associate the foregoing with the return of some Mus- [i/103]
lims who had taken part in the well-known emigration to Abyssinia. Upon
their arrival at Mecca they are supposed to have learned that Muammads
remark regarding the sublime gharnq had in the meantime resulted in
the conversion of many pagans. If the connection of the two events, as well
as their dating to the months of Raman and Shawwl of the fifth year
of the prophetic commission, as Ibn Sad alone contends,
120
were correct,
it can only follow that the two verses originated at that time. By the same
token, although they both correspond well to the other verses in terms of
length and rhyme, we have no guarantee that they might not have been
117
Muslimstudies, vol. 2. See now also Sezgin, Goldziher and hadith.
118
Regarding this important aspect of the Koranic theology see in particular Snouck
Hurgronje, de Islam, tweede deel, pp. 259sq. and 455sq., derde deel, p. 109; and also his
review Une nouvelle biographie de Mohammed, by H. Grimme, pp. 63 and 150.
119
Cf. now also Th. Nldekes review, L. Caetani, Annali dellislm, p. 299.
120
Ibn Sad (al-abaqt): Biographie Muhammads bis zur Flucht, vol. 1, part 1, p. , l 12.
introduced into a previously revealed sra. Verse 23, as well as 26 to 33, are
undoubtedly somewhat later than the rest of the sra, yet they neverthless
refer to the verses that Muammad dismissed and declared to be the work
of Satan when he recovered his senses. Verses 58sqq. make up a unique,
independent segment with different rhymes. Muir (vol. 2, p. 319) lists the
entire sra under his fourth stage either because of its greater length or
because of verses inserted later. Some writers hold verse 33
121
or verses 34
to 42
122
or the entire sra
123
to be of Medinan origin.
The beginning of sra 84 [When heaven is rent asunder] is a continua- [i/104]
tion of sras 82 (When heaven is split open), and 81 [When the sun shall
be darkened]. Verse 25 [theirs shall be a wage unfailing and identical with
95:6] is probably secondary, for the reason stated above on page 70.
Sra 100 some writers
124
falsely consider Medinan, thinking that the first
verses refer to the riding mounts
125
used by Muammad in campaigns.
Sra 79 consists of three parts: verses 1 to 14, 15 to 26, and 27 to 46. The
third group of these is probably somewhat later, a fact that, in addition to
the slightly greater length of the sra, apparently prompted Muir to include
the entire sra in the following stage of his list of sras.
According to one tradition,
126
sra 77 was revealed when Muammad
once found himself with several companions in a cave near Min. Verse 48
is inappropriately considered by some writers to refer to the tribe of Thaqf
who during the last years of the Prophet wanted to accept Islam only on
the condition that they be exempted from prayer.
127
In addition, the sra is
noteworthy for the refrain-like repetition
128
of the same verse (15, 19, 24, 28,
34, 40, 45, 47, and 49).
Sra 78:17 already seems to presuppose sra 77:12sqq. Judging by their
style, verses 37sqq. were probably added only in the second stage.
129
Hibat
Allh [Ibn Salmas] strange opinion that the sra was the very last Meccan
121
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; al-Itqn, p. 36.
122
123
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF].
124
Loc. cit.; al-Wid; Hibat Allh b. Salma; al-Zamakhshar; al-Bayw; al-Itqn, p. 30.
125
al-Suy, al-Itqn, p. 30; al-Wid; al-abar, Tafsr; al-Zamakhshar.
126
al-Bukhr in Kitb al-tafsr, s.v., according to Abd Allh b. Masd; al-Itqn, 45.
127
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; al-Suy, al-Itqn, p. 37;
al-Suy, Asbb al-nuzl.
128
Cf. above, p. 33.
129
Flgel has after
, verse 40, a hemistich, which is neither in agreement with good

tradition nor with the fact that beginning with verse 6 the rhyme is identical and in the
paenultimate syllable all end with .
sra revealed, on the day before the emigration to Medina, can be explained
only when verse 17 is considered to be a reference to this event.
130
According to Hibat Allh, sra 88 is said to originate from the year of the
conquest of Mecca, in 8/629.
Some exegetes hold sra 89 to be of Medinan origin.
131
In sra 75 we find some verses (16 to 19) that have no connection with the [i/105]
immediate context or withother verses of this sra. It cannot be determined
why they ended up in this place.
We also have a great deal of inaccurate information regarding sra 83.
Since by a stretch of imagination the first verses can somehowbe connected
withMedinanaffairs,
132
the first six
133
or the first twenty-eight
134
or all
135
verses
are sometimes held to be of Medinan origin. As we were able to observe
above, p. 48sq., others hold the sra to be either the last Meccan or the first
Medinan sra. Finally, both views find their intermediary in a third view,
maintaining that it originated between Mecca and Medina.
136
Muir assigns sra 69 to the following period, probably on account of its
greater length.
137
In sra 51, verses 24sqq. are likely to have been added later.
In sra 52, which already presents a somewhat broader description of
Paradise, there are verses from stage two: verse 21,
138
which does not fit into
the context and is disproportionately long, running three times the length
of the largest of the rest of the verse, and also verse 29sqq. As examples
of the difference of their diction from what is generally prevalent in the
first stage, one can list only the expressions, and , both
from verse 43 (regarding the recognition of other gods), as well as the
phraseology beginning at verse 48, all of which belong to Muammads
130
is the Final Judgement, but also means to go away, depart, e.g., in 12:94.
131
132
al-Suy, al-Itqn, p. 28sq.; al-Zamakhshar; al-Bayw.
133
134
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; Al al-Dn (Al b. Muam-
mad al-Baghdd AL-KHZIN).
135
al-Suy, al-Itqn, pp. 28 n. 55; al-Suyt, Asbb al-nuzl, according to al-Nas, Sunan,
and Ab Abd Allh IBN MJAH [EI
2
; EQ; Juynboll, Encyclopedia, pp. xxxxxxii, Sezgin, GAS,
vol. 1, pp, 147148], Sunan. Although according to another tradition in Al al-Dn (Al b.
Muammad al-Baghdd AL-KHZIN), verse 13 is held to be Meccan.
136
al-Suy, al-Itqn, p. 29; Al al-Dn (Al b. Muammad al-Baghdd AL-KHZIN).
137
The types of rhyme of the sra are very peculiar. Verses 1 to 29 rhyme with , verses 30
to 32 with
, and verses 33 to 52 mixed rhyme with m, n, n, l.

138
Cf. thereon, Fakhr al-Dn al-Rz, s.v., and al-Khab al-Tibrz, Mishct, K. al-mn
bi-l-qadar, fal 3, 5.
later diction.
139
This sra, which is somewhat longer, Muir reckons among
his fourth stage, probably on account of the later verses.
Sra 56
140
is Medinan according to al-asan al-Bar.
141
Others claim this [i/106]
only for the verses 74 to 81
142
or solely for verse 81,
143
which they apply to the
Hypocrites of Medina. Some others say the same thing of verses 94sqq. or,
finally, also of verses 1 to 3,
144
possibly because of the allusion to the Battle
of Badr. This sra can be considered to be homogeneous, although verse 74
seems to be the beginning of a newsection; both sections (verses 73 and 96)
end identically [Then magnify the Name of thy Lord, the All-mighty], and
the verses from 87 to the end show a distinct relation to the first part. But
it might just as well be conceivable that we are facing a special revelation
which Muammad composed with verses 1 to 73 in mind and therefore later
combined the two. In this case verse 96 (= verse 73) would be an editorial
matter.
The opening of sra 70 seems to be related to that of sra 56. Perhaps
a disbeliever mockingly asked Muammad to comment on those verses
and received a thundering reply in this sra. The sra first supplies a more
detailed description of the duties of believers. Weil
145
holds that verses three
andfour were addedlater. This guess makes sense but only for the later verse,
which indeed appears to be a gloss. Verses 30 to 32, and 34 are also found in
sra 23:59. Since verse 34 constitutes an almost literal repetition of verse 23
it could well be regarded as later. In this case, verses 30 to 32 are also likely
borrowed from sra 23.
Sra 55, with its almost playful manner, shows itself to be a somewhat
later product, for which reason I previously followed Weil in reckoning it
part of the second period. Most Muslim scholars dismiss
146
the opinion that
all of the sra, or part of it, but in any case verse 29,
147
is Medinan. The
139
Cf. above, p. 79sq. on sra 68:17sqq.
140
The verses 8 to 10 do not seem to have been transmitted intact. Since verse 10, where
analogically to verses 26 and 40 after probably is missing, can only be the intro-
duction to what follows, you should think that the sbiqn ought to have been mentioned
once before. In addition, the questions in verses 8 and 9 are here not quite fitting, and might
possibly come here from verses 26 and 40.
141
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF].
142
al-Suy, al-Itqn, pp. 36 and 44; al-Suy, Asbb al-nuzl; Fakhr al-Dn al-Rz on v. 81.
143
144
Cf. on these different opinions al-Suy, al-Itqn, p. 36.
145
Weil, Historisch-kritische Einleitung, p. 70, note.
146
Hibat Allh (Ibn Salma;) al-Suy, al-Itqn, p. 27.
147
Cf. [Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; al-Itqn, pp. 27 and 36;
al-Zamakhshar; al-Bayw.
peculiar style of this sra, already indicated above (p. 30), is the endless
repetition of the tiresome refrain
[Owhich of your Lords

bounties will you and you deny?] It is found from verse 12 to verse 21 in
every third verse, and from there to the end in every second verse, with the
exception of verses 25 to 28, and 43 to 45,
148
where, as in the beginning, two
verses come inbetween. It is hardtounderstandwhy the refrainwas not also
used in verses 2 to 10. The moral lesson of verses 7 and 8, it seems, was added
to verse 6 only later. Verse 33 is disproportionately long and lacks rhythmic
elegance, suggesting that probably only the last five words (starting with
) belong to the original form.
We follow up the discussion of the revelations of the first period with [i/107]
some short sras that serve as formulas of the creed and magical conjura-
tion. Although it is hardly possible to determine their exact date, since they
are too short and their whole character differs from all the others too much
to give any clue, still they seemmore likely to belong to Muammads earlier
period than to his later one.
Many scholars assign sra 112 to Medina because it was thought to be the
Prophets reply toits Jewishpopulationregarding the essence of God.
149
Muir
attributed it to the very earliest period, immediately following upon sra 96.
He probably arrived at this conclusion
150
through his erroneous assumption
that Muammad must have established a kind of creed immediately after
his prophetic commission.
Sra 109 contains the answer to the Meccans compromise to follow the
Prophet so long as he gave their gods their adequate due.
151
It can only
148
This exception would not apply if the verses 43 and 44 of Flgels edition of the
Koran were to form a single verse as a sound tradition (Ab Yay Zakariyy AL-ANR,
al-Maqad) requires. This would also eliminate the extremely offensive and isolated rhyme
with n (verse 43). Apart from this, conspicuous rhymes are only found in verses 16 and 17
(with ayn, everywhere else with n, r, and m), which, in addition, also disagree with the
tenor of the surrounding rhymes.
149
Ibn Hishm, p. 400; [Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF;] al-
Wid; Hibat Allh (Ibn Salma); al-Bayw. In addition we also learn from al-Suy, al-
Itqn, p. 30, that to this sra, as well as the first sra, some scholars attribute a double origin,
both Mecca and Medina.
150
Hirschfeld, Newresearches, pp. 35, 89, 143, is guided by the same principles and lists the
sra as the third oldest, immediately following upon sras 87 and 68. Sprenger, Leben, vol. 2,
p. 33sq., and Grimme, part 2, Einleitung in den Koran, p. 26, are more in agreement with
our point of view.
151
Ibn Hishm, p. 239; al-abar, vol. 1, p. 1191, and in his Tafsr; al-Wid; Fakhr al-Dn
al-Rz; al-Bayw; Weil, Historisch-kritische Einleitung, 1st ed., p. 60, 2nd ed., p. 69; Sprenger,
Leben, vol. 2, p. 34sq. We have seen above (in the case of sra 53, cf. also sra 106) that
have been revealed after Muammad had quarrelled with his countrymen
for quite some time, as they could not have demanded such conditions
otherwise. Some writers falsely hold the sra to be Medinan,
152
but after the
hijra his reaction to such a proposition
153
would have likely been different.
Far more difficult is the classification of sras 113 and 114, which Muslims [i/108]
call collectively the muawwidhatni. Not even Muir,
154
who in other cases
assigns a definite place to every sra, dares say anything in this respect.
According to a common tradition
155
they were revealed to relieve the Mes-
senger of Allhfromthe disease whichLabd, the MedinanJew, hadinflicted
on him by a magic spell. One cannot doubt that Muammad might have
been susceptible to such outright superstition. In this respect he never
ceased to share the views of his age and people, as proven by the numerous
episodes of credulity recordedby his biographers. The wording of bothsras,
however, is entirely general and has no specific subject at all. Weil
156
differs
with the tradition insofar as he claims that the Prophet used this conjura-
tion only at this time, although orthographically they must be older. Yet this
view is equally open to criticism, as language and style, ordinarily an excel-
lent aid in dating sras, are of no use in this instance. Incantations the world
over have an antiquated diction different from the common usage of the
era and individual. If Muammad had composed such an invocation, even
in his later years, it would certainly be far different from the usual style of
the Medinan sras, resembling more closely the archaic type of the pagan
averruncationformulae. We might evengo one stepfurther andsuspect that
Prophet did not freely invent these sras at all, instead slightly modifying a
Muammad did not always outright reject such proposals. This passage also shows us how
the pagans envisaged a compromise between Muammad and their belief. Since they were
already at that time contemplating some kind of monotheism, they merely wanted an
honourable status for their oldgods; but althoughMuammadat times displayedwillingness
to accept them in his heaven as subordinate beings, his rigid unity of God soon gained the
upper hand.
152
[Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; Hibat Allh.
153
We naturally need not worry too much about the details of form in which Islamic
tradition depicts these proposals of the pagans.
154
The Life of Mahomet, vol. 2, p. 320.
155
al-Wid; Fakhr al-Dn al-Rz; al-Suy, al-Itqn, p. 30sq.; Ibn Sad, ed. Sachau, [al-
abaqt] vol. 1, iv, p. 5sq.; Weil, Das Leben Mohammeds, p. 94. That these sras are held to
be of Medinan origin by some people is also mentioned by al-Zamakhshar, al-Bayw, and
al-Itqn, p. 20sq. That story is related also by al-Bukhr, K. al-ibb, 7; Muslim, K. al-ibb, 2
(= al-Qasalln, vol. 9, p. 19); al-Nasa, K. Tarm al-dam, 19; al-Khab al-Tibrz, Mishct,
bb al-mujizt, fal 1, 24, without mention of these sras.
156
Historisch-kritische Einleitung, 1st ed., p. 60 n. 2, 2nd ed., p. 69 n. 3.
traditional text and presenting it in Islamic garb. The three last verses of the
muawwidhatnithis is more than half of themindeed display a pagan
mark. The necessity of rewriting might have appeared quite early, since,
while Islamshared with paganismthe belief in misanthropic spirits, it could
not beseech help from any god other than the unique Allh. Furthermore,
if it is true that several sras referring to conjurations against the power
of Satan (sras 23:99, 16:100, 41:36 = 7:199) belong to the second and third
Meccanperiods, we may assume witha great likelihoodthe muawwidhatni
are older. Their positionat the endof the Koranis most likely due tothe same
superstition that prompts Muslims to this very day to begin every recitation
of the Koran with the formula: I seek refuge in God from the accursed
Satan.
Similar to the two sras of conjuration, the first chapter of the Koran is [i/110]
also out of place with the rest. Whereas the former sras serve to instruct
and admonish, the first sra contains nothing but enthusiastic glorification
of Allh, closing with request for guidance in the straight path. The partic-
ular Islamic ingredient recedes to such an extent that the prayer would fit
unnoticed into Jewish or Christian devotional literature. It is precisely for
this reason that it is so difficult to determine its age. Given these circum-
stances, it is wrong to consider it to be the earliest sra,
157
or even one of the
earliest,
158
merely because of its repute among Muslims fromtime immemo-
rial,
159
or because of it position in our current Koran. Although the frequent
157
al-Diyrbakr, Tarkh al-khams, Cairo ed., vol. 1, p. 10. Thus Weil, p. 364, note.
158
al-Diyrbakr, Tarkh al-khams, loc. cit.; al-Zamakhshar; al-Suy, al-Itqn, p. 54; al-
Wid, p. 11. Cf. Weil, loc. cit. The strange statement that the Ftia was revealed at Mecca
as well as at Medina (cf. [Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; al-
Zamakhshar; al-Bayw; al-Suy, al-Itqn, pp. 25 and 124; al-Diyrbakr, Tarkh al-khams
(Cairo ed., vol. 1, p. 11)) is based on the inaccurate interpretation of mathn as repetition.
Some writers tried to solve this discrepancy by having one part of the sra be revealed
at Mecca and the other part at Medina, but this explanation is rejected (Ab l-Layth al-
Samarqand and al-Itqn, p. 25). When a few ancient exegetes hold this sra to be Medinan
(Ab l-Layth al-Samarqand; [Ab l-Qsim] Umar b. Muammad [IBN ABD AL-KF]; al-
Wid; al-Suy, al-Itqn, p. 25) this is unlikely to conformto a tradition, rather it was done
because of the alleged antithesis to Jews and Christians contained in the last verse. Even
early traditions call the former and the latter . Al-Kalb in Ms. Sprenger 404
(= Ahlwardt, Verzeichnis, vol. 1, no. 732), al-Tirmidh in the Tafsr; al-abar, Cairo ed., vol. 1,
pp. 60sqq.
159
As the first chapter of the Koran it is called al-Ftia (the Opening, actually Ftiat
al-Kitb) and because of its unique content, umm al-kitb (al-Bukhr, Fail al-Qurn, 9;
al-abar, Tafsr, vol. 1, p 35, etc.). According to an alleged saying of the Prophet, the like
of it is to be found neither in the Torah, the Gospel, the Psalms nor the Koran (al-abar,
Tafsr, vol. 1, p. 36; al-Wid, p. 12sq., etc.). As a prayer it equals the Christian Lords Prayer.
Tradition knows of a saying of the Prophet that there is no valid religious service without
use of we
160
would, at least, indicate that Muammad already had a small
congregation around him at the time of sras origin, Muir
161
still thinks
that the sra belongs to his first stage, i.e. the period before Muammads
actual prophetic commission. However, it seems that the sra originates
fromthe end of the first period at the earliest, as it contains several remark-
able words and expressions that appear nowhere else in the first period but
frequently
162
in the second period. It is not easy to determine the lower limit,
since the literary relation of the first sra to the parallels mentioned below
in note 162 is by no means clear.
Of course, if these phrases were entirely of Muammads invention, their [i/111]
original placement would likely not be in a prayer, as a liturgical form with
rigid terminology could not possibly go back to an ecstatic period of reli-
gious foundation grappling not only with theological conceptions but also
with linguistic expression. Nevertheless, as we shall demonstrate below,
163
the Ftia (al-Bukhr, Adhn, 93, cf. 105; al-Tirmidh, alt 63; al-Nas, Iftit, 24; Ibn
Mjah, K. al-alt, bb al-iftita al-qira). At an early period it was used as a charm(Bukhr,
Fail al-Qurn 9 at the end, Kitb al-ibb, 33, etc.) An equally frequent name of the sra,
al-amd (e.g., Ibn al-Nadm, K. al-Fihrist, ed. Flgel, p. 26; al-Suy, al-Itqn, p. 150), has been
taken from the first word.
160
, , , twice.
161
Life of Mahomet, vol. 2, p. 59.
162
(verse 1) sras 18:1, 34:1, 35:1, 27:15, 17:111;
(verse 1) sras 37:182,

40:67, 39:75, 10:11; (verse 5) 43:42, 11:59, 7:15, 36:3, 42:52, 37:118. Regarding al-ramn
(verse 2) see below, the introductiontothe sras of the secondperiod; al-ramalone is found
in sra 52:28.
163
(I) is the exact equivalent to the Syriac and ,
respectively as well as the NewTestament . Next to it we also find =
, St. Luke 1:68, II Corinthians 1:3, a formula which is used even in the Old
Testament ( ) , Exodus 18:10, etc.; Tobith8:5 and, with a little change, became
prevalent in the Jewish liturgy ( ).
(II)
: Withthis the following phrases are tobe compared: , TargumQohel

7:3, 7:13, 9:7; Ruth 4:21; Genesis 9:6 Y[erushalmi]; Genesis 22:1, 22:5, 49:27; Exodus 12:11, 19:17;
Numbers 21:1, 21:14; Palmyrene, de Vogu, no. 73, 1 (Cooke, Textbook, p. 296).Mandaean
Sidrh Rabb, p. 1, l 21, p. 36, l 1 and 9. , Exodus 23:17,
O.J. [abbreviation not identified] and frequently. , very frequently in the
Midrashim and at the beginning of Jewish prayers. , Targum to Mic[a]h. 4:13;
Canticles 1:1. , Targum Qohel 8:3; Genesis 18:30sqq.; Numbers 23:19.
(frequently in the Midrash).In the Jewish liturgy there is most commonly
(already Jeremiah 10:10; but Targum to Isaiah 6:5; Zechariah 14:16, .) In the
NewTestament Apocr. 15:3, (other Mss. ).Ethiopic: egza (kuellu)
lamt (Liber jubilaeorum, ed. A. Dillmann, cap. 31, p. 112, penult.); Enoch, p. 81, l 10, amlka
lam(lamt) Liber jubilaeorum, cap. 12, p. 52, p. 52, l 1; amlkmulalam, Liber jubilaeorum,
cap. 25, p. 93, l 12sq.
(III) : That the name was new to the Meccans we can see from sras
17:110 and 25:61; cf. the commentaries Ibn Hishm, p. 747, l 11, and al-abar, vol. 1, p. 1546,
the largest part of the sra, namely the phraseology of the verses 1, 2, 3, and
5, is of Jewish or Christian provenance. In such circumstances the Ftia
might well be older thanthe above-mentionedparallels. This wouldbecome
l 9. But it was by no means totally unknown in Arabia. Its occurrence in the poetry of al-
Burayq al-Hudhal [Sezgin, GAS, vol. 2, pp. 263264] (Dwn der Hudhailiten, no. 165, l 6)
and Suwayd b. Ab Khil, [d. ca. 65/685; Sezgin, GAS, vol. 2, pp. 165166] (al-Mufaaliyyt,
no. 34, l 60) can be accorded little credence since these men were familiar with Islam; on
the other hand, we might here even have a Muslim correction like in Izz al-Dn IBN AL-
ATHR, al-Kmil, vol. 1, p. 450, l 2. It is more important that the prophet Musaylima indicates
his god (al-abar, vol. 1, p. 1933, l 12 and p. 1937, l 3), and that he himself is even called
the Ramn al-Yamma corresponding to his South Arabian rival Aswad, the Ramn of
the Yaman (al-Baldhur, p. 105, l 6; al-abar, vol. 1, p. 1935, l 14; Ibn Hishm, p. 200, l 3;
al-Zamakhshar on sra 1). If this were to be in imitation of Muammad, it would be diffi-
cult to understand why a name was chosen that is used for Allh only in the middle Meccan
period. We are fortunate to have six Sabaean inscriptions, authentic pre-Islamic documents,
mentioning ramn. (Mordtmann and Mller, Sabische Denkmler, no. 43, l 2, p. 96sq.;
F. Fresnel, Sabische Inschriften, in Denkschrift d. Wiener Akad. d. W., v. 33, p. 3, l 3; J. Halvy,
Sabische Inschriften, p. 63, l 7; D.H. Mller, Himjarische St., ZDMG, 671708; Mordtmann
& Mller, Monotheist. sabische Inschrift, WZKM, 285292; Ed. Glaser, Zwei Inschriften,
p. 618.) This consonantal form, , common to both texts, is usually considered a plural.
But from Ed. Glaser, loc. cit., p. 554, by comparing l. 32 and l. 81 sq.
it is evident that here as well as in other passages we are dealing with a singular. This
linguistic use cannot possibly have developed spontaneously, rather it must be based on bor-
rowing. Now, ramn is extremely rare in Christian Aramaic, e.g., in Ephrem (see P. Smith,
s.v.), and in Christian-Palestinian Aramaic. Peshitta interprets Old Testament as well as
and by . Onthe other hand, inHebrewliterature, starting withthe
Targums, was so popular that, for example, in both Talmuds it became a common name
of God. The ancient Arab lexicographers like al-Mubarrad and (Amad b. Yay) THALAB,
[EI
2
; Sezgin, GAS, vol. 9, pp. 140142,] claiming its Hebrew origin (
al-
Suy, al-Itqn, p. 321; Lisn al-Arab, vol. 15, p. 122), thus seem likely to be quite correct. In
the final analysis ramn seems to be the Syriac equivalent of the Assyrian rmnu; as an
epithet of various deities the word appears in the Palmyrene inscriptions of the first, second,
and third centuries (cf. Cooke, Textbook of North-Semitic inscriptions, pp. 295, 300, and 301).
Although is a truly Arabic form, its specific meaning of compassionate is based in
this form and all the others of this root on the assimilation of the North Semitic linguistic
usage. Muammad himself might have still used it withthe meaning of amiable, kind as it is
rendered by in the bilingual Heidelberg papyrus no. 21 (Papyri Schott-Reinhard,
vol. 1, edited and explained by C.H. Becker, Heidelberg, 1906, p. 103). On the other hand, the
close connectionof the twoexpressions makes it probable that the adjective ramwas added
to the noun ramn merely to intensify the noun. Apart from the basmala and sra 1:2 the
connection al-ramn al-ram is found only in some sras of the second and third Meccan
period (2:158, 27:30, 41:1) as well as once in a Medinan passage (29:22.)
(IV) : already Judith16:17; Testamentaxii Patriarch. inLevi [sic] at
the beginning, frequently in the NewTestament and later Peshitta , Ephrem
; , Targum Qohel. 3:15 and 17, 7:15, 12:14; Job 5:4; in the prayer of Elxai, Epiph.
Haeres. 19:4, cf. M.A. Levy, Bemerkungen, p. 712; Mekhiltha on Exodus 6:25.
Ethiopic elata kuenan abi, Enoch, c. 16: 1; elata dain, Sive Liber jubilaeorum qui idem a
Graecis, c. 4, p. 18, l 2.
even more likely if verses 1 to 5 had been available to the Prophet as a defi-
nite entity. However, if he had borrowed only single phrases and then freely
composed a prayer, sra 1 could also be from a later date.
The mystery of the date of composition would immediately resolve itself [i/114]
if the words of the seven of the mathni (sra 15: 87)
164
were indeed to refer
to the first sra, as many Muslimexegetes maintain.
165
But this is very doubt-
ful. The expression seven of the oft-repeated includes the presumption
that there had been other mathn. Muslimtradition, which tacitly replaces
it with the seven mathn, cannot be right. As far as the sense
of mathn is concerned, none of the transmitted meanings, such as rep-
etitions
166
or verses, is sound. In the only other passage of the Koran
167
in
whichwe findal-mathn (39:24), the meaning is alsonot clear. More accept-
able than any of those interpretations I consider A. Geigers assertion
168
that
I cannot find a reference to the phrase King of the Day of Judgement although the
kingdom of the Messiah is quite a common notion not only among Jews (e.g. Targum
Yerushalmi toNumbers 24:7, 17) but alsoamong Christians (St. Matthew2:2; St. Mark 15:2sqq.;
15:2sqq.; St. John 19:3sqq., etc.).
(V) corresponds as far as possible to Psalms 27:11 . But
this is not to say that Muammad can have borrowed these words only from Jews (cf. above,
p. 5).
We cannot say for sure whether the following two verses are the Prophets free invention
or only a traditional interpretation, although the somewhat clumsy diction might readily
be explained by the difficulty of translation. Calling the behaviour of the unbelievers going
astray as in the last verse of the sra is exceedingly often to be found in the Koran. in this
religious meaning corresponds to Aramaic and is common in Jewish as well as Christian
literature. In the case of , as time went on, Christians were increasingly thinking of
heretics rather than pagans.
164
We have given thee seven of the oft-repeated, and the mighty Koran.
165
al-Muwa, p. 28; al-Bukhr, K. al-Tafsr on sra 1 and 15:87, Fail al-Qurn, 9;
al-Tirmidh, Fail al-Qurn at the beginning, K. al-Tafsr onsra 15:87; al-Nas, K. al-Iftit,
26; al-Wid; al-Suy, al-Itqn, p. 124; the Commentators, particularly al-abar, and the
dictionaries.
166
The designationrepetitions is derivedfromthe frequent use of the Ftiainthe liturgy
of prayer, or from the endless repetition of the sra headings In the Name of God, which,
as we shall see later, many consider to be the first verse of the Ftia; or from the phrase
used twice in this sra, al-Ramn al-Ram. The meaning verse (=
) is justified by the
peculiarity of the verses to followone upon the other (
), or that
Allh thus distinguished Muammad from other prophets (cf. above, p. 90, n. 159):

. Cf. inthis context the Commentators, particularly al-abar on

sra 15:87 (vol. 14, pp. 3238) and al-Suy, al-Itqn, p. 124, where other strange explanations
are listed. Regarding the identification of Hebrew mishn and Arabic al-mathnt in adth,
cf. Ignaz Goldziher, Kmpfe und die Stellung des Had, pp. 866sqq.
167
Allh has sent down the fairest discourse as a Book, consimilar in its oft-repeated,
whereat shiver the skins of those who fear [their Lord.]
168
Loc. cit., p. 58, which Sprenger, Das Leben und die Lehre, vol. 1, p. 463, follows and
translates mathn by renewed revelation.
it was connected with Hebrew mishnit would be more appropriately
calledJewish-Aramaic mathnthtradition. This couldalso be the mean-
ing in sra 15:87.
169
Finally, the division of the first sra into seven verses creates enormous [i/115]
difficulties, since, in order to arrive at this number cannot form the
end of a verse for lack of topical break and rhymewe are obliged to follow
that type of enumeration according to which the heading
is considered to be one of the verses of the sra.
170
Most exegetes,
however, do not reckon this heading to be a verse in other sras, and the
heading does not represent an integral part of the Ftia, which, by analogy
with Jewish and Christian prayers, begins with the words Praise be to
God.
171
Therefore, if the first sra consists of only six verses, the seven
mathn of sra 15:87 cannot refer to it.
The words of introduction, in the Name of God, called by the Arabs [i/116]
tasmiya or, commonly, basmala, can be traced back to the diction of the
Bible.
172
When this expression occurs there, of course, it is always combined
with words of action, although phrases such as to invoke in the Name of
God, and passages like Colossians 3:17, presupposeas I pointed out on
169
Those Muslimexegetes who, when commenting on sra 15:87, orientate themselves on
sra 39:24, apply saban min al-mathn not to the Ftia but either to all of the Koran
arguing that its content consists of seven types (

, al-abar, Tafsr s.v., vol. 14, p. 36, l 9sq.), or that its individual narrations are oft-
repeatedor to the sevenlong sras whichto some peopleprobably most of themrefer
to sras 2 to 7 and 10, to others, to sras 2 to 8, and still others, to sras 2 to 7, with the remark
that the last of the seven were unknown to them. Cf. particularly al-abar, Tafsr on sra
15:87, and al-Suy, al-Itqn, p. 124.
170
Meccan and Kufan readers consider the basmala to be a verse. From among the latter
it is amza alone, but he does it only in the case of this sra. Others introduce a caesura
after . The division into seven verses is by far the most common but not the only one
as al-Zamakhshar, al-Bayw and others maintain. There are still others who reckon only
six verses, by disregarding the basmala in their count, yet they still do not make a break after
, or eight verses by reckoning those, and here presuppose anend of the verse, or evennine
verses by making a break also after . Cf. al-Suy, al-Itqn, p. 159sq.; Umar b. Muammad;
al-Zamakhshar; al-Bayw; al-Sajwand on pauses (Ms. Wien, Mxt., 717); Leiden, Ms 653
Warner; al-Itqn, p. 185sq.; Ab Yay Zakariyy [AL-ANR] al-Shfi F l-waqf wa-l-ibtid,
p. 14; Muslim, K. al-alt, bb, 15,
(al-Qasalln, vol. 3,
pp. 2628).
171
occurs also at the opening of other sras (6:1, 18:1, 34:1, and 35:1), thus all Meccan
passages. Regarding the occurrence of the religious term to praise in Sabaean inscrip-
tions cf. Johannes H. Mordtmann and David H. Mller, Eine monotheistische sabische
Inschrift, particularly p. 286.
172
Cf. in the Old Testament, and in the NewTestament. As far as
we nowknow, Greek usage did not produce this formula. Cf. A. Dieterich, Eine Mithrasliturgie
erlutert, p. 115.
pp. 5758the absolute usage of the formula. Thus, the two only passages
in the Koran in which the basmala occursapart fromthe sra headings
can unmistakably be retraced to Jewish sources. In sra (Hd) 11:43 it says:
Embark in it! (i.e. the Ark) In Gods Name. Sra 27:30 mentions a letter of
Solomon to the Queen of Sheba that begins with the words, bismillh al-
Ramn al-Ram. This verse is of particular importance. Apart from the
headings, it is not only the single passage where the expanded form of the
basmala occurs within a sra but alsoif the basmala does not belong
to the original text of the Ftiathe oldest passage with the formula
at all. Sra 27, however, originates from about the middle of the Meccan
period. The next certain evidence for the use of the formula by the Prophet
are the transmitted texts
173
of the Constitution of Medina,
174
the Pact of al-
udaybiyya, and the epistles to the pagan tribes, all of which belong to the
Medinan period. Even if it cannot be doubted that at some point Muam-
mad began to place the basmala at the head of sras, the dating of the
formula remains unknown. Tradition
175
even holds the basmala to be the
oldest revelation, although it is by no means certain that the Prophet ever
considered this formula a part of revelation.
173
Cf. Ibn Hishm, p. 342sqq., and p. 747; al-abar, vol. 1, p. 1546; J. Wellhausen, Medina
vor demIslam, p. 87sqq.
174
The English translation of Wellhausens text in A.J. Wensinck, Muammad and the Jews
of Medina, pp. 128138.
175
al-abar in the introduction to the Tafsr (Cairo ed.), vol. 1, p. 37sqq.; al-Wid in
the introduction (Cairo ed.), p. 10sq.; Leiden, Ms. 653, fol. 275
v
; al-Suy, al-Itqn, p. 54sq.,
184sqq., etc. According to a different tradition (al-Wid on sra 17:110, Cairo ed.), p. 223, the
basmala was used by Muammad only after the revelation of sra 27:30.
THE SRAS OF THE SECOND MECCAN PERIOD
General Comments
As has been observed above, on p. 52, these sras have no particular char- [i/117]
acter in common, some of them bearing resemblance to those of the first
period and others to those of the third period. We recognize the transition
from overwhelming enthusiasm to great serenity and on to the rather pro-
saic later sras. Gustav Weil
1
attributes the main purpose of this change
of style to Muammads endeavour not to appear as a poet or soothsayer.
This opinion, however, can be disregarded, as this transition did not hap-
pen suddenly, as would be expected if it had been based on a conscious
purpose, but rather came about gradually. It might be added that even in
later sras
2
Muammad still complains about reproaches directed against
boththe content andthe formof sras. Weils other criticismis quite serious,
even though presented with undue harshness. The first outburst of enthu-
siasm was bound to be dampened by the disappointment of reality. The
constant repetition of the same ideas which, again and again, fell on barren
ground, were bound to be detrimental to the artistic form of its presenta-
tion. Muammads fantasy hadtosacrifice elegance andoriginality the more
he was obliged to look after the practical needs of the young community.
This development is not surprising, since it follows the law of nature; and
in view of the final success it need not be regretted. Nevertheless, Muam-
mad was convinced of the outcome of his divine commission up to the
very end. Again and again he gathered fresh hope from this conviction, and
all the grandeur of the later revelations emanated from his never-ending
stamina.
Quiet reflection gradually replaced the enormously excited fantasy and [i/118]
enthusiasm of the first period. The Prophet endeavours to explain his sen-
tences with numerous examples fromnature and history. Since he heaps up
these examples rather than logically arranging them, however, he becomes
verbose, vague, and evenboring. His ability to reasonleaves something to be
desired, endless repetition leading to the intimidation of his opponents but
1
Das LebenMohammeds, p. 387; Historisch-kritische Einteitung indenKoran, 1st ed., p. 55,
2nd ed., p. 64.
2
Sras 23:72, 34:8 and 45, 7:183.
98 the sras of the second meccan period
never to their conviction. This is not to say that the later sras are without
beautiful and serene passages. The power of his thoughts, which made him
a prophet, is permanently evident. The traces of his poetical spirit, which
is everywhere apparent in the earliest sras, declines without ever disap-
pearing. Despite all the rambling in his presentation, these late revelations
not infrequently offer passages in which thoughts boldly bypass the con-
text, eveninthe narrative sections, whichgenerally containmany appealing
passages. At the same time, we must remember that the Koran was aiming
primarily at listeners and not readers, and that many things that appear bor-
ing to usbecause we are familiar with its original Biblical formleft an
entirely different impression on Muammads contemporaries.
In this second period, all these qualities of the later revelations gradually [i/119]
come to the fore. The diction at first attempts to maintain the level of
the early sras but the narratives become increasingly broad and more
dispassionate. The greater calmbecomes apparent in the gradually growing
length of both the verses and the individual revelations.
Fiery declamations give way tobroadelaborations ondogma, particularly
the recognition of God from the signs (
) prevalent in nature, and also to

long tales from the early prophets, which serve as proof of dogma, warning
to enemies, and consolation to followers. In so doing, Muammad usually
has the old messengers of God use his own personal style. On the whole, all
these prophets share with one another, and with Muammad, a great family
likeness that at times extends even to minor, unimportant characteristics.
The indications that the Koran offersnot so much about events but about
the general relationof the Prophet tohis followersare aptly supplemented
by many tales related. Incidentally, Muammad deals most frequently with
the history of Moses, to whom he then felt most closely related.
The change of style implies new idioms and the abandonment of old [i/120]
ones. For example, the stilted invocations (cf. p. 55sq.) so typical of the
early period gradually disappear. Sra 37 still sets out with a lengthy invo-
cation; thereafter we only find such short formulae as by the Koran, by
the Book,
3
etc. until, finally, the third period has none of this at all. On the
other hand, the Prophet starts in this period to assign formal headings
4
to
3

(sra 36);
(sra 38);
(sra 50); (sras 43,

and 44).
4
For example: Those are the signs of the Book and of a manifest Koran,
(sra 15); similar
(sra 26), and
(sra
27:1).
the sras of the second meccan period 99
those sras that emanate from more serious reflection, as attestation of
their divine origin, e.g.: This is the revelation of God, etc. Or he announces
himself as the speaker of the divine words by an explicit
speak! which is
totally lacking in the early sras, and precedes only formulae intended to be
used frequently by man, i.e. sras 112, 113, and 114but not the first sra. In
this connection we cannot consider it accidental that certain expressions
for to reveal occur only now and then in the Meccan sras but become
quite frequent later on.
5
The Divine Name Ramn
In this period, Muammad started to introduce the specific name, al-Ra- [i/121]
mn, the Merciful,
6
for his God, concurrently with Allh, which was famil-
iar alsotothe pagans. This name, whichwas previously usedonly once,
7
now
becomes in places even more frequent than the usual Allh.
8
al-Ramn,
on the other hand, disappears in the sras of the third period, apart from a
few exceptions,
9
and is completely lacking in the Medinan period.
10
What
prompted the Prophet to abandon the use of this name is only vaguely
known to us. It could possibly have been his intention to avoid being sus-
pected of worshipping two deities, Allh and al-Ramn. At least a fewMus-
lim Commentators on sra 17:110 maintain that such a silly defamation was
indeed once mentioned.
As stated above, the sras of this period are somewhat easier to place
into a kind of chronological arrangement. This, naturally, applies only in a
general sense, as even in this instance we are unable to assign a precise or
even a relative place to individual sras with any kind of certainty.
5
revelation and to reveal respectively, in the first Meccan period only three
times, 99:5, 53:4 and 10, but in the second period fifty-three times, and in the third period
thirty-three times; of the sending downof the revelationinthat first periodonly five times,
97:1 and 4; 56:79; 53:13; 69:43, but in the later Meccan sras exceeding one hundred.
6
Regarding the origin of this name cf. above, p. 91 n. 163 section (III).
7
Sra 55:1, from which the entire chapter derived its name, Srat al-Ramn. Sra
78:37sq. isas indicated above, p. 85most likely to be assigned to the second Meccan
period. Sra 1:2 does not belong here, even if the Ftia belongs to the first period since
al-Ramn is here not an independent proper name but an epithet of Allh.
8
On the whole some 50-odd quotations, most frequently in sra 19 (sixteen times).
9
Sras 13:29, and 41:1.
10
Sra 2:138 is according to what has been remarked below s.v. Meccan. Sra 59:22 is to
be interpreted like sra 1:2 (cf. above, foot-note 7).
Comments on Sras 54, 37, 71, 76, 44, 50, 20, 26, 15, 19, 38
An inaccurate interpretation of sra 54:1
11
led to an equally absurd tale
12
based on sra 94:1; many Muslims, however, correctly interpret the passage
as referring to the Final Judgement.
13
Gustav Weil is mistaken in thinking
that it comes fromanother sra.
14
Sra 21 alsodisplays a beginning tothe oth-
ers (cf. also 16:1). The first verse, which matches the others well, particularly
in its rare rhyme, is tied to the second verse, which, incidentally, does not
discuss ancient people, as Weil thinks, but rather the disbelieving contem-
poraries of the Prophet. We first encounter in this sra the history of several
former prophets side by side. Verse 45 is considered to be Medinan
15
some
also apply this to the verses 54sq.probably because it is generally
16
asso-
ciated with the Battle of Badr. Verses 47 to 49 are believed to refer to the
embassy of the Christians of Najrn to Muammad or even the sect of the
Qadarites.
17
Such untenable assumptions have led to the entire sra being
attributed to the Medinan period.
18
11
Regarding the artistic construction of sra fifty-four, and its double refrain (verse 16 =
18, 21, and 30; verse 17 = 22, 32, and 40); cf. above, p. 33, and David H. Mller, Die Propheten,
vol. 1, p. 53sq.
12
Neither in Ibn Hishm nor in Ibn Sad but in numerous other passages: al-Bukhr,
K. al-Tafsr, K. bad al-khalq, 98 (Bb sul al-mushrikn), 167 (Bb inshiqq al-qamar);
al-Tirmidh, K. al-Tafsr, abwb al-fitan, 13; al-Khab al-Tibrz, Mishct, Bb almt al-
nubuwwa at the beginning; al-abar in al-Tafsr; al-Wid, s.v.; al-Diyrbakr, Tarkh al-
khams, vol. 1, p. 298sq.; al-Qasalln, Mawhib laduniyya, vol. 1, pp. 465468, where, as it
is customary, also the dogmatic aspect is discussed. We here also learn that all philosophers
( ) beginning with Ab Isq (d. 188/802) denied a priori the possibility of such an
event. The real author of this tale seems to be IbnMasd; of the others inthis isnd Anas (Ibn
Mlik [d. 91/709, EI
2
; EQ; Juynboll, Encyclopedia, pp. 131134]) and udhayfa (Ibn al-Yamn
al-Abs), [EI
2
; EQ; Juynboll, Encyclopedia, p. 231, col 2; Goldziher, Schools of Koranic commen-
tators, p. 54] were Medinans, Ibn Abbs, at the time when the event had to have occurred,
not yet born, [Abd Allh] Ibn Umar [Ibn al-Khab (Juynboll, Encyclopedia, pp. 1011; Sez-
gin, GAS, vol. 8, p. 172, l. 4)], a little boy; also Ab Ad JUBAYR IBN MUIM [EQ; Juynboll,
Encyclopedia, p. 48sqq.; Sezgin, GAS, vol. 1, p. 258] must have been a youngster unless he had
lived almost to the age of eighty (d. 59/678). In general, we cannot accept this manwho is
also the authority of another tale (Sprenger, p. 138 [sic])as witness for Muammad since he
became Muslimonly in 8/629. Only Al (Ibn Ab lib)who incidentally, as far as I can see,
first appears in the Mawhib as an authority in this mattermight qualify as witness, but
likewise only as a youngster, as at his death in 40/660 he was likely no older than fifty-eight
years old.
13
Cf. also the beginnings of sras 81 and 82.
14
Historisch-kritische Einleitung in den Koran, 1st ed., p. 62, n. 2; 2nd ed., p. 71 n. 3.
15
16
Cf. the Commentators, already al-Wqid, p. 132.
17
al-Wid.
18
[AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF;] who could not refrain
from adding the critical to this statement.
In the rather long sra 37,
19
verses 1 to 70 contrast the disbelief of the [i/123]
Meccans with the certainty of resurrection and judgement. Verse 71 sq. leads
to the second
20
part (verses 73 to 148), which uses the history of seven Jewish
prophets to demonstrate that their contemporaries also remained largely
in disbelief. While verses 167 to 182 constitute a good conclusion for this
passage,
21
verses 149 to 166 on the polytheism of the Meccans
22
have a much
looser relationship with the whole. Nevertheless, the fact that this section
shares some phrases and two verses with the other two, as well a common
style, rhyme and rhythm, means that the unity of the whole cannot be
challenged.
Sra 71, in which Muammad makes the patriarch Noah warn against the [i/124]
idols of the Arabs, seems to be a fragment.
23
Sra 76 deals with the hereafter and the Final Judgement. Because of
a miserable tale depicting Fima, al-asan (Ibn Al b. Ab lib), and
al-usayn(IbnAl b. Ab lib),
24
some writers date the entire sra,
25
or parts
of it, namely verses 8 to 31
26
or verses 1 to 23
27
or verses 1 to 23 and 25 to 31,
28
to
19
Snouck Hurgronje, Het Mekkaansche feest, p. 31, suspects that sra 37 is later than sra
15 since the latter relates less of Ibrhm [Abraham] and his son. But the scanty content of
sra 16 need not necessarily lead to the conclusion that at that time Muammad did not
know more about these Biblical persons. In any case, the style of sra 37 leaves one with the
impression of being older than the style of sra 15.
20
Its unity is also assured by the great uniformity of the style which increases from the
repetition of phrases to entire verses, , verses 109, 120, and 130; verse 78 = verses 105,
110, 121, and 131; verse 79 = verses 111, 122, and 132. The homogeneity of the two parts, however,
is externally hardly at all indicated. This is the case only in verse 39 (72) = 128 (160, and 169).
21
Cf. verses 168 and 69. The verses 169 to 174, but particularly the verses 171 and 181, clearly
refer to earlier matter. Compare also the phrases verse 181 with verses 109, 120, and
130; verse 169 = verses 39, 72, and 128; , verses 174 and 178, and verse 148; ,
verses 171 and 181 with verses 36, 123, 133, and 139.
22
The verses 149sqq. are on the same level as the sras 53:19sqq., and 16:59. But the
verses 150 and 158 show that the reference is not only to the familiar triad of goddesses of
53:19sq. but also to other female spirits. Cf. also the remarks of R. Dussaud, Les Arabes en
Syrie avant l Islam, p. 121 sq., based on Hartwig Derenbourgs Le Culte et la desse al-Ouzza
en Arabie au IVe sicle de notre re, pp. 3340.
23
There can be no end of a verse after
(Flgel, verse 23) as the rhyming words of the

rest of the verses all have tass. The number of twenty-nine transmitted verses results from
the break inserted after
in verse 5.
24
al-Zamakhshar; al-Bayw on verse 12.
25
[AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF]; Hibat Allh (IBN SAL-
MAH), al-Itqn, p. 28, Al al-Dn (Al b. Muammad al-Baghdd al-Khzin).
26
[AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF].
27
Al al-Dn (Al b. Muammad al-Baghdd AL-KHZIN) at the beginning.
28
[AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF]; Al al-Dn (Al b. Mu-
ammad al-Baghdd AL-KHZIN).
the time after the hijra, whereas other writers consider verse 24 alone to be
of Medinan
29
origin. The same applies probably also to verse 23sqq.
30
According to some writers, verse 14 of sra 44 is Medinan because
was interpreted as referring to the long famine that God inflicted upon the
Meccans after Muammads emigration.
31
Verse 15, like so many others, was
interpreted as referring to the Battle of Badr.
32
In sra 50, verse 37 appears to be an objection to the Biblical view that
God rested after completing Creation. Since this was largely held to be a
polemic against the Jews, the verse was immediately considered to be of
Medinan origin.
33
Muir lists sra 20 in his final stage because of its length. The presentation
of the first fourteen, sixteen or seventeen verses is said to have prompted
Umar toaccept Islam. Althoughseveral early witnesses attest tothis
34
and
ina formthat, by and large, is not untrustworthywe are unable to produce
the evidence. Another tradition that connects Umars acceptance of Islam
with the early Meccan sra 69
35
is less well documented. The remaining
traditions cannot be considered, since they replace sra 20 with Medinan
passages, namely sra 61
36
or sra 57.
37
Against this account (Ibn Hishm,
29
al-Suy, al-Itqn, pp. 28 and 37.
30
31
32
Ibid. Cf. also Friedrich Rckert in the notes to his translation of the Koran. The break
in the verse after
(verse 36) in Flgels edition of the Koran is wrong.

33
[ABAL-QSIM] Umar b. Muammad [IBNABDAL-KF]; al-Itqn, p. 36; al-Wid;
Al al-Dn(Al b. Muammadal-Baghdd AL-KHZIN). According toal-Nasaf inal-Khzin
al-Baghdd (Tafsr), vol. 4, p. 188, also verse 38. For the opinions of Muslim theologians see
Goldziher, Die Sabbathinstitution.
34
Ibn Hishm, p. 226sq. (cf. the foot-note); Ibn Sad, [al-abaqt] ed. Sachau, Biographie
Muhammads bis zur Flucht (vol. 1, part 1), p. 192; al-abar, ed. Zotenberg, vol. 2, p 245;
A.I. Silvestre de Sacy, Mmoire sur l origine et les anciens monuments de la littrature parmi
les Arabes, p. 420]; cf. Weil, Das Leben Mohammeds, p. 60; Caussin de Perceval, Essai sur
l histoire des Arabes, vol. 1, p 396sqq.; Sprenger, Life, p. 187sq.; Sprenger, Das Leben, vol. 2,
p. 87sq.; Muir, Life of Mahomet, vol. 2, p. 168; Caetani, Annali, vol. 1, p. 284sqq.The notes
in Ibn Hishm, p. 76, al-Bayhaq, Masin, ed. Schwally, p. 71 sq., etc., also add sra 81:114 of
the first Meccan period. But I have not found anything about this either in the collections
of adths nor in the Commentators (al-Wid, Ab l-Layth al-Samarqand, Fakhr al-Dn
al-Rz, al-abar, Tafsr, al-Zamakhshar, al-Bayw).
35
Notes on Ibn Hishm, p. 76; Izz al-Dn IBN AL-ATHR, Usd al-ghba, vol. 4, p. 53; Ibn
ajar al-Asqaln, vol. 2, p. 1234; cf. G. Weil, Geschichte der Chalifen, vol. 1, p. 132, n. 2.
36
Izz al-Dn IBN AL-ATHR, Usd al-ghba f marifat al-aba, vol. 4, p. 54; G. Weil, Das
Leben Mohammeds, p. 69, note.
37
Hibat Allh (Ibn Salma) regarding this sra; al-Qasalln, al-Mawhib al-laduniyya,
vol. 1, p. 67; G. Weil, Das Leben Mohammeds, p. 69, note. It ought to be observed that sras 61
and 57 begin with the identical verse.
p. 227sq.), which differs markedly with all other reports, it must be observed
that the story does not fit in with Umars historically established character.
This conversion is said to have taken place at the end of the sixth year of
Muammads prophetic commission,
38
or when Umars sonwho in 2/623
was fourteenyears old, andin8/629twenty
39
hadnot yet reachedmaturity
( ),
40
or was only six years old,
41
i.e., in the year six before the hijra. When
we reckon the period of the Prophets activity before the emigration to have
been thirteen years,
42
the difference between the two statements becomes
negligible. For no reason at all some people consider verses 130sq. to be
Medinan.
43
Sra 26 has the first formal heading that confirms its character as a rev- [i/126]
elation.
44
Like so many verses that seem to refer to Jews, verse 197 is also
held by some to have originated from Medina.
45
Far more important is the
claimthat verse 214 alone, or verses 214sq., or verses 214 to 223, belong to the
very first part of the Koran, as they are considered to contain the first invi-
tation to the Prophet to preach Islam to his relatives.
46
This cannot be true,
38
Ibn Sad, [al-abaqt al-kabr,] ed. Sachau, 1 ed., vol. 3, part 1, p. 193, l 12, quoted by
al-Nawaw, Tahdhb al-asm, ed. Wstenfeld, p. 449; Sprenger, loc. cit.; Muir, loc. cit.; Caetani,
Annali dellislam, vol. 1, p. 285, n. 2.
39
Ibn Sad, [abaqt] ed. Sachau, vol. 4, part 1, p. 105, l 5, p. 126, l 25, s.v. .
40
Loc. cit., p. 105, l 5.
41
Ibn Sad, [al-abaqt] ed. Sachau: Biographien der mekkan., vol. 3, part 1, p. 193, l 14.
42
Ibn Hishm, who as usual also in this case does not supply a date, says at leastlike
al-abar, vol. 1, p. 1189, l 9sq.that Umars conversion occurred after the Abyssinian emi-
gration. In abars Persian edition, vol. 2, p. 403sqq., everything is mixed up: He has Umar
embrace Islamevenbefore year three (i.e., before the hijra), andconfuses the greater audacity
that the Muslims displayed upon Umars encouragement also in the exercise of the religious
service with the very first public appearance of the Prophet.
43
al-Itqn, p. 34; al-abar in the Tafsr; al-Wid; Al al-Dn (Al b. Muammad al-
Baghdd AL-KHZIN), etc. relate an event from the Medinan period in verse 131.
44
Cf. above, p. 75.
45
46
Ibn Hishm, p. 166; Ibn Sad, ed. Sachau, Biographie Muhammads bis zur Flucht (vol. 1,
part 1), pp. 42 and 133; al-abar, vol. 1, p. 1169; al-Bukhr, K. al-Way, 10, K. bad al-khalq,
83; Muslim, al-Qasalln, vol. 2, p. 181 sqq., K. al-mn, fourth last bb; al-Khab al-Tibrz,
Mishct, Bb al-Indhr wa-al-tandhr, fal 1 2, Bb al-Mabath, fal 1, 9; al-abar in the
Tafsr, vol. 19, 66 [sic], Ab l-Layth al-Samarqand; al-Zamakhshar; al-Bayw; al-Suy,
al-Itqn, p. 34, etc. The connection, not infrequent, of this informationwiththe one regarding
the occasion to sra 111 might be wanting in the original formof this tradition. That precisely
these verses were the first invitation to the conversion of the Prophets relatives is not
specifically mentioned by any of the sources.
Weil (Historisch-kritische Einleitung, 2nd ed., p. 65) thinks that the earliest part of the sra
consists of the verses 214 to 218. Hirschfeld (New researches, p. 143) separates the verses 221
to 228, and lists them in the chronological arrangement of the revelations on sixth place,
without supplying any reason on p. 63.
however. The more general and rather less forceful style of these verses is
in total agreement with the rest of the sra,
47
which makes it impossible to
give them the same date as sras 111 and 74. Naturally, we disassociate our-
selves fromthe totally untenable viewthat would assign only the creation of
verses 214sqq. to this time. Apart fromthis, verses 215 and 219 indicate quite
clearly the existence of a community, albeit small,
48
whereas such a group
did not exist at that time. Consequently, verse 214 can only be interpreted as
a renewed invitation to conversion. After all, we knowthat among Muam-
mads uncles Ab Lahab (IBN ABD AL-MUALIB) died a pagan in 2/623,
Abbs (IBN ABD AL-MUALIB) embraced Islam only after the Battle of
Badr, and Ab lib had to ward off his nephews proseletyzing efforts even
on his deathbed.
Since
(verse 227) cannot constitute the endof a verseverses 227 and [i/127]
228 of Flgels edition of the Koran must rather be brought togetherthe
verse becomes disproportionately long. Added to this formal reservation
is a conceptional one: the words from to weaken the main idea
excessively andif indeed the conclusion that starts with belongs
to the earliest partthe context is probably also interrupted.
49
As for the
particulars of the interpolation, we learn the following: the poets, assn b.
Thbit, AbdAllhb. Rawa al-Anr al-Khazraj,
50
andKabb. Mlik,
51
who
all put their talents to the service of Islam, one day came weeping to the
Prophet, complaining that Allh made such derogatory comments about
poets in verse 224, although He ought to know that they, too, were poets.
Although the details of this tradition are not the least bit reliable,
52
the
general tenor of the matter is probably correct.
Although our conjecture regarding the Medinan origin of the interpo-
lation is thus confirmed, there is no reason to believe that verses 224 to
47
The individual words, too, e.g., which in this sra are rather frequently con-
nected (in the monophonic refrain-like verses 8, 68, 104, 122, 140, 159, 175, 191, and 217) but
otherwise only three times, namely in the sras of the second and third period; and
(verse 220) which, as similar epithets of God, never occur in the first period.
48
Ibn Hishm, p. 166, admittedly speaks of many secret followers of Muammad around
that time who, reportedly, were won over before the time of his public sermonsprobably
by a kind of spell!
49
Cf. above, p. 80 and 85, the comments on sras 95:6, 103:3, and 84:25.
50
Juynboll, Encyclopedia, pp. 493 and 679; Sezgin, GAS, vol. 2, pp. 292293; vol. 9, p. 277.
His dwn has been edited in 1972 by H.M. al-Bjawda.
51
EI
2
; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 2, pp. 293294.
52
al-abar, Tafsr; al-Suy, Asbb al-nuzl. Slightly different, also regarding the names,
the traditions can be found in Ab l-Layth al-Samarqand, al-Zamakhshar, al-Bayw, [AB
AL-QSIM] Umar b. Muammad [IBN ABD AL-KF], and Fakhr al-Dn al-Rz.
226 must also be assigned to this late date,
53
even if, against all appear-
ances,
54
they did not belong to the preceding verses. It makes no difference
whether these verses refer to the polemics Quraysh poets might have waged
against the Prophet at an early date, or to the guild of the shuar in general.
The latter case is the more likely, as the other Koranic passages that men-
tion the shir have the entire class in mind when they passionately deny
that Muammad had any connection to them. Here (sras 21:5, 37:35, 52:30,
and 69:41) the shir is put on the same level as the khin, or soothsayer, and
is considered a man whose hotchpotch of nightmares (21:5) are ominous
andreveal Fates uncertainty (52:30). The jinnor demoninthe shir (37:35)
is not there to whisper nice words or ideas into his ears but rather to inspire
him when the clan turns to him for spiritual assistance.
55
The word poet,
which usually serves as translation of shir, is in this case, of course, not
quite accurate. In any case, the Meccan origin of verses 224 to 226 is assured
by the fact that no Medinan passage of the Koran refers to shir.
Additionally, even if it were certain that the dubious verses belonged to [i/128]
the same period as the rest of the sra,
56
it would still seem to me that the
literary unity has not been established. As D.H. Mller
57
in particular has
shown, verses 1 to 191 were composed in accord with an artificial scheme.
The introduction (verses 1 to 6), as well as the seven following sections,
which deal with the ancient prophets and the judgements on their godless
countrymen, have the same refrain.
58
Apart from the name, the first verses
of the last five sections have the same text.
59
This stylistic device ceases to
53
Hibat Allh; Fakhr al-Dnal-Rz; al-Zamakhshar, al-Bayw; [ABAL-QSIM] Umar
b. Muammad [IBN ABD AL-KF]; Al al-Dn (Al b. Muammad al-Baghdd AL-KH-
ZIN). After Abl-Laythal-Samarqand, al-Kalb holds some verses (
) at the endof this sra

tobe Medinan. Muqtil [IbnSulaymnd. 150/767; EI
2
; EQ; Juynboll, Encyclopedia, pp. 431432;
Goldziher, Schools, 38sqq.; Sezgin, GAS, vol. 1, 3637] on the other hand, considers the entire
sra to be Meccan.
54
The menof verses 210 and 221 inspired by the devil are likely to include also the shuar.
55
The profane literature of the Arabs contains numerous examples of the enormous
influence of the shir upon the actions of the clans. For the collection and elaboration of
this literature we are indebted to Goldzihers Abhandlungen zur arabischen Philologie, 1. Teil,
pp. 125. Cf. also Schwally, Die Vision in his Semitische Kriegsaltertmer, pp. 1820.
56
So also Muir, Life of Mahomet, vol. 2, p. 113, and Caetani, Annali, vol. 1, p. 239.
57
Die Propheten in ihrer ursprnglichen Form, vol. 1, pp. 3442.
58
[Surely in that is a sign, yet most of them are not believers.] Verses 78 = 6768,
103104, 121122, 139b-140, 158b-159, 174175, and 190191.
59
[ cried lies to the Envoys.] Verses 105109, 123127, 141145, 160164, and 176180. In
addition verse 108 [so fear you God, and obey you me] (= 126, 144, 163, and 179) is found still
three times more, verses 110, 131, and 150. Th. Nldeke does not recognize in these repetitions
a conscious artful literary device but simply a nave technique.
be used in the last part of the sra (verses 192 to 228), which would suggest
that another revelation was hastily attached to verse 191 by means of three,
introductory little words (wa-inna-hu la).
60
In sra 15 as well a few very old verses are thought to exist,
61
namely [i/129]
verses 89 and 94, which, like 26:214sqq., were allegedly the first to con-
tain invitations to propaganda. This view is inferred solely from the words
and , as if Muammad could not also have propagated the faith
later without fear. Moreover, these and the closely related verses deal with
the opponents whose mockery and persecution he had to suffer for a long
time. Finally, even the style and individual expressions do not correspond
to the earliest period.
62
Ibn Hishm,
63
incidentally, mentions a different and
better occasionina later periodfor verses 94 to96. For trivial reasons certain
exegetes consider verses 24 and 87 to be Medinan.
64
The first part of sra 19 the Muslims are said to have recited to the Chris- [i/130]
tian Najshi (Negs) of Abyssinia in the presence of the Quraysh envoys.
65
Verses 59
66
and 74
67
are unreasonably considered to be of Medinan origin.
Verses 35 to 41 Muammad can only have added later, at approximately the
beginning of the third or the end of the second period, i.e., as a dogmatic
and polemic supplement to the verses on Jesus, which differ in language
as well as in rhyme.
68
The rhyme, too, changes in verses 76sqq. In view of
the identical rhyme in verses 1 to 34, and 42 to 75, this fact alone suffices to
make us suspect a later addition, even if the loose structure of the homily
does not make the content of the text after verse 75 appear unreasonable,
60
In this case, the original revelation would have started with the word tanzl, a much
favoured beginning of sras, cf. sras 32, 39, 40, 41, 45, and 46, all of which belong to the late
Meccan period.
61
Ibn Hishm, p. 166; al-abar, vol. 1, p. 1169; cf. G. Weil, Das Leben Mohammeds, p. 51,
Historisch-kritische Einleitung in den Koran, p. 56, 2nd ed., p. 65sq.; Spr., 177 [sic, not listed
among the abbreviations: Sprenger, but which work?]
62
Cf. e.g. the idioms, ,
, etc. which are never used in the first period.

63
Ibn Hishm, p. 272; al-abar, Tafsr, vol. 14, p. 74.
64
al-Suy, al-Itqn, p. 33; al-Wid; al-abar, Tafsr, only verse 24 is related to events
after the hijra.
65
Ibn Hishm, p. 220, etc.; cf. Sprenger, Leben und Lehre, vol. 2, p. 182; Ab l-Layth
al-Samarqand; Caetani, Annali, vol. 1, p. 277.
66
al-Suy, al-Itqn, p. 33. [This and the following foot-note appear in the same line in
reverse order].
67
al-Bayw; al-Suy, al-Itqn, p. 59.
68
The rhyming words of the oldest part of this sra (verses 1 to 34, and 4275) end with
(once, verse 75
),
verses 10, 43, 61, and 68 or verses 15, 32, 60, and 67. The other
rhymes of verses 3 and26 of the Flgel editionmust be basedoninaccurate division. Verses 35
to 41 end with n, verses 76 to 98 (end) with d, b, and z.
content which is, incidentally, of the same period. Whether the reason for
the combinationsimilar to that of verse 75is to be found in the ending
of verse 98, or whether this verse was composed with that other one in
mind, remains to be seen. The sra
69
is the earliestor at least one of the
earliestto refer to the New Testament and saintly
70
persons such Mary,
Zechariah, John, and Jesus.
Comments on Sras 38, 36, 43, 72, 67, 23, 21, 25, 17, 27, 18
The first ten verses of sra 38, or verse 5 alone, are supposed to date from [i/131]
the time when the Quraysh tried to persuade Ab lib (IBN ABD AL-
MUALIB) not to protect Muammad any longer, or when the former was
on his death-bed.
71
But these are mere conclusions from the simple words,
. Verse 28 poses difficulties in context, namely, does it refer to David
and the revelation of the Psalms that is often mentioned in the Koran, or
to Muammad, which, according to passages like sras 6:92 and 156, 21:51,
7:1, 11:1, and 14:1, seems more likely. In this case the interpolation of the verse
would make even less sense.
Verses 67 to 87 (end) have a common rhyme with m, n, n, and m,
72
whereas in the rest of the sra the rhyme is exclusively with b, r, d,
etc.
73
Consequently, the assumption that the two parts originally did not
belong togethera view that comes easily to mindis not in contrast to
the content. Al-Suy says that one exegete holds this sra to be Medinan
in contradiction of the general view.
74
The same opinion also prevails in regard to sra 36.
75
Other exegetes
maintain this pertains only to both verse 11applying it to the Ban Sal-
ima, who intended to settle not far from the mosque of Medina,
76
and
verse 47,
77
because the prescribed charity is taken, as often in other cases, to
69
This sra, as well as all the following of the second period, with the exception of sra
67, Muir puts into his fifth stage.
70
From here on, they are repeatedly mentioned not only in the Meccan period (sras 21,
23, 43, 42, and 6) but also in the Medinan period (sras 2, 3, 4, 5, 9, 33, 57, 62, and 66).
71
al-Wid; al-Bayw.
72
The verses 75 and 76 constitute one verse in the Flgel edition of the Koran.
73
The verses 43 and 44 of the Flgel edition ought to be one single verse.
74
75
76
al-Itqn, p. 35 after al-Tirmidh, s.v.; al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Wid.
77
al-Itqn, p. 35; [AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF].
refer tothe community tax ( , ) institutedonly after the hijra. Between
verses 24 and 25 a few words may have been lost that mention the murder
of the only believer by the godless people.
Sra 43:44 is said to originate from Jerusalem
78
or heaven
79
and was ad-
dressed to the prophets who were assembled there in the so-called Night
Journey. The origin of this strange account is not difficult to find, and Weil
supplies the proper explanation.
80
The assertion that the verse originates
fromMedina
81
is perhaps based on an inaccurate interpretation of the afore-
mentioned account; the verses sound unlike a Meccan sra, so a Medinan
origin was simply deduced from this. If the consonantal text of 43:88 is not
damaged, a few words must be missing at its beginning, since ,
82
even
with changed diacritics, can hardly be connected satisfactorily with the pre-
ceding verse. Hirschfeld,
83
without supplying any sound reason, attributes
verses 1 to 24 and 25 to 89 to different periods.
Sra 72
84
is held to refer to the vision when Muammad learned of the [i/132]
jinn listening to his recitation of the Koran. According to the traditional
account, this happened when he was on his way home from al-if, where
the Prophet had gone after the death of Ab lib, and reached Nakhla.
85
Other writers agree on the place of this event but attribute it to a different
time, namely during the journey to the fair of Uk.
86
A third tradition
puts the event in the immediate vicinity of Medina.
87
Although we cannot
78
al-Itqn, p. 43; al-abar, Tafsr; al-Zamakhshar; Fakhr al-Dn al-Rz; Al al-Dn (Al b.
Muammad AL-KHZIN al-Baghdd).
79
80
Das Leben Mohammeds, p. 374.
81
82
The easiest explanation would opt for the nominative which, however, is not recog-
nized by the canonical readers and accepted only as isolated (
) reading. Even in this case,

there still remains the great problem of the change in person.
83
Newresearches into the composition and exegesis of the Qoran, p. 144.
84
The verses 22 and 23 constitute only a single verse in Flgels edition of the Koran.
Verse 26 (Flgel), however, can be divided into two parts.
85
Ibn Hishm, p. 281; Ibn Sad, [al-abaqt] ed. Sachau, vol. 1, part 1, p. 141 sq.; al-abar,
vol. 1, p. 1202sq.who evensupplies the names of the sevenjinnsand inthe Tafsr. Cf. Weil,
Leben Mohammeds, p. 69; Sprenger, Life, p. 187sq., Leben und die Lehre, vol. 2, p. 246sq.; Muir,
Life of Mahomet, vol. 2, p. 204; Snouck Hurgronje, inde Gids, 1886, part 6, 267; August Mller in
Fr. Rckerts translation of the Koran, p. 525; Fr. Buhl, Muhammeds Liv, p. 187; Caetani, Annali,
vol. 1, p. 311.
86
al-Bukhr, K. al-Adhn 103, Tafsr, s.v.; Muslimal-Qasalln, vol. 3, p. 88sqq., K.
al-alt 33 (Bb al-jahr bi-l-qira f l-ub); al-Tirmidh, Tafsr; al-abar, Tafsr; Fakhr al-Dn
al-Rz.
87
al-Tirmidh, Tafsr; al-abar, Tafsr; Muslim, al-Qasalln, vol. 3, p. 91 sqq. This is a
ascertain the historicity of any of these accounts, we do have it from other
sources that Muammad believed quite seriously that he had a mission to
the jinn. One day, on his way to Tabk (9/630), a large, strong, male snake
approached him and remained in front of him for a long time as he halted
on his camel, then turned away from the path and raised up. Do you know
who this is, Muammad asked. It is one of the jinn who wanted to hear
the revelation.
88
In some Koranic passages the sermon is directed to the
assembly of jinn (6:128, and 130, 55:33).
Sra 67, according to one tradition, is held to be Medinan,
89
probably only [i/133]
because its length is similar to Medinan sras 57 to 66, which it follows in
our Koran.
Verse 78 of sra 23, which is considered to be Medinan on the basis of a
mistaken identification with the Battle of Badr,
90
some writers maintain for
unknown reasons to be the very last Meccan sra.
91
In sra 21, some people consider verse 7 to be of Medinan origin.
92
Sra 25:47, according to one traditionwhich al-Suy distrustswas
promulgated at al-if.
93
If this were the case, the related verses ought to
be of identical origin, although there is nothing to support this.
94
Verse 68,
andpossibly alsothe following verses,
95
are attributedtoMedina, since there
appears in these as well as in other verses of similar context (atonement and
forgiveness) anallusiontoWash [Ibnarbal-abash
96
], whointhe Battle
of Uud killed amza (Ibn Abd al-Mualib), Muammads uncle, but later
became a Muslim.
97
According to others, these verses, although composed
local tradition at Mecca (al-Azraq, p. 424), where until this day this particular place (Masjid
al-Jinn) is showntopilgrims (RichardBurton, Personal narrative of aPilgrimage toal-Madinah
&Meccah, vol. 3, p. 353), and for this reason the least reliable.
88
al-Waqid (i.e. Wellhausen, Muhammed in Medina), p. 400; Wellhausen, Reste arabis-
chen Heidentums, 2nd ed., p. 153.
89
90
al-Itqn, p. 34; cf. the commentaries which apply the verses 66 and 79 to things that
happened after the hijra.
91
al-Wid in the introduction, Cairo edition, p. 8; al-Itqn, p. 55 (end).
92
93
al-Itqn, p. 43; al-Diyrbakr, Tarkh, vol. 1, p. 12; from it G. Weil, Das Leben, p. 374.
94
Weil is wrong if he thinks that the words here used by Muammad are allegorical.
95
Incidentally, these verses are attributedto the manskht whichwere abrogated by 4:35
(al-abar, Tafsr, vol. 19, 25sq.; al-Wid).
96
EI
2
; EQ; Juynboll, Encyclopedia, p. 73, col. 1.
97
Ab l-Layth al-Samarqand; [AB AL-QSIM] Umar b. Muammad [IBN ABD AL-
KF], al-Itqn, p. 34; EI
2
; Juynboll, Encyclopedia, p. 73, col. 1, sqq.
at Mecca, were later dispatched fromMedina to Wash at Mecca.
98
ak
[Ibn Muzim al-Hill, d. 105/723
99
] considers the entire sra to be Medi-
nan.
100
Since verses 25:64 to 25:77 (enumerationof the marks of the true believer) [i/134]
lack topical relation to the preceding text, and also have a somewhat differ-
ent rhyme,
101
the question might be allowed as to whether they are now in
their original place. As pointed out above, on p. 12, verses 5 and 6 are of fun-
damental importance for the early history of Islam, because we learn there
of Muammads zeal for procuring copies of ancient holy texts.
Sra 17:1 refers to Muammads Night Journey fromMecca to Jerusalem.
102
Tradition considers this journey a miracle, although this is in conflict with
several passages of the Koran (e.g. 13:8 and 27, 17:95, 25:8sqq., and 29:44)
in which the Prophet explicitly rejects miracles, wanting to do nothing
other than warn and preach. We therefore have to assume that Muammad
wanted to communicate a dream.
103
However, this assumption can only be
reconciled with the first verse if the dream was a real experience for him
and not an illusion.
104
His extreme fantasy here resembles the nave belief of
primitive races that visits toor fromstrange persons ina dreamcanbe a real-
ity. That verse 62
105
relates tothis dreamis possible but cannot be proven, not
98
Ab l-Layth al-Samarqand.
99
EI
2
100
101
The verses from 1 to 63 throughout rhyme with l or r because Flgels verses 3 and 4
as well as 20 and 21 must be brought together, and in verse 18 is perhaps to be replaced
by
. On the other hand, the verses 64sqq. usually rhyme with m, and only verses 71 with
b, and 73 with n.
102
Cf. the Commentators.
103
This is also Muslimopinion as can be seen fromthe introductory words to the relevant
traditions: as well as , cf. al-Bayw, IbnHishm,
pp. 263266, al-abar, Tafsr on verse 1. In other traditions (Ibn Sad, loc. cit.; al-Yaqb,
Historiae, ed. Houtsma, vol. 2, p. 25; Muslim, al-Qasalln, vol. 2, p. 63, K. al-mn, 72) this
is not quite as apparent.This dreamof Muammad is possibly somehowinfluenced by the
familiar vision of the Hebrew Prophet Ezekiel when the spirit took him by the lock of his
head and lifted himup between the earth and the heaven and brought him[fromBabylonia]
to Jerusalem (Ezekiel 8:3).
104
It is out of the questionthat it is anunblushing forgery (Sprenger, Life, p. 124; Sprenger,
Leben, vol. 1, p. 306, vol. 2, p. 528).
105
Nearly all Commentators relate the verse with the Night Journey (isr) in addition
to Ibn Hishm, p. 265; Ibn Sad, [al-abaqt] ed. Sachau, vol. 1, part 1, p. 144; al-Bukhr,
K. al-Qadar, 10. There are only a few who relate it with the dream of the conquest of
Mecca (Ab l-Layth al-Samarqand, Fakhr al-Dn al-Rz, al-Zamakhshar, Al al-Dn (Al
b. Muammad al-Baghdd AL-KHZIN), Cairo ed., vol. 3, p. 177; al-Qasalln, Mawhib,
Maqad, 5, at the beginning), for which reason it is held to be Medinan in al-Itqn, p. 33.
even if it was originally part of the same homily as the first verse. The con-
text would instead suggest a vision with eschatological information. Ruy,
incidentally, refers not only to a dream but also to a daytime vision.
106
On
no account can verse 95 be included in this context, as it speaks only hypo-
thetically of an ascension to heaven. Even if, as some think,
107
this verse was
an allusion to Muammads ascension, verse 1 deals exclusively with the
Night Journey to Jerusalem. Although both events are usually related to one
another,
108
the ascent is accorded such importance and independence that
its absence inthe first verse is not at all self-evident. However, since Muam-
mads ascension is not mentioned anywhere in the Koran, this tale can have
originated only after the death of Muammad, probably influenced by the
heavenly journeys of Ecstatics
109
in early Christian literature.
That the first verse of this sra cannot be connected with the following [i/136]
one is so obvious that it needs no proof. In a sra with a totally uniform
rhyme of ,
110
the isolated rhyme of r in the first verse is already suspicious.
An explanation of the actual state of the text, however, cannot be offered
with any degree of certainty. It is possible that some verses are missing after
106
Cf. e.g. al-Bayw; al-Bukhr, K. al-iyl 27.
107
G. Sale in his translation of the Koran. On the other hand, this relation is, as far as I
know, nowhere maintained in Islamic tradition.
108
Even in the earliest traditions, largely going back to Anas b. Mlik (d. 93/711) [EQ; EI
2
,
d. 91/709], and are traced back from him to Ab Dharr [al-Ghifr, d. 652/1254; EI
2
], and oth-
ers: Ibn Hishm, p. 268; al-Bukhr, al-Tirmidh, and al-abar in the Tafsr on verses 1 and 62;
al-Bukhr, K. bad al-khalq, 174, bb al-mirj; Muslim, K. al-mn, 72; Ibn Hishm, p. 268;
al-Yaqb, Historiae, vol. 2, p. 28; al-Khab al-Tibrz, Mishct, p. 518 (526), bb al-mirj; al-
Nas, K. al-alt, beginning; Ibn Sad, [al-abaqt] ed. Sachau, Biographie Muhammads bis
zur Flucht (vol. 1, part 1), p. 162sq., relates first the ascent, and then p. 143sqq., the air journey,
without connecting them. Al-abar in the Annales, vol. 1, p. 1157, does not even mention the
air journey, and places the ascent at the beginning of Muammads prophethood, something
that also happened to the air journey in a tradition in Muslim, al-Qasalln, vol. 2, p. 63. Al-
Bukhr, in places other than his Tafsr, accords detailed treatment almost only to the ascent
(K. al-alt at the beginning, and K. Bad al-khalq, 5). It can be observed that interest is
increasingly focusedonthe latter subject. Cf. inadditionSprenger, Life, pp. 126136, his Leben,
vol. 2, p. 527sqq., vol. 3, p. lvi; Wm. Muir, vol. 2, pp. 219222; Caetani, Annali, vol. 1, p. 229sqq.
A precise and critical study of the stories of the ascent, from the earliest traditions down to
the embellishments of the Persian and Turkish poets would be very instructive.
109
II Corinthians 12: 1 sqq.; Ascencio Isaiae; Apocalypses of Baruch, Sophonias and Abra-
ham; Talmud, aggah, fol. 14b, 18, regarding Rabbi Ab; Teshvt ha-genm (Rabbi
Ismael). Cf. D.W. Bousset, Die Himmelsreise der Seele, pp. 136sqq. and 229sqq., and A. Die-
terich, Eine Mithrasliturgie, p. 180.
110
In the Flgel edition the verses 9 and 10 as well as 26 and 27 must be united to one verse
each since rhymes with r and n are impossible in the sra. The same applies to verses 48
and 49 since in all the other verses the rhyming word in the penultimate has a long vowel.
the first verse, which itself segued naturally into the second verse, or that
the first verse is totally out of context and was intentionally placed here
because people applied it to verse 62. In this case, the original introduction
to the second and following verses must have been lost. Completely unlikely
is Weils
111
assumptionthat 17:1 was fabricatedafter the deathof Muammad
or erroneously included in the Koran. Al-Bayw considers verse twelve
to be Medinan but this is wrong, as his source, al-Zamakhshar, merely
mentions this verse in an account of an event after the hijra without ever
once saying that it originated from that date. Verses 23 to 41, which briefly
summarize the duties of a Muslim, and verses 34sqq. are considered by
asan al-Bar
112
to be Medinan. Two other traditions include verses 28 and
31 in this provenance as well.
113
Weil agrees at least as far as verse 35 of this sra is concerned.
114
However, [i/137]
we would expect first of all in such an enumeration of the duties of Muslims
the interdiction of murder. It is not at all necessary, as Weil thinks, that the
words we have appointed to his next-of-kin (i.e. his avenger) authority is a
reference to the Medinan passage, sra 2:173sqq., since Muammad had no
executive power until Medina. As with all ancient peoples, the blood feud
was so deeply rooted among the Arabs, and so sacred, that Muammad con-
sidered it a godlike law. That he mentions it here, when he is merely estab-
lishing moral principles, is no more astonishing than his recognizing it as a
lawinsra 2. Similar reasons advancedby Weil might alsoserve toprove that
verse 36, among many others, could not have originated in Mecca.
115
There
are greatly varying accounts regarding verses 75 to 82. Many hold themto be
Medinan,
116
considering verse 75 to be a reference to the Ban Thaqfwho,
in 9/630, were prepared to accept Islam only under conditions contrary to
111
Historisch-kritische Einleitung in den Koran, 2nd ed., p. 74.
112
113
al-Suy, Asbb al-nuzl in the margin of Jallayn (Cairo, 1301/1883).
114
Weil, Das Leben Mohammeds, p. 377; Historisch-kritische Einleitung, 1st ed., p. 64, 2nd
ed. p. 74.
115
Qatda [Ibn Dima] [EI
2
; EQ; Juynboll, Encyclopedia, pp. 438449; Sezgin, GAS, vol. 1,
pp. 3132] is said to have declared verse 45 to be Medinan, as did Muqtil [Ibn Sulaymn]. Cf.
Al al-Dn (AL-KHZIN AL-BAGHDD), Tafsr, introduction to sra 17.
116
Ab l-Layth al-Samarqand; al-Bayw; [AB AL-QSIM] Umar b. Muammad [IBN
ABD AL-KF]. According to al-Suy, al-Itqn, p. 83, the verses 78 to 80 are allegedly
Medinan, according to al-abar Tafsr, the verses 75, 78, and 82, according to al-Wid,
verses 75 and 78, after al-Nsbr (in the margin of al-abar, Tafsr), verse 75, after al-Farr
[d. 200/822; EI
2
; Sezgin, GAS, vol. 9, pp. 131134], verses 75 and 78, according to Qatda in
al-Khzin al-Baghdd, loc. cit., verses 75 to 77, after Muqtil [Ibn Sulaymn], ibid., verses 76,
77, and 82, after al-Suy, Asbb al-nuzl, verses 75 to 78, and 83.
Muammads interest, and had nearly obtained his consent
117
and verse 78
to the Jews of Yathrib,
118
i.e., the story that the Prophet one day was prompted
by a stratagem of the Jews to go to Palestine, only soon to return.
119
Others
see verse 82 as referring to the conquest of Mecca,
120
or originating between
Mecca and Medina, considering it a reference to the entry into the cave
[of Thawr] (cf. 9:40)
121
or the marching into Medina.
122
Still others appro-
priately find in verses 75
123
and 78
124
merely a reference to the Quraysh, and
in verse 32on whose inaccurate and literal interpretation all those fan-
tasies are basedsimply a general meaning.
125
Weil will not even admit that
verse 78 refers to the Quraysh.
126
But it is not improbable that there had
been an earlier attempt to oust Muammad from Mecca, without consid-
ering that his followers were to accompany himally themselves to a strange
clan, and eventually make war against his native town. The verse cannot
refer to the Jews as even their initial attempts to use force against Muam-
mad ended with their expulsion. Also the language of the verse conforms to
that of the rest of the text.
127
Incidentally, some writers find inverse 75, or inverses 75 to 77, anallusion
to the words inserted in sra 53:14: these are the sublime gharnq.
128
But
it is quite obvious that these verses must be of a much later date. (AB
117
al-abar, Tafsr, vol. 15, p. 83; Ab l-Layth al-Samarqand; Fakhr al-Dn al-Rz; al-
Wid; al-Zamakhshar; al-Bayw; [ABAL-QSIM] Umar b. Muammad [IBNABDAL-
KF]; al-Khzin al-Baghdd; al-Nsbr; al-Suy, Asbb al-nuzl.
118
The same.
119
Ab l-Layth al-Samarqand; Fakhr al-Dn al-Rz; al-Wid; al-Zamakhshar; al-Ns-
br in al-abar, Tafsr, in the margin of vol. 15, p. 72; al-Suy, Asbb al-nuzl; al-Khzin
al-Baghdd.
120
al-Zamakhshar; al-Bayw; Fakhr al-Dn al-Rz; al-Khzin al-Baghdd.
121
al-Tirmidh, Tafsr; al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Khzin al-Baghdd; al-
Zamakhshar; al-Bayw.
122
Ab l-Layth al-Samarqand; al-Wid; [AB AL-QSIM] Umar b. Muammad [IBN
ABD AL-KF]; al-Bayw; al-Nsbr, loc. cit.; al-Suy, Asbb al-nuzl.
123
al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar; al-Bayw;
al-Suy, Asbb al-nuzl; al-Khzin al-Baghdd.
124
al-abar, Tafsr; al-Wid; al-Zamakhshar; al-Bayw; August Mller in Friedrich
Rckerts translation of the Koran, note, s.v., p. 488.
125
al-abar, Tafsr; al-Wid; al-Zamakhshar; al-Bayw; al-Nsbr; Al al-Dn (AL-
KHZIN al-Baghdd).
126
Historisch-kritische Einleitung in den Koran, 1st ed., p. 64sq, 2nd ed., p. 74.
127
Cf.
verses 78 and 75;
verses 78, 77, 75, 44, 102;
verses 78 and 66, but

nowhere else in the Koran.
128
Ab l-Layth al-Samarqand; Ibn Sad, [al-abaqt]: Biographie Muhammad bis zur
Flucht, p. 137; al-abar, Annales, vol. 1, p. 1195, but not in the Tafsr; al-Suy, Asbb al-nuzl;
Aug. Mller in Fr. Rckerts translation of the Koran, p. 488; cf. also above, pp. 7071.
AL-QSIM) Umar b. Muammad IBN ABD AL-KF alleges that verse 80,
like 28 (because of alms; see above), and verse 59, are considered to be
Medinan by asan al-Bar. Verse 87 allegedly contains the answer to one
of the three questions which the Jews, or Quraysh, once put to the Prophet
at the instigation of the Jews. This event, and consequently also the verse, is
sometimes placed at Mecca and sometimes at Medina, other details varying
considerably. All of this is so fanciful that we should think little of it. There is
even less reason to conclude that sra 18, which is supposed to contain the
answer to the two questions, shares the same dating as this passage, even if
it is quite possible.
129
Thus, not even a single verse has been proven to be of Medinan origin. [i/139]
Although there is uniform rhyme from the second verse on,
130
the internal
coherence of the various parts is so weak, and the external correspondence
so inadequate,
131
that scepticism about the unity of the sra is not unwar-
ranted. However, given the absence of sound standards, it would hardly be
possible to arrive at a sound conclusion. The great uniformity of language
and style alone argues against Hirschfeld,
132
who would like to attribute
verses 1 to 8, 103 to 111, 87 to 102, and 9 to 86 to three different periods.
133
Some words in sra 27 must be missing, as in verse 42 the words that [i/140]
follow canrefer only toSolomonor his retinue; a transitionindicating this
is certainly necessary. Before verse 93 should to be supplied according to
context, or simply be considered missing.
In sra 18, some verses are occasionally considered Medinan: Verse 27
either in toto or up to ,
134
verses 1 to 7, and 107sqq.,
135
as well as verse 82
136
for the same reason as sra 17:87. I would not venture to maintain with cer-
tainty that the two strange partstelling us how Moses recognized Divine
129
Cf. in this matter Ibn Hishm, p. 192sq.; al-Bukhr, K. al-Ilm, 48; al-Wid; al-Suy,
Asbb al-nuzl; and the Commentators in general.
130
Cf. above, p. 111.
131
E.g., the connectionof verses 22 and23, of 41band41 is difficult. Then, too, verse 41bsqq.
does not seemto presuppose verse 23. The great importance attached to the code of conduct
in verses 23 to 40 would suggest that it once formed the core of a revelation.
132
Newresearches into the composition and exegesis of the Qoran, pp. 70 and 144.
133
Apart from the idioms listed above (pp. 99100) I draw attention to verses 8, 53,
and 81; verses (1), 95, 108, 45 (46);
verses 19 and 41;
verses 9, 43, 47, 49, 62, 84,

90, and 91; verses 44, 53, 58, 86, 87, 90, 95, 98, 108, 110, and 111; verse 100 b = verse 52.
134
[AB AL-QSIM] Umar b. Muammad [IBN ABD AL-KF]; al-Bayw; al-Suy,
al-Itqn, p. 33.
135
136
Ibid.
Providence and his own weakness (verses 59 to 81), and howDh l-Qarnayn,
i.e., Alexander the Great,
137
crossed the world and set up a barrier against
Gog and Magog (verses 82 to 98)originate from the same time as the
preceding verses. As suggested above in the comments on sra 17:97, there is
little tobe saidfor their homogeneity except that their beginning, the legend
137
Muslims produced much fantastical material about the name Dh l-Qarnayn. Here
I draw attention only to the most important sources: al-abar, Tafsr, vol. 16, p. 6sqq., as
well as the other Commentators; Ibn Qutayba, al-Marif, ed Wstenfeld, p. 26; al-Masd,
Prairies, vol. 2, p. 248sq.; Ahlwardt, Verzeichnis, no. 10163 (fol. 1104), no. 7019 (fol. 105144)
= I Petermann, [Ms without title by Abd al-Ramn b. Al IBN AL-JAWZ]; Ibn al-Khab
al-Dahsha, Tufa, ed. Mann, p. 52; al-Suy, al-Itqn, p. 807.
This appellation is likely to refer to Alexander the Great, as also Muslims assume. As
far as we know, the first mention of the horns of Alexander (The History of Alexander the
Great, ed. E.W. Budge, pp. 257 and 274) I know that you had me grow horns in order to
destroy the empires of this world, whichaccording to Nldekes Beitrge zur Geschichte
(cf. S. Fraenkels review of Nldekes Beitrge zur Geschichte des Alexanderromans)dates
from ad514 or 515 (826 Seleuc.), but not from 626 as Carl Hunnius has it. The expression
the horned man, of course, does not occur in the Syriac legend. But if it is really the first
source of Muammad or his authority, the Arabic designation must be a creation based on
the transmitted characteristics. After all, many an Arabic epithet is made up of dh with
following dual. (Cf. al-Mubarrad, al-Kmil, ed. W. Wright, p. 777sq., and the comprehensive
compilation in Goldzihers Ueber Dualtitel, p. 321 sq.) Dh l-Qarnayn, in particular, is also
found as epithet of the Lakhmid King Mundhir III (cf. Nldeke, Geschichte der Perser und
Araber, p. 169 n. 3, and Goldziher, Abhandlungen, vol. 2, p. 26, note 13, regarding no. 28. [See
further EI
2
.]) The name horned man might have conceivably been used in an unknown
recension of the legend, and fromthere found its way into Arabic. Syriac bel might possibly
correspond to arn or arnm, since this dialect no longer has a dual.
An equivalent to the Arabic idiom has been known from Hebrew literature, namely the
apocalyptic Daniel 8:21, where the Persian Empire appears as a ramhaving two horns (
), while Alexander the Great is presented as a goat with the great horn between
his eyes, and Midrash Rabb to Genesis, Par. 99:2, here Edom (i.e. Rome) is called ,
cf. also , Ibn Hishm, p. 187, note. Fromthese sources it seems that the horns of
Alexander are derived fromthe apocalyptic literature, being the symbol of invincible power.
The problem, however, becomes complicated as according to the historians (Curtius Rufus
4, 29, 5sqq., and Flavius Arrian 3, 4 [sic, the authors inconsistent page references]) the oracle
of Jupiter Ammon recognized Alexander the Great as son of this god, and that the horn
originates from this very deity, which is imagined to be a ram, and which ornaments the
head of Alexander on the Ptolomaeic and Lysimachic coins. (Cf. J. Bernouilli, Die erhaltenen
Darstellungen, plate viii, figure 4; Theod. Schreiber, Studie ber das Bildnis Alexanders des
Groen, plate xiii, figure 5.) Since all these are graphic reproductions in profile, only one
horn is visible. The often cited statement of Athenaeus, of Naucratis in Egypt, after Ephippus,
of Athens (vol. 12, p. 537) that Alexander at times dressed himself as God Ammon, can be
disregarded since it is possibly nothing but learned conclusion from the numismatic find.
The fish of the Moses legend in verse 18:60 corresponds exactly to the fish which, accord-
ing to the Alexander legend, becomes alive again in the well-head of life (al-abar, vol. 1,
p. 428; Th. Nldeke, loc. cit, p. 25). If this is a confusion of images, this might have happened
since Moses shining face appears in the Biblical story (Exodus 34:29, 30, and 35) in Aquila,
of the Seven Sleepers (verse 8sqq.),
138
as well as their end (Dh l-Qarnayn),
are thought to refer to the three questions of the Jews. Nowhere else in
the three sections mentioned is there any reference to the others. It is far
more significant that the legends touched upon here belong to the timeless
storehouse of contemporary world literature.
139
The uniformity of rhyme in
the verses
140
cannot therefore be accidental; it must be assumed that those
parts were intended a priori
141
to be united in a single sra. It is conceivable
that Muammad also closely followed the tradition in terms of sequence.
and the Vulgata as horned (cornuta), by bringing together the word radiate in the Hebrew
text (aran) with qeren horn. (Cf. C. Hunnius thesis, Das syrische Alexanderlied, p. 27.)
The remainder of the Moses legend belongs to a legendary cycle of the Orient still
shrouded in obscurity, which from there spread to the European Occident. (Cf. the bibliog-
raphy in Joh. Pauli, Schimpf, p. 550sq., Gaston Paris, La posie du moyen ge, chapter L Ange
et l ermite, pp. 150187.) Apart fromthe Koran, its oldest source is a Jewish work of the tenth
century. The Midrash itself might be considerably older because here one of the two wan-
derers has been identified with a historical personage of the first half of the third century,
the famous Palestine Amorite Joshua ben Levi (cf. Encyclopedia Judaica, 2nd ed., s.v. Elijah,
p. 125, col. 2). Still, a direct dependence is excluded as the stories are too different. As the
well-head of life, which the Koran mentions in this connection, plays an important role in
Babylonian mythology, the entire legend might originate from there. If we consider that in
Paradise not only eternal life butas can be seen from the Biblical form of the storyalso
super-human knowledge was to be gained, it follows that between the apparent disparate
parts of the Koranic legend of Moses there is an inner connection.
138
Cf. I. Guidi, Testi orientali; Nldekes review of Testi orientali; de Goeje, De Legende der
Zevenslapers; J. Koch, Die Siebenschlerlegende; B. Heller, Elments parallles et origines de
la lgende.
139
Cf. also Wellhausen, Reste arabischen Heidentums, 2nd ed., p. 236.
140
The rhyme regularly ends with , because the verses 21 and 22 as well as 97 and 98 of
Flgels edition of the Koran must be combined to one verse each. The rhyming letters (raw),
however, are very different and cover the entire scale of the alphabet; most frequently d (47
times), r (22 times), h (17 times), and l (13 times.)
141
Among the frequently used expressions common to several parts I note: , my Lord,
verses 21, 23, 34, 36, 38, 94, 97, 98, 109, and , 70, 73, 76, 84, 89, 92 and 95.
THE SRAS OF THE THIRD MECCAN PERIOD
General Comments
The style, language, and treatment of subjects that developed gradually dur- [i/143]
ing the second Meccan period appear fully developed in the third period.
The language becomes drawn-out, dull, andprosaic. The endless repetitions,
in which the Prophet does not hesitate to use almost identical words, a line
of argument devoid of sharpness and clarity, which convinced only those
who had already been converted, and the monotonous narratives all often
make the revelations downright boring. Someone not interested in the lan-
guage of the original, or in historical religious problems, finds it difficult to
readthe later parts of the Koranfor a secondtime.
1
Of course, one shouldnot
imagine that the ardent spirit of the first revelations never appears again; it
does, albeit in isolated sparks. The prosaic, longwinded diction is unsuited
to providing a dignifiedgarbfor his fantasy, whenever it does appear. Closely
related to the style, which becomes increasingly prosaic, is the growing
lengthof the verses, withthe result that nothing remains of the poetical form
but the rhyme. Althoughthe rhyme does still frequently leave animpression
as a forceful conclusion to ideas, it is also oftenannoying, employed casually
and reduced to the simplest forms, such as n, n, etc. The sras themselves
are at times exceedingly long, although some of these long sections may be
made up of shorter ones without readily apparent joints. Another peculiar-
ity of the third period is the address, O, you people ( ). Just as
an Arab usually uses an address when speaking at an assembly,
2
Muam-
mad now does so more often when speaking prosaically. The earlier sras,
which are poetically or, even more, rhetorically motivated, lack this figure
of speech.
1
Muammad was an average stylist at best. The literary importance of the man is
based on his originality to create a new Bible-like style for the documents of his new reli-
gion.
2
For example,
.
118 the sras of the third meccan period
Comments on Sras: 32, 41, 45, 16, 30, 11, 14, 12, 40, 28, 39
Since the sras of the third period display virtually no apparent develop- [i/144]
ment, it is even more difficult than for the earlier periods to establish any
kind of chronological order.
In sra 32,
3
the words of the twenty-third verse, were
certainly interpolated, since they do not fit into the context in any way.
4
Verse 16,
5
or verses 18 to 20,
6
are incorrectly considered to be of Medinan
origin, the former because of a traditionthat appliedit tothe poor emigrants
or the helpers, and the latter because it was cited in connection with an
event during the Battle of Badr.
With sra 41:13 Muammad is supposed to have tried to convert Utba
b. Raba,
7
a respected Meccan. Even if this were true, all it teaches us
is that the sra antedates the conversion attempt. Ibn Hishm dates this
to immediately after the conversion of amza b. abb al-Taym, yet Ibn
Hishm is known to have paid virtually no attention to precise chronology
of the events before the hijra. It must be added that we have no reliable
information on amzas conversion.
8
As for the external form of the sra, it
is noteworthy that verses 1 to38 regularly rhyme withnor n, less frequently
(verses 1, 11, 32, and 34 to 36) with m. The caesuras at the end of verses 12 and
26 (Flgels edition) are inaccurate. Thereafter, in verses 39 to 54, n and n
respectively disappear altogether; m occurs only once and is replaced by a
great variety of other rhyming letters ( )
9
This, however,
3
This sra, like several other shorter ones, Muir assigns to his fourth stage, not the fifth
stage. In Flgels edition of the Koran the verses 9 and 10 constitute one verse.
4
All attempts at explanation by Muslim exegetes are futile as is the case in Muslim
al-Qasalln, vol. 2, p. 75 (bb al-isr, K. al-mn, 72) the in refers to Moses. The true
sense of these words in the original place becomes apparent from passages like 32:10 and 14,
and 41:54.
5
al-Wid; al-Itqn, pp. 34 and 19; Al al-Dn (AL-KHZIN al-Baghdd).
6
al-abar, Tafsr; al-Nasaf; al-Wid; [AB AL-QSIM] Umar b. Muammad [IBN
ABD AL-KF]; al-Zamakhshar; al-Itqn, p. 19sq.
7
Ibn Hishm, p. 186; cf. Sprenger, Das Leben, vol. 2, second ed., p. 7sq.; [EI
2
; Juynboll,
Encyclopedia, p. 460, 531; Wensinck, Muammad and the Jews of Medina, p. 111].
8
Ibn Hishm, p. 227 and Ibn Sad (al-abaqt al-kabr): Biographien der mekkanischen
Kmpfer (vol. 3, part 1), p. 192, presuppose that amza embraced Islam before Umar did.
Al-abar, vol. 1, p. 1189, explicitly says so. Ibn Sad, ibid., p. 4, puts the conversion into year 6
of Muammads prophetic commission. Ibn ajar al-Asqaln, vol. 1, no. 1818; and Izz al-Dn
IBNAL-ATHR, Usdal-ghbaf marifat al-aba, vol. 2, p. 46, claimthat this had takenplace
already in the second year after Muammads commission.
9
Cf. below on sra 40.
the sras of the third meccan period 119
is no reason divide the sra, particularly as verse 39 (rhyme with r) belongs
together with verses 34 and 38 (rhyme with m), while verse 44 seems to
refer to the first verse.
According to al-Wid, sra 45:13 was occasioned by a campaign against [i/145]
the Ban Mualiq, or some other event in Medina (cf. also al-Itqn, p. 35).
Umar, who plays a noteworthy role in this affair, appears also in traditions
supporting the Meccan origin. That a man of the Ban Ghifr
10
appears in
some of these traditions as Umars opponent perhaps stems originally from
that verse is containing the word .
In sra 16 we find some verses that were not promulgated until Med-
ina. Verse 43sq. could be taken to refer to the emigration to Abyssinia, but
verse 111 clearly speaks of those who have emigrated after persecution and
then struggled against the infidels. Since the two afore-mentioned verses
bear a great resemblance to this one, we might assume an identical origin.
Incidentally, this verse pertains here to emigrants in general and not to any
particular band, as al-Wqid, p. 111, and al-Wid, s.v., report. Verses 115 to
118 could be considered Meccan if it were certain that sra 6:119 referred to
them.
11
In contrast, verse 119 must have originated at Medina if, as seems
likely, it does have 6:147 in mind. The same applies to verse 120, which is
connected with it and displays similarity to verse 111, as well as verse 125,
which deals with the Jewish Sabbath. The Meccan origin of verse 124 is
doubtful,
12
if only because most of the verses, like this one, consider Islam
to be the religion of Abraham (millat Ibrhm.)
13
Sras 2:134 and 129, 3:89,
4:124, and 22:77 are surely Medinan on the basis of their context. This sus-
picion is strengthened by internal evidence. In the beginning Muammad
is convinced to bring to the Arabs what the Christians received from s
(Ibn Maryam, Jesus), and the Jews from Ms (Ibn Imrn, i.e., Moses) etc.;
and in dealing with the pagans, he optimistically makes a reference to the
10
Fakhr al-Dn al-Rz; al-Zamakhshar; Al al-Dn (AL-KHZIN al-Baghdd).
11
It is wrong to think it refers to 5:4, one of the latest verses of the entire Koran.
12
The arguments for the following evidence I adoptedfromSnouckHurgronjes Mekkaan-
sche feest, pp. 2840. Also in his subsequent articles did this scholar quite rightly emphasize
again and again the importance of Ibrhm for the development of the attitude vis--vis
Muammads early revelations. Cf. his De Islam, tweede deel, pp. 460 and 466; and his
review, Une nouvelle biographie de Mohammed, by H. Grimme, p. 64sqq.
13
In other passages of the Koran milla refers to the religion of the Jews and Christians
(once, 2:114) as well as pagans (four times; in 38:6 the meaning is ambiguous). Its origin from
Aramaic is beyond doubt (mellth word), but the Koranic meaning religion is not to be
found there. However, it seems to me that the usage of this word among the Arabs is older
than Islam.
enlightened (sras 16:45, and 21:7), who merely have to be asked to have
the truth of his teaching confirmed. The disappointment comes at Medina,
where the People of the Book refuse to recognize the Prophet. He is thus
obliged to find for himself an instance that still does not contradict his early
revelations. Given this situation, he reaches for the older prophets whose
communities cannot cross him. In the final analysis, though, this tendency
is only expressed in sra 2:129. It comes as no surprise that Muammad later
felt himself most closely related to Abraham, since this patriarch was for
bothChristians andJews the perfect example of justice andobedience tothe
faith, the Father
14
of all pious men who God took for a friend.
15
Muam-
mads preference for Abraham is closely connected with the idea of sras
2:119 and 2:121, which present him as the founder of the Meccan sanctu-
ary.
16
Moreover, Muammad might not even have adopted this view until
Medina, for still in the late Meccan sras he was of the opinion that no pre-
vious warner hadcome tohis contemporaries fromAllh(sras 32:2, 34:43,
36:5). One is tempted to assign the entire section fromverses 111 to 125 to this
period, since, besides verse 124, verses 111, 119, and 120 are definitely Medi-
nan, as well as, possibly, verses 113 to 118.
17
On the other hand, it is wrong to
consider verses 96 (starting with ) to 98, or verses 97 to 99, to be Med-
inan
18
by interpreting (verse 97) to refer to agreements concluded
with various clans after the hijra. By the same token, the preceding verse
might be assigned to after the hijra. Additionally, the division of verse 96
into two parts cannot be justified in any way. Weil
19
declares verses 103 to
14
E.g. Bershth Rabb, Par. 39, beginning; Matthew 3:9; Luke 16:24; Romans 4:1, 4:16, etc.
This is presumably where also sra 22:77 belongs. The idea that Abraham is the patriarch of
the Arabs need not be the basis of this passage.
15
In the Koran only sra 4:124 (khall), but quite common in adth. The idea is expressed
already in the Old Testament (hb, Isaiah 4:8, II Chronicles 20:7). In later Hebrew writing
Abraham is called yedd, e.g., Talmud Babl, Menhth 53 b, and Shabbth 137 b (other
passages see B. Beer, Das Leben Abrahams, notes 427 and 950) or rem, e.g., in an Aramaic
synagogical liturgy for the minor Atonement Day. In early Christian literature it is called
(Jacob. 2:23; Ep[istle of] Clem[ent] 10:1 and 17:2).
16
This legendis perhaps not Muammads invention, rather the product of ArabJewishor
Christian brains who did not want to renounce the religious celebration at the Kaba. Snouck
Hurgronje in his Mekkaansche feest, p. 28, writes that he [Muammad] shunned the ajj
festivities because this included the presence of polytheists and that it is not unlikely,
and considered a fact by tradition, that also Christians participated, which explains the swift
riding through Wd Muassir [Juynboll, Encyclopedia, p. 634] where in early days Christians
observed the wuqf (cf. Muammad bid, Hidyat al-nsik, p. 112).
17
Thus Grimme, Mohammed, vol. 2, p. 26, although without giving a reason.
18
Umar b. Muammad IBN ABD AL-KF; Al al-Dn (AL-KHZIN al-Baghdd).
19
Historisch-kritische Einleitung in den Koran, 1st ed., p. 64, 2nd ed., p. 74.
105 to be Medinan, yet the assumption that Muammad did not abrogate
or change verses before the hijra is wrong; one merely needs to recall what
was said above regarding sra 53.
20
Verse 105, in which it says that only a
mortal is teaching him, as well as verse 103, according to which the unbe-
lievers openly call hima mere forger, do not reflect the conditions after his
emigrationfromhis native town. Finally, the verses inquestionare, it seems,
connected with their surroundings. Totally worthless is the argument that
verse 105 points to the Persian Salmn, who did not embrace Islam until
Medina.
21
It seems to have emanated from the inaccurate interpretation of

, a name which was later predominantly applied to Persians. Many tra-

ditions name other people, all obscure persons, slaves with names as well as
unknown ones (e.g., , , , ). Verse 108, according to a tradition
of Ab l-ajjj MUJHID[Ibn Jabr al-Makk
22
] refers to those believers who
were afraidto followthe example of the Prophet andturntheir back ontheir
native town.
23
All other traditionists correctly apply the verse to Muslims
without means or reputation who before the hijra were the object of much
persecution on the part of the Meccans. The verses starting with 126 are
Meccan in content as well as form.
24
Tradition throughout interprets them
as prohibition against Muammads taking revenge on the Meccans for the
death of amza in accordance with his vow.
25
Some writers add that these
verses were not occasioned immediately after the Battle of Uud but only
during the conquest of Mecca,
26
where Muammad was diplomatic enough
not to act on the opportunity for vengeance. These inaccurate views might
be based on the fact that the text of that earlier revelation was referred to
by the Prophet during one of these occasions,
27
but, more likely, they are
fabrications of the exegetes. Moreover, since the last half of sra 16 contains
20
See above, p. 82sqq.
21
al-abar, Tafsr, vol. 14, p. 111, l 5; al-Zamakhshar; al-Bayw; Weil, Das Leben Mo-
hammeds, p. 369, note, Weil, Historisch-kritische Einleitung, 2nd ed., p. 74.
22
EI
2
; EQ; Juynboll, Encyclopedia, pp. 430431; Sezgin, GAS, vols. 1, p. 29, vol. 6, p. 10, vol. 7,
p. 365, vol. 8, p. 22.
23
Cf. al-Wid; al-Suy, Asbb al-nuzl; Al al-Dn (AL-KHZIN al-Baghdd).
24
Cf. (verse 126); (v. 128); (v. 128); all indications are
that the Prophet found himself in the situation of the underdog and could not even consider
resistance, least of all open warfare.
25
Ibn Hishm, p. 584sq.; al-abar, Annales, vol. 1, p. 1420sq., and his Tafsr; al-Wqid,
p. 283; al-Tirmidh (Tafsr), al-abar (Persian), vol. 3, p. 38; al-Wid; (AB AL-QSIM)
Umar b. Muammad IBN ABD AL-KF; al-Aghn, vol. 14, p. 22sq.; al-Zamakhshar; al-
Bayw; al-Suy, al-Itqn, pp. 19, 33, and 42; al-Suy, Asbb al-nuzl.
26
al-Tirmidh, loc. cit.; al-Suy, al-Itqn, p. 42; al-Suyu, Asbb al-nuzl.
27
Cf. Weil, Historisch-kritische Einleitung in den Koran, 1st ed., p. 64, 2nd ed., p. 74.
several genuine or allegedly Medinan verses, some people take the easy way
out and consider the verses 42 or 41 to the end,
28
or the entire sra, to be
completely Medinan.
29
The first verses of sra 30 must have been promulgated after the Byzan- [i/149]
tines (al-Rm) had fought the Persians with little fortune in a neighbouring
country of Arabia.
30
It is very difficult, however, to determine which of the
many Byzantine defeats dating until after the hijra
31
is meant, particularly as
the early Muslim writers,
32
who supply confusing and incomplete accounts
of these events, cannot be confirmed by Byzantine reports. The common
view holds that this sra concerns a Byzantine defeat at Adhrut
33
and
Bur in Mesopotamia or in Palestine. The Persian translator of al-abar
(Chronique, Zotenberg, vol. 2, p. 306sq.), who mentions in this context all
sorts of confused details about the dethronement of Mauricius ( ), etc.,
says that the Koran speaks of the capture of Jerusalem. It is difficult to
deny that the sra pertains to an important event that took place either
in Palestine or its vicinity. Yet we cannot say for sure if that capture hap-
pened in June, ad614,
34
in accord with the most reliable source (Chron[icon]
pasch[ale]), or at a later date.
35
Perhaps Muammad had no single event in
mind at all. Gustav Weil
36
is wrong to separate the first verses from the fol-
28
(AB AL-QSIM) Umar b. Muammad IBN ABD AL-KF; Hibat Allh.
29
Ibid.
30
The readings
and are, of course, oldandalready mentionedinal-Tirmidhs

Tafsr on sra 30:1; but they are based on less reliable authorities than the common reading
and must be rejected because they emanate only from the defeats later inflicted upon the
Byzantines by the Muslims. This, however, Muammad cannot have anticipated at the time.
Al-abar says in his Tafsr:
) .
31
Cf. Bar Hebraeus, Chronicon, p. 100.
32
al-abar, vol. 1, p. 1003sqq., and Nldeke, Geschichte der Perser, p. 297sqq.; al-Wid;
Fakhr al-Dn al-Rz; al-Qurub; and al-Bayw.
33
In this battle the Greek military leader was, according to al-Wid
(Yuannis),
about whom I have been unable to find anything. The Persian leader , however, is also
mentioned by the Byzantines ( etc), by the Armenians (after LeBeau, Histoire du
Bas-Empire), and Bar Hebraeus . Cf. in particular Th. Nldeke, Geschichte der Perser,
p. 292.
34
Nldeke, Geschichte der Perser, p. 297, and Aufstze zur persischen Geschichte, p. 126.
35
As far as this subject is concerned, the Meccans were rather indifferent regarding the
defeat of either the Persians or the Byzantines, for the viewthat they as idolaters sympathized
with the Persians as the Muslims relate is missing the point. But Muammad was indeed
interested in the Christians, with whomat that time he nearly identified himself. For himthe
victory of the Byzantines over the Persians must have been equivalent to the victory of the
monotheists over the disbelievers, and for this reason he enabled his opponents to reproach
him because his friends had been defeated, and that his god had apparently been unable to
help them.
36
Historisch-kritische Einleitung in den Koran, 1st ed., p. 67, 2nd ed., p. 76.
lowing ones, with which they are closely connected. Verse 16sq.is held by
some to be Medinan on the grounds that they refer to liturgies which, natu-
rally, had already been in service before the emigration.
37
For no good reason whatsoever some people hold verse 5 of sra 11
38
to [i/151]
have been promulgated at al-if.
39
A different opinion, suspecting here a
reference to the Hypocrites of Medina,
40
has already been rejected by al-
Bayw. Verses 15,
41
20
42
(because of its mentioning the Jews), and 116
43
(be-
cause it fixes the times for prayer) some hold to be Medinan. The indi-
vidual parts of the sra are generally coherent.
44
Still, some irregularities
in the composition must be pointed out. In the stories of N [Noah]
(verse 27sqq.), Hd (verse 52sqq.), li (verse 64sqq.), and Shuayb (verse
85sqq.),
45
verses 72 to 84 are divorced fromthe scheme adopted in the intro-
duction. Verse 85 is more readily understood as a continuation of verse 71.
Verses 112 to 123 unmistakably refer to afore-mentioned generations
(11:118), cities (11:119), and messengers (11:121), although verses 102 to 111
certainly appear to be a concluding recapitulation. The mention of Moses
in verse 112 is conspicuous in view of verse 99.
37
Umar b. Muammad IBN ABD AL-KF; al-Zamakhshar; al-Bayw. Even with a
stretch of language you arrive at only four prayers; but the verses 16 and 17 are probably
parallel. The five daily prayers are nowhere explicitly instituted in the Koran. Cf. above, p. 45.
38
Verse 5 of Flgels edition concludes with , which is against its sense and all good
traditions. Cf. thereon Umar b. Muammad IBN ABD AL-KF, and Ab Yay Zakariyy
AL-ANR al-Shfi [d. 926/1519], K. al-Maqad, s.v.
39
al-Bayw, cf. Al al-Dn (AL-KHZIN al-Baghdd).
40
Al al-Dn (AL-KHZIN al-Baghdd), s.v.
41
al-Itqn, p. 32; Al al-Dn (AL-KHZINal-Baghdd) in the introduction after Muqtil.
42
Ibid.
43
Ibid.; al-abar, Tafsr, vol. 12, p. 75; Umar b. Muammad IBNABDAL-KF; al-Wid;
al-Suy, Asbb al-nuzl; Al al-Dn(AL-KHZINal-Baghdd) after IbnAbbs and Qatda;
al-Qasalln to al-Bukhr, K. mawqt al-alt, 4.
44
Cf., for example, the omission of , verses 52, 64, and 85, because the phrase had
already been used in verse 27.
45
For the first time, the people of Shuayba name still not properly identified, and
previously always known by the genuine Arabic name al-Aykaare here called Madyan,
a name which can have reached Muammad only through Jewish channels. According to
him, both names indicate an identity because (1) they have only one prophet, which never
happens in the case of two peoples; (2) once the name Madyan is introduced it never
reappears; (3) both peoples are accused of filling up the balance unjustly (sras 26:181 sq.,
7:83, and 11:86). For the reasons, both one and two, some Muslims presuppose the identity
of both peoples (al-Suy, al-Itqn, p. 795). It is, of course a different question whether the
identity of Shuayb and the father-in-lawof Moses, and the related problemof his people and
Madyan is original. Cf. Nldekes article Midian in Encyclopdia Biblica, vol. 3, col. 3080.
[EQ; Juynboll, Encyclopedia, p. 650, col. 2.]
Sra 14:33 and 34 several exegetes falsely consider to refer to the Quraysh [i/152]
fighting at Badr.
46
Inverse 38sqq., Ibrhm[Abraham] asks his Lordto make
the sacred territory of Mecca secure and keep his sons fromserving idols; he
then praises God for having given him two sons, Jacob and Ishmael, despite
his old age. For the same reasons explained above regarding sra 16:124, as
well as the arguments of Snouck Hurgronje,
47
these verses also ought to be
considered Medinan. Fromthenonthe patriarchs are no longer mentioned
without including Ishmael between Abraham and Isaac. At a later stage
Ishmael advances to become a joint founder of the Kaba ([raised up the
foundations of the House,] sra 2:121).
The entirety of sra 12, with the exception of the final few verses, which
are nevertheless still connected with the others,
48
is distinguished from
all the other large sras by its focusing on only one subject,
49
the life of
Joseph.
50
We have it fromtwolater writers
51
that Muammaddispatchedthis
sra with the first men fromYathrib converted near Mecca. Even supposing
that this was entirely certain, it merely follows that the sra dates from
before this event, and not that this was the occasion of the revelation itself,
as Weil seems to believe.
52
Regarding this point of view, which considers
verses 1 to 3 to be Medinan,
53
al-Suy
54
correctly says that this is untenable
and baseless. The same applies to the tradition that ascribes a Medinan
origin to sra 12, verse 7.
55
Sra 40:58sq. is unjustly considered to refer to the Jews and, thus, held to [i/153]
be Medinan.
56
The verses from59 to the end (v. 87) stand out in so far as they
all rhyme withnor nwhile inthe preceding rhymes anenormous diversity
46
al-Wqid, p. 133; (AB AL-QSIM) Umar b. Muammad (IBN ABD AL-KF); al-
abar, Tafsr; al-Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 33; Al al-Dn (AL-KHZIN
al-Baghdd). The verses 11, 12, 13, 14 and 24, 25 of Flgels edition consitute only one verse
each.
47
Snouck Hurgronje, Het Mekkaansche feest, p. 40, l 1723. Cf. above, p. 119.
48
See verse 109sqq., but particularly verse 111.
49
This also corresponds to the rhyme which throughout ends with n, m, and n, once
each only with r and l. The rhymes with r (v. 39) and r (v. 96) are based on the inaccurate
division of the verses.
50
Regarding the Jewish sources of the Koranic version cf. Geiger, loc. cit., p. 139sqq., as
well as Israel Schapiros comprehensive thesis.
51
al-Suy, al-Itqn, p. 39; al-Diyrbakr, Cairo ed., 1283, juz 1, p. 13.
52
Das Leben Mohammeds, p. 380.
53
(AB AL-QSIM) Umar b. Muammad IBN ABD AL-KF.
54
55
56
Ibid.; al-Suy, al-Itqn; al-Suy, Asbb al-nuzl; Al al-Dn (AL-KHZIN).
prevails.
57
Additionally, since the two parts of the sra are incoherent inter-
nally, we may suspect different origins. Their combination possibly owes to
the fact that the pet idiom of the first part,
(verses 4, 5, 37,
and 58) appears also once in the second part of this sra (verse 71).
58
Sra 28:52 is just as falsely considered to refer to the Christians who
came toMuammadat Medina.
59
Among other reasons, howcouldMuam-
mad still claim after his sad experiences with the Jews that those who had
received the Scripture believed in the Koran. Verses 76 to 82 look like a seg-
ment inserted at the wrong place, since it is difficult to connect with either
the preceding or the following text, particularly as 28:83 is more suitable to
follow 28:75. Given the usual, frequently-jumping style of the Koran, how-
ever, this is not the issue. One can thus consider verse 83 as a contrast to the
whole story of Qrn [the Biblical ora], who believes in his own strength
and does not worry about God and the hereafter.
60
On account of a literal
and, in this case, totally inappropriate interpretation
61
of the words
, verse 85 is saidto have originatedduring the emigrationto Jufa, a place
between Mecca and Yathrib.
62
It is likely to be nothing but a misinterpreta-
tion that makes this a Medinan sra
63
or considers the entire sra to have
been revealed between Mecca and Medina.
64
Regarding sra 39,
65
the verses 54, or 54 to 56, or 54 to 61, are supposed [i/154]
to have been sent from Mecca to Medina on account of Wash or other
57
The principle rhyme is with following b, d, r, q, l, , , altogether forty-one times;
with following m, n, r, l, b, twenty-one times; with following d, n, r, twenty-two times. The
conspicuous rhyming word
in verse 39 (Flgels edition) is certainly occasioned by an

inaccurate division of the verses.
58
Given a different context, this evidence would opt for original uniformity.
59
al-abar in the Tafsr after al-ak; and Fakhr al-Dn al-Rz after Muqtil b. Sulay-
mn; Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 34. According to Al al-Dn (AL-KH-
ZIN al-Baghdd) also the following eight verses are Medinan.
60
Also in the diction of the verses 77 and 83 there are some points of contact, e.g.,

.
61
This is also Weils view in Historische-kritische Einleitung, 1st ed., p. 66, 2nd ed., p. 76.
62
Fakhr al-Dn al-Rz; al-Zamakhshar; al-Bayw; al-Suy, Asbb al-nuzl; Al al-Dn
(AL-KHZIN al-Baghdd). Cf. Weil, Das Leben Mohammeds, p. 373. There are, however, also
different explanations of these words in the commentaries. For a strange interpretation see
al-abar, vol. 1, p. 2942, and Weil, Geschichte der Chalifen, vol. 1, p. 174.
63
64
65
The problems of the rhyme of this sra display great similarity with those of sra 41.
Against Flgel the verses 3 and 4 are only one verse. The words in verse 9,
which appear also in sras 6:164, 17:16, 35:19, and with a slight change (
) also in 53:39,
might be the result of interpolation.
noted criminals, with the result that the verses are largely considered to
be Medinan.
66
Other writers also date verse 13
67
to the time after the hijra,
probably by mistake, and likewise verse 24
68
for no good reason.
Comments on Sras 29, 31, 42, 10, 34, 35, 7, 46, 6, 13
Sra 29:110 many writers rightfully consider Medinan.
69
Verses 7 and8 must
be included, although the commentators, giving some other explanations,
generally regard this passage, as well as sras 31:13 and 46:14, as referring to
Sad b. Ab Waqqs,
70
one of the first believers. Yet these passages refer to
those men of Medina who, obeying their parents wishes, declined to par-
ticipate in the campaigns of the Prophet. These ten verses, however, must
originate from the time after Muammad had already completed several
campaigns, certainly after the Battle of Badr and probably after the Battle
of Uud.
71
The explanations of these stories
72
produced by tradition are of
little use. Verse 45 in its current form is certainly Medinan, since here the
Muslims are permitted to deal withstubbornJews otherwise thanthe fairer
manner, i.e., not with words but with violence. Muammad could not use
such expressions before the hijra. Furthermore, this is in contradiction to
the Meccan verse 46, where it says that those to whom We have given the
Book believe in it; and some of these believe in it.
73
However, the words
(otherwise only in sra 2:145) look like a later insertion, as

66
Ibn Hishm, p. 320; Ab l-Layth al-Samarqand on sra 4:51; al-Wid; al-abar in his
Tafsr; Umar b. Muammad IBNABDAL-KF; al-Zamakhshar; al-Bayw; al-Itqn, pp. 20
and 35; al-Suy, Asbb al-nuzl; Al al-Dn (AL-KHZIN al-Baghdd).
67
al-Suy, al-Itqn, p. 35; Al al-Dn (AL-KHZIN al-Baghdd).
68
Ibid.
69
Cf. the Commentators; Fakhr al-Dn al-Rz; al-Suy, Asbb al-nuzl; al-Wid. A
misunderstanding reverses this andconsiders only the tenverses tobe Meccan. (Hibat Allh.)
70
EI
2
; Juynboll, Encyclopedia; Sezgin, GAS, vol. 8, p. 171.
71
Cf. the word , verse 10, which is not yet found in sra 2. H. Grimme, Mohammed,
vol. 2, p. 26, andAugust Mller inhis editionof Fr. Rckerts translationof the Koran, p. 509sq.,
also consider the verses 11 and 12 to belong to this context, while Weil, Historisch-kritische
Einleitung, 2nd ed., p. 76, fixes only the verses 9 and 10and with less certainty verse 5to
Medina. Sprenger, loc. cit., vol. 2, p. 132sq., is trying toestablishthe Meccanoriginof the entire
sra, dating it at the time of the emigration to Abyssinia, a view which is shared by many of
the early Muslim authorities. Hirschfeld (New researches, p. 144) concurs with him, with the
exception that he assigns individual parts to different classes, verses 1 to 12 to the sixth class,
13 to 42 to the fourth, and 43 to 69 to the fifth.
72
See the Commentators and al-Wqid, p. 68 (Wellhausen, MuhammedinMedina, p. 55).
73
Cf. above, p. 125, the comments on sra 28:52, [those to whomWe have given the Book
believe in it.]
the subordinate clause starting with does not presuppose them, and
the double exception with within the same sentence is not only awkward
but also appears only here in the Koran. If this passage is omitted, verse 45
says that the People of the Book may only be opposed with words but not
force. The provenance of the text thus shortened is unknown. In favour of
its Medinan origin speaks the phrase, ahl al-kitb, the People of the Book,
which Meccan sras do not contain, instead employing cumbersome para-
phrases,
74
whereas all the other Koranic passages with the phrase
(sras 6:53, 16:126, and 23:98) are generally considered to be Meccan.
In verse 56 it is suggested to the believers even to leave their country in the
interest of the newreligion. Yet this alone is no reasontoconclude that these
verses belong tothe periodshortly before the emigrationtoYathrib. After all,
we know that some Muslims, as well as even Muammad, had already left
the city beforehand. Verse 69 might have been added at Medina, although it
is also possible that here simply means to bear misfortune or persecu-
tion courageously rather than to fight, with the result that the verse also
fits in with the Meccan circumstances.
75
Because of a legend, which we also
have fromal-Wid, verse 60 is also held to be Medinan.
76
Another opinion,
based on the isolated Medinan verses, even applies this to the entire sra,
although there is unlikely to be yet another passage besides verse 67 that
points more clearly to its origin in the inviolable territory of Mecca. As far
as verses 18 to 22 are concerned, it might easily seemparticularly because
of the word qulthat they are out of place here. All the same, we must not
consider words like those insra 11:37 to be addressedto Muammad; rather
it is the Prophet whose sermon is being communicated. Only the historical
reference that God said to him needs to be added. It is not quite clear
for what reason this sra has been considered on more than one occasion
to be the last one promulgated before the hijra;
77
perhaps it is because of
verse 56. The verses have uniform rhymes (n, m, r, n). The glaring excep-
tion in verse 51 () is occasioned by an inaccurate division. Verses 51 and 52
in Flgels edition of the Koran ought be combined.
Sra 31:3 is considered by some to be Medinan because of the apparent [i/157]
reference to the community tax.
78
Verses 13sq., dealing with foolish parents,
74
E.g.
, verse 46.
75
Cf. Weil, Historisch-kritische Einleitung, 1st ed., p. 67, note; 2nd ed., p. 76, note.
76
77
See above the lists of the sras and the introduction to al-Wids Asbb al-nuzl, p. 8;
and al-Diyrbakr, Tarkh al-khams, juz 1, p. 10.
78
al-Bayw; al-Suy, al-Itqn, p. 19.
are most likely not in the proper place; they might rather be located after
verse 18 to serve as a contrast to Luqmn al-akms
79
wise sayings to his son.
Still, like 29:7, they probably belong to the Medinan period (cf. above, s.v.)
Before verse 15 something has most likely been omitted, since can hardly
do without the noun to which it refers. Similar cases are not infrequently
the result of interpolations. Since verse 19 can more readily follow verse 10,
the entire pericope of Luqmn might have been inserted later. Verses 16 to
18, like so many others, are said to be directed against the Jews of Medina
and thus promulgated there.
80
In sra 42, too, several verses are declared to be Medinan for no apparent
reason, namely verse 26,
81
or verse 22 b (starting with ) and 26,
82
or verse 22
(from the beginning) and 23,
83
or verses 22 b to 26,
84
or the verses 23 to
26,
85
and, finally, verse 35
86
or the verses from 37 to 39.
87
In sra 10, we also find several verses erroneously considered to have [i/158]
been created at Medina, namely verse 41,
88
which writers take to refer to the
local Jews, verse 59
89
and verse 94 or 94sq. or 94 to 96
90
or verses 59 and
60
91
which, incidentally, are the oldest verses of the Koran according to
Hibat Allh (Ibn Salma)or verse 41 right to the end,
92
or even the entire
sra.
93
The same happens occasionally with sra 34:6 because the Jews are
mentioned.
94
79
Cf. Luqmn, Fables de Loqman le Sage, ed. J. Derenbourg, introduction; EI
2
.
80
Fakhr al-Dn al-Rz; al-Wid; Umar b. Muammad; al-abar, Tafsr; al-Suy, al-
Itqn, p. 35; al-Suy, Asbbal-nuzl; Al al-Dn(AL-KHZINal-Baghdd); al-Zamakhshar;
al-Bayw. Flgels verses 32 and 33 constitute only one verse; this is in accordance with
sound tradition because the rhyming word is in this sra impossible.
81
Fakhr al-Dn al-Rz; al-Bayw.
82
al-abar, Tafsr; al-Zamakhshar; al-Wid.
83
Al al-Dn AL-KHZIN al-Baghdd.
84
(AB AL-QSIM) Umar b. Muammad; al-Suy, Asbb al-nuzl; Al al-Dn AL-
KHZIN al-Baghdd.
85
86
al-abar, Tafsr; al-Zamakhshar; al-Bayw. Flgels verses 50 and 51 constitute only
one verse.
87
al-Itqn, p. 35; Al al-Dn AL-KHZIN al-Baghdd.
88
Umar b. Muammad; al-Suy, al-Itqn, p. 32; Al al-Dn AL-KHZIN al-Baghdd.
89
90
Cf. note 86. Hibat Allh b. Salma seems to mean these verses when he says that except
for one or two verses this sra is Meccan.
91
al-Khzin al-Baghdd after Muqtil.
92
al-Suy, Itqn, p. 32.
93
al-Suy, Itqn, p. 26. Flgels verses ten and eleven must be combined to one verse.
94
al-Suy, Itqn, p. 35.
Sra 35, verse 37 to the end (verse 45), have a common rhyme with
that differs from the rest. This, however, is no reason to consider them late
additions, particularly as verse 37 continues verse 33 quite well and the
phraseology shows some other points of contact.
95
Sra 7 can be divided into five sections: verses 1 to 56 (the temptation of
Adam and admonition addressed to the children of Adam), verses 57 to 100
(the sending of the ancient prophets, N, li, and Shuayb), verses 101
to 173
96
(Moses and the subsequent fate of the Jews), verses 174 to 185 (on
an anonymous enemy of God), and, finally, verses 186
97
to 205 (on the Last
Hour). Although there is no close relation between these sections it is still
conceivable that Muammad himself made this combination. The first part
probably dates from a pilgrimage celebration at Mecca, because it is an
attackonthe customof circumambulating the Kaba inthe nude, andfasting
during the time of the pilgrimage (verse 29). Verses 29sq. (cf. verses 127sq.)
seem to indicate that shortly before this time there had been a scarcity of
provisions at Mecca. Verse 163, to which some of the following verses are
occasionally added, some writers consider Medinan,
98
probably because of
a false inference from
(verse 163), which was taken to refer to the Jews

of Yathrib. It is rare that verse 198
99
or verse 203
100
are considered Medinan,
although in verse 156 there are several indications that indeed betray a
Medinan origin:
is to be found only in Medinan passages, where it

fits better, since the contrast between the prophets, who originated from
pagans, and the People of the Book was of less importance at Mecca; the
Torahand the Gospel are never mentioned inthe Meccansras. And, finally,
95
Cf. verse 2:39; , verses 11 and 41; and
, verses 14 and 44. Verse 42 looks

like a variant to the last five words of the preceding verse. According to sound tradition, there
is no end of the verse after Flgels
(verse 44.)
96
The verses 139, 140, 143, 144; 146, 147, and 157, 158 (Flgels edition of the Koran) actually
constitute one verse each as isolated rhymes with are inadmissible in this sra.
F. Rckert in the notes to his translation of the Koran, p. 157sq. considers the last sentence
from verse 142 to verse 148 an allusion to the content of verse 149, and whatever there is in
between, to be false and, in any case, useless, but without supplying sufficient evidence.In
accordance with sound tradition, verse 166 ends after .
97
Following the example of several passages (7:93, 12:107, 26:202, 29:53, 43:66), andinorder
toproduce a rhyme, we must addinverse 186after something like .Verse 199
= sra 41:36.
98
Umar b. Muammad; al-Suy, al-Itqn, verse 163; Hibat Allh, verses 163166; Al
al-Dn AL-KHZINal-Baghdd, verses 163167; al-Zamakhshar and al-Bayw in the intro-
duction, al-Suy, al-Itqn, p. 32; Al al-Dn AL-KHZIN al-Baghdd, verses 163170.
99
al-Bayw in the introduction.
100
al-Wid; al-Suy, Asbb al-nuzl.

unmistakably indicate the anr. We must therefore consider

both this verse and the following verse, 157, i.e., verses 156 to 158,
101
which
also happento interfere withthe development of the subjectto be a Medi-
nanappendix, possibly put here by Muammad himself. Tradition
102
usually
considers verses 174 to 182 to refer either to the Biblical Balaam or to the
aforementioned Umayya b. Ab l-alt. A more recent exegete,
103
however,
believes they refer to the Jewish poet Kab b. al-Ashraf and therefore con-
siders the section to be of Medinan origin.
Sra 46 is likewise considered to be Medinan because the Jews are men- [i/160]
tioned.
104
Sunnites trace verse 14 back to Ab Bakr. It cannot be determined
whether this tradition was fabricated to justify his caliphate or whether this
was the result of less partisan motives. The text of the verses 14 to 16, how-
ever, has no particular person in mind,
105
instead emphasizing in a general
sense that reverence for ones parents is among the signs of the true Muslim.
Like the sectionfromverses 14 to16, verses 34sq.,
106
whichactually constitute
only one verse, are also occasionally considered to be of Medinan origin for
no apparent reason. Verses 20 to 31 certainly were not inthis place originally,
as they interrupt the continuity of verses 32sq. and 19, although they do still
belong to the same period. Even the earliest tradition attributes verse 28 to
that same situation as sra 72.
107
Even if this were not the case, it is certain
that Muammad believed that he was in contact with the world of the jinn.
In sra 6 we find serious changes of subject after verses 45, 72, 90, 117, [i/161]
134, 141, and 154. Still, the sra displays an extraordinary stylistic as well as
lexical uniformity.
108
This phenomenon can best be explained by the conjec-
101
I have no idea why H. Hirschfeld in his New researches, pp. 132 and 145 considers also
the remainder to verse 172 to be Medinan.
102
The Commentators; al-Wid, etc.
103
H. Hirschfeld, Newresearches, p. 95.
104
The verse is thought to refer to Abd Allh b. Salm (Umar b. Muammad IBN ABD
AL-KF); al-abar, Tafsr; al-Zamakhshar; al-Bayw; Fakhr al-Dn al-Rz; Al al-Dn AL-
KHZIN al-Baghdd; al-Suy, al-Itqn, p. 36; al-Suy, Asbb al-nuzl; Ibn ajar al-Asqa-
ln, vol. 2, p. 782; Izz al-Dn IBN AL-ATHR, Usd al-ghba f marifat al-aba, vol. 3, p. 176,
although certainly no single man is meant.
105
Al al-Dn (AL-KHZIN al-Baghdd) on verse 16:

.
106
al-Suy, al-Itqn, 36; Al al-Dn (AL-KHZIN al-Baghdd).
107
Ibn Hishm, p. 281; al-abar, vol. 1, p. 1202; Ibn Sad, Biographie Muammads, p. 142;
al-Diyrbakr, vol. 1, p. 303; and the Commentators. Cf. above, p. 108sq.
108
Cf. , verse 12, 20, 31, and 141;
, verses 21, 93, 145; ,

verses 24, 112, 138, 139;
, verses 46, 65, 105; , verses 46 and 158, three times, but
nowhere else in the Koran; , verses 3, 69, 120, and 129; , verses 22, 94, 137, and 139;
,
verses 17 and 49; , verses 9, 65, 82, and 138; , verses 92 and 156;
,
ture that the majority of the individual parts originate froma strictly limited
period. Without sufficient grounds some writers consider verse 20 as Med-
inan,
109
presumably because of its reference to the People of the Book. We
find this view
110
even more frequently applied to verse 93, because it is con-
sidered to refer to the false prophets (Musaylima, etc.) or to Abd Allh b.
Sad b. Ab Sar, [d. 57/676677],
111
who is said to have falsified the revela-
tions. A better case can be made about verse 91, whose date is fixed to after
the emigration,
112
as the direct charge against the Jews of writing down their
holy books (and thereby suppressing a great deal, including the passages
referring to Muammad) is more likely to have been made at Medina than
at Mecca. Since verse 93 was also considered Medinan, some writers sim-
ply give verses 92
113
and 94
114
the same dating. Verses 118 to 121 can hardly be
in their proper place; instead, they must be seen as a fragment that natu-
rally bears a great resemblance
115
to the section minutely discussing dietary
laws and other prescriptions in verses 135 to 154. Verse 142, recommending
alms,
116
and verses 152 to 154
117
are falsely considered Medinan. Before the
section that starts with verse 155, something seems to have disappeared.
verses 55, 97, 98, 119 and 126; and
respectively, verses 80, 83 and 150;
, verses 43,
108, 122 and 138; , verses 78 and 135; , verses 112, 128 and 130; , verses 113 and
120; , verses 32 and 69.The idiom , verse 147, which refers to the Jews is
elsewhere found only in Medinan passages, sras 2:59, 4:48 and 158, 5:45, 48, and 73, 16:119,
22:17, and 62:6.
109
Umar b. Muammad; al-Suy, al-Itqn, p. 31; al-Khzin al-Baghdd [EI
2
].
110
See above, p. 37sq.
111
al-abar in his Tafsr; al-Zamakhshar; al-Bayw; Al al-Dn AL-KHZIN al-Bagh-
dd; Ab l-Layth al-Samarqand; al-Wid; al-Suy, al-Itqn, p. 31; al-Suy, Asbb al-
nuzl. EI
2
.
112
Ibid.
113
Fakhr al-Dn al-Rz; al-Suy, Itqn, p. 31.
114
al-Suy, al-Itqn, p. 31; Al al-Dn AL-KHZIN al-Baghdd.
115
Verse 119 refers either to sra 16:116 or 6:146.
116
Umar b. Muammad IBN ABD AL-KF; Ab l-Layth al-Samarqand; al-Zamakhshar;
al-Bayw; Al al-Dn AL-KHZIN al-Baghdd.
117
Umar b. Muammad IBN ABD AL-KF; Ab l-Layth al-Samarqand; al-Zamakhshar;
al-Baydw; Fakhr al-Dnal-Rz; al-Suy, al-Itqn, p. 31; AL-KHZINal-Baghdd. Al-Suy,
al-Itqn, p. 31 sq., and Al al-Dn AL-KHZIN al-Baghdd in the introduction, who, among
the writers accessible tome, supply the best informationregarding the Medinanverses of this
sra, bothhanddownthree different views each. Medinanare according toal-Suy, al-Itqn
a: vv. 152154, 93, 94, 20, and114; according toal-Suy, al-Itqnb: v 91 sq.; according toal-Itqn
c: vv. 152 and 153; according to Al al-Dn AL-KHZIN al-Baghdd a: verses 152154, 91, 93,
and 94; according to Al al-Dn AL-KHZIN b: vv. 152154, 91, 93, 94, 114, and 20; according
to Al al-Dn AL-KHZIN c: vv. 91 and 142.
According to common interpretation, sra 13:13 and 14,
118
or verse 14 [i/162]
alone,
119
refer to mir b. ufayl and Arbad b. Qays,
120
heads of the Ban mir
b. aaa, who apparently wanted to kill the Prophet in 9/630 or 10/631 and
met an early death as punishment. For the same reason other verses con-
nected with verse 13sq., namely verses 11 and 12,
121
11, 12 and 15,
122
or 9 to
12,
123
are inferred to be Medinan. The fact is that these men negotiated with
Muammad in vain because of their association with the Medinan religious
state, and shortly thereafter met a strange end: mir fromplague and Arbad
by lightning strike.
124
Although the latters cause of death is supported by an
elegy of his stepbrother, the famous poet Labd b. Raba,
125
it is improper
to cite verses 13 and 14 in connection with this. They merely state the gen-
eral idea that Allhoccasionally has people die fromlightning. The simplest,
andtherefore likely the earliest, mentionof Arbad
126
does not say anything of
this revelation, nor do the many later accounts,
127
no matter how fancifully
they might be presented. We find still other accounts explaining that verse,
yet they too cannot be trusted. Verse 29 is dated 6/627, when the Meccans
turned down the request of the Muslims to make Bismillh al-Ramn al-
Ramthe preamble of Pact of al-udaybiyya, as they didnot knowthe word
al-Ramn.
128
Other scholars just as erroneously think that verse 31 dates
118
119
Ibn Qutayba, Liber posis, p. 151, l 10; al-Bayw.Hibat Allh [Ibn Salma] tells the
story without mentioning the verse.
120
EQ.
121
Fakhr al-Dn al-Rz.
122
al-Wid.
123
Ibn Hishm, p. 940 (not from Ibn Isq); al-abar, Annales, vol. 1, p. 1745sqq., Fakhr
al-Dn al-Rz, al-Wid, Hibat Allh, al-Maydn, ed. G.W.F. Freytag, vol. 2, p. 172sq.
124
Weil, Das Leben Mohammeds, p. 256sq.; Sprenger, Das Leben, vol. 3, p. 401. Caetani,
Annali, vol. 2, part i, p. 90sq., basedona comparisonof the accounts by IbnSad(Wellhausens
Medina vor dem Islam, p. 152) and al-Wqid (Wellhausens Muhammed in Medina, p. 306)
dates the embassy of the Ban mir before Jumd II of 8ah.
125
Die Gedichte des Labd ibn Rabah, ed. by Brockelmann and Huber, no. 25; Ab Tam-
mms amsa, p. 468. Cf. Ibn Hishm, p. 941, l 9; Ibn Qutayba, Liber posis, p. 151, l 9; al-
Aghn, vol. 15, p. 139, l 22. The scholiast is wrong when he thinks that the Arbad of the Dwn
der Hudhailiten, no. 106, verse 5 is Labds brother.
126
Ibn Sad in Wellhausens Skizzen und Vorarbeiten, Heft 4, p. 151 sq., is the only source I
knowof that is so ignorant as not to knowthat the two chieftains went to Muammad to kill
him.
127
Ibn Hishm, p. 940; al-abar, Annales, vol. 1, p. 1745sq.; Fakhr al-Dn al-Rz; al-Wid;
Hibat Allh [Ibn Salma]; al-Maydn in Georg W.F. Freytags Arabum proverbia, vol. 2,
p. 172sq.
128
al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Wid; al-Diyrbakr, Tarkh al-khams, juz
from this period and apply it either to the Muslim force camping outside
Mecca
129
or toMuammads campaigns ingeneral.
130
There is finally verse 43,
which is occasionally considered to be Medinan because of the expression
shahd, since both here and elsewhere shhid (sra 46:9) is applied to the
Jewish convert Abd Allh b. Salm.
131
The apparent Medinanoriginof isolatedverses has temptedsome writers [i/164]
toapply this tothe entire sra.
132
Inconnectionwiththis attitude verse 30sq.,
or verse 31which, as mentionedabove, some authorities consider tobe the
sole Medinan verse in an otherwise totally Meccan srais declared to be
of Meccan origin.
133
1, p. 12, and following this, G. Weil, Das Leben Mohammeds, p. 375. In al-Wid we find still
other explanations according to which the verse originates from Mecca.
129
Ab l-Layth al-Samarqand, and less precise al-Zamakhshar; and al-Bayw.
130
al-abar, Tafsr; Al al-Dn AL-KHZIN al-Baghdd.
131
Umar b. Muammad IBNABDAL-KF; Fakhr al-Dn al-Rz; Al al-Dn AL-KHZIN
al-Baghdd; al-Suy, al-Itqn, p. 26. In Fakhr al-Dn al-Rz, and al-abar, Tafsr, this
tradition is pronounced false.
132
Apart fromthe lists of sras (see above, p. 48sq.), also Umar b. Muammad; Hibat Allh
[Ibn Salma]; al-Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 32 (Qatda).
133
Umar b. Muammad IBN ABD AL-KF; al-Suy, al-Itqn, p. 32 (Qatda); Al al-
Dn AL-KHZIN al-Baghdd. In al-Suy, al-Itqn, pp. 26 and 32, and in the introduction
to Al al-Dn AL-KHZIN al-Baghdd we find the following list of different views regard-
ing sra 13: First.It is completely Meccan (al-Suy, al-Itqn, p. 26, and Al al-Dn AL-
KHZINal-Baghdd). Second.Completely Medinan (Itqn, p. 26, al-Khzin al-Baghdd).
Third.Meccan except verses 914 (al-Itqn, p. 32). Fourth.Meccan except verses 43 and
914 (al-Itqn, p. 26). Fifth.Meccan except verses 13 and 14 (Al al-Dn AL-KHZIN al-
Baghdd). Sixth.Meccan except verses 31 and 43 (Al al-Dn AL-KHZIN al-Baghdd);
Medinan except verse 31 (al-Itqn, p. 31). Seventh.Medinan except verses 30 and 31 (Al
al-Dn AL-KHZIN al-Baghdd).
THE MEDINAN SRAS
General Comments
Before starting with the discussion of these sras it might be a good idea to [i/164]
review briefly the Prophets circumstances before and after his emigration,
as well as his political position in Medina vis--vis the various parties. The
difference of those parts of the Koran that were revealed at Medina is
occasioned by the change of historical reality.
Political and Religious Conditions at Yathrib before the Hijra
At Mecca Muammad played the unenviable role of a prophet who ap-
pealed to only fewmen, mainly fromthe lowest strata of society, and whom
most people considered a fool or impostor, and who was protected against
personal libel only by his relatives in deference to indivisible family ties.
With his emigration he suddenly became a recognized spiritual and, soon
thereafter, temporal leader of a large community. In spite of Julius Well-
hausens great studies on pre-Islamic Medina,
1
it is not quite clear what
caused this remarkable change. For decades before the hijra Medina was the
scene of violent feuds among the two great tribes, the Aws and Khazraj. The
final great encounter in this struggle, the Battle of Buth, failed to establish
the supremacy of the victorious Aws and did not lead to an actual peace; on
the contrary, the insecurity of the city became even greater than before, as
the various blood-feuds were not officially settled and were left to personal
vengeance. That the inhabitants of Yathrib later became so quickly accus-
tomed to the supremacy of a stranger is certainly among the consequences
of an anarchical situation that must have become increasingly unbearable
as time went on. Nevertheless, contrary to what Leone Caetani argues,
2
this
cannot lead to the conclusion that the men of Medina, who had established
contacts with Muammad at Mecca, were politically motivated to pacify
their city, even if it is quite possible that such matters had been discussed.
1
Skizzen und Vorarbeiten, Heft 4, pp. 183; Das arabischen Reich und sein Sturz, pp. 115.
2
Annali dellIslam, vol. 1, p. 334.
136 the medinan sras
Althoughtradition
3
explicitly maintains the latter point of view, it empha- [i/165]
sizes above all the religious aspect, saying that Muammad had presented
his ideas to a group of men from Medina who had come to visit the Kaba
andhadencounteredreceptive hearts. After their returntoMedina, the men
presented such a lively propaganda for Islamthat within less than two years
a respectable community had been formed that was prepared to offer the
vindicated Prophet a new homeland.
The Unprecedented Success of Islamic Propaganda at Yathrib
In order to explain the unprecedented success of Islam at Yathrib it has
been pointed out that the Medinans must already have been familiar with
the main tenets of Islam, thanks to the large number of Jews living in the
city, as well as to the Christian Arab tribes residing in the vicinity, to whom
they were related in part. It is even noted that religious reformers such
as the Khazrajite
4
Aws b. mir al-Rhib appeared among them and had
followers.
5
This is undoubtedly correct. Even though similar people existed
in Mecca, and the religions of the People of the Book were not unknown
there,
6
nevertheless we must assume an incomparably stronger influx of
Biblical ideas at Yathrib. The Medinan peasants could well have been more
receptive to religion than the merchants of Mecca.
Muslimtradition thus remains correct in its estimation that the religious [i/166]
atmosphere at Medina was the decisive moment for the acceptance of
Muammad. In this case, the mutation from spiritual authority to political
leadership was not a preconceived idea but rather a result of the prevailing
conditions brought about by the clever manipulations of the Prophet, whose
political ability achieved here its first success. Not even two years after his
emigration he was able to dare to dictate a kind of constitution
7
to the
inhabitants of Yathrib. What he seemedto have hadinmindwas a theocracy
similar to that of Moses, where Allh and Muammad were the last resort
in all conflicts.
3
Ibn Hishm, p. 287, l 1; al-abar, vol. 1, p. 1210, l 6sqq.
4
It is nomere accident that the great majority of the first Medinanconverts toIslamwere
members of the Khazraj tribe. At the second meeting in al-Aqaba, fifty-eight participants
are said to have been from the Khazraj, whereas only eight from the Aws were present.
Cf. L. Caetani, Annali dellIslam, vol. 1, p. 321 sq.
5
Wellhausen, Medina vor demIslam, pp. 1517.
6
Cf. above, pp. 6 and 13; J. Wellhausen, Reste arabischen Heidentums, 2nd ed., p. 238.
7
Cf. A.J. Wellhausens basic study, Muammads Gemeindeordnung, pp. 6783, [and its
translation in A.J. Wensinck, Muhammad and the Jews of Medina, pp. 128138].
the medinan sras 137
The Waverers (munfiqn)
Among the factions we assume were at Medina only the true Muslims [i/167]
were unconditonally allied with him. This group consisted primarily of
the Meccan emigrants (muhjirn) and a not insignificant number of the
inhabitants of Yathrib, who had enthusiastically embraced Islam and dis-
tinguished themselves sufficiently to be named the Helpers (anr) of the
Prophet.
8
Still, many inhabitants of Medina harboured less than friendly
sentiments towards Muammad, neither recognizing him as a prophet nor
being inclined to accept himas a ruler. Because of his great, enthusiastic fol-
lowing they did not dare to take position against himopenly but rather met
him with a passive resistance that more than once crossed his plans. Their
influence was such that he had to treat them with respect and occasion-
ally even give in. This party of the munfiqn, the Hypocrites and waver-
ers,
9
was not particularly clearcut and simple. Even many who believed in
Muammad remained far from strict obedience, as the bonds of blood and
family, uniting its members and making them subservient to the authority
of an innate or chosen head, were extremely strong among the contempo-
rary Arabs as well as among all people under a patriarch. The reputation of
Abd Allh b. Ubayy b. Sall,
10
the most famous of the Khazraj, who outnum-
bered the Aws, especially counteracted this. Even after this man had lost his
direct political power, his influence was still large enough that Muammad,
who must have hated him dearly, was obliged until his death to treat him
with consideration and nearly as an equal. Otherwise Muammad would
easily have had the whole clan against him,
11
including the believers. The
8
The undeniable bravery displayed by the relatively small band of Muslims in their
struggle becomes obvious when it is realized that the only choice the emigrants had was
to win or to die, and that many of them, particularly the non-Quraysh, had an axe to
grind, but that the Medinans, used to intestine warfare between the Aws and Khazraj,
were accustomed to war and therefore more than a match for the Quraysh merchants in
their safe sanctuary. Added to all this was the growing religious fanaticism as a powerful
incentive.
9
Regarding the meaning and etymology of munfiq cf. above, p. 73.
10
EI
2
; Juynboll, Encyclopedia, pp. 505, 506, 721.
11
The best example of the semi-pagan, semi-Islamic mind of the new converts, torn
between unconditional obedience and irresistible force of family ties and the related blood-
feud is best demonstrated by the account of how, one day, the son of this Abd Allh (Ibn
Ubayy), a good Muslim, asked the Prophet for permission to kill his own father because of
a dishonourable remark; for, he said, if another were to do this, I cannot guarantee that I
might not be overcome by the bigotry of heathendom and avenge my fathers blood, be the
former a believer and the latter a pagan. (Ibn Hishm, p. 727sq.; al-abar, vol. 1, p. 1514sq.;
expression munfiq is occasionally extended also to include true believers,
if they became disobedient or lax in the performance of obligations for any
reason. The word very likely also referred to the multitude of those whoas
the great rabble always doessupported the Prophet in his glory yet were
inclined to desert him when things went wrong. The same must apply also
to the Arab tribes who, since the Pact of udaybiyya, and particularly since
the take-over of Mecca, went over to Muammad. Some of them became
true believers, but the greater part, among them the heads of the Quraysh,
particularly the entire Ban Umayya b. Abd al-Shams,
12
accepted Islamonly
forcedly or for personal advantage, although all of them were recognized as
Muslims for political reason only.
The Pagan Population and the Jewish Tribes
In their dealings with the Prophet, the energy displayed by the waverers [i/169]
[munfiqn] was far exceeded by that of the Jewish tribes living in Yathrib
proper or in the nearby oases. In addition to their mental superiority over
the Arabs, whichthey derivedfromanancient literary traditionregardless
of how little one may value their scholarship
13
there were also martial
bravery and other qualities
14
that enabled them, in the wonderful way of
all Jews, to become fully integrated without sacrificing their own identity.
In the beginning, Muammad placed great hopes on them as a people
who had already become familiar with the revelation. However, as they
were not inclined to give up their established view for the sake of the new
al-Wqid, ed. Wellhausen, p. 181 sq.; [Muir, Life of Mahomet, vol. 3, p. 240, note ]; al-abar,
Tafsr; and Fakhr al-Dn al-Rz on sra 63, etc.). It is conceivable that initially Abd Allh
did not stand up to the Prophet boldly enough or that he even supported him. An allusion
to this are the words which he later used with reference to Muammad and his followers:
fatten your dog and it will devour you. (Ibn Hishm, p. 726; al-abar, vol. 1,
p. 1512; al-Wqid, ed. Wellhausen, p. 179sq.; and the Commentators on sra 63. Cf. G. Freytag,
Arabum proverbia, vol. 1, p. 609). When his reputation then declined as the Prophets rose,
and the members of his own clan went over to the Prophet, he lamented his sorrowwith the
beautiful verses which we find in Ibn Hishm, p. 413.
12
These people were called
those whose hearts are brought together (in the

Koran only in 9:60, but frequently in traditions). Cf. also H. Lammens, Mowiya, p. 222; EI
2
.
13
The Jews appear totally Arab in their poems, of which somepartly very beautiful
fragmentscan still be reconstructed from the Kitb al-Aghn. Cf. Nldeke, Beitrge zur
Kenntnis der Poesie, pp. 5286. It may be mentioned in passing that a great many of these
Jews were convertedArabs; cf. Wellhausen, Medinavor demIslam, p. 15. Cf. alsoJ.A. Wensinck,
Mohammed en de Joden te Medina, pp. 4144, and its translation, Muammad and the Jews of
Medina, pp. 2931.
14
Cf. Wellhausen, Medina vor demIslam, p. 12sq.
prophet,
15
whose great deviationfromtheir ownfaithwas for themfar easier
to discern than for him, the discord increased and did not cease until all the
Jews were either killed, banned, or subjugated. The hostility of the Jews was
all the more dangerous because they hurt Muammad not only by means of
war andpolitics but alsoby mockery andbiting interrogationabout religious
matters.
16
If their various tribes had not been divided into two parts by the
fightings of the Aws and Khazraj, Muammad would have hardly been able
to decimate them one after the other.
Content and Style of the Medinan Sras
The pagans, against whom open warfare raged during the Medinan period, [i/170]
are nowonly rarely the target of verbal attacks. The Christians too, who after
all were living far from Yathrib, and with whom Muammad had hostile
encounters only during his last years, he seldom mentions, although when
he does it is rather congenially, accompanied merely by the disapproval of
certain dogmas. Muammads assaults upon the Jews, on the other hand,
are quite violent. After the hijra, this takes the formof his endeavour to show
that they had always been obstinate and are therefore damned by God. The
waverers [munfiqn], too, are often severely reproved, although Muam-
mad frequently had to be more considerate when dealing with them. In
the Koran he gives free rein to his feelings, albeit without divulging names.
Moreover, the Prophet addresses here almost exclusively the munfiqn of
Medina. The other Arabs, who likewise hadadoptedIslamonly superficially,
he attempted to win over with kindness instead of discouraging them with
strong measures and words.
Finally, these sras are directed at the Muslims, although seldom with a
view towards lecturing on dogmatic or moral articles of faith, subjects they
were sufficiently familiar with from the Meccan sras. Instead, he speaks
15
With the exception of some very few like Abd Allh b. Salmwho must therefore
serve Muslims in general as the example of a faithful Jewlike Abd Allh b. Ubayyas the
example of a Hypocrite [munfiq],or like Ab Jahlas idolateretc., and therefore often
mentioned at the wrong place.
16
They, for example, hurled interjections against the Prophet.
(al-Kalb inal-Wid onsra 13:38). (Sothey asked

him: God created the world; who, then, created the creator?) Of course, they themselves
had a captious answer ready and only wanted to find out whether Muammad could match
it. Muslims, however, see in such questions nothing but the disbelief and wickedness of the
Jews. This question is typical and is put into the mouth of disbelievers and sceptics; cf. Ab
Dwd (al-Sijistn), Sunan, vol. 2, p. 178.
on home ground and as their leader in the field, rebuking or praising, as
the case may be, particularly after victory or defeat, putting events in their
proper perspective, planning the future or issuing orders and laws. These
legal revelations are of particular importance. Some of themare intendedfor
the moment, while others are for eternity. They settle civil andritual matters
without precisely defining them. Just as the Koran follows the dictates
of the moment rather than a defined system, so also many of these laws
evolve from decisions on contentious issues. Muammad often added to
the verdicts further regulations regarding issues that could possibly arise.
It is unlikely that such a statute-book
17
evolved totally without concrete and
actual issues. Several laws and ordinances also refer to the domestic affairs
of the Prophet.
The new subjects which appeared on the Prophets horizon, and which [i/171]
were dealt with in the sras, would have had to entailone would think
considerable deviations from the style of the last Meccan period. Never-
theless, this is generally not the case. New expressions and idioms were
nearly never applied unless absolutely required by the subject. This is most
evident in those laws in whose formulation all rhetorical embellishment
is avoided. It is only the rhymeconsisting here often of totally superflu-
ous additions, therefore at times annoyingto which Muammad remains
enslaved. Since he seldomas at Meccaaddresses the people in general
but rather the respective parties separately, it is very rare here to find the
address O, you people! In contrast, we frequently find O, you believers!,
and less frequently O, you Jews!, O, you waverers [munfiqn], etc. Inci-
dentally, we also find in these sras the odd powerful and even poetic pas-
sage.
18
In general, the Medinan revelationswhich consist of rather brief
laws, addresses, orders, etc.are originally of a smaller volume than the
majority of those from the late Meccan period, which tend to be lengthy
lectures. On the other hand, the uniformity of the content resulted in dis-
proportionately many single, Medinan revelations being put together to
produce a single, collective sra, so that the current Medinan sras became
the longest in our Koran.
The development of the linguistic parlance as it appears before the emi- [i/172]
gration can later be demonstrated only in isolated instancesif at all. How-
ever, this very unreliable device can easily be done away with, as content,
17
This origin explains on the one hand the simplicity and the common sense usually
embedded in these laws and, on the other hand, the contradictions which, naturally, cannot
be avoided in systematic codifications either.
18
Cf., e.g., sras 2:16sqq., and 2:266sqq.
steady reference to established events or circumstances, and close connec-
tion with the development of the new state supply much better guidance.
Anyone studying the history of Muammad realizes immediately the differ-
ence between the transmission of events before and after the emigration. In
the former case, only a few reliable memoirs from a small circle are avail-
able, with uncertain chronology and many legends, whereas in the latter
case, pure history predominates, enabling us to trace the events from year
to year. We can thus establish a chronology of the Medinan sras with accu-
rate details. Of course, much still remains uncertain; for many a period we
have but approximate chronological limits of origin, and there remain still
others where it can only be stated that they originate from the Medinan
period.
19
Comments on Sras 2, 98, 64, 62, 8, 47, 3, 61, 57, 4, 65, 59
Even if it is conceivable that some passages revealed soon after the hijra [i/173]
have later disappeared or were discarded by the Prophet himself, we must
concur with Muslims that from among the remaining sras the second
sra is the earliest of the Medinan sras.
20
Its greater part originates from
the second year after the hijra, i.e., from the period before the Battle of
Badr. The first part, verses 1 to 19 (up to ) is the only one of all the
Medinanrevelationssimilar to many a late Meccansrathat starts with
the words .
21
Muslims are divided as to whether or not these
verses refer to the Jews or the munfiqn,
22
but that the latter are meant
becomes evident from the verses 7sqq. Yet as Muammad does not say
19
Muir presented his view regarding the Chronological order of the suras revealed at
Medina in his Appendix to vol. 3, pp. 311313, which, however, is extremely brief and
disregards details; he does not present anything new. He arranges the Medinan sras as
follows, without concealing the fact that this is only an approximate arrangement, and that
some sras contain sections originating fromentirely different years: Sra 98 (8 verses); sra
2 (287 verses); 3 (200 verses); 8 (76 verses); 47 (28 verses); 62 (11 verses); 5 (120 verses); 59 (24
verses); 4 (175 verses); 58 (22 verses); 65 (13 verses); 24 (65 verses); 63 (11 verses); 57 (29 verses);
61 (14 verses). He assigns only the following to the last five years: Sra 48 (29 verses); 60 (13
verses); 66 (13 verses); 49 (18 verses); 9 (131 verses). Muir omitted sra 33.What needs to be
said regarding the classifications of Grimme and Hirschfeld see above, p. 61 sq.
20
See above, p. 48, the list of sras; al-Suy, al-Itqn, p. 56.
21
Cf. sras 7:1, 11:1, and 14:1.
22
Particularly Muammadibnal-Sibal-Kalb mentions the Jews inthe individual verses.
Cf. the Commentators, in particular Ab l-Layth al-Samarqand, and al-abar, Tafsr; less
precise is Ms. Sprenger, no. 404 [= Ahlwardt, no. 732: Abd Allh b. Abbs al-Hshim, Tafsr
al-Qurn].
that they refused either to fight or pay the community tax
23
generally the
main complaint about themit is probable that the verses originate, if not
indeed fromthe very first period, still rather early, possibly the beginning of
2/623. The following part, verses 19 (starting with ) to 37, shows no
obvious traces of a Medinan origin; rather it contains indications of a Mec-
can origin. In the first verses the Prophet is speaking against the idolaters,
which many Muslims also recognize.
24
The following verses deal with sub-
jects that are frequently toucheduponinthe Meccansras but never appear
in the Medinan sras. This part, howeverwhere, as frequently before the
hijra, the story of creation and the fall of man are relatedappears as the
introduction to a larger Medinan piece that tries to demonstrate to the Jews
their godlessness from time immemorial. Although this could only have
originatedsome time after the hijra, whenthe ill will of the Jews hadbecome
apparent, there is nothing to indicate that Muammad had already fought
against them. Some verses clearly point to the time when the direction of
prayer was changed from Jerusalem to Mecca.
25
This is in agreement with
23
Also here, as in the entire sra, the word still has not yet appeared.
24
Cf. the Commentators, none of them directly claims that this verse is Meccan; all they
say is that it is addressed to the Meccans (al-Wid).
25
The information from the traditions about this event differ from one another by only
one month. Many fix it as follows:
1. In the month of Rajab, 2/623 (Ibn Hishm, p. 381; Ibn Sad, cod. Gotha, I, 261 [sic]; al-
abar, Tafsr, vol. 2, p. 3; al-Zamakhshar; al-Bayw; Fakhr al-Dn al-Rz; Hibat Allh
b. Salma, al-Nsikhwa-l-manskh, where he declares this date to be the commonone;
Izz al-Dn IBN AL-ATHR, Usd al-ghba f marifat al-aba, vol. 1, p. 22, between the
months of RajabandShabn; al-alab, Cairoed., 1280, vol. 2, p. 297, variously between
Rajab, Shabn, and Jumd II), or
2. At the end of the 16th or beginning of the 17th month after Muammads arrival in
Medina (al-abar, vol. 1, p. 1280, and his Tafsr; Ibn Sad, cod. Gotha, vol. 1, 263, IX
[sic]; al-Bukhr, K. al-alt, 31; Muslim, K. al-alt, 50; al-Nas, K. al-alt, 24,
K. al-Qibla, 1; al-Muwa, p. 68, with the addition two months before the Battle of
Badr, thus, as above in Rajab, cf. also al-abar, Tafsr, loc. cit.), or
3. In the beginning of the 18th month (al-Azraq, p. 265; Ms. Sprenger, 404; al-abar,
Tafsr, vol. 2, pp. 3 and 8; Izz al-Dn IBNAL-ATHR (al-Kmil) vol. 2, p. 88, whereas Ibn
Hishm, p. 427, l 12, al-abar, vol. 1, p. 1279sq., and al-Yaqb, Historiae, vol. 2, p. 42,
add the name of the month, Shabn. This month only is also given by al-Masd, K.
al-Tanbh wa-l-ashrf, p. 237, l 1).
4. In the beginning of the 19th month (al-Masd, Les Prairies d or, vol. 4, p. 141; al-abar,
Tafsr, vol. 2, pp. 3 and 12; al-Yaqb, loc. cit.; Hibat Allh b. Salma, Cairo ed., p. 40; Ab
l-Layth al-Samarqand).
5. Others vary between sixteen and seventeen months: (al-Bukhr, K. al-alt, 31, and
K. al-Tafsr; Muslim, loc. cit.; al-Tirmidh in K. al-Tafsr and K. al-alt, 139; al-Nas,
K. al-alt, 24; Fakhr al-Dn al-Rz; Ab l-Layth al-Samarqand).
Only some isolated traditions fix the duration of the Jerusalem qibla at Medina:
the entire passage, and we may therefore fix it at the time when Muam-
mad decided on the change, i.e., the first half of 2/623. Verse 59 expresses
the idea that all depends on faith, in which respect a Jew is no better than
a Nazarene or Sabian. Such a version fits well into the context. Yet those
ideas are so vaguely expressed in the transmitted version that one is easily
tempted to suspect an interpolation. This, however, is difficult to accept, as
nearly the exact same verse is found in a conspicuously similar context in
sra 5:73.
Verses 2:70 to 76 are addressed to the Muslims, however with reference to
the Jews. Verses 2:88 to 90, and 91 to 97, which aim at the ungodly sayings
6. To fifteen months after the completion of the mosque (al-alab, loc. cit.) and after
the arrival in Medina (Izz al-Dn IBN AL-ATHR (al-Kmil) vol. 2, p. 88) respectively,
or
7. To fourteen months (al-alab), or
8. To thirteen months (al-abar, Tafsr, vol. 2, p. 3; Hibat Allh b. Salma, al-Nsikh
wa-l-manskh, Cairo ed., p. 40), or
9. To nine or ten months (al-abar, Tafsr, loc. cit.). This information is possibly based
on an inaccurate text; or, finally,
10. Totally general to ten and some months (al-abar, Tafsr, vol. 2, p. 13, l 9).
Here we cannot dwell on the genesis of these different calculations which occasionally
include the date of the month and the day of the week. Suffice it to say that the number
of sixteen or seventeen months is best documented.
The reason for the change of the Jerusalem direction of prayer is based on the new
positionregarding the earlier revealedreligions, whichMuammadgradually came torealize
at Medina. Whereas he formerly felt closest related to Jews and Christians, the futility of the
propaganda among them prompted him to look for another connection, which he finally
discovered in the religion of Ibrahim whom a revelation associated with the Kaba (cf.
above, p. 119). The pagan place of worship became the sacred Islamic territory and as such
suitable for a direction of prayer, as Jerusalem was for the Jews. The Meccan qibla, which
according to his theory of the religion of Ibrhm, Muammad must have considered to be
the only correct direction, not only increased Muslim self-confidence by the erection of a
newbarrier against the anti-Islamic Jews, but it also facilitated propaganda among the pagan
tribes.
According to a prevalent opinionWeil, Das Leben Mohammeds, p. 90; Muir, vol. 3,
p. 42sqq.; Grimme, Mohammed, vol. 1, p. 71, andhis Mohammed, die welt-geschichtliche Bedeu-
tung, p. 64; Caetani, Annali, vol. 1, p. 466sq.; Fr. Buhl, loc. cit., p. 212the Jerusalemqibla was
not instituted until Yathrib in order to win over the hearts of the numerous Jewish popula-
tion ( ). Although the many quotations in the first part of the foot-note are unimportant
for this question since nearly all of them are limited to the information from the Medinan
period, we do find this tendency clearly expressed in the Commentators on sra 2:136sqq.,
and some historical works, e.g., al-Diyrbakr, Tarkh, vol. 1, p. 367; al-alab, vol. 2, p. 297sqq.;
al-abar, Chronique, vol. 2, p. 477. These accounts, however, are not reliable because they con-
tradict our oldest and best sources (Ibn Hishm, pp. 190, 228, 294sq.; al-abar, vol. 1, p. 1280;
al-Azraq, p. 273 according to al-Wqid; Ibn al-Athr (al-Kmil), vol. 2, p. 88) according to
which Muammad had turned in prayer towards Jerusalem or Syria even before the hijra.
The Koran is silent on the matter because the Meccan passage, sra 10:87, proves only that
of individual Jews,
26
might also be attributed to the same time. Verses 2:98 to
115 probably date fromthe time shortly before the institution of the Meccan
qibla, because 2:100 probably refers to the abrogationof previous legislation.
Verse 2:108 refers to the Medinan opponents of the Prophet, who were dis-
turbing the Muslims in the exercise of their ritual service, even attempting
to demolish their meeting-places. Verse 109, which states how meaningless
is the direction of prayer
27
for the believers, seems to attack the qibla of
the Jews, yet verse 110 can only refer to Christians.
28
That the Kaba and the
religion of Abraham is to be given much preference to Judaism
29
Muam-
mad tries to prove in verses 116 to 135. What was more or less alluded to in
these and the preceding verses Muammad finally says openly in verses 136
to 145, which prescribe to the Muslims turning the face during all prayers
in the direction of the Kaba. At the same time, the Koran predicted that
the concept of qibla had been known to Muammad at that time. But the credibility of those
traditions is supported by internal evidence. Since Muammad derived not only the name
for the ritual prayer (alt) but also several of its formulae and rites from the older revealed
religions, it would be a surprise if he had not followed their example andconcurrently,
or certainly very soon thereafterinstituted a direction of prayer. Sprenger (Leben, vol. 3,
p. 46 n. 2) and, more recently, Wensinck, Mohammed en de Joden te Medina, p. 108 [and its
Englishtranslation, p. 78], concur withthe conclusionthat Muammadhadalready adopted
this custom at Mecca and completely followed the Jewish practise by using Jerusalem as the
qibla. We knowthat the Jews turned in prayer towards Jerusalem(I Kings 8:44 and 48; Daniel
6:11; III Ezra 4:58; Mishna, Berkht pereq, 5 and 6; Ibn Hishm, p. 381), the Christians of the
first centuries, on the other hand, turned towards the east (cf. J. Bingham, vol. 5, pp. 275280;
H. Nissen, Orientation, vol. 2, pp. 110sq. and 247sq.). But this is not to say that the Jerusalem
qibla was adopted by Muammad as a specific Jewish institution. It is conceivable that he
encountered it also among the Arabian Christian communities who are known to have been
under considerably Judaic influence.
During Muammads Meccan period the Kaba can hardly have served as ritual direction
of prayer sinceas we have seen aboveit did not become a recognized sanctuary of
Islam until his Medinan period. Equally little convincing is the tradition, which looks like
an attempted compromise, claiming that the Jerusalem qibla was observed only by those
Medinans who had been converted shortly before the hijra (Ibn Sad, cod. Gotha, vol. 9 [sic],
s.v. Kab b. Mlik; al-Baldhur, Fut al-buldn, p. 2; Ab l-Layth al-Samarqand).
26
Cf. the Commentators.
27
That verse 109is aiming at the qiblais alsomaintainedby many Muslims; cf. al-Tirmidh,
K. al-Tafsr; [Abd Allh b. Abbs al-Hshim, Tafsr al-Qurn,] Ms. Sprenger, no. 404 [ =
Ahlwardt, no. 732]; Hibat Allh b. Salma; Ab l-Layth al-Samarqand; al-Wid; al-abar,
Tafsr; al-Zamakhshar; al-Bayw. They all supply also all sorts of other explanations.
28
Like all the passages dealing with Christians, the verses 107 and 110against all proba-
bilityare brought in connection with the embassy to Najrn. (Ab l-Layth al-Samarqand;
al-Wid; al-abar, Tafsr; al-Zamakhshar; al-Bayw.)
29
Cf. Snouck Hurgronje, Het Mekkaansche feest, p. 33sqq., as well as the discussion above
on sra 16:124.
many people would object to this.
30
Verses 148 to 152 are of a later date.
Some exegetes apply them to the believers
31
killed at Badr. Since it is indi-
cated in verse 150, however, that the Muslims fortune in war was low, this
point of viewis unlikely, and it is preferable to accept the viewof al-ak
b. Muzim, who thinks this refers to the casualties of the Battle at the
Uud near Bir Mana.
32
Verses 154 to 157 canbe connected withthem, since
those who persevere and receive Gods blessings (verse 152) are compared
with the disbelievers who are cursed by all (verses 154 and 156), both those
who die in the way of truth, and those who wither away in their malice
(verses 146 and 156). The detached verse 153 fits better after the pericope
2:185199, concerned with the ceremonies of the pilgrimage. As it readily
follows from the text, this refers to the Muslims reservations about partic-
ipating in the traditional run between the hills of af and Marwa.
33
We, as
well as tradition, cannot say on what occasion it was revealed, but there
is something that would point to the minor ajj of 7/628. Verses 158 to
162 are of Meccan origin, probably from the opening of a sra, with the
result that only a fewwords or verses would be missing fromthe beginning.
With them we might have to connect verses 196 (starting with ( ))
to 198, and 200 to 203,
34
which are frequently inaccurately interpreted by
Muslims.
35
To the Meccan period also belong verses 2:163166, which are
directed against the idolaters who follow such things as [they] found their
fathers doing. This part precedes the Medinan verses 167 to 171, since both
of them deal with prohibited food. These verses could possibly contain a
veiled retaliation against the Jews, who demanded the observation of the
Mosaic dietary prescriptions from the Muslims, and would thus, like many
other parts of our sra, fit well into the time when Muammad dismissed
Jewish customs. Verse 172 is to suggest to those who object to the change
of the direction of prayer that the outward demonstration of such habits
is far less important than true piety; we must assume the verse to have
30
Verse 2:136.Regarding the connection of 2:118sqq. with 2:136sqq. cf. Snouck Hur-
gronje, loc. cit., p. 38sq.
31
So Muammad b. al-Sib al-Kalb with reference to Ab l-Layth al-Samarqand, and
[Abd Allh b. Abbs al-Hshim, Tafsr al-Qurn], Ms. Sprenger, no. 404 [ = Ahlwardt,
no. 732]; cf. al-Bayw.
32
33
al-Bukhr, K. al-ajj, 80 and 162; Muslim, K. al-ajj, 39; Ab l-Layth al-Samarqand;
al-Wid; al-abar, Tafsr; al-Zamakhshar; al-Bayw, etc.; cf. Snouck Hurgronje, loc. cit.,
p. 117.
34
Especially fitting is the expression ( ) , verses 160, 196, and200, also the rhyme.
35
See Ibn Hishm, p. 642, and the Commentators.
been revealed shortly after this event. This is followed by three laws in
verses 173 to 181, which are not only of the same length, i.e., three verses
each, but also of the same beginning, i.e., the words
so that we can
hardly doubt their original homogeneity. Their origin must be fixed at the
time immediately before Raman 2/623,
36
when the second law instituted
fasting during this month for the first time,
37
i.e., essentially at the same
time as the above-mentioned parts of the sra. Verse 182 constitutes the
end of these laws. Verse 183 must certainly belong to a later period, because
not only is it far more precise than the rest of the laws but it mentions
in particular that Muslims repeatedly exaggerated abstinence during the
month of fasting. We must consequently consider this a later addition to
that law. Verse 184 seems to be a fragment of a larger revelation. Verses 185
36
Ibn Sad, cod. Gotha, I, p. 261 and 266 [sic]. The traditions supplied by al-abar in his
Tafsr mention a general situation only. It is totally wrong in al-abars Chronique, vol. 3,
p. 126, to fix verse 181 at the time of the conquest of Mecca, which happened in the month of
Raman.
37
According to the unanimous claim of tradition, the fasting during Raman replaced
the shr fasting. As far as the latter is concerned, the traditions are divided. Some peo-
ple(Muwa, p. 91; al-Bukhr, K. al-awm, beginning and end, K. Bad al-khalq, 157,
bb ayym al-jhiliyya, K. al-Tafsr; al-Tirmidh, K. al-awm, 47, Shamil al-nab, 43; the
Commentators)recognize it as an old Meccan festival, others see it as an innovation,
which Muammad did not adopt from the Jews until his Medinan period (al-Bukhr, K. al-
awm, beginning and passim; al-abar, vol. 1, p. 1281; Ibn al-Athr, Chronicon, vol. 2, p. 88;
al-Diyrbakr, Tarkh al-khams, vol. 1, pp. 360 and 368; al-Khab al-Tibrz, Mishct, bb
iym taawwu, fal 3, beginning). In favour of the first view is the fact that Muarram is
an old sacred month, during the first ten days of which still modern Muslims observe all
sorts of superstitious customs. There is no reason to doubt the great age of this institution
since also the first ten days of Dh l-ijja are known for their particular sacredness. On the
other hand, it is extremely unlikely that already the pagan Meccans observed the fast of the
Day of shr as the above-mentioned traditions claim. This we learn particularly fromthe
linguistic observation according to which the word shr represents in Arabic an isolated
nominal formation, corresponding exactly to the Jewish sr tenth day, with the Aramaic
determinative ending . Quite rightly shr is identified with the Jewish Day of Atonement
on10Tishr as the most sacredof days of fasting inthis context. But since generally the first ten
days of Tishr are considered days of atonement for times immemorial, the above-mentioned
sacredness of the corresponding days of the Islamic calendar might possibly be also of Jewish
origin.
As European biographers of the Prophet suspectGustav Weil, Das Leben Mohammeds,
p. 90; W. Muir, vol. 3, p. 47sq.; Sprenger, Das Leben und die Lehre, vol. 3, p. 53sq.; Fr. Buhl,
Muhammeds Liv, p. 212; H. Grimme, Mohammed, vol. 1, p. 55; L. Caetani, Annali, vol. 1, p. 431 sq.,
and 470sq.when adopting the Jewish fasting as well as the qibla, Muammad was guided
by the purpose of winning over the Jews to his religion. This is not improbable, but not
necessarily dependent on the question whether this fasting was introduced already at the
end of his Meccan period or only at the beginning of his Medinan period.
The idea to replace the fasting during Muarram by another one in Raman might be
related to the well-known Night of Power (laylat al-qadr) of this month. But what made
to 199with the exception of the possibly Meccan verses from 196 b
to the end of 198constitute a conglomerate of ordinances, all relat-
ing to the sacred territory of Mecca. This puts their Medinan origin beyond
doubt but does not say anything about their chronology. For the time being it
can only be stated that, on the one hand, they belong to the period after pas-
sages like 16:124, and 2:119 and 121, where the basic position of Islamtowards
the Kaba is established. On the other hand, they can be seen as referring
to the pilgrimages (ajj) and visits (umra) Muammad made in the years
ah6, 7 and 10. Verse 185 recalls an ancient custom observed during the ajj.
If we disregard the origin of the actual contextin the case of this verse as
well as the verses 193, 196b, and 199, which lack a contemporary reference
then we must abandon any attempt to fix a date for them.
38
Verses 186 to
189 recommend religious war and permit the believers to defend them-
selves with weapons against the Meccans, even in the sacred territory of
the Kaba. Doubts arise as to whether this part belongs to the time shortly
before the conquest of Mecca or to the campaign of al-udaybiyya, since
during the preparations for the pilgrimage of 6/627 Muammad was ready
for violent complications. However, according to Snouck Hurgronjes
39
keen
Muammad replace a one-day fast with a fast lasting an entire month? This unbelievable
extention, like the change of the qibla, cannot be regarded as either emanating fromthe basic
Islamic tenets, or frompagan institutions, or as a free and arbitrary invention. Sprenger (Das
Leben und die Lehre, vol. 3, p. 55) sees in it an approach to the Christian Quadragesima, cf.
also L. Caetani, Annali, vol. 1, p. 471). As far as the time is concerned, it pretty much coincides,
provided the institution of the fast of Raman took place either in ad624 or 625 (Sprenger,
loc. cit.), but it does not, if this happened already in ad623 as it would seem from a note in
al-Wqid, p. 41 (J. Wellhausen, Muhammed in Medina, p. 46, cf. G. Jacob, Der muslimische
Fastenmonat Ramadn, p. 5). This, of course, presupposes that the fast of Lent did indeed
last forty days, but it is doubtful that this duration prevalent in the greater Church applied
to the obscure sects of the Arabian Peninsula. There is a fundamental difference in the type
of fasting. The Church requires only abstinence of certain food, whereas Islam prescribes
absolute abstinence during the daytime but no fasting at night. This particular type, as far
as I know, can be identified only among the Christian sect of the Manichaeans whoafter
Flgels edition of al-Fihrist, p. 333sqq., when the newmoon begins to shine, the sun is in the
sign of Aquarius (about 20th of January), and eight days of the month have passedfast for
thirty days, however, breaking the fast daily at sunset; (cf. Flgel, Mani, p. 97, andnote, p. 245;
K. Kessler, Manichaeer, p. 213). Perhaps there were Christian sects in Arabia who observed
the fast of Lent in the same way.
38
Cf. Snouck Hurgronje, Het Mekkaansche feest, pp. 49sq., 80, and 135. Al-Azraq, p. 124
and al-abar, Tafsr, vol. 2, p. 105, l 29sqq., put verse 185 into the year of al-udaybiyya.
39
Snouck Hurgronje, loc. cit., pp. 51 sqq., 115sq., 14.Verse 190 has been revealed accord-
ing to Ibn Hishm, p. 789, l 2 (not from Ibn Isq) during the minor ajj 7/628 (called
,
or ). Verse 192according to al-Wqid, edited by Wellhausen, p. 244; al-

Bukhr, K. al-ajj, 179, K. al-Maghz 27; al-Tirmidh, K. al-Tafsr; al-abar, Tafsr, vol. 2,
p. 131, l 2; al-Wid; Hibat Allh b. Salmain the year of al-udaybiyya.
elaborations, it seems certain that the verses 190 and 192a belong to the
periodshortly before this campaign. As they primarily deal withthe ajj, and
as it is obvious fromthe mention of the sacrificial animals, a reference to the
umra of 7/628 is excluded. For good reason Chr. Snouck Hurgronje
40
holds
verse 192b to be an addition from the time of the so-called Farewell Pil-
grimage (10/631), when Muammad performed the ajj as well as the umra,
using the completion of the latter as a welcome chance to relinquish the
state of ritual consecration (irm) and satisfy his desire for a woman. The
most respected of his companions, primarily Umar, disapproved so vio-
lently that Allh had to justify His Messenger with a new revelation.
Since verse 2:207 is explicitly presented as a question to be put to the [i/182]
Children of Israel, tradition applies verses 204sqq. to Muslims wanting to
observe Jewish laws.
41
This passage could thus be concurrent with verses 100
and 179sqq., where Jewish customs are rejected. The text, however, speaks
only in very general terms of the fall and temptation of Satan, whereas the
allusionin2:206is unfortunately incomprehensible.
42
Verses 221 and214219
contain answers to all sorts of questions addressed to the Prophet. Of these,
only verses 216 to 219 (Flgel) are certainly from the same time, since they
actually constitute only one verse.
43
Verse 211, which starts with this very
question with which verse 216 (Flgel) closes, must belong to a different
period. It seems as if traditions properly apply verse 214 to Abd Allh b.
Jash (of Ban Asad b. Khuzayna)
44
and his men, who on one of the last
40
Loc. cit., pp. 49, 8392. Cf. al-Bukhr, K. al-ajj 34; Muslim al-Qasalln, verses 274,
277, 281, 285, 288, 289 (K. al-ajj 16), and the other passages supplied by Snouck Hurgronje.
41
al-abar, Tafsr; al-Zamakhshar; al-Bayw; Ab l-Layth al-Samarqand; al-Wid.
42
For this reason Hirschfeld, New researches, p. 144, might be correct when he considers
these verses to be Meccan.
43
Verse 216 is generally held to be the oldest passage of the Koran that prohibits wine.
Frowning uponthe use of intoxicants is anascetic andpietist dispositionoriginating fromthe
Christian environment (Severians, Manichaeans) and is also found in Musaylima (al-abar,
vol. 1, p. 1916, bottom). The later unconditional prohibition, whichis muchstronger expressed
in tradition than the Koran, has become an immense blessing for the Islamic world. In
the afore-mentioned verse Muammad is still an opportunist, possibly even more than the
present text permits, because is conspicuous in the context, and seems to have been
insertedlater. Inany case, shortly after the Battle of Badr; but before the expulsionof the Ban
Qaynuq, some Helpers [ ] came together for a cheerful drinking-bout inthe house of the
female singer ( ) in the course of which the Prophets uncle, amza b. Abd al-Mualib, got
quite drunk and performed pranks (al-Bukhr, K. al-Maghz, 12; Ibn Badrn, ed. Dozy,
p. 139sq.). As sra 4:46 seems to indicate, the consumption of wine was originally prohibited
only before a religious service, cf. below on sra 4:46, and I. Goldziher, Muslimstudies, vol. 1,
p. 28sqq.
44
Encyclopedia of Islam.
days of the sacred month of Rajab ambushed a Qurayshite caravan near
Nakhla and killed the leader.
45
Verse 222, formally related to those questions,
now stands detached. According to tradition, it is also directed against a
custom of the Jews, causing them finally to say: This man really does not
want us to keep any of our institutions; he objects to everything.
46
Not
much attention needs to be paid to this, as the principles on menstruation
developed in this verse are entirely identical with the Jewish counterparts.
The introductory marital ordinances of verses 220sq., 223 to 238, and 241 to
243, offer no references for chronological purposes. Verse 241 might have
originated still before the Battle at the Uud (4/625) if it were certain that
sra 4:12sqq. is an alteration of it. Yet perhaps this concerns only a special
case that is not at all considered in sra four.
47
Nothing can be said about
verses 239sq.; they probably originate from the time before the institution
of the prayer of danger,
48
i.e., before 4/625. Verses 212sqq., which according
to sense and traditionconstitute but one single verse, might possibly have to
be connected with verses 245sq. Thus, yet another lawis created, consisting
of three verses beginning with
, which we may fix to the same

time as verses 173 to 182 above.
49
The time shortly before the first battle is
also quite suitable for verses that consist of the concise command to fight.
Connectedwiththis are certainly verses 244, and247 to 257, whichincite the
Muslims tocourage andobedience by way of examples fromIsraelite history.
Obviously Muammad realized that open conflict with his countrymen
could no longer be avoided. Possibly connected with this are the stories
about verses 260 to 262, which, similarly to verse 244,
50
and with reference
45
Ibn Hishm, p. 423sq.; al-Wqid, p. 8sqq. (Wellhausen, p. 34 and Vorbemerkungen,
p. 11 sq.); al-abar, vol. 1, p. 1272sqq.; al-abar, Tafsr, as well as other Commentators on this
passage and on sra 4:46; Weil, Das Leben Mohammeds, p. 98sqq.; Sprenger, Das Leben, vol. 3,
p. 105sq.; Caetani, Annali, vol. 1, p. 463sqq. The poem in Ibn Hishm, p. 427, l 5sqq., is a
poetical paraphrase of that verse (sra 2:214).
46
Muslim, K. al-ay, 3; al-Tirmidh, Tafsr; al-Khab al-Tibrz, Mishct, Bb al-ay,
beginning; al-Nas, K. al-ay, bb 4. According to Ibn ajar al-Asqaln, vol. 1, no. 873,
al-Wid, etc., Thbit ibn Dad, [Juynboll, Encyclopedia, p. 495, col. 1] who died either at
the Uud or in 6/627, consulted the Prophet in this matter.
47
This problem can only be clarified in connection with a detailed study in the marital
regulations of the Koran. The most recent thesis on the subject (R. Roberts, Familienrecht im
Qoran) ignored the subject.
48
See below on sra 4:102sq.
49
One might argue that the content of verse 246 is not closely enough related to that of
verse 245 in order to reckon them to be one short law. However, the expenditure for war and
the fight itself are often brought into closest connection with one another in the Koran. Both
is , the former , the latter .
50
Worth noting is also .
to Gods resurrecting power, incite them to defy death. Verses 253sq. seem
to complete that section. Verses 255 to 259
51
cannot easily be attributed
to any specific situation, because the axiom that there is no compulsion
in religion (2:257) could have been established in a state of psychological
depression as well as in the expectation of greatest victory. Moreover, in
the Medinan period such an utterance was practically of little importance,
since in this case purely religious propaganda was subservient to politics
aiming at the recognition of sovereignty. Equally vague remains the dating
of verses 263 to 281, containing a request for alms and a prohibition of
usury. Some traditions apply verses 278sqq. to the outstanding debts of the
inhabitants of al-if to some rich Quraysh; others hold verse 278 or 281
and verses 278 to 281
52
respectively for the last part of the entire Koran,
revealed at the Farewell Pilgrimage,
53
because of the money Abbs (IBN
ABD AL-MUALIB) and others had borrowed against the payment of
interest. But all this lacks sufficient evidence. At best, one can agree that the
verses were once applied to the respective cases. Verses 282 to 284, which
deal in a very awkward way with the procedure of borrowing, can hardly
originate from the early years of the hijra. Verses 285sq. might equally be of
Meccan or Medinan origin.
54
Approximately concurrent with the greater part of this sra might be a [i/185]
few short ones.
51
Verse 256 is called yat al-kurs, i.e., Verse of the Throne, and is held by Muslims
to be the Prince of the Koran. As far as its use in prayer is concerned we refer to the
Commentators, E.W. Lane, Manners and customs, vol. 1, pp. 9091, and [Goldziher, Schools
of Koranic commentators, p. 95]. Great magic power is imputed to this verse (al-Bukhr,
K. Fail al-Qurn, 10; al-Tirmidh, loc. cit., 2, etc.). Of the nine short sentences that
constitute this verse, the text of 1 (= sra 3:1), 2, 3, and 9 is to be found literally in the Old
Testament and in ancient Jewish literature. Cf. Targum to II Samuel 22:32, and Psalms 18:32,
Mekhiltha on Exodus 15:26 and 16:18, Jac. Levy, Chaldisches Wrterbuch ber die Targumim,
s.v. ; Psalms 121:3; I Chronicles 29:11; the Hebrew encyclopedias, s.v. and . The
phrases from4 to 8 are entirely on the level with the Biblical linguistic usage, cf., for example,
7 to Isaiah 66:1. We are thus possibly dealing with nothing but an Arabic translation of a
Jewish or Christian hymn.
52
al-abar, Tafsr, vol. 3, p. 70.
53
Cf. Ibn Hishm, p. 275sq.; al-Bukhr, K. al-Buy, 24; the Commentators; al-Khab
al-Tibrz, Mishct, Bb al-rib, fal 3, 7; al-Wid, s.v., and in the introduction; al-Qurub,
vol. 1, fol. 23
v
; al-Shshw, cap. 1; al-Suy, al-Itqn, p. 59, etc. Tradition varies greatly as to
the days left to the Prophet after the revelation of the verse: 81, twenty-one, and seven days
respectively (al-Bayw, al-Zamakhshar, al-Nasaf, Madrik al-tanzl); nine days (al-abar,
Tafsr); and even only three hours (al-Bayw, al-Zamakhshar, al-Nasaf.)
54
These twoverses calledkhawtmal-baqarawere apparently revealedtothe Prophet on
his ascension(al-Nas, K. al-alt, 1 end; al-Khabal-Tibrz, Mishct, bbal-mirj towards
the end; Ab l-Layth al-Samarqand).
Sra 98
55
is heldby most people to be of Medinanorigin, less frequently to
be Meccan,
56
probably because it is listed only among early Meccan sras.
In support of the former view, however, there is the fact that in the verses
one and five the disbelievers among the People of the Book are mentioned
inone breathwiththe idolaters. Wm. Muir (vol. 3, p. 311) lists the sra among
the Medinan ones, albeit conceding that an exact dating is impossible.
Sra 64 is similar to the Meccan ones and is therefore occasionally [i/186]
counted among them;
57
yet verses 14sq. are undoubtedly of Medinan
58
ori-
gin even though their date cannot definitely be determined. This is likely to
apply also to verses 11 to 13.
59
For this reason several writers hold the last
verses to be Medinan.
60
The opinion that the entire sra might be Medi-
nan
61
has something to be said for it. This probably applies generally to all
the musabbit, i.e., the sras 57, 59, 61, 62, and 64, beginning with sabbaa
and yusabbiu respectively.
The first part of sra 62, aimed at the Jews, seems to originate from the
same period as the majority of sra 2. According to some exegetes,
62
the sec-
ond part (verse 9sq.) refers to Daya al-Kalb who, before being converted,
once entered Medina with a noisy mob during a Friday prayer. Even if this
description is correct, it would not lead to a more accurate fixing of the date.
All we know is that at the Battle of the Trench (end of 5/626) Daya (Ibn
Khalfa al-Kalb) had already become a Muslim, whereas some writers have
him fight with the believers already at Uud.
63
55
The verses 7 and 8 (Flgel) constitute one verse only. On the other hand, verse 2 can be
divided into two parts as it is the case in the best transmission.
56
Umar b. Muammad (IBN ABD AL-KF); al-Zamakhshar; al-Bayw; al-Suy, al-
Itqn, pp. 21 and 29sq.; Hibat Allh b. Salma; Hirschfeld, New researches, p. 143; Grimme,
Mohammed, part 2, p. 26, varies.
57
al-Zamakhshar; al-Bayw; Al al-Dn AL-KHZIN al-Baghdd; al-Suy, al-Itqn,
p. 28; Tafsr al-Jallayn. This is also the opinion of Weil (Historisch-kritische Einleitung in den
Koran, 1st ed., p. 63, 2nd ed., p. 72); and Muir.
58
al-abar, Tafsr on 64:14 (vol. 28, p. 74); al-Wid.
59
Cf. the idiom which is otherwise found only in Medinan passages.
Cf. in addition verse 11. Starting with verse 11 the rhyme, too, begins to change slightly.
60

Umar b. Muammad [GAS, vol. 1, p. 16], al-Suy, Itqn, pp. 20 and 36.
61
Hibat Allh (b. Salma); al-Zamakhshar; al-Bayw; Al al-Dn AL-KHZIN al-Bagh-
dd; al-Suy, K. al-Nsikh wa-l-manskh; al-Suy, al-Itqn, p. 28; and the list of sras.
Particularly noteworthy is the beginning, which is frequently found in Medinan revelations
but never in those from Mecca.
62
al-abar, Tafsr; Fakhr al-Dn al-Rz; and al-Wid. The earliest exegetic tradition
(al-Bukhr andal-Tirmidh inthe tafsr) alsosays that verse 10 refers toa trade caravanwhich
once entered Medina on a Friday, but does not mention names. [Cf. M. Pickthall on vv 911.]
63
Ibn ajar al-Asqaln, vol. 1, no. 2378; Izz al-Dn IBN AL-ATHR, Usd al-ghba, vol. 2,
p. 130.Ibn Sad, ed. Sachau (al-abaqt, vol. 4, part 1): Biographien der Muhgirn und
As an old tradition
64
maintains, not all of sra 8 but the greater part of [i/187]
it refers directly to the victory at Badr. Since the historians report that it
took about a month to distribute the spoils of war completely,
65
Muammad
would have made public most of the sra within this short time. The begin-
ning seems to be somewhat earlier than verses 29 to 46;
66
it is beyond doubt
that verse 42, which fixes the final
67
distribution of the spoils of war, is later
than verse 1. Also, verse 27 contains a warning not to steal anything fromthe
spoils of war. Some writers
68
see in it a reference to Ab Lubba [Ibn Abd al-
Mundhir
69
], who had warned the Jewish Ban Quraya by sign of the hand
(in 5/626) that their life was at stake even if they surrendered to the Prophet.
Verses 30 to 35,
70
or verse 30
71
alone, some incorrectly hold to be from the
Meccan period; they merely aimto remind the triumphant Prophet and his
believers howweak and helpless they had previously been at Mecca.
72
Some
think that verse 36 was revealed on the Day of Uud.
73
Verses 47 to 64,
Anr, p. 184 (= al-Nawaw, [Tahdhb al-asm], p. 239), however, says (without isnd) that
Daya had become a Muslim certainly at a very early time, but did not participate in
the Battle of Badr; but such pre-dating of the conversion of people who did not become
acquainted with the Prophet until Medina (like Ab Dharr [al-Ghifr]) can be ignored. But
it seems to be certain that Daya was a trader who had been in foreign countries. He presents
Coptic garments tothe Prophet (Ibnajar al-Asqaln, Ibnal-Athr, Usdal-ghba), andwants
to teach him mule breeding, which was certainly unknown to the Arabs (Ibn ajar, loc. cit).
Because of his familiarity with foreign countries Muammad appointed him envoy to the
Byzantine Emperor (Ibn Hishm, p. 971; Ibn Sad (al-abaqt), ibid., vol. 4, part 1, p. 185; and
Wellhausen, Seine Schreiben, und die Gesandtschaften, p. 98).
64
IbnIsqinIbnHishm, p. 476, l 6 =al-abar, vol. 1, p. 1354, l 17; cf. Caetani, Annali,
vol. 1, p. 497.
65
According toIbnHishm, p. 539, l 16sq., andal-abar, vol. 1, p. 1363, l 10, Muammaddid
not finish settling these affairs until the last days of this month or even the following month;
according to al-abar, loc. cit., on the day before returning to Medina, which was the 25th or
26th of Raman.
66
Flgels verses 43 and 44 constitute only one verse since
, verse 43, represents a

rhyme which occurs nowhere else in this sra.
67
Some writers consider this verse to refer to the spoils of war carried off about a month
later from the Ban Qaynuq (al-abar, ed. Zotenberg, vol. 3, p. 4; and al-Zamakhshar).
68
Ibn Hishm, p. 686sq. (not according to Muammad IBN ISQ [EI
2
; Juynboll, Ency-
clopedia, pp. 419423; Sezgin, GAS, vol. 1, pp. 288290]); al-Wqid (ed. Wellhausen), p. 213sq.;
al-abar, ed. Zotenberg, vol. 3, p. 70; al-Wid; Fakhr al-Dn al-Rz; al-abar, Tafsr; al-
Zamakhshar; al-Bayw. Cf. Weil, Das Leben Mohammeds, p. 428; Caussin de Perceval, Essai
sur l histoire des Arabes, vol. 4, p. 144; Muir, Life of Mahomet, vol. 3, p. 272.
69
70
al-abar, Tafsr; Umar b. Muammad b. Abd al-Kf; Fakhr al-Dn al-Rz, who appro-
priately adds,
; cf. also al-Suy, Asbb al-nuzl.

71
72
Verse 31 is to be translated: and when Our signs were being recited to them, quum (=
quoties) legebantur.
73
al-abar, Tafsr.
which require to fight all enemies most zealously, are a self-sufficient sec-
tion
74
but belong tothe same period(cf. verse 57 withverse 22). Occasionally,
and for no reason, verses 60sqq. are applied to the Ban Qaynuq,
75
who
were attacked soon after the Battle of Badr. Verse 65 is dated either after
Umars conversion
76
or, more likely, shortly before that attack.
77
Verse 66, in
spite of its air of expectant victory, might have been revealed before the final
outcome. Verse 67, certainly in its current form, is not the original continu-
ation of verse 66, as it looks instead like a different recension or a qualifying
postscript. According to Weil,
78
verse 68sq. was not revealed until after the
defeat at Uud. This view, however, is erroneous the passage does not state
that those who had spared the prisoners had now really
79
been punished.
Rather, it states that God exempted them from punishment through reve-
lation. Connected with this is verse 70, which deals with the spoils of war
acquired at Badr. These verses, as also verses 71 sq.which, however, are not
a continuation of the precedingwe must consider to be concurrent with
the greater part of the sra. Verse 75 and the following verses seem to be
from about the same time. Yet when it is said in verse 76 that those related
by blood are nearer to one another, this refers to the bond which Muam-
mad had established between the inhabitants of Yathrib and his menwho
were largely destitutebut which he abrogated after the Battle.
80
It is false
whenexegetes see inverse 76anabrogationof verse 73 because, eventhough
this compact was annulled, according to verse 76 intimate friendship and
mutual support should last forever.
74
Fakhr al-Dn al-Rz; verse 54 = sra 3:9 (apart from some small differences). Verse 56
looks like a different recension of verse 54.
75
al-Wqid, pp. 131, 178, 181; al-abar, vol. 1, p. 1360; cf. L. Caetani, Annali, vol. 1, p. 521.
Others mention instead the Ban Quraya and the Ban Nar, who were attacked in 5/626
and 4/625 respectively (al-Wqid, p. 131 on verse 63; al-abar, Tafsr; Fakhr al-Dn al-Rz;
al-Bayw). J. Wellhausen, Muammads constitution of Medina, in: Wensinck, Muam-
mad and the Jews of Medina, p. 136 n. 4, establishes that these different accounts need not
at all exclude one another as even after the destruction of the respective tribes several Jews
remained at Medina.
76
al-Wid; al-Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 32. For this reason, here
and there the verse is held to be Meccan. Cf. Umar b. Muammad [IBN ABD AL-KF].
77
al-Wqid, p. 131; al-abar, Tafsr; al-Wid; al-Zamakhshar; al-Bayw.
78
Historisch-kritische Einleitung, 1st ed., p. 72, 2nd ed., p. 82.
79
As Weil thinks, by this defeat. Al-abar, vol. 1, p. 1355, l 17. Ibn al-Athr (al-Kmil), vol. 2,
p. 105, however, considers the day of Uud as a punishment for that sparing of the prisoners.
80
Ibn Hishm, p. 344sq.; Ibn Sad, cod. Gotha, vol. 1, p. 257 [sic]; al-Bukhr, K. al-Fari,
16; al-Nas, K. al-Nik, towards the end, etc. Cf. Gustav Weil, Das Leben Mohammeds,
p. 83sq.; Caussin de Perceval, Essai, vol. 3, p. 24sq.; Wm. Muir, Life of Mahomet, vol. 3, p. 17sq;
Sprenger, Leben, vol. 3, p. 26.
Sra 47 must have been revealed shortly after the Battle of Badr
81
because [i/189]
the second part attacksapart from the Hypocrites [munfiqn] (verses 31
and 22)the irresolute fellow combatants of the Prophet who, despite the
victory (verse 37), wanted to make peace with the Meccans. Some attribute
the sra to the Meccan period, while others disagree.
82
A third group of
exegetes considers verse 14 to have been revealed when the Prophet was
emigrating and looked back crying over his native town.
83
It is rather uncertain when the first part of sra 3 (verses 1 to 86) was
revealed. If it belongs to only one period, it must be later than the Battle of
Badr, as verse 11 unmistakably refers to it. It is very dangerous to fix the lower
limit of the compositiontothe year 6/627 or 7/628. Evenif verse 57 is referred
to in the Prophets letters to the Emperor Heraclius
84
and the Coptic Patri-
arch of Egypt,
85
the authenticity of the text of these letters is open to most
serious doubts.
86
The historical explanations of tradition regarding the indi-
81
So also Weil, Historisch-kritische Einleitung, 1st ed., p. 72sq., 2nd ed., p. 82.
82
Cf. Umar b. Muammad(IBNABDAL-KF); Hibat Allh(b. Salma); al-Zamakhshar;
al-Bayw; al-Suy, al-Itqn, p. 27. Sprenger, Leben und die Lehre, vol. 2, p. 376, mentions
Meccan fragments before verse 15, but without elaborating.
83
Umar b. Muammad (IBN ABD AL-KF); al-Suy, al-Itqn, p. 43. Quite different
Wm. Muir, vol. 3, p. 308.Of the forty verses of this sra thirty-six rhyme with umusually
with preceding in the antepenulta rhyme which is only sporadically encountered in
other sras (e.g. 79:33, 80:32, 88:26, 99:6, and 109:4). For other differences see Rudolf Geyers
review of K. Vollers Volkssprache und Schriftsprache im alten Arabien, p. 40 of the off-print.
The conspicuous rhymes of the verses 4, 16, and 22 in Flgels edition are definitely based on
an inaccurate division; the incomplete rhymes in Flgels verses 11, and 26, on the other hand,
might be original.
84
al-Bukhr, K. Bad al-way, and K. al-Tafsr; Muslim, K. al-Jihd, 23 (al-Qasalln,
vol. 7, p. 380).
85
al-Suy, usn al-mudara, vol. 1, cap. 18, which is identical with the alleged original
published by F.A. Belin in Journal asiatique, 1854 on p. 482. Missing fromS[uy] are only the
words
, and instead of there is only .

86
Cf. J. Wellhausen, Seine [Muammads] Schreiben und die Gesandtschaften an ihn, p. 90;
Caetani, Annali, vol. 1, p. 725sqq. An apparent original of the letter to the Muqawqis, writ-
ten on parchment, was discovered in 1852 by the French archaeologist tienne Barthlemy
in a monastery not far from Akhmm, Upper Egypt, and in 1854 published by F.A. Belin. The
original later came to Constantinople, and was added as a precious gem to the relics of the
Prophet at the OldSerail. Recently Jurj Zaydnpublisheda photographic reproductionof the
original inhis Hill (vol. 13 no. 2, Cairo, November 1904, pp. 103sq). The document, however, is
definitely a forgery. Official documents of the period probably consisted of far less Kfic char-
acters. By the same token, instead of the signature a clay seal was used at the time, but not a
coloured stamp. Finally, such an official document must not only mention the issuing secre-
tary of the document, but also the accredited envoy must clearly be named.Regarding the
clumsy forgery of a letter from Muammad to the Persian governor of Barayn, al-Mundhir
b. Sw, see H.L. Fleischers devastating review, ber einen angeblichen Brief.The tradi-
tion(Commentators, al-Wid, etc.) that the embassy of the Christians of Najrnoccasioned
vidual verses are of little use to us. As is said by several writers,
87
only verse 10
refers to the Jewish Ban Qaynuq. This would establish a more solid date.
Approximately to this time point also verses 58sq., which present vis--vis
the People of the Book the religion of Ibrhm as the only true religion.
After the exposition, above, p. 119sq., this idea makes sense only under the
precondition that the Prophet had already fallen out with the Jews, and had
abandoned any hope of their voluntary conversion. Verses 25sq., which do
not fit in with the rest, are panegyrics that, particularly in the second half,
are unmistakably of Jewish origin in content and form. Their origin is com-
monly dated 5/626, when a trench was dug at Medina,
88
or, more seldom,
after the conquest of Mecca
89
in 8/629; but such statements can be ignored.
Verse 79 is considered to be late,
90
since in it all unbelievers are threatened
with eternal torment. This, however, is no evidence because the termunbe-
lievers does not necessarily have to apply to all non-Muslims. It is known
to us that in the first years of his Medinan activity, Muammad reckoned
true Christians among the believers.
91
We have several references for dat-
ing the verses 87113, which, incidentally, have much in common with the
first part. The unique esteem of the religion of Abraham (millat Ibrhm) in
verse 89 and the inclusion of the pilgrimage to Mecca in the Islamic cult
(verse 90sq.) certainly point to the time after the Battle of Badr, as has been
demonstrated earlier. Verses 93sqq. are generally applied to Shas b. Qays,
one of Ban Qaynuq,
92
whose poetry tries to inflame again the old rivalry
betweenthe tribes of Yathrib, the Aws andKhazraj.
93
Since this Shas appears
the beginning of the sra is without support. This event is not dated either in Ibn Hishm,
p. 944sq. or in al-Bukhr, K. al-mn, 40, but for internal reasons it can belong only to
Muammads final years. How could he have possibly imposed conditions earlier on a tribe
whose settlements were so far from Medina? Indeed Ibn Sad fixes it to this time, and Well-
hausen(Seine Schreiben, p. 155sq.), Sprenger (Das Leben, vol. 3, p. 372sqq.), andCaetani (vol. 2,
part 1, p. 198sq.) follow suit, fixing the embassy to 8/629, while al-abar, vol. 1, p. 1736sqq. as
well as Al al-Dn al-KHZIN al-Baghdd prefer 10/631, and Muir (vol. 4, p. 181), 9/630. The
latter has the verses 52 to 57 revealed on this occasion (vol. 2, p. 302sq., vol. 3, p. 312), whereas
Sprenger, vol. 3, p. 490sqq., chooses still others (verses 30 to 51).
87
Ibn Hishm, pp. 383 and 545; Tafsr [sic]; Ab l-Layth al-Samarqand; al-Zamakhshar.
Against all probability other writers mention Ban Nar or Ban Quraya. Cf. the Commen-
tators.
88
Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar; al-Bayw.
89
al-Zamakhshar; al-Wid.
90
Weil, Historisch-kritische Einleitung, 1st ed., p. 73sq., 2nd ed., p. 83.
91
Cf. above, p. 119sq., and Snouck Hurgronje, Het Mekkaansche feest, p. 42sq.
92
Ibn Hishm, p. 352; R.A. Nicholson, Literary history of the Arabs (London, 1907), p. 125.
93
Ibn Hishm, p. 385sq.; al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid, al-Za-
makhshar; al-Bayw.
in one of the poems of either Kab b. Mlik or Abd Allh b. Rawa as
one of the men who were hard hit by the defeat of Ban Nar,
94
this event
would have to have happened before I Rab 4/625. It was perhaps because
the Jews had behaved so provocatively after the Battle of Uud, in which
the Muslims suffered a severe defeat. This might explain verse 107, which
mentions the insult ( ) of the Jews who apparently could not be deterred
from open war. This fits only too well into the picture when the believ-
ers, distressed by misfortune, were exposed to the malice of their ene-
mies. These verses might therefore originate from the time shortly before
the war with Ban Nar. But since the text of verses 92sqq. is quite gen-
eral, suggesting a connection neither with the Jew Shas b. Qays nor any
particular person in general, that story is probably not based on an old
accompanying tradition but rather on scholastic interpretation of the scrip-
ture. Nevertheless, the general situation is likely to be accurately described.
From about the same time originate verses 114sqq., in which it says that
the Jews make no bones about their enmity against the Muslims who are
smitten by evil (verse 116). We can therefore connect this passage with
the lost battle at Mount Uud (Shawwl 3/624), which undoubtedly had
not happened much earlier. Verse 123, according to many accounts, is an
inspiration the Prophet received when he was lying wounded on the bat-
tlefield.
95
However, even if it is in fact possible that he had such thoughts
in such a situation, the verse, which is connected with the rest of them,
can have originated only later. Moreover, tradition knows of other events
that prompted a revelation during or shortly after the battle.
96
Verses 125
to 130, which are of indeterminate provenance, separate this part from
another one, 131154, which refers to the same battle but belongs more
to the time immediately thereafter. Verses 155 to 158 link a tradition with
the statement that one day a precious garment from the booty of Badr
94
Ibn Hishm, p. 661.
95
Ibn Hishm, p. 571; al-Wqid, p. 242; Muslim, K. al-Jihd, 32; al-Tirmidh, K. al-Tafsr;
al-Aghn, vol. 14, p. 18, l 22; al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Zamakhshar;
al-Bayw. Similarly reported in al-abar, Chronique, ed. Zotenberg, vol. 2, p. 505, that sra
8:12 and other verses had been revealed in the Battle of Badr.
96
al-Wqid, pp. 311 and 341; al-Bukhr, K. al-Maghz 22 etc.; Muslim, K. al-alt, 95
(al-Qasalln, vol. 3, p. 363); al-Tirmidh, K. al-Tafsr; al-Nas, K. al-Tabq, 21; al-Khab al-
Tibrz, Mishct, K. al-Qunt, beginning; Tafsr [sic].; Ab l-Layth al-Samarqand; al-Wd;
al-Zamakhshar. According to a tradition in Muslim-al-Qasalln, vol. 3, p. 364, and al-
Qasalln, vol. 6, p. 303sq., onal-Bukhr, al-Maghz, 22, the verse purports toapologize for
the curse on the traitors of Bir Mana. Muammads curse is certainly historical (cf. Muslim,
loc. cit.; al-Qasalln, vol. 3, p. 364sqq; al-Wqid, p. 341, etc.) but not the connection which
the Koranic verse is associated with.
was missing, and Muammad was suspected of having kept it for him-
self.
97
At first sight, this conjecture leaves the impression of historical reli-
ability, particularly because of its content which concerns the Prophet.
Still, it seems to arouse suspicion, as it was too easy an interpretation to
reach. Other writers
98
apply it tothe archers who, fearing Muammadwould
deprive themof their share of the spoils of war, left their formation, thereby
precipitating the defeat. We see how an event which is quite likely true can
be artificially connected with this verse, and later even modified.
99
Con-
nected with verse 154 are verses 159 to 176, in which Muammad remembers
his faithful followers who, onthe morning after the Battle, followedthe Mec-
cans to amr al-Asad.
100
Verses 177 to 181 are the replies to the mocking
speeches of a Jew. There are variant accounts of the particular circumstances
that cannot be ascertained.
101
The general situation might be identical with
that of the following verses, 182sq.,
102
whichwe might have todate tothe time
not long after the Battle of Uud. This wouldindicate the depressedmoodof
the Muslims whenthey were commandedpatiently tosuffer misfortune and
insults (verses 200 and 183); also mentioned are the Muslims killed for their
faith (verse 194) and the pagans insolence after their victory (verse 196).
Sra 61, like several of the shorter Medinan sras, is occasionally attri- [i/194]
buted to the Meccan period.
103
According to Weil,
104
part of it belongs to
97
al-Wqid, pp. 97 and 316; al-Tirmidh, K. al-Tafsr; al-Wid; al-Zamakhshar; al-
Bayw; Sprenger, Das Leben und die Lehre, vol. 3, p. 128.
98
Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar, al-Bayw. Cf. Ibn Hishm,
p. 570; al-Wqid, p 226sq.; al-abar, vol. 1, p. 1440sq., etc.
99
The explanation that the suppression refers to the concealing of revelations (Ibn
Hishm, p. 602, etc.) we cannot recognize as an authoritative explanation.
100
Verse 166sq. Cf. Ibn Hishm, pp. 588sq. and 606; al-Wqid, p. 330; al-Bukhr, K.
al-Maghz 27; al-abar, vol. 1, p. 1355, 1427 and inthe Tafsr, s.v.; al-Aghn, vol. 14, p. 25; Ab
l-Laythal-Samarqand; al-Zamakhshar; al-Bayw. Cf. Caussinde Perceval, vol. 3, p. 112; Weil,
Das LebenMohammeds, p. 130; Muir, Life of Mahomet, vol. 3, p. 193; Sprenger, Das Lebenunddie
Lehre, vol. 3, p. 180; Caetani, Annali, vol. 1, p. 566sq.; Wellhausen, MuhammedinMedina, p. 169,
attributes verse 166sq., [and] al-Yaqb, Historiae, vol. 2, p. 69, verse 167sq., to the campaign
against Badr al-Mawid in 4/625. Muir, Life of Mahomet, vol. 3, p. 222, attributes the verses 166
to 169 to the same campaign.
101
Apart from the Commentators. Cf. Ibn Hishm, p. 388sq.; Wellhausen, Muhammed in
Medina, p. 219.
102
Sprenger, Das Leben und die Lehre, vol. 3, p. 19, fixes verse 183 to the time shortly after
the construction of the first mosque in Medina.
103
Umar b. Muammad IBN ABD AL-KF; al-Bayw; Al al-Dn (AL-KHZIN al-
Baghdd).
104
Weil, Historisch-kritische Einleitung, 1st ed., p. 76, 2nd ed., p. 86sq.Muir, Life of Maho-
met, vol. 3, p. 313, in his Chronological order of the sras revealed at Medina lists sra 61 in
the sixth last place.
the unfortunate pilgrimage to al-udaybiyya, particularly verse 13, which
promises a speedy victory, and the first four verses that reproach the believ-
ers for not keeping their word, which probably refers to their staying behind
on the march on Mecca. Tradition,
105
more accurately, is thinking of the Bat-
tle of Uud, where many had abandoned their positions and did not stand
like a solid edifice. The victory mentioned in verse 13, which the Muslims
were earnestly longing for, particularly as fortune had left them, is either
none in particular or Muammad is already thinking of the attack on Ban
Nar. This certainly does not include a definite reference to the conquest
of Mecca. Nothing definite can be said about verses 5 to 9. That they are of
Medinan origin can be seen from verse 89, namely that before his emigra-
tion Muammad cannot easily have enunciated the idea of a final victory of
Islam over all other religions with such conviction.
106
The entire
107
sra 57, or at least its first
108
or last
109
part, is attributed to [i/195]
the Meccan period. The main theme, as is the case with many other sras,
consists of admonitions tocontribute tothe wars of religion, andcomplaints
about the Hypocrites who refuse to contribute money. Verse 10 is often
attributedtothe conquest of Mecca,
110
yet this is not appropriate, since inthe
entire passage Muammad does not appear as confident as he actually was
after this great success. If we are not all wrong, verse 22sq.
111
would indicate
that at the time of its composition, Muammad was afflicted by misfortune.
We would therefore best date the sra to the time between the Battle of
Uud and the Battle of the Trench. The victory (fat), which verse 10 refers
to, is likely the Battle of Badr.
The greater part of sra 4 seems to belong to the time between the end of
3/624 and the end of 5/626.
112
This period is more or less clearly identified by
105
al-abar, Tafsr; al-Wid; al-Zamakhshar; al-Bayw.
106
We would have a clue to a more accurate fixing if we could show when the contempo-
rary poets first used Amad of verse 7 instead of Muammad; but the vague transmission of
these poems, and the mass of the fictious ones, where especially the name Amad is partic-
ularly popular, will make verification extremely difficult.Verses 8 and 9 = sra 9:32 and 33;
only verse 32 shows some differences. Verse 9 appears with a different final formula also in
sra 48:28. Verse 14 comes close to sra 3:45.
107
Hibat Allh (b. Salma); al-Bayw; al-Suy, al-Itqn, p. 27; al-Nasaf, Madrik.
108
al-Itqn, p. 27.
109
al-Itqn, p. 36.
110
al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Bayw; so also Weil, Historisch-kritische Ein-
leitung, 1st ed., p. 73, 2nd ed., p. 83.
111
Compare, for example, 3:93 to sra 3:147.
112
Weil, Historisch-kritische Einleitung, 1st ed., p. 71, 2nd ed., p. 81 fixes this sra to the
early period after the emigration. That this is incorrect is quite obvious as a great part of
it is directed against the Hypocrites. Already a tradition in al-Nas, K. al-alq, bb Iddat
various passages in the sra and fits with most of its parts. In the first part,
verses 1 to 18,
113
the Muslims tell all sorts of stories, but those who try to fix
chronology do so to the time soon after the Battle of Uud. Verses 8 and 12
respectively are said to refer to a woman who complained to the Prophet
that, according to ancient Arab custom, she was excluded frominheritance.
The name of the woman is either not mentioned at all, or she is named
Umm Kua.
114
Her deceased husband is known by the following names in
different reports:
1. Rifa,
115
without exact genealogy. His son is called Thbit.
116
Several of [i/196]
the Prophets companions are called Rifa. Ibn ajar al-Asqaln says
nowhere that one of them is meant. It could well be possible that it is
Rifa b. Amr
117
or Rifa b. Waqash,
118
bothof whomwere killedat Uud.
2. Sad b. al-Rab, who was killed at the Battle of
119
Uud.
3. Aws b. Thbit al-Anr,
120
the brother of the poet assn b. Thbit, [i/197]
who died at that battle.
121
The possibility that this man is not meant
al-mil, towards the end, more or less correctly fixes this sra to a time later than sra 2.
[Pickthall, in his introduction to sra 4, puts it between the end of the third year and the
beginning of the fifth year.]
113
Flgels verses 5 to 7 and 13 to 16 constitute one verse each; this applies also to the
verses 29 and 30. Cf. also Rudolf Geyer.
114
So correctly in al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Zamakhshar; al-Bayw; al-
Wid; al-Jawhar, Tj al-lugha wa-al-i [GAS, viii, 215]; Lisn al-Arab; Ibn ajar al-As-
qaln, vol. 1, no. 315 end, and 921; against that Ibn ajar has in vol. 4, p. 946, the explicit
reading (
) next to the variant ( ); is to be

found also in Ibn al-Athr, Usd al-ghba f marifat al-aba, vol. 5, p. 611, vol. 3, p. 402, l 2,
and al-Dhahab, Tajrd asm al-aba, vol. 2, p. 349.
115
al-Wid on verse 5; Ab l-Layth al-Samarqand on v. 8; Fakhr al-Dn al-Rz on v. 5.
116
Briefly mentioned in al-Wid; Ibn ajar al-Asqaln, vol. 1, no. 877; Ibn al-Athr, Usd
al-ghba f marifat al-aba, vol. 1, p. 223, without date.
117
Ibn Hishm, p. 609; al-Wqid, p. 297; Ibn ajar al-Asqaln, vol. 1, no. 2661; Ibn al-Athr,
Usd al-ghba, vol. 2, p. 184.
118
Ibn Hishm, p. 607; al-Wqid, pp. 230 and 293; Ibn ajar, vol. 1, no. 2666; Ibn al-Athr,
Usd al-ghba, vol. 2, p. 185, where he mentions as his brother the above-cited Thbit.
119
al-Wqid, p. 320, which undoubtedly indicates verse 12; al-Tirmidh, Fari, 3; Ibn
ajar al-Asqaln, vol. 1, no. 2734; Ibn al-Athr, Usd al-ghba, vol. 2, p. 277sq.; and al-Nawaw,
[Tahdhb al-asm] p. 271 sq., who all refer to verse 12; Ibn Hishm, p. 608; Ibn ajar, vol. 4,
p. 945, bottom.Ibn ajar, vol. 2, no. 4048, mentions verse 175 or verse 38, and as his wife,
Amra bint azm. On the other hand, the biographers of women (Ibn Sad (al-abaqt):
Biographien der Frauen (vol. 8), p. 328; Ibn ajar, vol. 4, p. 704sq.; Ibn al-Athr, Usd al-ghba,
vol. 5, p. 509), do not record anything regarding revelations. [Juynboll, Encyclopedia, p. 483,
col. 2.]
120
Ab l-Layth al-Samarqand; Fakhr al-Dn al-Rz on verse 8; Ibn ajar, vol. 1, no. 315.
121
This is ascertainedby the testimony of his ownbrother (Dwn
assn b. Thbit, Tunis ed., p. 27; Ibn ajar al-Asqaln, vol. 1, no 315.) Cf. Ibn Hishm, p. 608.
here and that it is, instead, another man with the same name men-
tioned nowhere else, Ibn ajar al-Asqaln excludes for totally insuf-
ficient reasons.
4. Instead of Aws b. Thbit some mention Abd al-Ramn b. Thbit,
122
another brother of the poet who, Ibn ajar emphasizes, is otherwise
unknown; still others call the person killed at Uud Aws b. Mlik.
123
5. Thbit b. Qays [Ibn Shamms],
124
who is also said to have ended his life
at Uud.
125
6. Regarding verse 8, someone mentions Aws b. Suwayd,
126
about whom
we know nothing more.
Even if some of the names mentioned might be a confusion, there still [i/198]
remain three or certainly two names, as in options 4, 5, and 6 above. Yet
given the well-known tendentiousness of the exegetic tradition, one cannot
even rely on these names. Only one detail is credible, namely that it con-
cerns men killed at Uud. In addition, it is quite likely that these minute
laws regarding orphans, and the inheritance of deceased husbands, were
promulgated when many heads of families had died at the same time; this,
however, fits best with that great defeat. Verses 19 to 22, regarding the inde-
cency of men and women, might also be fixed to this time. In any case, they
are older than the passage 24:2, which seems to date from6/627. Verses 23 to
122
al-abar in the Tafsr on verse 12; Ibn ajar, vol. 4, p. 946, vol. 2, no. 9459.
123
Ibn ajar al-Asqaln, vol. 1, nos. 315 and 350 (vol. 4, p. 946). This name is not listed
among the men killed, neither in al-Wqid nor in Ibn Hishm. al-Zamakhshar and al-
Bayw have instead . This man is quite a different Anr who is said to have
been the cause of the revelation of sra 58, and who died a long time after the Prophet, prob-
ably under Caliph Uthmn (Ibn Sad, [al-abaqt]: Biographien der medinischen Kmpfer
(vol. 3, part 2), p. 93; Ibn ajar, vol. 1, no. 338; Ibn al-Athr, Usd al-ghba, vol. 1, p. 146sq.;
al-Nawaw, [Tahdhb al-asm] Biographical dictionary, p. 168sq. The confusion of the two
brothers is also the basis of the inaccurate report that the former, Aws b. Thbit, survived the
Prophet by many years (Ibn Sad, ibid., vol. 3, part 2, p. 63; Ibn ajar, vol. 1, no. 314, Ibn al-Athr,
Usd al-ghba, vol. 1, p. 141, all according to al-Wqid).
124
EI
2
; Juynboll, Encyclopedia, p. 72, col. 2, p. 142, col. 2, p. 380, col. 1.
125
So al-Wid on verse 12 in a cumbersome account; Ibn ajar al-Asqaln, vol. 4, p. 946;
Ibn ajar, vol. 1, no. 984, does not approve of this, he is missing fromthe lists in al-Wqid and
Ibn Hishm. The many men listed under this name by Ibn ajar all died after Muammad
except one, about whose death nothing is known (no. 900).
126
Ibn ajar, vol. 1, no. 336, vol. 4, p. 946, l 13, is to be replaced by
, of whom the former was the husband of Umm Kua, the latter his brother.
Cf. also Ibn ajar, vol. 1, no. 935; in no. 315 Suwayd appears in place of Khlid or Qatda as
the brother of Urfua who, according to some writers, was the brother, according to others,
the cousin of Aws ibn Thbit, cf. Ibn ajar, vol. 1, no. 2140, vol. 2, no. 9877; Ibn al-Athr, Usd
al-ghba f marifat al-aba, vol. 2, p. 85, vol. 3, p. 401 sq., vol. 4, p. 194.
32 might be concurrent with the first verses, as verse 23 is of nearly identi-
cal content, leading Muslims to supply the same stories to explain it. The
remainder of the verses, too, dealing with marital matters, fits well into this
time period when there were a large number of widows. Verse 28 men-
tions muta, the temporary marriage, which was later abrogated during the
siege of Khaybar (7/628).
127
Verses 33 to 45, in which Muammad begins
to fight the Hypocrites (verse 40sq.), also belong roughly to this period
(cf. 4:36sq.) It is difficult to express an opinion regarding the composition of
verse 46. It is certain that this verse, which prohibits prayer under the influ-
ence of alcohol, must have beenpromulgatedbefore the general prohibition
of wine.
128
Since this prohibition was apparently
129
promulgated
130
during
the campaign against Ban Nar in I Rab 4/625, this verse would have to
127
al-Muwa, p. 196; al-Bukhr, K. al-Nik, 31; Muslim, K. al-Nik, 3, K. al-ayd, 4;
al-Nas, K. al-Nik, 61; al-Tirmidh, Nik, 27; Hibat Allh (b. Salma). From among the
different traditions on the subject it seems to appear that shortly after the conquest of Mecca
the muta was again permitted for a short time; cf. Muslim, Nik, 3, note on Ibn Hishm,
p. 758; Weil, Das Leben Mohammeds, note on p. 357. According to Ibn Sad, Biographien der
Genossen, ed. Julius Lippert, p. 68 bottom, the muta marriage was prohibited in the Farewell
Pilgrimage, but it is not stated that it was the first time this happened. The editor of the
Arabic text, J. Lippert, did not understand the meaning of muta as is evident from his note
s.v. (p. xcix). After

Ibn Abbs is said to have added
, al-abar,
Tafsr; Lisn al-Arab, vol. 10, p. 205sq. Contrary to the Sha, all conservative authorities of
the sunna reject the so-called temporary marriage; cf. al-Sharn, Mzn (Cairo, 1317), vol. 2,
p. 107; J. Wellhausen, Die Ehe, p. 464sq.; Th.W. Juynboll, Mohammedaansche Wet, 39.
128
Muslims accurately supply the chronological order of the passages dealing with wine
as follows: sras 16:69 (Meccan); 2:216 (as we have been able to see above on page 127,
shortly before the Battle of Badr); 4:46; 5:92 (al-Tirmidh, Tafsr, s.v.; al-Nas, K. al-Ashriba,
beginning; al-abar, Tafsr on sra 2:216 (vol. 2, p. 203sq., and 5:92 (vol. 7, p. 20sq.)); Ab
l-Layth al-Samarqand on sra 5:92; Hibat Allh (b. Salma); al-Zamakhshar; al-Bayw on
sra 2:216; al-Suy, al-Itqn, p. 58). It is incomprehensible that Weil holds this prohibition
to be later than sra 5:92. If this had been the case, Muammad would have reacted quite
differently against the ritual prayer under the influence of alcohol as in this instance. Then
there is the added argument that the above-mentioned writers perhaps rightfully point
out that this concerns Abd al-Ramn b. Awf, one of the earliest and most faithful of the
Prophets companions, who is unlikely to have done such a thing if there had been a previous
general interdiction. [EI
2
; Juynboll, Encyclopedia, p. 155, etc.; Sezgin, GAS, vol. 8, p. 171, no. 5
(8).]
129
Ibn Hishm, p. 653; following this, also Weil, Das Leben Mohammeds, p. 139; Caussin de
Perceval, vol. 3, p. 122; Caetani, Annali, vol. 1, p. 586.
130
The above-mentioned authorities (page 148) relate that the consumption of wine was
prohibited on occasion of a quarrel at a drinking-bout of Sad b. Ab Waqqs, [EI
2
; Juynboll,
Encyclopedia, p. 25sqq.] but without reference to a military campaign. Al-Wqid, p. 261,
last line (Wellhausen, Muhammed in Medina, p. 125 = al-Bukhr, K. al-Maghz, 18, centre)
presupposes that at the time of the Battle of Uud wine had not yet been prohibited.
be fixed to the time before that campaign. This is contradicted by traditions
onthe originof the secondhalf of the verse, however,
131
whichpermit, incase
of need, to have recourse to wholesome dust
132
instead of water for the rit-
ual ablution. Some attribute this promulgationtoanunidentifiedcampaign;
the localities mentioned in this context are all in the vicinity of Medina,
where Muammads forces were often in action, namely Dht al-Jaysh
133
or
Ult al-Jaysh,
134
and al-Bay.
135
Al-Wqid
136
fixes the revelation to the cam-
paign against Ban Mualiq which, according to him, occurred in Shabn
5/626, and, according to Ibn Isq,
137
in the same month but of the following
year. Far less frequently this is attributed to the Dht al-Riq
138
campaign of
Muarram 5/626. It is quite possible that verse 46 does not represent the
original text of a uniform revelation but is rather the subsequent literary
digest of two divine commands that may have been revealed on entirely
different occasions.
139
Verses 47 to 60 can be attached to verse 45. Tradi-
tion applies verse 54 to the Jews, who incited the Quraysh to fight against
the Prophet by derogating Islamas an innovation fromthe pagan cult of the
ancient Arabs. If these verses, as some claim,
140
refer to Kabb. al-Ashraf, they
would predate the Battle of Uud, as he was killed in I Rab 3/624.
141
This
is not very probable, however, as in this battle the Jews did not assist the
131
The verse of tayammum is from
to identical with sra 5:9.

132
This regulation is not at all an invention of the Prophet, rather it goes back first of all to
Jewish (Talm., Berkht, fol. 15a, top ) and
the resulting Christian ritual respectively.
133
al-Muwaa, p. 18sq.; al-Bukhr, K. al-Tayammum, 1; Muslim, K. al-ay, 27; al-
Nas, K. al-ahra, bb 106; al-Wid. This entire tradition is in some aspects, including
the locality, conspicuously similar to the infamous story of ishas lost necklace; see below
on sra 24. Juynboll, Encyclopedia, p. 390.
134
al-Frzbd, al-Qms; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 473, bottom.
135
See above, note 128.
136
Translated by Wellhausen, Muhammed in Medina, pp. 184 and 188; al-Diyrbakr, ibid.;
the glosses on Mlik ibn Anas, Muwaa, loc. cit.; al-Suy, al-Itqn, p. 41 sq. Cf. Weil, Das
Leben Mohammeds, p. 159; Caussin de Perceval, vol. 3, p. 161; Caetani, Annali, vol. 1, p. 604.
137
Ibn Hishm, p. 725; al-Tabar, vol. 1, p. 1510.
138
The glosses on Muwaa, ibid.; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 464; cf. below
on verse 102sq. Juynboll, Encyclopedia, pp. 67 col. 2, 673, col. 1.
139
Cf. Muir, vol. 3, p. 301; Sprenger, Das Leben und die Lehre, vol. 3, p. xxxi sq.
140
al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar; al-Bayw.
Verse 50 al-abar attributes in his Tafsr to other Jews, Rifa b. Zayd b. Sib (Tbt), and
Mlik b. ayf, verse 47 to the former alone. This is the same person who in Ibn Hishm, p. 397,
is connected with sra 5:62sq. Such associations are inconsequential.
141
Wellhausen, Muhammed in Medina, pp. 184 and 188; al-abar, vol. 1, p. 1368sq.; Ibn
Hishm, p. 548sqq. This fact is supported by several poems on the war against Ban Nar,
where Kabs death is mentioned (Ibn Hishm, p. 656sq).
Quraysh. Instead, some of Ban Nar are probably meant here,
142
who after
losing their dwelling-places, formed a coalition of the Quraya Jews with
the Quraysh and their followers as well as the widely-dispersed Ghaafn,
nearly bringing Muammad to destruction in 5/626.
143
Verses 6272 must
refer to a dispute which a Muslimdid not want to be settled by the Prophet.
The stories alluded to in verses 63 and 68, however, are neither sufficiently
reliable nor conclusive for the chronology. As this section is similar in con-
tent and language to the previous one, they might both be from the same
period. Verses 7385 were undoubtedly revealednot very long after the great
defeat, as they clearly indicate that at that time the pagans were consid-
erably stronger than the Muslims. Verses 8695 can have obviously been
revealed only after the Muslims had already concluded pacts with vari-
ous tribes (cf. verses 92 and 94). Verse 94 is held to refer to Ayysh b. Ab
Raba of the Makhzm who, without being aware of the conversion of al-
rith b. Yazd (or Ibn Zayd) al-aram,
144
slew him on account of an old
quarrel. Following another tradition,
145
this happened after the Battle of
Uud. Others have instead of rith an anonymous man, or they call the
murderer Ab l-Dard.
146
That verse 90 refers to those who deserted the
Prophet at Uud
147
cannot be proven, since this verse deals only with the
Hypocrites in general. Verses 96 to 105 have much in common with the pre-
ceding passages. Verses 102sq. even mentions the so-called prayer of danger,
alt al-khawf,
148
which according to some sources
149
refers to the campaign
142
Ibn Hishm, p. 669; al-abar, vol. 1, p. 1464; Fakhr al-Dn al-Rz on sra 33:9; al-Wid.
That this happened after the Battle of Uud is also reported by al-Wid who, thus, is
contradicting himself when he mentions Kab who was killed before that battle. It must be
added that this whole matter, though historically established, is marred by all sorts of silly
additions.
143
The common tradition regarding verse 56 has been correctly interpreted already by
Weil (Historisch-kritische Einleitung, 1st ed., p. 72, note 2, 2nd ed., p. 81, note 8).
144
F. Sezgin, Geschichte des arabischen Schrifttums, vol. 1, p. 342; Juynboll, Encyclopedia, s.v.
145
Ibn ajar, vol. 1, no. 1503, end.
146
al-abar, Tafsr.
147
al-Bukhr, K. al-Maghz, 17, end, K. al-Tafsr; Muslim, K. al-ift al-munfiqn, 1;
al-Tirmidh, K. al-Tafsr; al-abar, Tafsr; al-Zamakhshar; al-Wid; al-Bayw. The Com-
mentators offer still other explanations.
148
Also this regulation (see above, page 161) is of Jewish origin. Mishna Berkht 4:4
andTalmudBabli, Berkht, fol. 29b [cf. Tosefta Berkht,
3] .
149
al-Bukhr, K. al-Maghz, 33; al-abar, vol. 1, p. 1454; al-Masd, Prairies d or, v. 4,
p. 156sq.; al-Wqid translated by Wellhausen, Muhammed in Medina, p. 172. Some do not say
it straight away that this then happened for the first time: al-Muwaa, p. 64; Ibn Hishm,
to disperse certain tribes of the Ban Ghaafn ( ), i.e., in I
Jumd 4/625
150
or Muarram 5/626.
151
Verse 96, which is closely related to
the latter, Muslims mostly take to refer to the murder of a man by Muam-
mads favourite, Usma b. Zayd,
152
during a campaign in 7/628. This cannot
be trusted any more than other stories that, according to tradition,
153
occa-
sioned this revelation. Verses 106 to 115 and 116 are generally applied to
a man from Medina found guilty of theft whom Muammad nearly par-
doned on behalf of the intercession of his clansmen, the Ban afar. In
the various individual accounts
154
there are can be found some tendentious
p. 662; Muslim, K. al-Fal al-Qurn, 18; al-Nas, K. alt al-khawf; al-Khab al-Tibrz,
Mishct, ibid. Also other military campaigns and places are mentioned, where this prayer
might have been observed without explicitly denying that it might have happened even
earlier, namely Usfn and anajn: (campaign of al-udaybiyya, 6/627); al-Tirmidh, K.
al-Tafsr; al-abar in Tafsr, vol. 5, pp. 144 and 152; al-Wid; Dh Qarad: al-Nas, loc. cit.
(ah6); against some of the Juhayna [Juynboll, Encyclopedia, p. 79 col. 1]; Muslim, loc. cit.
That the verse was also revealed at Usfn is to be found in al-abar, Tafsr, vol. 5, p. 145.
Al-Wqid; translated by Wellhausen, Muhammed in Medina, p. 245; al-Suy, al-Itqn, p. 41.
Cf. also Muir, Life of Mahomet, vol. 3, p. 224; and Caetani, Annali, vol. 1, p. 596.
150
Ibn Hishm, p. 661; al-Masd, loc. cit.
151
al-Wqid, p. 4 and Wellhausen, Muhammed in Medina, p. 172.
152
al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar; al-Bayw.
Without mentionof the verse, the story is related inIbnHishm, p. 984; IbnSad (al-abaqt):
Biographiender Muhgirun, p. 48; J. Wellhausen, MuhammedinMedina, p. 297sq.; Muslim, K.
al-mn, 41; al-Tirmidh, K. al-Tafsr; al-Khab al-Tibrz, Mishct, k. al-Qi, fal 1, 5; Ibn
al-Athr, Usd al-ghba f marifat al-aba in the article Usmah. [EI
2
; Juynboll, Encyclope-
dia, p. 25sqq.]
153
Thus have, e.g., Ibn Hishm, p. 987 (cf. the accompanying notes) and al-Wqid, trans-
lated by Wellhausen, Muhammed in Medina, p. 325, an event that is said to have occurred
shortly before the fall of Mecca. Cf. Caetani, Annali, vol. 2 part 1, p. 116.
154
Ibn Hishm, p. 359; al-Tirmidh, K. al-Tafsr; al-abar, Tafsr, vol. 5, pp. 157160; Ab
l-Layth al-Samarqand; al-Wid; al-Zamakhshar; al-Bayw; Ibn al-Athr, Usd al-ghba,
vol. 2, p. 179sq; Ibn ajar, vol. 1, no. 2651; assn b. Thbit, Dwn, Tunis edition, p. 64sq.,
scholium according to which the thief stayed at Mecca with ( ). The name of
the thief is given as [Ab uma b. Ubayriq] (so al-Wid, a variant in Ibn
Hishm, and the best tradition in Ab l-Layth al-Samarqand) or commonly ( )
(Ibn al-Athr, Usd al-ghba, vol. 3, p. 52; Ibn ajar, vol. 2, no. 8734) or ! (Ibn
Hishm; Usd, vol. 3, p. 52 [sic]) or (al-Tirmidh). Inassn[IbnThbits] verse he is
simply called Sriq al-Dirayn, and in its accompaying historical explanation

. For uma, the best ascertained form of the namecf. also amsa,
p. 452; Ibn Khab al-Dahsha, Tufat dhaw l-Arabwe occasionally find ima (variants in
Ibn Hishm and Fleischers al-Bayw). It is less certain whether to read Bashr or Bushayr.
Ibn ajar, vol. 1, no. 686, varies, but the notes in Ibn Hishm, p. 359, and Ibn al-Athr, Usd
al-ghba, vol. 1, p. 184, hold the diminutive to be correct. Totally incorrect is the distinction in
cod. Sprenger, no. 282 (= Ahlwardt, no. 10169) between
and a/ima b. Ubayriq

(sic!).
traits,
155
apart from many harmless ingredients. Most sources lack the least
chronological clue, while only a few state that the thief died in disbelief at
Mecca.
156
Ibnal-Athrs Usdal-ghbaf marifat al-aba
157
has a note, which
cannot be confirmed, saying the theft occurred before I Rab 4/625. The
event itself is indeed quite credible, as its historicity is substantiated by a
satirical poem of assn b. Thbit referring to it. Its connection with those
verses of the sra, however, is not supported by the text of the Koran and
certainly belongs to the great realm of exegetic fancy. Verses 116125, and
the verses apparently connected with them, 130133,
158
fight idolatry. The
importance of the religion of Abraham (millat Ibrhm), which verse 122
expounds, would suggest that none of this be dated to before the Battle
of Badr.
159
Verses 126129 should very likely be interpreted as a supplement
to the laws appearing at the beginning of the sra. Verse 134 may be con-
nected with verse 61. Verses 135142, which apparently were revealed at the
same time as verses 143152, belong to the period after the Battle of Uud,
as verses 136 and 146 show that the Muslims even at that time were bat-
tling with changing fortunes.
160
In verses 152 to 168 Muammad summarizes
everything that he harbours against the Jewish people. These bitter senti-
ments can have hardly developed before he clashed openly with the Jews
of Medina. Closely connected with this are the verses 4:169 to 174, where
the Christians are also reproached with several false doctrines.
161
Tradition
is divided as to the origin of verse 175 (lawof inheritance). Some hold that it
was revealed on a campaign,
162
whereas others say that this happened when
the Prophet was once visiting Jbir b. Abd Allh while ill.
163
Finally, some
say that this verse originated from a response to a question from this same
Jbir during the Farewell Pilgrimage,
164
and was the very last of the entire
155
For example, when it is said that someone dealing in stolen goods, or even the thief
himself, had been a Jew.
156
al-Zamakhshar; al-abar, Tafsr.
157
Vol. 1, p. 184, s.v. .
158
Attention ought to be paid to
in the verses 125, 130 (twice)

and 131.
159
Cf. above, p. 155sq. on sra 3:89, and the comprehensive comments on sra 16:124 on
p. 119sq.
160
It is wrong that the Commentators attribute verse 135sq. to the Jews instead of the
Hypocrites.
161
Cf. verse 168, which closes the first part, as well as does verse 174, the second one.
162
al-abar, Tafsr; al-Suy, al-Itqn, p. 41.
163
al-Bukhr, K. al-Fari, 1; al-Tirmidh, K. al-Fari, 4; al-abar, Tafsr; al-Zamakh-
shar; al-Bayw; al-Wid; Fakhr al-Dn al-Rz; al-Nsbr; al-Nasaf.
164

. Fakhr al-Dn al-Rz; al-Zamakhshar; al-Khzin al-Baghdd;

Koran.
165
Althoughthis statement is oldandquite common, it deserves as lit-
tle confidence as all the rest, whichwiththe exceptionof one, andbecause
of the impossibility of fixing its timeare rather useless and totally defy
verification. Giventhat the Farewell Pilgrimage occurredinthe beginning of
March632ad, that traditionis incontradictiontoyet another one, according
to which the verse was revealed in summer.
166
Even this tradition, however,
looking ostensibly unsuspicious, is deprived of all credibility when we real-
ize that verse 15, with its slightly changed lawof inheritance, was transposed
by verse 175 to winter.
167
We must therefore be satisfied with the general con-
clusion that verse 175 was formulated some time after verse 15. When fixing
the dates of Koranic legal passages one should always consider that many of
themprobably do not go back to particular cases, for as soon as Muammad
was confronted at Medina with the task of leading and legislating, it must
have been in his interest to recast particularly offensive and objectionable
pagan institutions as quickly as possible and not anticipate the appearance
of specific cases.
Because of its similar content we now turn our attention to sra 65,
168
[i/205]
which complements sra 2:225sqq. If the first verse really referrs to Ibn
Umar, who had divorced his wife when she had her monthly courses upon
her,
169
the sra could hardly be fixed before 8/629, as at that time he was
only twenty years old. Yet this contradicts a tradition according to which,
on this occasion, Muammad recited only this particular verse.
170
Moreover,
al-Nsbr; al-Suy, al-Itqn, p. 49. Also the regulation of other matters of inheritance
tradition fixes to this time; cf. al-Bukhr, K. al-Maghz, 78 towards the end, J. Wellhausen,
Muammad in Medina, p. 432sq.
165
al-Bukhr, K. al-Fari, 14, and K. al-Tafsr; Muslim, K. al-Fari; al-Qasalln, vol. 7,
p. 62; al-Tirmidh, K. al-Tafsr; al-abar, Tafsr; al-Bayw; Fakhr al-Dn al-Rz; al-Wid
in the introduction, Cairo ed., p. 9, top; al-Qurub, fol. 23
v
; Abd al-Aad b. Muammad
al-arrn, fourth last leaf; al-Suy, al-Itqn, p. 60; al-Shshw, chapter 1; Al al-Dn
AL-KHZIN al-Baghdd.
166
al-Muwa, p. 328; Muslim, K. al-alt, 65 towards the end, K. al-Fari(al-Qasalln,
vol. 7, 59); al-Tirmidh, K. al-Tafsr; al-abar, Tafsr, vol. 4, p. 26; Fakhr al-Dn al-Rz; Al
al-Dn AL-KHZIN al-Baghdd; al-Suy, al-Itqn, pp. 49, and 924; al-Diyrbakr, Tarkh
al-khams, vol. 1, p. 12; Hibat Allh b. Salma on sra 4. It is precisely for this reason that
the verse is called the Verse of Summer (
). It is totally wrong when al-Itqn, p. 49,

connects the two traditions and even claims that all the verses revealed during the Farewell
Pilgrimage are summerly (
).
167
al-Diyrbakr, Tarkh al-khams, vol. 1, p. 12.
168
In Flgels verse 2 the end of the verse ought to be put after
, and the rest to be

united with verse 3; the same applies to the verses 10 and 11.
169
al-abar, Tafsr; al-Wid; al-Bayw; [Juynboll, Encyclopedia, p. 222, col. 1; G. Sale,
sra 65, foot-note].
170
Muslim, K. al-alq, 1, towards the endal-Qasalln, vol. 6, p. 257; Fakhr al-Dnal-Rz;
different versions of the origin have been transmitted,
171
all of which are
unreliable. One source attributes this sra to the Meccan period,
172
possibly
because of its ending, which resembles the style of the Meccan period.
Sra 59 refers largely to the subjection and expulsion of the Jewish tribe [i/206]
Nar in I Rab 4/625,
173
for which reason it is explicitly called the Sra of the
Nar.
174
It is difficult to say anything definite about the time of verses 18sqq.;
there is nothing against the assumption that they were revealed concur-
rently with the first part, particularly as verse one reappears verbatimat the
end of the sra.
Comments on Sras 33, 63, 24, 58, 22, 48, 66, 60, 110, 49, 9, 5
Sra 33 consists of several pieces. Verses 927 certainly originate from the
endof 5/626,
175
after the powerful allies, the Quraysh, Ghaafn, andQuraya,
had abandoned the siege of Yathrib, which was defended by a trench ( ),
in all the other recensions of this transmission(Fakhr al-Dn al-Rz, but particularly
in Muslim as well as al-Bukhr, al-Tirmidh, al-Nasaf, al-Muwaa; al-Khab al-Tibrz,
Mishct, at the beginning of K. al-alq)the Prophets reply refers to this verse only.
171
al-abar, Tafsr; al-Wid; al-Suy, Asbb al-nuzl.
172
In Umar b. Muammad (IBN ABD AL-KF).
173
This is the general point of view(Ibn Hishm, p. 653; al-Wqid, vol. 4. 353; Wellhausen,
Muhammed in Medina, p. 163; al-Bukhr, K. al-Maghz, 14, beginning; al-Baldhur, ed.
M.J. de Goeje, p. 18; al-abar, vol. 1, p. 1448; al-Masd, Prairies d or, vol. 4, p. 157; the
Commentators. Cf. G. Weil, Das Leben, p. 135sq.; Caussin de Perceval, vol. 3, p. 121; Muir, vol. 3,
p. 215sq.; Sprenger, Das Leben, vol. 3, p. 160sq.; Caetani, Annali, vol. 1, p. 584). Totally isolated
is Zuhrs tradition via al-Bukhr, loc. cit., 14, beginning; al-Diyrbakr, Tarkh al-khams,
vol. 1, p 460 top; this event occurred six months after the Battle of Badr, i.e., also in I Rab, but
in 3/624. There are still other accounts against this. The historians (Ibn Hishm, pp. 650 and
652; al-Wqid, p. 354; al-abar, vol. 1, pp. 1444 and 1448) suggest as cause of the campaign
against Ban Nar an event which was precipitated by the conflict of Bir Mana in Safar
4/625. In addition, a poem of a Jew on the death of Kab b. al-Ashraf (in Ibn Hishm, p. 659)
which cannot really have been composed after the expulsion of Ban Nar, mentions the
Battle of Uud:

*

Like Sakhrs power (Ab Sufyns) which you felt at the Uud where you did not have a
saviour!
174
al-Bukhr, K. al-Maghz, cap. 14, 3; Fakhr al-Dn al-Rz; al-Khzin al-Baghdd.
175
Ibn Hishm, p. 668sqq, particularly p. 693sq.; al-abar, vol. 1, p. 1463sqq.; al-Yaqb,
vol. 2, p. 50 (fifty months after the hijra); al-Wqid, p. 4sq. and 157 (Wellhausen, Muhammed
in Medina, p. 210); al-Baldhur, ed. de Goeje, p. 21, etc. The year seems certain and fits
the course of events much better than the mere tradition that gives Ibn Umars age as
fourteen at the Battle of Uud, and fifteen at the time of the Battle of the Trench (al-Bukhr,
K. al-Maghz, 31, beginning), assuming the encounter to have taken place in 4/625 (Ibn
Qutayba, p. 80, al-Diyrbakr, Tarkh al-khams, vol. 1, p. 479 bottom). Cf. al-Bukhr, loc. cit.,
and the Jewish Ban Quraya were shortly thereafter defeated by Muam-
mad. Verses 3640
176
belong to approximately the same time. They refer to
Zaynab bt. Jash,
177
whom Muammad wanted to marry, and who was the
divorced wife of Zayd,
178
a freedman and adopted son of the Prophet. This
divorce is fixed to 5/626,
179
and it is precisely this date that the statement of
this new marriage refers to as being before the campaign against the Ban
Mualiq.
180
Those verses can further be connected with: (1) verses 1 to 3, a kind of
introduction; (2) verse 4sq., in which Muammad explains that adopted
sons are not true sons in order to legitimate his marriage with the wife of his
adopted son; (3) verses 6 to 8, regarding the relation of the Prophet and his
wives to the believers; and, finally (4) verses 28 to 35, stipulations regarding
Muammads wives.
181
Verses 53 to 55 are commonly taken to refer to the guests at Zaynabs sec- [i/207]
ond wedding, where they stayed longer than the Prophet cared for.
182
Still,
some sources
183
also list other reasons, and we must admit that there were
occasions in the life of Muammad that might have prompted such a reve-
lation. Verse 59, a regulation about womens dress, is perhaps a later addi-
tion, in any case before 8/629. In this year Muammads daughter, Umm
where inadditionthe month(Shawwl) is supplied(cf. Muir, vol. 3, p. 156). This month, which
agrees with Ibn Isq, is probably to be preferred to that of al-Wqid (Dh l-Qada), adopted
by Wellhausen, Muhammed in Medina, p. 17sq., Caetani, Annali, vol. 1, p. 611 sq.; and F. Buhl,
Muhammeds Liv, p. 265.
176
al-Bukhr, K. al-alq, 8; al-Nas, K. al-Nik, 25; the Commentators. Cf. also the
following foot-notes numbers 179 and 180.
177
EI
2
; EQ; Juynboll, Encyclopedia, p. 155sqq.
178
EI
2
179
al-abar, vol. 1, p. 1460sqq.; al-Masd, Prairies d or, vol. 4, p. 157; Ibn Sad (al-abaqt):
Biographien der Frauen (vol. 8, p. 81).Also 3/624 is mentioned in Ibn al-Athr, Usd al-ghba
f marifat al-aba, vol. 5, p. 463; Ibn ajar al-Asqaln, vol. 4, p. 600; al-Nawaw, Tahdhb,
p. 842; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 500.
180
This is evident from the role played by Zaynaband even more so her sisterin the
story of the libel against isha. See below, foot-note 195 for citations.
181
With reference to verse 28sq., Muslims have much to say regarding the great dissention
between Muammad and his wives, but generally without precise citation (
.) The
adorned and legend-like story does not make it clear how the conflict originated. Cf. al-
Bukhr, K. al-alq, 5; Muslim, K. al-alq, 4; Fakhr al-Dn al-Rz; and the Commenta-
tors.
182
al-Bukhr, K. al-Nik, 47, K. al-Aima, end, K. al-Istdhn, 10; Muslim, K. al-Nik,
15; al-abar, Tafsr; al-Wid; Fakhr al-Dn al-Rz. Cf. Weil, Das Leben, p. 229, foot-note;
Caussin de Perceval, vol. 3, p. 151; Muir, vol. 3, p. 228sq.; Sprenger, Das Leben, vol. 1, p. 400sq.;
Caetani, Annali, vol. 1, p. 610sq. Less specific al-Tirmidh, K. al-Tafsr, etc. The so-called verse
of ijb of al-Nas, K. al-Nik, 25, establishes a different relation to Zaynab.
183
al-abar, Tafsr; al-Wid.
Kulthm,
184
died, leaving only Fima, so that an address to his daughters
(bantika), as in this verse, was no longer possible. For the definition of the
period of verses 49 to 51, which granted the Prophet inter alia a slave concu-
bine, it must be remembered that the first time he made use of this permis-
sion was the case of the slave girl Rayna,
185
taken prisoner in the campaign
against the Quraya in 5/626. Verse 52 certainly belongs to the last years of
Muammads life.
186
Verse 48 regarding a particular case of divorce is a com-
plement to sra 2:237 and seems to date from about the same time as the
majority of the sra. Verses 4147, 5658, and 6073 deal with the Prophets
relation to believers and unbelievers. Also, the language of these verses has
something incommon; cf. verse 42 with56, andthe word inthe verses 47,
57, 58, 59, and 69, which otherwise appears only in verse 53. The arrange-
ment of the individual parts of this sra is irrelevant because the manifold
regulations of family and property law(verses 48, 2840, 48, and 5355) are
carelessly interspersedwithremarks onthe Prophet andhis contemporaries
(4147, 5658, and 6073), while the rambling address regarding the Battle
of the Trench (verses 927) does not fit with any part of the sra and merely
results in the separation of identical subjects. Despite this confusion, the
address vacillates regularly between
(verses 9, 41, 48, 53, and

69 (apart fromthe sole exception of verse 70)) and (verses 1, 28, 44,
49, and 50). This cannot be an accident but must be the intention of either
the writer or the editor. Since it wouldappear that most of the parts are close
to one another in time, however, it is conceivable that the sra received its
present form from the Prophet himself.
187
Sra 63 is fixed to the time shortly after the campaign against Ban [i/209]
Mualiq, a branch of Ban Khuza, and refers to the rebellious words that
Abd Allh b. Ubayy used on this occasion.
188
Verses 9 to 11 might also belong
to a different period, as they are not connected with the content of the rest.
184
Cf. IbnSad, Biographiender Frauen (vol. 8, p. 25); Ibn Qutayba, ed. Wstenfeld, p. 69sq.;
Ibn ajar, vol. 4, p. 949, etc. [EQ; s.v.; Juynboll, Encyclopedia, p. 427.]
185
Ibn Hishm, p. 963; al-abar, vol. 1, p. 1497sq.; Ibn Sad (al-abaqt): Biographien der
Frauen (vol. 8) p. 92sqq.; Ibn ajar, vol. 4, p. 591 sq. Cf. Weil, Das Leben, p. 170sq.; Muir, vol. 3,
p. 272; Sprenger, Das Leben, vol. 3, p. 77sq.; Caetani, Annali, vol. 1, p. 634.
186
Weil (Das Leben, p. 358sq., and Historisch-kritische Einleitung, 2nd ed., p. 86) presented
evidence for this against the strange views of Muslims (cf. the Commentators.)
187
Cf. thereon Sprengers remarkable exposition in his Das Leben, vol. 3, p. xxxii sq.
188
Ibn Hishm, p. 726sq., and 360, where it says that the entire sra ( ) was revealed
at that time. Muslim, K. al-ift al-munfiqn, beginning; al-Tirmidh, K. al-Tafsr; al-Bukhr,
ibid.; al-abar, vol. 1, p. 1512sq.; Ibn al-Athr (al-Kmil), vol. 2, p. 147; Wellhausen (al-Wqd),
Sra 24, verses 1 to 3 deal with the sin of prostitution as well as wedlock [i/210]
with prostitutes, verses 4 to 10 with fornication and the penalty for those
who falsely accuse women of this offence. According to tradition, verses 6
to 9 go back to Uwaymir b. rith
189
or Hill b. Umayya,
190
who suspected
their wives of extramarital relations. The name of the guilty man, Shark b.
al-Sam,
191
is mentioned as frequently in connection with Hill as im b.
Ad is with his friend Uwaymir. Nowhere in the older sources
192
does Shark
appear as the seducer of the woman. By the same token, it is very late works
that mention the name of the frivolous woman.
193
The date of the offence
Muhammed in Medina, p. 179sqq.; al-abar in the Tafsr; al-Wid; al-Zamakhshar; al-
Bayw. Cf. Caussin de Perceval, vol. 3, p. 162sq.; Weil, Das Leben Mohammeds, 148sqq.;
Muir, vol. 3, p. 240; Sprenger, Das Leben, vol. 3, p. 193; Caetani, Annali, vol. 1, p. 602sq.
As far as the time of this war is concerned we can fix it to Shabn 6/627: Ibn Hishm,
p. 661 and 725; al-Bukhr, K. al-Maghz, 34; al-abar, vol. 1, p. 1520; Ibn al-Athr, Usd al-
ghba, vol. 1, p. 22; less accurately al-Yaqb, vol. 2, p. 53; al-Masd, Prairies d or, vol. 4.,
p. 143. There is something to be said for this date as Muammads wiveswho during
the Battle of the Trench in 5/626 had not yet been living in retreat from the world(Ibn
Hishm, p. 687; al-abar, vol. 1, p. 1489; Fakhr al-Dn al-Rz on sra 33:9; al-Bukhr, loc.
cit., 36) were totally segregated during this campaignas is evident from the tradition
on sra 24 which is cited below. On the other hand, another tradition (al-Wqid, p. 4,
and Wellhausen, Muhammed in Medina, p. 175; Ibn Qutayba, p. 80; al-Diyrbakr, Tarkh al-
khams, vol. 1, p. 470)which, incidentally, Weil, Das Leben, p. 143sq.; Muir, Life of Mahomet,
vol. 3, pp. 233237, and Caetani, Annali, vol. 1, p. 600 followmentions the year 5/626. Al-
Diyrbakr adds that the campaign took place five months and three days after the one on
Dmat al-Jandal [Juynboll, Encyclopedia, p. 538, col. 1, n. 1]. However, Ibn Qutayba, who
concurs with Ibn Isq and the other sources mentioned above in so far as he fixes the
Battle of the Trench (according to him in 4/625) before our campaign. Dated back still
one more year (4/625)[sic] is this event in the frequently cited (al-Bukhr, loc. cit., 34,
beginning; al-Diyrbakr, vol. 1, p. 470) Maghz of Ms b. Uqba, [EI
2
; Juynboll, Encyclope-
dia, p. 283, etc.; Sezgin, GAS, vol. 1, pp. 286287] on which perhaps also al-Masd, Prairies
d or, vol. 4, p. 157, is based, and which Sprenger (Das Leben, vol. 3, p. 192) is inclined to fol-
low.
189
al-Muwa, p. 206; al-Bukhr, K. al-alq, 4, K. al-Tafsr; Muslim, K. al-Lin, al-
Qasalln, vol. 6, p. 315sqq.; al-Nas, K. al-alq, Bb 28; Ibn Qutayba, p. 170; Ab l-Layth
al-Samarqand; Fakhr al-Dn al-Rz; al-Khab al-Tibrz, Mishct, Bb al-lin, beginning; Ibn
ajar al-Asqaln, vol. 3, p. 88; Ibnal-Athr, Usdal-ghba, vol. 4, p. 158sq.; al-Zamakhshar; EI
2
.
190
Muslim, K. al-Lin; al-Qasalln, vol. 6, p. 326; al-Tirmidh, K. al-Tafsr; al-Nas, K.
al-alq, Bb 30; al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid; Fakhr al-Dnal-Rz;
al-Bayw; al-Zamakhshar; al-Khab al-Tibrz, Mishct, loc. cit.; Ibn ajar, vol. 3, p. 1250,
and vol. 2, p. 414sq.; Ibn al-Athr, Usd al-ghba, vol. 2, p. 397sq., and vol. 5, p. 66; al-Nawaw,
ed. Wstenfeld, p. 609; Juynboll, Encyclopedia, p. 182, col. 2.
191
Juynboll, Encyclopedia, p. 182.
192
Fakhr al-Dn al-Rz; Ibn al-Athr, Usd al-ghba, vol. 4, p. 158; al-Nawaw, p. 491.
193
Fakhr al-Dn al-Rz; al-Zamakhshar.The name of the woman, Khawla, is probably
derived from the traditions on sra 58.
some writers
194
fix to Shabn 9/630, after Muammad had returned from
Tabk. Very little of this exegetic scholarship will pass critical examination.
Even if in the family of Uwaymir or Hill [Ibn Umayya] such a case of
fornication might have occurred, its connection with our Koranic passage
is probably a fabrication.
Whereas the previous verses are presented in a general vein, verses 10
to 20 focus on an unmistakably specific event. The conspicuously uniform
transmissioninterprets this torefer toishas well-knownadventure during
the previous campaign against Ban Mualiq, and suspects her of having
commited an offence with a strange man.
195
In this case there is no rea-
son to challenge the reliability of the tradition. As it seems, this passage
revealed approximately one month
196
after the campaignwas inserted in
the present context only later. The composition of the latter would there-
fore belong to an earlier time. Verses 27 to 33 and 57 to 61, although both
containing regulations governing respect for decency and manners inhouse
and family, are apparently of a different origin.
197
Inserted in between is now
a piece of an entirely different character, belonging to the most lofty and
poetical passages of the Medinan sras, namely verses 34 to 44, which jux-
tapose Allh, the Light of the World, and the darkness of unbelief; the same
applies to verses 45 to 56, which have the identical beginning
198
but are com-
posed in a more simple style. They take exception to the behaviour of the
Hypocrites (verse 49) and the unreliable elements who, despite their sworn
obedience, refuse to followthe Prophet loyally on his campaigns (verse 52).
Such polemics must belong to a time when Muammads fortune was at a
lowpoint, possibly in the period between the Battle of Uud and the end of
the Battle of the Trench. The old tradition
199
that regards verses 62 to 64 as
194
Ibn al-Athr, Usdal-ghba, vol. 4, p. 158; al-Nawaw, ed. Wstenfeld, p. 491; al-Nawaw on
Muslimal-Qasalln, vol. 6, p. 316, all with reference to al-abar, probably his Annales, vol. 1,
p. 1705, but where it is merely stated that Hill remained back in Medina during the Tabk
campaign. Cf. also al-Diyrbakr, Tarkh al-khams, vol. 2, p. 133.
195
Ibn Hishm, p. 731 sq.; al-abar, vol. 1, p. 1517sqq.; Wellhausen, Muhammed in Medina,
p. 184sqq.; al-Bukhr, K. al-Shahdt, 15, Maghz, 36, Tafsr; Muslim, K. al-Tawba, 11;
al-Tirmidh, K. al-Tafsr; Ab l-Layth al-Samarqand; al-Wid; Fakhr al-Dn al-Rz. Cf. Weil,
Das Leben, p. 151 sqq.; Caussin de Perceval, vol. 3, p. 164sqq.; Sprenger, Das Leben, vol. 3,
p. 63sqq.; Muir, Life of Mahomet, vol. 3, p. 244sqq.; Aug. Mller, Der Islam, vol. 1, p. 133sq.;
Caetani, Annali, vol. 1, p. 604sqq.; Buhl, Muhammeds Liv, p. 275sqq.
196
Cf. the sources cited in the previous note.
197
Each part has a separate introduction, and verse 61 is parallel to verses 2729.
198
) (
.
199
Ibn Hishm, p. 670; al-abar, vol. 1, p. 1465sq.; al-Zamakhshar.
referring to the digging of the entrenchment outside Medina might be
correct, even though it is almost certainly based on exegetic conjecture
alone.
Sra 58 deals with subjects similar to those in sra 24. The text of the [i/212]
first verses (58:15) makes it clear that it is occasioned by a particular event.
It is unanimously agreed that all details refer to Aws b. al-mit, who had
separated from one of his wives, called Khawla or Khuwayla,
200
with the
pagan formula, be as my mothers back,
201
but later regretted and wanted
to resume his marital intercourse without further ceremony. Later sources
202
date this event shortly after the return from al-udaybiyya, i.e., at the end
of 6/627 or the beginning of 7/628. Earlier sources, however, lack this infor-
mation. The rules for due respect for the Prophet fit in with the period of
sra 24; they certainly do not originate from the first years after the hijra.
Verses 6 to 9, and 15sqq. are addressed to the Hypocrites. The separate parts
(verses 15, 69, 10, 11, 12, 13sq., and 15sq.) belong chronologically close to
one another, as is also evident fromthe phraseology.
203
Some regard verses 1
to 10
204
or 9 to 11
205
to be of Meccan origin for no good reason.
Sra 22, which is commonly regarded as Meccan but occasionally also [i/213]
as Medinan,
206
is primarily important for the Medinan parts which it con-
tains, despite the fact that it was largely revealed during the third Mec-
can period before the hijra. Of Meccan origin are verses 1 to 24of which
verses 5 to 7 do not fit in with the contextand the verses 43 to 56, 60
to 65, and 67 to 75. Of these verses several are unjustly regarded as Medi-
nan: verses 1 sqq. are said to have been revealed on the campaign against
200
In Ibn al-Athrs Usd al-ghba f marifat al-aba, vol. 5, p. 417; Ibn ajar al-Asqaln,
vol. 1, no. 338, vol. 4, p. 503. In al-Nawaw, ed. Wstenfeld, p. 839, we find the variant,
, but this is most likely only an error. [G. Juynboll, Encyclopedia of canonical adth,
p. 698.]Behind
, verse 3, Flgel erroneously has the end of the verse.

201
Ibn Qutayba, p. 131; Ibn Sad (al-abaqt): Biographien der medinischen Kmpfer, p. ,
and his Biographiender Frauen, p. 274sq.; al-abar, Tafsr; al-Tirmidh, K. al-Tafsr; al-Wid;
Fakhr al-Dn al-Rz; al-Zamakhshar; al-Bayw; Ibn ajar al-Asqaln; Ibn al-Athr, Usd
al-ghba f marifat al-aba, and al-Nawaw under the relevant headings.
202
al-alab, Insn al-uyn, vol. 3, p. 140sqq. Already al-Diyrbakr, Tarkh al-khams,
vol. 2, p. 25sq., a work which is a bit older, mentions this, but without listing authorities.
Weil, Das Leben Mohammeds, p. 184, and note on p. 280.
203
Cf. verse 5 to verses 6, 21, and 22; verses 13sq. deal with like verses 6
to 9; the beginning of verse 21 is identical with that of verse 6; verses 8, 9, and 15.
204
al-Bayw.
205
206
Cf. Umar b. Muammad; al-Suy, al-Itqn, p. 26sq., and the Commentators.
Ban Mualiq,
207
possibly because Muammad once recited them at that
time; verses 11 sq. some people fix to after the emigration, because they were
regarded as a reference to those Arab tribes who accepted Islam only later,
or to things which happened after the hijra.
208
By the same token, in the case
of verse 15, some commentators had Arab tribes in mind when they inter-
preted the word yanuruh ( ) as rendering victorious.
209
Quite common
is the view that verses 20 to 22 or 20 to 23 or 20 to 24 refer to individual
encounters of Al and some followers with prominent Quraysh in the Battle
of Badr.
210
This explanation is based solely on the literal interpretation of the
word khamn ( ).
211
Verses 18 and 19 must have been originally out of
context, as their rhyme does not agree with the other verses, and combining
them with verse 20 is neither documented nor permissible. Verse 17 is obvi-
ously inserted later. For its Medinan origin speaks the expression alladhna
hd, used by the Jews, which certainly does not occur in the older pas-
sages of the Koran.
212
Most of the remaining Medinan verses (2538) seem
to date from the period of the pilgrimage of 6/627 or 7/628, as they contain
a guide for pilgrims. We cannot date them to the time after the conquest of
Mecca because verse 25 reproves the disbelievers not only for insufficient
upkeep of the Kaba but also for barring believers from Gods way and the
Holy Mosque, and verse 31 advises the believers to eschew the abomina-
tion of idolsall remarks which became superfluous after the occupation
of the Holy City. Verse 66, which begins as does verse 35, must likely be
attributed to the same period. If, as tradition has it,
213
verses 3942 were the
first to permit to Muslims open warfare with the enemies of the true reli-
gion, they wouldhave tooriginate fromthe time before the Battle of Badr. On
the other hand, this, Muammads permission to fight, cannot possibly have
207
Fakhr al-Dn al-Rz; al-Zamakhshar; al-Suy, al-Itqn, p. 43; al-abar, Tafsr, only
says that Muammad recited these verses or when he returned from
(i.e. the campaign against Tabk; cf. al-Bukhr, K. al-Maghz, 79).
208
Umar b. Muammad (IBN ABD AL-KF); al-Wid; Fakhr al-Dn al-Rz; al-abar,
Tafsr.
209
Fakhr al-Dn al-Rz.
210
al-Wqid, p. 64; Muslim in the very inadequate K. al-Tafsr at the end; al-abar,
Tafsr; Ab l-Layth al-Samarqand; Fakhr al-Dn al-Rz; al-Zamakhshar; al-Bayw; Umar
b. Muammad (IBN ABD AL-KF); al-Suy, al-Itqn, p. 19.
211
However, the Commentators supply also different interpretations of the passage.
212
References to this phrase: sras 2:59, 4:48 and 158, 5:45, 48 and 73, 6:147, 16:119, 62:6; all [i]
are Medinan. Cf. also Hirschfeld, Newresearches, p. 125.
213
al-Yaqb, Historiae, vol. 2, p. 44; al-Nas, K. al-Jihd, beginning; al-abar, Tafsr;
Ab l-Layth al-Samarqand; al-Wid; al-Zamakhshar, al-Bayw; al-Suy, al-Itqn, p. 58;
G. Weil, Leben Mohammeds, p. 94; Sprenger, Das Leben und die Lehre, vol. 3, p. 100.
been issued before the hijra.
214
Verse 51 is regularly
215
applied to the Meccan
goddesses, al-Lt, al-Uzz, and Mant, whose veneration the Prophet was
inclined to tolerate at a weak moment. This explanation, however, is based
on the conjecture that the words tamann and umniyya meant qaraa to
recite, qira recitation.
216
The verse couldactually very well be Meccanif it
were not connectedwithverse 52, whichamong the enemies clearly empha-
sizes the Hypocrites.
217
Verses 57 to 59 must be later than the Battle of Badr,
as they already mention believers who had been killed in the war.
218
Further-
more, an old exegete justifiably regards verse 76sqq. as Medinan.
219
Because
of the call for holy war, their composition must be dated to before the Bat-
tle of Badr at the earliest,
220
and because of the mention of the religion of
Abraham, near the time of the first wars against the Jews.
221
Sra 48 must be fixed at the time after the Pact of al-udaybiyya (in Dh [i/215]
l-Qada 6/627), althoughverses 1 to17 alone date fromshortly after its conclu-
sion, most likely still before Muammads return to Medina,
222
a date which
214
Ibn Hishm, p. 313; cf. cod. Sprenger, no. 207.
215
See the references to sra 53, above, p. 82sq.
216
This meaning is unknown to the Koran, although some want to find themalso in other
Koranic passages, e.g., Ibn Hishm, p. 370, from Ab Ubayda in
, sra 2:105; they also do

not seem to occur in ancient poets. Muslims, of course, cite the following reference:

(Ibn Hishm, p. 370; Ab l-Layth al-Samarqand; al-Zamakhshar on sra 22:51; cf. Shar
al-shawhid, s.v.; al-Bayw on sra 2:73, 22:51, and Lisn, vol. 20, p. 164 have
, a
confusion with the following verse, of which only half is cited in al-Zamakhshar), and

(Fakhr al-Dn al-Rz; Ab l-Layth al-Samarqand; al-Zamakhshar on sra 2:73; Majd al-Dn
IBN AL-ATHR, Nihya f gharb al-adth, vol. 4, p. 111; Lisn al-Arab, loc. cit.; Ibn Hishm,
p. 371, with a variant reading instead of ), the latter verse is said to refer to the death of
Uthmn (Fakhr al-Dn al-Rz; al-Zamakhshar; Lisn and Nihya, ibid.; Shar shawhid). In
any case, it is likely that the present unique meaning of is derivedfromthat misinterpreted
Koranic passage.
217
Those in whose hearts is sickness; this, according to standard Koranic usage, refers to
the munfiqn.
218
Even if was passive (man qutila) who is being killed or when someone is being
killed, these verses might still have beenrevealedbefore the actual battle, but
shows,
excluding conditional meaning, the completed fact, those who have been killed.
219
Umar b. Muammad (IBN ABD AL-KF).Flgels verses 77 and 78, for syntactical
reasons, can constitute only one verse, a fact which is confirmed by tradition.
220
Cf. above on verse 39sqq.
221
Cf. above, p. 119sqq. on sra 16:124.
222
See verses 11 and 15: Those who were left behind will say to thee (after your return).
Cf. Muir, The Life of Mahomet, vol. 4, p. 30sqq.
many holdfor the entire sra.
223
These verses demonstrate more clearly than
the accounts of historians that Muammad was planning to attack Mecca
already at that time but that the Bedouins who were allied with him dis-
appointed him. Nevertheless, on account of their large number and with
their help two years later, he was able to take the city nearly without a
blow. For this reason he abandoned his plan to force his entry into the Holy
City, and came to a compromise with the Quraysh, which, apart from other
advantages, guaranteedhimanunmolestedpilgrimage inthe following year.
That this armistice was a masterpiece of his politics and a true victory
224
is
best demonstrated by its result. Verses 18sqq. originate from the time of
the submission of the Jews of Khaybar and its vicinity (beginning 7/628),
whose wealth Muammad had promised his followers after returning from
al-udaybiyya.
225
Verses 19, 20, and 27 cannot be explained otherwise. After
this success, he was entitled to regard the result of al-udaybiyya in retro-
spective; this is why also in these sections he speaks much about it, most
clearly in verse 23, where he tries to convince the Muslims that God sup-
ported him there as much as He did at Khaybar. It is wrong of tradition to
fix verse 27 after the pilgrimage of 7/628.
226
Traditionconnects the first verses of sra 66witha scandal inthe house of [i/217]
the Prophet.
227
It once happened that Muammad was using the roomof his
wife afa [Bint Umar] for a rendezvous with the Coptic slave Mary(am).
This not only constituted an offence against good manners but was also a
223
Ibn Hishm, p. 749sqq.; al-Wqid (Wellhausen, Muhammed in Medina, p. 260); al-
Bukhr, K. al-Maghz, 37, and K. al-Tafsr; al-Wid, al-Bayw. Less explicitly al-Mu-
wa, p. 71. One tradition in Muslim (K. al-Jihd, 29, al-Qasalln, vol. 7, p. 424sq.) says
this only with regard to the first five verses. Cf. Muir, The Life, vol. 4, p. 36sqq.; Sprenger, Das
Leben, vol. 3, p. 251 sqq.; Buhl, Muhammeds Liv, p. 285; Aug. Mller in Fr. Rckerts translation
of the Koran.Hirschfeld, Newresearches, p. 127, regards the verses 1 to 17 as referring to the
period after the conquest of Mecca. That this follows from verse 12, as he claims, is hard to
believe.
224
Thus, verse 1 is easily explained (cf. the Commentators), so that we need not put the
first verses after the campaign against Khaybar.
225
Vv. 15 and 20. As regards the large spoils of war cf. Ibn Hishm, p. 773sqq.; al-abar,
vol. 1, p. 1588sqq.; al-Wqid (Wellhausen, Muhammed in Medina, p. 284sqq.); al-Baldhur,
p. 25sqq.; Caussin de Perceval, vol. 3, p. 202; Muir, Life of Mahomet, vol. 4, p. 73sqq., Sprenger,
Das Leben, vol. 3, p. 274sqq.; Caetani, Annali, vol. 2, part 1, p. 38sqq.
226
al-abar, ed. Zotenberg, vol. 3, p. 111.That the words You shall enter [the Holy
Mosque,] if God wills can refer only to the future and need no additional evidence.
227
al-Nas, K.Ishrat al-nis, 4 (only a brief reference); Fakhr al-Dn al-Rz; al-Wid;
al-abar, Tafsr; al-Zamakhshar; al-Bayw. Cf. Weil, Das Leben Mohammeds, p. 274sqq.;
Caussin de Perceval, vol. 3, p. 268; Sprenger, Das Leben, vol. 3, p. 85sq.; Muir, Life of Mahomet,
vol. 4, p. 160sqq.; Caetani, Annali, vol. 2, part 1, p. 236sqq.
serious trespass. afa,
228
returning home unexpectedly, surprised the two
and, reproaching the Prophet bitterly, incited isha and the other wives
against him. The mistake of their leader must have caused the gravest anxi-
ety, or he would not have considered it prudent to be exculpated by a special
revelation. This tradition bears the guarantee of its historicity in itself. An
episode showing Muammads character insuchanunfavourable light Mus-
lims can have neither fabricated nor derived from the rumors of the unbe-
lievers. Another traditionhas it that because of fondness for afa, who dur-
ing the Prophets visits regularly treated him to honey, he had neglected his
other wives for some time.
229
Because of its harmlessness, this silly account
is often referred to as an explanation of the first verse of this sra, although
it could not displace the other interpretations. All that can be said regard-
ing the period is that it probably happened before the birth of Ibrhm(Ibn
Muammad). If this had not been the case, Mary(am)s merit for having
borne the Prophet his first child after twenty-five years, and a son at that,
would not have remained unmentioned in the accounts of this scene of
jealousy.
230
Verses 6 to 8, which alternatively are addressed to believers and
unbelievers, as well as verses 8 to 12, which describe models of believing and
unbelieving women, defy close identification on account of their general
content. Verse 9, which alone in this sra makes particular mention of the
Hypocrites (munfiqn), is identical with 9:74. Since the verse is necessarily
before 9:75, it seems to have its original place in sra 9 but not sra 66:10.
The beginning of sra 60 (verses 1 to 9) warns Muslims not to be friends [i/218]
with people who at that time had driven Muammad and his followers into
emigration, although leaving open the possibility of a later reconciliation
228
Juynboll, Encyclopedia, pp. 155, 162, 183, 184, 197, 232 etc.
229
Ibn Sad (al-abaqt): Biographien der Frauen, ed. by Brockelmann, p. 76; al-Bukhr,
K. al-alq, 8, K. al-Aymn, 23; Muslim, K. al-alq, 3; al-Nas, K. al-alq, bb 10,
K. al-Aymn, 20, K.Ishrat al-nis, 4; al-Khab al-Tibrz, Mishct, bb al-alq, fal 1,
5; Fakhr al-Dn al-Rz; al-Wid; al-abar, Tafsr (brief allusion only); al-Zamakhshar;
al-Bayw.
230
Muir, loc. cit., and Caetani, loc. cit., both fix the time of this event between Ibrhm
(Ibn Muammads) birth and the campaign against Tabk. According to Weil, Das Leben,
p. 274sqq., and Historisch-kritische Einleitung, 1st ed., p. 78, it is even still later, which in its
2nd ed., p. 88, is founded on the fact that the verses 6 to 8 refer to those who remained
behind during this campaign, and that there is no reason also to fix the first verses in this
time. But this apparent relation is by no means clear. The slave girl Mary(am) is assumed to
have been part of the presents which the ruler at Alexandria sent to the Prophet in response
to his embassy. (al-abar, vol. 1, p. 1561; Ibn Sad, Wellhausen, Seine Schreiben und seine
Gesandtschaften, p. 99sq.) Unfortunately the year of the embassy cannot be determined. Cf.
Wellhausen, loc. cit., and Caetani, Annali, vol. 1, p. 730sq.Verse 5, which has a rhyme (r)
totally different from the other verses, seems to have lost its original end.
(verse 7). Tradition fixes this passage to the time shortly before the conquest
of Mecca in Raman 8/629 and attributes it to the secret message from
ib b. Ab Baltaa
231
to the Quraysh, informing them of Muammads
imminent assault on Mecca.
232
This conjecture might be correct but no
evidence can be procured. The only thing certain is that the passage belongs
to the time before that conquest.
233
Verse 10sq. must originate from the
time shortly after the Pact of al-udaybiyya
234
and not, as tradition will have
it,
235
still at al-udaybiyya itself. It is inconceivable that after the conclusion
of the Pact the Prophet would have handed over the men who sought his
protectionwhile he wouldhave givenhis protectiontothe womenwhowere
more firmly tied to their families by tribal law (than the men). Verse 12 can
easily be connected with this. The view that it originates from the time of
the conquest of Mecca cannot be substantiated; it is to be found in very late
writers only.
236
Verse 13, it seems, relates tothe same subject as the first verses
and might have been revealed at the same time.
Sra 110, which looks like a fragment, is found in the Koran among all [i/219]
sorts of Meccan sras, and for this reason it is frequently reckoned to be
among them.
237
However, the optimistic idea that people would flock to the
231
Juynboll, Encyclopedia, pp. 271, 599.
232
Ibn Hishm, p. 809sq.; al-abar, vol. 1, p. 1627; al-Wqid in Wellhausen, Muhammed
in Medina, p. 325; al-Bukhr, K. al-Maghz, 48, K. al-Tafsr; Muslim, K. al-Fail 71;
al-Tirmidh, K. al-Tafsr; Khab al-Tibrz, Mishct, Bb al-jmi al-manqib, fal 1, 27; al-
Wid; Fakhr al-Dn al-Rz; al-abar, Tafsr; al-Zamakhshar; al-Bayw; Ibn al-Athr, Usd
al-ghba f marifat al-aba, vol. 1, p. 361; Ibn ajar al-Asqaln, vol. 1, no. 1532. Cf. Weil, Das
Leben, p. 209sq.; Caussin de Perceval, vol. 3, p. 221 sqq.; Muir, vol. 4, p. 113sq.; Caetani, Annali,
vol. 2, part 1, p. 117.
233
Cf. also Muir, Life of Mahomet, vol. 4, p, 114, n. 1.
234
This assumption corresponds well to the following reports: Ibn Hishm, p. 754; al-
Wqid (Wellhausen, Muhammed in Medina, p. 263); al-abar, vol. 1, p. 1552sq.; Ibn Sad
(al-abaqt): Biographie der Frauen, p. 168; Ibn ajar al-Asqaln, vol. 4, p. 953; Ibn al-Athr,
Usd al-ghba f marifat al-aba, vol. 5, p. 614.
235
al-Bukhr; K. al-Shur, 15; Hibat Allh b. Salma; al-Khab al-Tibrz, Mishct, Bb
al-ul, fal 1 , 1; al-Wid; Fakhr al-Dn al-Rz on sras 48 and 60; al-abar, Tafsr;
al-Zamakhshar; al-Bayw; al-Suy, Asbb al-nuzl. Cf. Weil, Das Leben, p. 183; Muir, Life,
vol. 4, p. 44sqq.; Caetani, Annali, vol. 1, p. 723.
236
The verse, of course, occurs in many traditions, mentioning the homage paid to the
women after the assault of Meccaal-Khab al-Tibrz, Mishct, bb al-ul, fal 1, 4;
al-Bukhr, K. al-Akm, 49; al-Nas, K. al-Baya, 18; al-abar, Tafsr; al-Zamakhshar;
al-Muttaq al-Hind; al-Nasaf; al-Suy, Asbb al-nuzlbut it is seldom that we find an
indication that it was revealed on this occasion, e.g., in al-Bayw, and in the Persian
al-abar, ed. Zotenberg, vol. 3, p. 138; Fakhr al-Dn al-Rz is rather vague.
237
Umar b. Muammad (IBN ABD AL-KF); Hibat Allh b. Salma; Muir, vol. 2, p. 319;
Sprenger, Das Leben, vol. 1, p. 560, was tempted to this view by a certain relationship of the
beginning of the sra with sras 26:118 and 32:28.
true religion (al-dn) speaks rather in favour of the later Medinan period.
On the other hand, it is doubtful that the words al-nar wa-l-fat (verse 1)
need to be interpreted as applying to the conquest of Mecca
238
and, there-
fore, in agreement with most traditions, be dated to the time around this
event.
239
Others go still further and have the sra predict the approaching
death of the Prophet, even considering it outright as the very latest revela-
tion.
240
Sra 49 consists of several parts. Nearly all commentators regard verses 1 [i/220]
to 5 as referring to the deputation of the Ban Tamm,
241
who arrived in
Medina in 9/630 or 10/631
242
to negotiate the release of prisoners; when
Muammad did not meet them immediately, they became noisy and pro-
voking. This agrees only too well with the text of verses 2 to 5, and we would
have to regard this tradition as absolutely reliable if it were certain that
its form was not influenced by the Koran. Less obvious is the situation on
which verses 6 to 8 are founded. We are told that they refer to the Umayyad
Wald b. Uqba b. Ab Muay,
243
who at the same time had set out to col-
lect taxes fromBan Mualiq but returned empty-handed, falsely accusing
the tribe of having refused to pay the tax and of having made a personal
assault on his life.
244
Although the text of the revelation can be interpreted
this way, the traditionis suspect, particularly because it concerns a manwho
238
Cf. above on sra 61:13.
239
Muslim, K. al-Fail al-Qurn 2, says that Muammad recited (qaraa) the sra in the
year or even on the very day of the conquest of Mecca. Al-Wid has it revealed (nazala) on
the return from the Battle of unayn.
240
al-Bukhr, K. al-Tafsr; Muslim, K. al-Tafsr (al-Qasalln, vol. 10, p. 487); al-Tirmidh, K.
al-Tafsr on sra 5 at the end, and on sra 100, etc.; al-abar, Tafsr; Ibn Hishm, p. 933, note;
al-Itqn, pp. 45 and 61; Leiden Cod. 653 Warner, fol. 275
v
; Ibn Qutayba, p. 82.
241
Explicitly referring tothe verses 1 to5 only inal-Bukhr, K. al-Maghz, 68, K. al-Tafsr;
W. [unidentified abbreaviation of the German authors]; and Fakhr al-Dn al-Rz; less explic-
itly in al-Tirmidh, K. al-Tafsr; al-abar, Tafsr; al-Zamakhshar, and al-Bayw; only verse 2
mentioned in: assn b. Thbit, Dwn, ed. Tunis, p. 113, and al-Wqid (Wellhausen, Muham-
med in Medina, p. 386); only verse 4 in: Ibn Hishm, p. 939. l 4; Ibn Sad (Wellhausen, Skizzen
undVorarbeiten, Heft 4, no. 78); al-abar, vol. 1, p. 1717; al-Aghn, vol. 4, p. 9. However, verses 2
to 5 are closely related, whereas this is doubtful in the case of verse 1. Cf. Weil, Das Leben
Mohammeds, p. 244sqq.; Caussin de Perceval, vol. 3, p. 271; W. Muir, Life of Mahomet, vol. 4,
p. 171 sq.; Sprenger, Das Leben, vol. 3, p. 366sq.; Caetani, vol. 2, part 1, pp. 219sq. and 449sq.
242
Ibn Sad (Wellhausen, Skizzen und Vorarbeiten, Heft 4, p. 137sqq.) and al-Aghn, vol. 4,
p. 8, do not supply a year. The general heading in Ibn Hishm, p. 933, indicates that 9/630
is meant. Al-abar, vol. 1, p. 1710, mentions 9/630; less explicitly, al-Waqid (Wellhausen,
Muhammed in Medina, p. 386).
243
EI
2
; Juynboll, Encyclopedia.
244
Ibn Hishm, p. 730sq.; Ibn Qutayba, p. 162sq.; al-Wid; Fakhr al-Dn al-Rz; al-abar,
Tafsr; al-Zamakhshar; al-Bayw; Ibn ajar al-Asqaln; Ibn al-Athr, Usd al-ghba, al-
Nawaw, ed. Wstenfeld, s.v. Wald was the brother of the future Caliph Uthmn.
later became a notorious Muslim. The nickname, al-fsiq the evil-doer,
by which he is often known, naturally goes back to the exegetic tradition.
Verses 9 and 10 speak of the fight among the Muslim clans; verses 11 and
12 prohibit backbiting and nicknames. The question of whether these two
parts were connected from the beginning with each other as well as with
verse 6sqq. cannot be determined.
245
Verse 13 develops the idea that in Islam
the noblest among you in the sight of God is the most godfearing of you.
Tradition, which considers it a reference to the arrogance of the Quraysh
after the occupation of Mecca,
246
is remarkable, as the verse can be applied
to many other circumstances.
247
The verse has no connection with the previ-
ous one, although also originally not with the following one either. Verses 14
to 17 properly characterize the Bedouins who accepted Islam outwardly
because belief had not yet entered their hearts. This part is commonly
held to refer to the Ban Asad b. Khuzayma,
248
who, during the famine of
9/630,
249
appeared in Medina and, by pointing out their voluntary accep-
tance of Islam, demandedfood. Another tradition
250
applies the verses tothe
Bedouin tribes who did not follow Muammad to al-udaybiyya. But since
these verses merely served to demonstrate the main characteristics of all
Bedouinsproud and arrogant character next to superficial conversion
and since any particular allusion is wanting, none of those traditions can be
verified; both of them are probably based on conjectures.
The most important verses of sra 9 are those which the Prophet asked [i/222]
Al to recite before the assembled Arabs at the ajj celebration of 9/630 at
Mecca. The traditions regarding the scope of this proclamation vary con-
siderably.
251
With some degree of certainty only verses 1 to 12 might belong
245
The repetition of fusq in the verses 7 (cf. verse 6), and 11 does not prove anything as
these facts might only have given rise to the editorial combination.
246
Fakhr al-Dn al-Rz; al-Wid; al-Zamakhshar; and al-Suy, Asbb al-nuzl, con-
sider it to have beenrevealed onthis occasion. According to IbnHishm, p. 821, and al-abar,
vol. 1, p. 1642, Muammad merely recited ( ) it at that time; cf. L. Caetani, vol. 2, part 1, p. 130.
247
al-Wid; Fakhr al-Dn al-Rz.
248
249
Ibn Sad (Wellhausen, Skizzen und Vorarbeiten, Heft 4, no. 77); al-abar, vol. 1, p. 1637,
and in the Tafsr; al-Wid; Fakhr al-Dn al-Rz; al-Zamakhshar; al-Bayw. Cf. Caetani,
Annali, vol. 2, part 1, p. 227.
250
Ibn Sad (Wellhausen, Skizzen und Vorarbeiten, Heft 4, no. 77); al-abar, vol. 1, p. 1687.
Regarding the famine of 9/630 we have information also in Ibn Hishm, p. 894, and al-abar,
vol. 1, p. 1693.
251
Ibn Hishm, p. 921; and al-Tirmidh, K. al-Tafsr; present the beginning of the sra
without further ado. Al-Diyrbakr, Tarkh al-khams, vol. 1, p. 11, mentions twenty-eight
verses; Mujhid in al-Zamakhshar, thirteen, al-abar, vol. 1, p. 1720sq., and in the Tafsr
to this: there the Muslims are ordered to attack all idolaters after the holy
months, unless there was a definite agreement for a limited time. The very
first words,
252
an acquittal, from God and His Messenger unto the idol-
aters with whomyou made a covenant, set the stage for the entire content.
Verse 28 belongs evidently to the same time, even though it was never part
of this acquittal. Far less certain is the date of verses 36 and 37, which now
appear completely detached and out of context, since they deal with the
basic rule of the Islamic calendar, the number of months, and the prohi-
bition of their change.
253
Verses 13 to 16 are probably best dated to before
the conquest of Mecca, for it is not far-fetchedas a wide-spread tradi-
tion has itto attribute the breach of treaty to the violation of the Pact of
al-udaybiyya on the part of the Quraysh. In this case, however, the expres-
sion hamm does not merely indicate the intention of the enemies, which
is not followed by action, but rather the bigotry which preceded the real
act. This estimation
254
also fits in with the content of the verses 17 to 22,
which do not readily lend themselves to being separated from the preced-
ing one, in which the idolaters are repeatedly referred to as the sustainers of
the Kaba.
255
The larger part of the sra deals with the campaign of Rajab
(vol. 10), p. 41, forty, al-Zamakhshar, al-Bayw, thirty or forty verses. The sra is simply
mentioned vaguely, as e.g. al-Masd, Prairies, vol. 9, p. 54; al-Tirmidh, K. al-Tafsr 4; cf.
Caetani, vol. 2, part 1, p. 294; Muir, vol. 4, p. 208sq., mentions the verses 1 to 7, p. 28; Sprenger,
vol. 3, p. 478sqq.; verses 1 to 28 in Snouck Hurgronje, Het Mekkaansche feest, pp. 6365,
v. 112, 28 and most likely also v. 36sq.Grimme, Mohammed, vol. 1, p. 128sqq., vol. 2, p. 29,
attributes the verses 1 to 24 to the campaign against Mecca in 8/629, but this opinion is
doomed by the expression al-ajj in the third verse. This expedition might have been called
a umra, but not a ajj. Cf. Nldeke, [Review] H. Grimme, Mohammed.
252
When a persons protection was revoked ( ) people first publicly announced
at the Meccan Kaba:
(cf. Joshua 2:19, 20: ) or something similar,

renouncing liability for protection and blood-feud. Cf. Goldziher, Muslim studies, vol. 1,
pp. 6970, and Abhandlungen zur arabischen Philologie, vol. 1, p. 32sq.; O. Procksch, ber die
Blutrache, p. 34; tim T, Dwn, ed. and transl. F. Schulthe, p. 63, foot-note 2.
Like many other sras also this one is named after the initial words, bar (
). Its
alternative name, al-Tawba, is related to the frequently occurring verb tba, to repent, and its
derivations (vv. 3, 5, 11, 15, 27, 103, 105, 107, 113, 118, 119, 127). As regards the many other names
of the sra see al-Zamakhshar, and al-Bayw at the beginning.
253
It is completely unimportant that these verses appear in an address by Muammad on
the occasion of his Farewell Pilgrimage in 10/631 (Ibn Hishm, p. 968; al-abar, vol. 1, p. 1754;
al-Wqid (Wellhausen, Muhammed in Medina, p. 431), Fakhr al-Dn al-Rz).
254
As Snouck Hurgronje rightfully points out inHet Mekkaansche feest, p. 50 n. 1, the reason
that the pieces, verses 1 to 12, and 13sqq., are merged must be that both are dealing with the
fight against the unbelievers. Outwardly, perhaps, also the phrase , occurring inboth,
verse 12 and verse 13, might have been of importance.
255
There are the most divergent traditions regarding verse 19; see Fakhr al-Dn al-Rz;
al-Wid; and the Commentators.
9/630 against the Byzantines and their Arab allies on the Syrian frontier,
in which many Medinans and Bedouins did not participate. Muammad
uses this opportunity to reproach strongly the Hypocrites and the lukewarm
Muslims. The verses, however, were not revealed in one piece; instead, this
happened either before the departure or during the campaign itself or after
the return home. The following verses can be placed before the beginning
of the campaign:
256
verses 23 and 24, which denounce the excuses of those
Muslims who stayed behind; verses 25 to 27, which serve to demonstrate
with the help of the Day of unayn (Shawwl 8/629) that only Allhs help
guarantees victory; and verses 28 to 35, which enjoin fighting the Christians
until they pay fealty. This fits in with the same period, particularly as even
earlier, in I Jumd 8/629 at Muta, the Muslims had come to blows with
Christian troops. To this time belong also verses 38 to 41, of which verse 41,
according to the note in Ibn Hishm, p. 924, is the earliest of the entire
sra; this applies also to verses 49 to 57, which becomes particularly appar-
ent from verse 49. On the other hand, verses 58 to 73in which Muam-
mad rejects the accusation of unfair distribution of alms (adaqt) as well
as other reproaches of the Hypocriteshave no obvious relation to that
campaign. During the campaign, verses 42 to 48 and 82 to 97 were promul-
gated, of which verse 85if it is really referring to the death of Abd Allh
b. Ubayy
257
must have been added later. Verses 74 to 81, and 98 to 107, can
be attributed to the time after the return. Verses 103 and 107 mention those
Muslims whorepentedhaving stayedbehind
258
as well as some others
259
who
256
Here we disregard the many legends which the commentators cite in support of the
particular verses, for example, the persecution of the Hypocrites, etc. Nothing of this is to be
found in Ibn Hishm. But it remains an enigma that so disproportionally many legends could
be attached to this and some other sketches. Cf. Ibn Hishm, al-Wqid, Wellhausen, etc.
257
Ibn Hishm, p. 927; al-Wqid (Wellhausen, Muhammed in Medina, p. 414); al-Bukhr,
K. al-Tafsr and K. al-Kusf, 203; Muslim, K. ifat al-munfiqn, 1; al-Tirmidh, K. al-Tafsr;
al-Nas, K. al-Janiz, 67; al-Wid; Fakhr al-Dn al-Rz; al-abar, Tafsr; al-Zamakhshar;
al-Bayw. Cf. Weil, Das Leben, p. 433, note, and p. 429.
258
As it is reported about themsimilar to Ab Lubba (cf. the references on sra 8:27)
that they had tied themselves to a column until God forgave them, this passage is also
considered to refer to him, or says that he belonged to these people, that is to say that in9/630
he did what he had already done in 5/626. Cf. Ibn Hishm, p. 687; al-Wqid (Wellhausen,
MuhammedinMedina, p. 416); al-abar, Tafsr; al-Zamakhshar; al-Bayw; al-Wid; Fakhr
al-Dn al-Rz. In al-Zamakhshar we even find the remark that it was a formal custom of
penitents to tie themselves to columns. Cf. Goldziher, Sulenmnner, p. 505, note 4.
259
With reference to Kab b. Mlik, Hill b. Umayya, and Murra b. al-Rab tradition
connects the verses 107 and 119. Cf. Ibn Hishm, p. 907sqq. (tradition of Kab b. Mlik);
al-Wqid (Wellhausen, Muhammed in Medina, pp. 411 sqq. and 416); al-abar, vol. 1, p. 1705
are deferred to Gods commandment. The verses from 108 to 111 are said
to have been revealed shortly before Muammad returned to Medina. They
are aimed at members of Ban Slim, who were secret followers of the anf
Ab mir and had built a mosque not far from Medina.
260
Connected with
this might be verses 112sq., whichdescribe the image of true Muslims, as well
as verses 114 to 117, which alleviate Muslims fromthe duty to ask pardon for
the idolaters, even if they are next of kin. Tradition attributes this portion
either to the death of Ab lib (IBNABDAL-MUALIB)
261
still before the
hijra, or to Muammads visit to his mother mina
262
(Bint Wahb)s grave
in al-Abw,
263
where Allh is said to have prohibited him to pray for her.
The first interpretationis impossible, if only for chronological reasons, while
the second one could be considered if the passage were aimed generally at
one particular instance. However, since it expresses only a general idea that,
particularly because of its generality, fits well into the Medinan situation,
both views are nothing but untenable exegetic fancy.
264
In verses 118 and 119,
three of the men who were left behind are pardoned.
265
Verses 120 to 123
reprove those of the Medinans who stayed behind as well as the Bedouins
who dwell around them, although with an important exception to this
general reproach, namely that not all Muslims necessarily had to go forth
totally, for Allhwouldhave beenpleasedif only some (firqa) of every group
(ifa) had participated. Verses 124 to 127 require the Muslims mercilessly
to fight the unbelievers who are near to them. The revelation is certainly
of late Medinan origin, and because of its position in the Koran probably
sqq.; al-Bukhr in K. al-Tafsr and al-Maghz; Muslim, K. al-Tawba, 10; the Commentators;
Ibn ajar; Ibn al-Athr, Usd al-ghba, under the relevant names. Cf Weil, Das Leben, p. 414,
note; Caussin de Perceval, vol. 3, p. 287; Muir, vol. 4, p. 197.
260
Ibn Hishm, p. 906sq.; al-Wqid (Wellhausen, Muhammed in Medina, p. 410sq.); al-
abar, vol. 1, p. 1704, and in his Tafsr; al-Zamakhshar; al-Bayw; al-Wid; Fakhr al-Dn
al-Rz. Cf. Weil, Das Leben, p. 267; Muir, vol. 4, p. 198sq.; Sprenger, Das Leben, vol. 3, p. 33sq.;
Caetani, Annali, vol. 2, part 1, p. 271 sqq.
261
Ibn Sad (abaqt, vol. 1, pt. 1): Biographie Muammads bis zur Flucht p. 78; al-Bukhr,
K. al-Tafsr, and K. Bad al-khalq, 171; Muslim, K. al-mn, 9; al-Wid; Fakhr al-Dnal-Rz;
al-Suy, al-Itqn, p. 32; al-Zamakhshar; al-Bayw; Ibn ajar, vol. 4, p. 214.
262
EI
2
: mina bt. Wahb b. Abd Manf, umm Muammad.
263
Cf. the Commentators. This happened in 6/627. See Ibn Sad (al-abaqt): Biographie
Muammads bis zur Flucht, vol. 1, pt. 1, p. 74, where the opinion is refuted that this happened
after the fall of Mecca. In al-Azraq, p. 433, it is reported as if the Prophet had the tradition of
the Meccans in view, according to which mina was buried in Mecca itself (see Burchardt,
Travels in Arabia, p. 173; Burton, Personal narrative of a pilgrimage, vol. 3, p. 352; Snouck
Hurgronje, Mekka, vol. 2, p. 66). But al-Azraq correctly writes al-Abw.
264
These verses might be seen as an allusion to the death of Abd Allh b. Ubayy, directed
to his son, if it were not inadmissible to call the former an idolater (mushrik, verse 114).
265
See above on verse 107.
dates fromthe same time as the previous one. Verse 128 has the same begin-
ning as verse 125,
266
although this is no evidence of their original literary
homogeneity. Verse 129sq. some regard as Meccan.
267
It is obvious to inter-
pret the expression min anfusikum in this passage, now there has come to
you a Messenger from among yourselves, as referring to the Quraysh. But if
inthis case the Prophet was only thinking of his Arab originingeneral, there
wouldbe nothing to prevent the two verses frombeing regardedas Medinan
and connected with the preceding text. That others regard them to be the
very latest verses
268
is connected with a tradition about the collection of the
Koran,
269
which we will discuss below (under Uncanonical Promulgations,
see page 189). The composition of sra nineas also in other casesis
obscure.
270
The numerous, conspicuous lexical and phraseological contacts
between their various parts
271
readily lead one to imagine that their dates of
composition could not have been far apart.
Although a great many regard sra nine as the last one,
272
there are others [i/227]
whoconsider sra 5 tobe still later,
273
probably because a fewof its important
verses were revealed later than all the others.
Among these, tradition explicitly counts only verse 5,
274
claiming that
Muammad recited it to the believers during the so-called Farewell Pil-
grimage in 10/631, only a few months before his death. Yet we may add
266
The words
are also found in verse 87.

267
al-Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 32; Al al-Dn (AL-KHZIN al-
Baghdd), al-Diyrbakr, Tarkh al-khams, vol. 1, p. 13, as also the Commentators indicate,
the predicates and are normally used only by Allh. For this reason something must
have been lost after of verse 129; cf. verse 118.
268
al-Wid in the introduction to the Cairo edition, p. 8; Fakhr al-Dn al-Rz; al-abar,
Tafsr; al-Zamakhshar; al-Bayw; al-Suy, al-Itqn, p. 32; al-Shshw, chapter 1.
269
Cf., for example, al-Bukhr, K. al-Tafsr on sra 9 at the end.
270
A suspicion relating to this context has been mentioned above on verse 13.
271
Cf. verses 3, 5, 11, 15, 27, 103, 105, 107, 113, 118, 119, 127; verses 69, 73, 78, 112, 115;
verses 8, 24, 53, 54, 68, 81, 85, 97; verses 42, 56, 63, 65, 75, 96, 97, 108 (otherwise only
five times more in the Koran); verses 3, 14, 26, 34, 39, 55, 62, 67, 69, 75, 80, 86, 91, 102,
107; verses 58, 60, 80, 104, 105; verses 58, 80 (otherwise only two more in the Koran);
verses 25, 48, 75, 118, 129;
verses 87, 125, 128.

272
al-Bukhr, K. al-Maghz, 67, K. al-Tafsr; Muslim, K. al-Fari; al-abar in the intro-
duction to the Tafsr, Cairo edition, vol. 1, p. 34; Cod. Lugd. 653 Warner, fol. 6b; al-Bayw;
al-n, p. 55sq. Cf. the list of sras, above, p. 48sqq.
273
al-Timidh, K. al-Tafsr on the sra, at the end; al-Zamakhshar on sra 9:2; al-Shshw,
chapter 1. Cf. the list of sras above, p. 48sqq.
274
al-Bukhr, K. al-mn, 33, K. al-Tafsr; Muslim, K. al-Tafsr at the beginning; al-Tir-
midh, K. al-Tafsr; al-abar, Tafsr; al-Zamakhshar; al-Bayw; al-Wid; Fakhr al-Dn al-
Rz; in Cod. Lugd. 653, third last leaf; and al-Yaqb, ed. Houtsma, vol. 2, p. 43sq., verse 5 is
called the very last revelation.
to our count also verses 1, 4, and 7,
275
because they appropriately comple-
ment that verse in content and are stylistically related. The time of trans-
mission mentioned is quite plausible; the emphatically repeated today
(al-yawma, verses 4, 5, and 7) betrays a particularly important situation,
and the Prophets rapturous emotions with respect to the success of his
mission, as expressed in the beginning of the fifth verse,
276
correspond per-
fectly with his last year of life. All transmissions are agreed that verse 6 is a
reply to the question of two respected ayyi,
277
Ad b. tim (al-)
278
or
Zayd al-Khayl,
279
who were not converted until Muammads last years.
Zayd
280
headed a delegation from his tribe to the Prophet; the other man,
281
also a leader of a tribe, plays a role in Als expedition during which the
idol of Fuls (Fals) was destroyed. Ibn Sad
282
mentions in this connection an
inquiry addressed to Muammad regarding venisonthe inquirer is here
called Amr b. al-Musabbibut the reply (
) is unre-
lated to anything in the Koran. The possibility thus cannot be excluded that
this episode was only later brought in connection with verse 6, and that
the little known Amrregardless of what kind of great Nimrod he might
have beenis replaced by a greater name. In such circumstances, no fair
judgement can be passed on verse 6. According to tradition, verse 2 was
revealed either during the pilgrimage of 7/628
283
or the abortive one of 6/627
275
Verse 1 (Flgels edition) Muslimswith the exception of Kufansdivide into two
parts. Flgel, against all traditions, makes an incision.
The order of the verses belonging to that revelation might have originally been more or
less as follows: verse 1, verse 4 up to , verse 5, starting with
, verse 7, and verse 4

from to
.
276
Today I have perfected your religion for you, and I have completed My blessing upon
you and I have approved Islam for your religion.
277
al-Bukhr, K. al-ayd, 7; al-abar, Tafsr; Fakhr al-Dn al-Rz; al-Wid; Ibn ajar
al-Asqaln, vol. 1, no. 2421.Muslim, K. al-ayd, 1; al-Nas, K. al-ayd, 1 sqq.; al-Bukhr,
loc. cit., 8, relates the affair without specifically mentioning the verse. Al-abar, Tafsr, and
al-Wid supply also other reasons.
278
EI
2
; Juynboll, Encyclopedia.
279
EI
2
280
Ibn Hishm, p. 946sq.; Ibn Sad (Wellhausen, Skizzen und Vorarbeiten, Heft 4, no. 103).
Al-abar, vol. 1, p. 1747sq.; al-Aghn, vol. 16, p. 48sq.
281
Ibn Hishm, p. 947sq.; al-Wqid (Wellhausen, Muhammed in Medina, p. 390sqq.); Ibn
Sad, loc. cit.; al-abar, vol. 1, p. 1706sqq.; al-Aghn, vol. 16, p. 97; Sprenger, Das Leben, vol. 3,
p. 386sqq.
282
Loc. cit. Cf. al-abar, vol. 3, p. 2362; Ibn al-Athr, Usd al-ghba, vol. 4, p. 131.
283
al-Wid; Hibat Allh (b. Salma).Other works (al-abar, Tafsr; Fakhr al-Dn al-
Rz; Al al-Dn (AL-KHAB al-Baghdd), al-Suy, Asbb al-nuzl), although reporting
the same story, do not indicate the year of the Medinan period.
(al-udaybiyya).
284
Since the closely
285
related verse 3 is always fixed in the
year 6/627,
286
however, the latter date is apparently better documented. For
traditional criticism everything revolves around the interpretation of the
words (mmn al-bayt) of the second verse. If this refers to pagan pilgrims,
then the verses belong to the period before 9/630, when Muammad per-
manently cut off relations with the idolaters.
287
Yet if those words refer to
Muslims, little can be said against a later composition. Verses 11 to 13 prob-
ably belong to the same period.
288
It is nearly impossible to determine the
date of verse 14,
289
as it mentions a danger for the Muslims
290
only in very
general terms. The tales of Muammads problems in private life, which tra-
dition brings forward,
291
are useless. As far as the time of verses 15 to 38
292
is
concerned, the only clue that might serve as a guide is verse 37, whichunmis-
takably preaches open and merciless war. Since the wording of the passage
indicates that the fight had started earlier, the pericope falls in the time after
the expulsion of Ban Qaynuq in Shawwl 2/623, and naturally before the
final assault against the power of Arabian Jewry, the occupation of Khaybar
in I Jumd 7/628. Presumably they are pretty close to the latter date, since
most parts of sra 5 apparently belong to 6/627 and 7/628. Verses 39 to 44,
similar instyle withthe preceding verses,
293
must inany case be datedbefore
the occupation of Mecca if it is true that on the way there Muammad pun-
ished
294
a female thief by cutting off her hand in accordance with verse 41,
284
al-Wid.
285
Flgels verses 2 and 3 actually constitute a single verse.
286
al-abar, Tafsr; Ab Layth al-Samarqand; al-Zamakhshar, al-Bayw; Al al-Dn
(AL-KHAB al-Baghdd); al-Suy, Asbb al-nuzl.
287
Cf. the beginning of sra 9.
288
The idiom
is to be found only in the verses three and eleven.

289
Observe the phrase verses 10, 14, and 23.
290
when a certain people purposed to stretch against you their hands
291
Ibn Hishm, pp. 392 and 663; al-Wqid; al-abar, vol. 1, p. 1456; al-alab, vol. 2,
p. 403; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 415; Fakhr al-Dn al-Rz; al-Wid; the
Commentators. Cf. Caetani, Annali, vol. 1, pp. 538 and 596.
292
We can disregard the tradition which has verse 27 cited in an address delivered before
the Battle of Badr (Ibn Hishm, p. 434; al-Wqid, p. 43; al-abar, vol. 1, p. 1300). No one would
want to attach documentary relevance to such talk. Hirschfeld, New researches, p. 71, allots
verses 23 to 38 to the Meccan period, which cannot apply after what has been said above.
293
Cf.
verse 40, and three times verse 30;
verse 37, and verse 42.

294
al-Bukhr, K. al-udd, 13; Muslim, K. al-udd, 2; al-Nas, K. Qa al-sriq, 6.
In all these passages the woman is generally identified as a member of the Makhzm.
al-Bayhaq, al-Masin, p. 395, she is called the daughter of Sufyn b. Abd al-Asad. In awil
literature (e.g. Ibn Qutayba, p. 273; Ibn Rustah, al-Alq al-nafsa, p. 191; al-Thalib, Laif
al-marif, p. 8) it is claimed that Wald b. al-Mughra introduced the penalty for theft by
and that this penalty was the first of its kind and not based on ancient Arab
common law. To the same period might belong also the tradition
295
that this
refers to Ab uma b. Ubayriq, although the name is somewhat suspicious,
as the man already served as model of a thief.
296
What caused Weil to allot
the verse to the last pilgrimage
297
is beyond anyone. Verses 45 to 55 deal
with a controversy among Jews that we have in quite different versions.
298
If
there is the least truth to it, the verses must be older than the extermination
of Ban Quraya, since this clan is mentioned in some traditions. Some-
times the accused is a Qura and sometimes it is the judge, so that basically
everything remains uncertain and unreliable.
299
Verses 56 to 63 would date
from 3/624 if indeed they were to refer to Abd Allh b. Ubayy, who, upon
his pleading, secured permission for the captured Ban Qaynuq to emi-
grate.
300
Uponcloser examinationwe see that the text reveals anunfortunate
time when some men suggested a pact with the Jews (verse 56) for fear of
the enemies but receivedthe Prophets response that Allhmight sendthem
victory or some other favourable change of fortune. These circumstances do
not at all fit in with a time when, in short succession, the Meccans as well
as the Jews suffered defeat. Another tradition attributes verse 56sq.and
thus the entire passageto the advice given to the Prophet after the Battle
cutting off the hand for this delict in the time of the Jhiliyya. Also other considerations opt
for the assumption that this is indeed an innovation on Arab soil, at least as far as freedmen
are concerned. In the case of slaves this might have always been permitted. Otherwise one
might think of a borrowing from Abyssinia, where this barbarian custom is still in use, while
it is foreign to Jewish as well as Greco-Roman law. On its spread in the Occident throughout
the Middle Ages see L. Gnther, Die Idee der Wiedervergeltung, vol. 1, pp. 200, 253, and 294.
295
Ab Layth al-Samarqand, and al-Wid.
296
Cf. above, p. 164, on sra 4:106.
297
Historisch-kritische Einleitung, 1st ed., p. 79sq., 2nd ed., p. 90.
298
Ibn Hishm, p. 393sqq.; al-abar, Tafsr; al-Zamakhshar; al-Bayw; Ab l-Layth al-
Samarqand; Fakhr al-Dnal-Rz; al-Wid.al-Muwaa, p. 347, and Mishkt, K. al-udd,
fal 1 4, without referring to this verse in particular. Cf. Sprenger, Das Leben und die Lehre
des Mohammad, vol. 3, p. 37sqq.
299
Others interpret verse 46 or verse 45 as a dispute of Ban Quraya and Ban Nar
(Ibn Hishm, p. 395; al-Nas, K. al-Buy, 111; al-abar, Tafsr; al-Zamakhshar; al-Bayw;
Ab l-Layth al-Samarqand; Fakhr al-Dn al-Rz). A tradition in al-Wqid (Wellhausen,
Muhammed in Medina, p. 215); al-abar, Tafsr, vol. 6, p. 134, attributes verse 45 to the
clemency for Ab Lubba, cf. Caetan, vol. 1, p. 629sq.Let us mention the following lexical
and factual points of contact of the verses 45 to 55 and other parts of the sra:

verses 45 and 16; verses 48 and 18; verses 48 and 4;
verses 45 and 57, 67; the last words of verse 45 pretty much similar to those of verse 37;
verses 50 and 82, 109, 112, 116, and in a different form, verses 19, 76,
and 79; there is no sra which mentions the name Jesus so frequently.
300
Ibn Hishm, p. 546; al-abar, Tafsr; al-Zamakhshar, al-Bayw; Fakhr al-Dn al-Rz;
al-Wid. Cf. Weil, Das Leben, p. 159, foot-note; Caussin de Perceval, vol. 3, p. 81 sq.
of Uud that he try to obtain help from the Jews against the pagans.
301
This
explanation is much more plausible, yet it still does not deserve particular
credibility, as it is likely nothing more than exegetic speculation.
302
Verses 64
to 88 presuppose that several wars had already beenwaged. When, based on
the passage, it is concluded
303
that the battles of Muta and Tabk had not
yet taken place, this is by no means beyond doubt. The friendly opinion of
Christians, particularly their priests and monks in verses 73 and 85, is purely
theoretical and fundamental and need not necessarily refer to those skir-
mishes. We would of course arrive at a different conclusion if Muammads
instruction to the Muta warriors to slaughter the parsons and leave the her-
mits inpeace
304
was indeedhistorical. Unimportant for us are the interpreta-
tions of verse 71, e.g., that it is one of the earliest Meccan verses.
305
Verses 89
to 91 must date from 7/628 at the latest, since already sra 66:2 clearly
refers to it. They might possibly originate from the same time as verses 92
to 94, where permissible and prohibited matters are juxtaposed. We have
already been able to see on p. 144 that the period of these verses cannot
be determined with any degree of accuracy, and that they might belong to
the year 4/625, and certainly not after 6/627. Verses 95 to 97 are occasion-
ally fixed to the year of al-udaybiyya;
306
verses 98 to 100, dealing with the
Meccan sanctuaries, would fit very well into this time. According to some
informants, verse 101 refers to a man who, when the order for the pilgrim-
age was issued, asked whether people were expected to participate every
year; he received Muammads irritated reply: If I were to affirm this ques-
tion, you would have to obey, but since you would not be able to do so,
you would again become an unbeliever; so stop asking questions!
307
Oth-
ers connect the verse with matters that the Prophet was either unfamilar
with or that he disdained.
308
Closer is the connection with the immediately
301
302
Here Hirschfeld, New researches, p. 119, is thinking of the expulsion of the Jewish Ban
Nar and the pact with Ban Quraya. But also this situation does not really fit.
303
Weil, Historisch-kritische Einleitung, 1st ed., p. 80, 2nd ed., p. 90.
304
al-Wqid (Wellhausen, Muhammed in Medina, p. 310).
305
al-abar, Tafsr. More explicitly al-Wid and al-Suy, Asbb al-nuzl (
).
306
307
al-Khab al-Tibrz, Mishct, K. al-Mansik, fal 1, 1 and 15; al-abar, Tafsr; al-Za-
makhshar; al-Bayw; Ab l-Layth al-Samarqand; al-Wid. Cf. Muslim, K. al-ajj, 30.
Traditions are not totally agreed regarding names and circumstances.
308
For example, many wanted to know where they would find their camel which went
astray; still others, who had been unknown to the Prophet, asked him for the name of their
father. Al-Bukhr, K. al-Itim, 4, K. al-Tafsr; Muslim, K. al-Aqiya, 5, K. al-Fail, 30;
following verses 102 and 103, which are directed against all sorts of pagan
superstitions yet cannot be fixed chronologically. Verse 104 fits better into
the preceding verse than the following. Verses 105sqq. are not concurrent
with sra 2:176 but were revealed quite some time later in order to elaborate
on this short law. The verses are generally assumed to refer to two Muslims
who embezzled a golden beaker from the property of their travel compan-
ion.
309
If this was an historical event, it must have happened after the occu-
pation of Mecca, as people and families of the Quraysh, who had only then
accepted Islam, are mentioned. These problematic verses, however, cannot
be explained in a satisfactory way from the situation, nor do they look like
a revelation for any particular occasion.
310
The literal contact of tradition
and these verses is of no importance. The period of verses 108sqq.
311
like-
wise cannot be determined. This narrative passage, with interesting details
from the legend of Jesus (miracles of birds and the table) has probably
been placed here because the sra had frequently mentioned the Son of
Mary(am) (verses 19, 50, 76, 79, and 82). It is difficult to believe that the
other parts should have been put together by sheer accident as well. Verses 1
to 7 and 89sqq. are of related content insofar as they both treat prohibited
food, hunting, and the sanctity of Mecca. Scattered within the section from
verse 15 to verse 58 we find discourses on the People of the Book (verses 15
to 38, 45 to 56, and 64 to 88), two admonitions to the believers (verses 39 to
44), and the Hypocrites (verses 56 to 63).
al-Tirmidh, K. al-Tafsr; al-abar, Tafsr; Ab l-Layth al-Samarqand; al-Wid. This is the
context of traditions according to which Muammad considers the endless questions (
) to the things Allh hates most: al-Muwa, p. 388; Muslim, K. al-Aqiya, 5; al-Khab
al-Tibrz, Mishct, Bb al-birr wa-l-ila, fal 1, 5. Cf. also Goldzihers review of A. de Vlieger,
Kitb al Qadr, p. 393.
Even if all the traditions regarding verse 101 are fabrications, they still emanate from a
correct general point of view. This applies particularly to the sound tradition in the previous
note 307, which unsurpassably depicts a man driven to despair over importunate questions.
309
al-Bukhr, K. al-Way, 36; al-Tirmidh, K. al-Tafsr; al-abar, Tafsr; al-Zamakhshar;
al-Bayw; Ab l-Layth al-Samarqand; al-Wid; Fakhr al-Dn al-Rz; Ibn ajar al-Asqa-
ln, vol. 1, nos. 608 and 832; Ibn al-Athr, Usd al-ghba, vol. 1, p. 169, and vol. 3, p. 390.
310
See above, p. 166.
311
Inverse 109, Flgels editionerroneously has anincisionafter
. Sucha rhyme is found

only in verse 2 (cf. thereon, above, p. 185, n. 285). The verses 19, 35, and 52, however, are based
everywhere on the confusion of pause and end of verse. In spite of all individual variety, the
rest of the verses in this sra consist of a closed syllable with long vowel as rhyme. Cf. also
Rudolf Geyers review of Volkssprache und Schriftsprache im alten Arabien, by Karl Vollers,
p. 27sqq., of the off-print. The words
are nearly repeated literally in sra

3:41.
MUAMMADS UNCANONICAL PROMULGATIONS
The revelations, which up to this point we have traced back to their origin, [i/234]
have been taken without exception from the Koran. Tradition, however,
knows of still other revelations of the Prophet.
In the first instance, we are concerned with those revelations that have
been preserved in their literal formand that tradition explicitly considers to
be part of the original Holy Book.
1
There is a very frequently
2
mentioned passage that varies so much in dif-
ferent recensions that we must list the most important versions individually:
(a) al-Tirmidh, K. al-Manqib, s.v., Ubayy b. Kab according to Ubayy, al-
Suy, al-Itqn, p. 525 (c)
3
from Ab Wqid al-Layth, Hibat Allh [Ibn
Salma] (in the introduction of the Cairo edition, 1315/1897, p. 11), Ab Abd
Allh Muammad Muammad IBN AZM (K. al-Nsikh wa-l-manskh, in
the margin of Jallayn of the Cairo edition, 1311/1893, vol. 2, p. 148), accord-
ing to Anas b. Mlik:

5

6

7

8

10

11

12

13

.
1
This is according to Hibat Allh (Ibn Salma, Cairo edition, p. 9) the first class of the
manskh or the abrogated passages,
.
2
The traditions we quote are based on the following authorities: Ubayy b. Kab, Anas b.
Mlik, Ab Ms al-Ashar, Ab Wqid al-Layth [Juynboll, Encyclopedia, p. 377], Ibn Abbs
and Ibn Zubayr. In addition al-Tirmidh mentions in his Abwb al-zuhd, bb, 20, Ab Sad
al-Khudr [Juynboll, Encyclopedia], isha, Jbir b. Abd Allh, and Ab Hurayra.
3
The different recensions are here designatedaccording to the arrangement inal-Suy,
al-Itqn, as a, b, and c.
4
, Ibn azm.
5
, Ibn azm; , Hibat Allh (Ibn Salma), Cairo edition, 1315/1897.
6
, Hibat Allh (Ibn Salma); al-Suy, al-Itqn, omits .
7

, al-Suy, al-Itqn.
8
, Hibat Allh (Ibn Salma) and Ibn azm.
9
, Hibat Allh (Ibn Salma) and Ibn azm; in al-Itqn.
10
al-Tirmidh, (
) ; al-Suy, al-Itqn ( ) .
11
, Hibat Allh and Ibn azm; , al-Suy, al-Itqn.
12

, al-Itqn.
13

, Hibat Allh (Ibn Salma) and Ibn azm; in al-Suy, al-Itqn.

190 mu
hammads uncanonical promulgations

If man (literally son of Adam, cf. below, p. 194 n. 62) had a vale of treasures
he wouldwant anadditional secondone, andif he hada secondone, he would
want anadditional thirdone; but only dust will fill mans belly, Allh, however,
turns to those who turn to Him.
In Itqn these words are preceded by:

[i/235]
) (
14
Allh says: Truly, We sent downtreasures to performthe prayer
and give alms, and would have, etc.
(b) Not muchdifferent fromthis is the formthat (according toal-Bukhr, al-
a (c),
15
and al-Mabn li-nam al-man, vol. iv, Berlin Ms., I Wetzstein,
no. 103, fol. 34
v
) Ibn Zubayr recited:

16

17

18

19

.
20
(c) In addition those which al-Itqn, p. 525 (b), from Ubayy b. Kab, and
al-Mabn li-namal-man, part iv, fol. 36
v
sqq. from Ikrima:

21

22

23

24

25

26

.
(d) Those listed from Ubayy (Ibn Kab) in al-Mabn, part iv, fol. 34
r
:

.
(e) Kanz al-umml f sunan al-aqwl wa-l-afl of Al al-Dn Al b. usm
(al-MUTTAQ) al-Hind (died 975/1567), Hyderabad, 1312/18941315/1897,
vol. 1, no. 4750, from Ubayy (ibn Kab):

.
14
Another introductory formula to a variant text has been listed below, p. 192.
15
All of al-Bukhrs versions of this passage are to be found in K. al-Riqq, cap. 10; we
designate them according to their sequence as a, b, c, d.
16

, al-Mabn li-namal-man.
17

al-Mabn (with following ); marginal reading in the Leiden Ms. no. 356 with a
, and al-Qasalln, vol. 9, p. 250 from Ab Dharr [al-Ghifr].
18
, al-Mabn li-namal-man.
19
, al-Mabn.
20
Unless indicated, the ending after
is everywhere the same as in (a).

21

, al-Mabn.
22
, al-Mabn.
23
, al-Mabn.
24
, al-Mabn.
25
Missing from al-Mabn.
26
, al-Mabn.
mu
hammads uncanonical promulgations 191

The following versions are shorter:
(f) al-Bukhr (d), Muslim, K. al-Zakt 26 (b),
27
al-Tirmidh, Abwb al-zuhd,
bb 20, all fromAnas b. Mlik, al-Suhayls commentary onIbnHishm(note
to p. 650) without reference to an authority:
28

29

30

31
32

33
34

35

36

37

.
(g) A b. Ab Rab al-Qurash, from Ibn Abbs in al-Bukhr (b) and
Muslim (c):

38

39

40

.
(h) al-Bukhr is a hybrid text, consisting of d, e, and f:
. This
27
The different recensions we designate according to their arrangement in Muslim (Ibn
al-ajjj) as a, b, c, and d.
28
This is the version of the verse already used by George Sale in The Koran, commonly
called the Alkoran of Mohammed, preliminary discourse, section iii.
29
, al-Tirmidh and Muslim b. al-ajjj.
30

, Muslim b. al-ajjj.
31
al-Suhayl points out that other scholars read . This reading must be consideredolder
and better because in the parallel text fromthe Syriac Ahiqar story it reads treasures.
See below, note 40.
32

, Muslim and al-Bukhr; but in the margin of the Leiden cod. 356 of al-Bukhr as
well as in al-Qasalln, vol. 9, p. 221, according to Ab Dharr,
.
33
Missing from Muslim b. al-ajjj.
34
instead
, al-Suhayl.
35

, Muslim;
both, al-Tirmidh and al-Suhayl.

36
, in the margin of the Leiden cod. 356, al-Tirmidh, and al-Suhayl.
37

, al-Suhayl, adding that sometimes is read (cf. the g recensions below),

sometimes . Muslim here has .
38
[milu], marginal reading in both, Leiden cod., 356, and Muslim.
39
, Muslim.
40
, Muslimwho also here has .. eye (cf. note 37) is undoubtedly the
more meaningful reading. But that this is also the earlier one is evident from the use of the
phrase in The Story of Ahikar which, regardless of what one thinks of its date of composition,
is in any case many centuries older than our Arabic sources. Cf. the relevant text according
to the best recension in Ahikar, The Story of Ahikar from the Syriac, p. 34 (40),
Mans eye is like a fountain-head,
it is not satiated by treasures until it is filled with earth. Apart from the Syriac version, this
dictumis extant only in a Slavic version. The reading is also known fromArabic literature,
e.g. Qub al-Dn al-Nahrawl, Geschichte, p. 302, where the words from to are
interwoven in the story of an avaricious grand vezir. The same recension is also the basis of Le
Synaxaire arabe-jacobite, p. 542 [466]:

. Finally, we must mention the modern Meccan proverb collected by Snouck Hurgronje,
Mekkanische Sprichwrter, no. 46:
.
192 mu

tradition traced back to Ibn Zubayr does not purport to be a revelation but
a adth. Cf. below, p. 194.
(i) IbnAbbs inal-Bukhr (a), andal-Khabal-Tibrz, Mishct, bbal-amal
wa-l-ir, fal 1 5, Ab Ms al-Ashar in Muslim (d), al-abar, Tafsr on
sra 2:100 (vol. 1, p. 361), al-Mabn li-nam al-man, part iv, fol. 39
r
, 40
r
,
al-Itqn, p. 525sq. (c), Ubayy (Ibn Kab) in al-Mabn li-nam al-man,
part ii, fol. 15
r
, Anas (Ibn Mlik) in Muslim (a):
41

42

43

44

45

46

47
.
In al-Suy, al-Itqn, and al-Mabn li-namal-man, part iv, fol. 39
r
, the
beginning according to Ab Ms al-Ashar reads as follows:
48

49

50
.
Amid this plethora of variants it is hardly possible to identify any par- [i/239]
ticular form as being older or more original than the rest, as through rem-
iniscences they are constantly interwoven and thus create new forms. By
comparison with the Syriac Ahiqar story, on the other hand, we are able to
determine that the readings treasure and eye are to be preferred to other
variants.
51
Just as different as the texts are the particulars regarding their divine
originand their original place inthe Koran. InHibat Allhb. Salma it reads:

52

53

41

, al-Mabn li-namal-man; al-Suy, al-Itqn; Tafsr.

42
, al-Mabn, al-Itqn, Tafsr.
43
, al-Mabn, part ii, and al-Muttaq, Kanz al-umml, vol. 1, no. 4755.
44
, al-Itqn; al-Mabn, parts ii and iv, fol. 40
r
.
45
Missing fromal-Tirmidh., al-Khabal-Tibrz, al-Mishct, al-Mabn li-namal-man,
fol. 39a instead
, al-Mabn, part ii , al-abar, Tafsr.

46
, al-Mabn, part iial-Muttaq, Kanz al-umml, vol. 1, no. 4755 reads instead

only . From to the end everything is missing from al-Tirmidh.

47
from Muslim from Ab Ms al-Ashar.
48

, al-Itqn.
49
al-Itqn has falsely, , cf also sra 3:71.
50
These are the menas shall be explainedbelowwho fought for Muammadwithout
believing and therefore do not share Paradise. The words are quite common as a adth.
Al-Bukhr, K. al-Jihd 18, has them in a slightly different form:

, al-Bukhr, K. al-Maghz, cap. 39 (ghazwat Khaybar) 8 with the variant,
.
51
See p. 170 n. 31 and p. 171 n. 40.
52
Thus, i.e. (we comparedthem) as readboththe LeidenMs. 411 as well as the Berlin
Ms. Sprenger 397 [Ahlwardt, Verzeichnis, no. 419]. Thus also the of the Leipzig Ms. of
Hibat Allh (Ibn Salma) undoubtedly ought to be changed against H.L. Fleischers in
the Catalogus librorummanuscriptorumqui inBibliotheca civitatis Lipsiensis asservantur. The
Cairo edition, p. 10, reads .
53
Thus in the Berlin Mss.; the Cairene edition reads
.
mu

. A tradition from Ab Ms AL-ASHAR [EI
2
; EQ; G. Juynboll; Sezgin,
Geschichte, vol. 1, pp. 61, etc.] in Muslim (Ibn al-ajjj), loc. cit., which is
nearly identical withthis one, adds by way of anexplanation
, of which only corresponds to the original meaning. Also in
al-Mabn li-namal-man, and al-Suy, al-Itqn, Ab Ms AL-ASHAR
says the same with less words; they had read a sra, or ,
only that also in this case an inaccurate tamyz,
, is added.
In al-Suhayl this reads completely differently:

[i/240]

54

, i.e. this verse is said to have been

placed originally behind sra 10:25. In this instance, however, its meaning
might tolerate such an addition, if need be, yet since the rhyme with b
is unthinkable in sra 10,
55
and most authorities refer to entirely different
sras, this statement of al-Suhayl is not reliable.
In al-Mabn li-nam al-man, part iv, fol. 34
v
, Ikrima says

[= d]
. Thus also Ubayy b. Kab in al-Itqn, p. 525, regards

this verse to be a portion (
) of sra 98. The earliest source for this

information is the following adth in al-Tirmidh.
56

57

.
And he (the Prophet) recited before him (Ubayy) sra 98, and therein
(the verse): Verily, the religion, etc., and he recited before him, If he had,
etc. Although it is not said here directly that this verse belongs to sra 98,
the words probably have to be interpreted this way.
58
That Ubayy considered this verse to be part of the Koran is also demon-
strated in al-Mabn li-namal-man, parts 2 and 4. Ab Wqid (al-Layth)
does not say this quite as clearly (al-Itqn, p. 525) with the following words:
The Prophet said:
. It so happens that occasionally

adths are also indicated like this, which the Prophet attributed to Gods
own words.
59
Ibn Abbs in al-Bukhr and in Muslim (ibn al-ajjj) as well
as Anas (Ibn Mlik) in Muslim (ibn al-ajjj) express their doubt that this
54
I.e. Ab Ubayd al-Qsim b. Sallm (d. 222/836); cf. thereon the chapter below, Fr.
Schwally, The Collecting and Editing of the Koran. [EI
2
; Sezgin, GAS, vol. 3, p. 367, and vol. 9,
pp. 7072, d. 224/839].
55
Since the last words of the alleged fragment of the Koran are identical in all recensions,
they must not be changed.
56
Kitb al-Tafsr on sra 98.
57
Here follows the verse quoted in section 2 on p. 172.
58
The order inal-Suy, al-Itqn, is reversedsothat the first passage here citedis assumed
to come from sra 98, and the other passage is mentioned afterwards.
59
Cf. below, p. 204sq.
194 mu

passage belongs to the Koran: For example,
; but their conclusion is wrong, because

sra 102 must be much earlier than this verse, and also the Medinan Anas b.
Mlik cannot have been present when sra 102 was promulgated.
In many instances these words are simply presented as a adth fromthe [i/241]
Prophet, such as, for example:
60

.
There is no ready answer to the important question of the reliability of these
traditions. Since the Prophets words in this case operate on the level of
Koranic thinking and expression,
61
they might well belong to a long lost
revelation. The passage might have remained in the memory of particular
contemporaries because the Prophet frequently referred to it. The consid-
erable differences of the recensions as well as the variations of tradition
regarding the origin would thus become quite plausible. The available facts,
however, lend themselves with equal probability to the opposite conclusion
as well, namely that those words are an original adth that, because of its
resemblance to Koranic diction, might have been considered a revelation.
Indeed, it cannot be ruled out that even the adth is spurious. Above all,
there is the fact that the idiom ibn dam, man,
62
is foreign to the language
of the Koran and would thus argue against Muammads authorship.
(2) al-Tirmidh, K. al-Tafsr, on sra 98; al-Mabn li-namal-man, part iv,
fol. 37
r
; al-Suy, al-Itqn, p. 525. Ubayy b. Kab read in sra 98:

63

64
65

66

67

68

69

.
70
60
al-Bukhr; cf. above p. 169sq. under (h); al-Mabn li-namal-man.
61
Cf., for example, the words , , .
62
It corresponds to the Aramaic , but more particularly to the Hebrew
which first appears in Ezekiel. The Koran always uses insn instead. Only the plural, ban
dam, occurs a few times, sra 7:25, 26, 29, 33, 171; and sras 17:72; and 36:60. But that the
singular, ibndam, was not foreigntocontemporary literature seems tobe evident, e.g., Labd
b. Raba, Gedichte (1891), no. 32, verse 10.
63
al-Itqn and al-Muttaq, Kanz al-umml, vol. 1, no. 4750 add .
64
Missing from al-Itqn, and Kanz al-umml; , al-Tirmidh.
65
, al-Itqn.Kanz al-umml inserts .
66
al-Mabn without article which is here indispensable.
67
The same. Al-Tirmidh adds
.
68
al-Tirmidh without .
69
, in both, al-Itqn and Kanz al-umml.
70
Kanz al-umml, .
mu

Truly, religion is for God the tolerant anafite School, neither Judaism, nor
Christianity; whoever does good shall not remain without reward.
If these words, whose rhyme fits somewhat with that of sra 98, were in- [i/242]
deed really Koranic, the original form still must have been slightly
changed, as the words
are foreign to the Koran.

71
There
are, however, ample instances where they could have been used.
The beginning of the text appears in several divergent versions as a
adth:

or
.
72
(3) According to a tradition listed in al-Itqn, p. 526, Maslama b. Mukhallad
al-Anr recited before his friends the following two verses, which purport
to be in the Koran but are wanting from the authorized text:
73

74

.
Those who believe, and have emigrated, and struggled withtheir possessions [i/243]
and their selves in the way of God, do rejoice, you are the happy people! And
they who accepted and aided them, and defended themfromthe people with
whom God is angry: No soul knows what comfort is laid up for them secretly,
as a recompense for what they were doing.
In the case of these two verses as well, no definite opinion can be supplied.
In support of their authenticity there is not only the Koranic character
of the diction throughout,
75
but also the common grammatical change of
person that occurs frequently in the Koran. On the other hand, however,
71
The only abstract term derived from a noun ending with that occurs in the Koran
is
paganism (four times). The original meaning of the word is likely to be state of
ignorance, comparable to the New Testament (Acts of the Apostles 17:30; I Peter 1:14).
Goldziher holds a somewhat different view in his Muslimstudies, vol. 1, pp. 201208.
72
al-Bukhr, K. al-mn, 29; Ibn Sad, vol. 1, part 1 (Biographie Muammads bis zur
Flucht) p. 128, l 13. Majd al-Dn IBN AL-ATHR, Nihya, s.v. . Another form occurs in a
frequently quoted early Islamic verse ascribed to Umayya b. Ab l-alt:

.
al-Aghn, vol. 3, p. 187; Ibn ajar al-Asqaln, al-Isba f tamyz al-aba, vol. 1, p. 263,
whereas Ibn Hishm, Sra, p. 40, has the incorrect reading . Cf. also Umayyas alleged
words in Ibn ajar al-Asqaln, vol. 1, p. 262, l 4:
. That the root is not

found in the Koran would seem to be unimportant.
73

.
74
One might consider changing
to but this is hardly feasible particularly as also

the second verse is in the third person.
75
However, with the imperative does not occur in the Koran.
196 mu

the text sounds somewhat like a combination of sras 8:73 and 32:17, which
the Medinan Maslama (Ibn Mukhallad) might actually have composed in
order to elevate the old Companions of the Prophet above the ruling dy-
nasty.
76
(4) It is related
77
that when the caliph Umar asked Abd al-Ramn b. [i/244]
Awf if he was familiar with the verse,
, fight
like you fought the first time (earlier!), he replied that the verse is one
of the abrogated verses of the Koran (
). No one
would deny that these words could be Koranic, yet it is quite legitimate
to question the survival of a corresponding tradition regarding the origin
of a passage devoid of all originality. Still another transmission (al-Mabn,
part iv, fol. 40
r
) presents the verse in this version:

.
The authenticity of this recension, however, is suspicious, because the ex-
pression zamn for time is foreign to the Holy Book. Even if one does not
attach great importance to this argument, the other remark in al-Mabn,
claiming that this is how the words appeared in sra 44, can only be cor-
rect if jhada were to refer not to the actual struggle for religion but meant
instead the zealous espousal of it in general.
78
Because of the strong empha-
sis on the first jihd, however, it is more likely that in this case we ought
to think of the Holy War. This interpretation would point to the Medinan
period at the earliest, namely after the Battle of Badr, since, according to
the text, one jihd had already taken place, whereas it is well-known that
sra 44 was promulgated at Mecca. Moreover, the sentence is probably to be
interpreted eschatologically and represents the classic period of the estab-
lishment of Islam in a distant future. This idea, of course, presupposes that
a longer period had passed after the death of Muammad.
(5) Muslim, K. al-Zakt, 26 at the end, and al-Itqn, p. 526. Ab Ms [i/245]
AL-ASHAR said:

79

80

.
76
The words , those with whom God is angry, could refer to the
Umayyads. Maslama did not die until the reign of Muwiya.
77
al-Mabn li namal-man, part iv, fol. 40
r
, al-Suy, al-Itqn, p. 526, al-Muttaq, Kanz
al-umml, vol. 1, no. 4749.
78
Cf. above, p. 127, on sra 29:69.
79

missing from al-Suy, al-Itqn.

80
al-Itqn, .
mu

We recite a sra that we consider similar
81
to one of the musabbit
82
which
fell into oblivion; I retained only the following: O, you who believe, wherefore
do you say what you do not do? A testimonial will be written on your neck,
and on the Day of Resurrection you will have to explain.
Since the second part of the citation is identical to sra 61:2, one is tempted
to regard the entire passage as a fragment from this sra, although this is
contrary to the rhyme, which in all the musabbit ends with n, n and
similar syllables, not to mention the explicit statement that the related sra
was lost. For this reason it is a hopeless undertaking from the start to try
to give the transmitted text of the fragment a different rhyme, be it by
rearrangement
83
or extension.
In al-Mabn, chapter iv, fol. 40a, the tradition reads a little differently:

.
The obvious fact that the entire quotationis identical here withsra 61:2 to 3 [i/246]
definitely suggests that we should give preference to the first recension over
this one. Otherwise, no objections can be raised against the reliability of the
tradition.
(6) Anas b. Mlikrelates that a verse of the Koranhadbeenrevealedreferring
to those who were killed at Bir Mana (afar 4/625) but was later abrogated
by God:

84
or similar.
85
The verse
can be found with some variants in four adths in al-Bukhr,
86
three in Ibn
Sad, Feldzge Muammads (vol. 2, part 1), p. 37sq., Ibn Sad, Biographien der
medinischenKmpfer (vol. 3, part 2), p. 71 sq., two in al-abar, Annales, vol. 1,
p. 1447 and 1448, and one each in al-Wqid, p. 341, Muslim, K. al-alt 93,
al-Suy, al-Itqn, p. 527, al-Diyrbakr, Tarkh (Cairo ed.) vol. 1, p. 14, and in
al-Suhayl in the note to Ibn Hishm, p. 650, as follows:
81
As far as length is concerned.
82
Sras 57, 59, 61, 62, 64, that begin with
.
83
Perhaps in .
84
al-Bukhr, a.d (cf. note 39); al-Diyrbakr, vol. 1, p. 14.Muslim reads instead .
85
Fromamong the numerous variants of this text most remarkable are those that say that
for quite some time it was recited as Koranic: Ibn Sad, Feldzge, p. 38

; al-abar, vol. 1, p. 447,
.
86
One in K. al-Jihd 183, here designated as d; the others in K. al-Maghz 30 are
designated as a, b, c.
198 mu

87

88

89

90

91
.
Informour people that we met our Lord and that He is pleased with us as we [i/247]
are also pleased with Him.
In other traditions this text is put in the mouth of those Muslims who died
at Uud or Bir Mana and immediately went to the glory of Paradise;
92
or
the Prophet relates in an address to his Companions the request of these
martyrs. Thus in al-Bukhr (in a passage between b and c) it reads:

.
in Muslim, K. al-Imra, 39:

.
93
in al-Tirmidh, K. al-Tafsr on sra 3:163:

(var.: )
.
in al-abar, al-Tafsr on sra 3:163 (vol. 4, p. 108):

.
It is not easy to determine whether the original passage was fromthe Koran
or adth. Although the phraseology of our text is undoubtedly Koranic,
94
it is precisely this peculiarity that might have caused the adth to be
87
Before
Ibn Hishm, Ibn Sad, Biographien der medinischen Kmpfer, Muslim, and
al-Itqn read
; al-Bukhr (d), . In addition, Ibn Hishm and Muslim read instead .

88

is missing fromal-Bukhr (c, d), al-Wqid, Ibn Hishm Ibn Sad, ibid., and Muslim.
The entire beginning from to is missing in al-Bukhr b.
89
Instead al-abar, vol. 1, p. 1448, l 3, reads . Ibn Sad (al-abaqt): Feldzge
Muhammeds, p. 37, puts before
.
90
Instead Ibn Hishm and Muslim read ; al-Bukhr d ; al-Bukhr c ;
al-Wqid, al-Itqn; Ibn Sad (al-abaqt): Die Feldzge Muhammeds, simply .
91
Insteadof al-Bukhr, a, b, d, IbnSad, vol. 2, part 1, p. 38, andal-Suy, al-Itqn,
read . But al-Diyrbakr, Tarkh al-khams, also lists the other variant readings.
92
This message from the hereafter has a remarkable parallel in Luke 16:27sq.
93
A greater deviation in al-Baghw on sra 3:163:

Al al-Dn Al b. Muammad AL-KHZIN al-Baghdd, Tafsr on sra 3:163 as follows:

94
Cf. ) ( , sras 6:31 and 155, 10:46, 13:2, 18:105 and 110, 29:4 and 22, 30:7, 32:10;
, sras 5:119, 9:101, 58:22, 98:8.
mu

considered a revelation. If, however, it was indeed from the Koran, the
introduction stating that these were the words of the men killed must have
disappeared.
(7) Very famous is the so-called Verse of Stoning (
), which, according
to traditions from Umar, was considered part of the Koran:
95

96

97

98

99

100

101

102
Do not long to be away fromyour fathers
103
for this is by you godlessness; and [i/249]
if an elderly man and woman
104
commit adultery then stone them definitely
as penalty from God All-Mighty and Wise.
Most writers simply say that this verse belongs to the manskht, the abro-
gated passages of the Koran.
105
Al-Qurub, al-Mabn li-nam, cap. iv,
95
Let us refer to only a few sources where this text appears complete or in parts: Ibn
Hishm, 105; al-abar, vol. 1, 1821; al-Muwa, 349; Ibn Sad, Biographien der mekkanischen
Kmpfer (vol. 3, part 1), p. 242; al-Yaqb, vol. 2, p. 184; al-Mabn li-nam al-man in three
passages, part 2, fol. 16
r
, part 4, fol. 34
v
; al-Qurub; al-Nsbr in the margin of al-Bukhrs
Tafsr, vol. 21, p. 81 on sra 33; al-Zamakhshar on sra 33, at the beginning; al-Nasaf in the
margin of Al al-Dn AL-KHZIN al-Baghdd, vol. 3, p. 472, on the same sra; Ibn azm in
the margin of the Jallayn, vol. 2, p. 148; al-Suy, al-Itqn, with three variants, p. 524 (a, b)
and p. 528 (c); Hibat Allh (IbnSalma), p. 13; W. Ahlwardt, Verzeichnis, no. 478; al-Diyrbakr,
al-Tarkh al-khams, vol. 1, p. 14; al-Suhayl, loc. cit., etc. Innumerable writers maintain that

has been part of the Koran.
96

missing from al-abar, vol. 1, p. 1821, cod. c.

97
Hibat Allh b. Salma:
.
98
The words up to here are missing frommany of the sources, for example, Muwaa, 349;
Ibn Sad; al-Yaqb; al-Mabn li-namal-man, part iv, fol. 33
v
; al-Nsbr; al-Zamakhshar;
al-Nasaf; al-Diyrbakr; Tarkh; al-Suy, al-Itqn, a, b, c; al-Muttaq, Kanz al-umml, vol. 1,
n. 4751. After they are combined with in al-Suhayl with the variant
; Ibn Hishm
and al-abar have the additional
, but omit everything else. According to

al-Mabn li-nam al-man, iv, fol. 32, some (e.g. al-Bukhr, K. al-Mubirn, 17) read

.
99
Ibn Sad, ; al-Suhayl, .
100
Some people read , al-Suy, al-Itqn, p. 524 (a); al-Mabn li-nam
al-man, iv, fol. 34
v
; al-Yaqb and al-Suy, al-Itqn omit .
101
All of the text that follows is missing from Muwa; Ibn Sad; Ibn Mjah, Bb al-rajm;
al-Nsbr, al-Suy, al-Itqn. al-Itqn, p. 524 (b) has instead the gloss .
102
The last three words are found in al-Suy, al-Itqn, p. 524 (a), al-Qurub, al-Zamakh-
shar, Hibat Allh b. Salma, al-Nasaf, Ibn azm. However, al-Yaqb and al-Diyrbakr,
Tarkh al-khams, have instead of . The words
appear identi-
cally in sra 5:62.
103
I.e., do not endeavour for false pride to belong to a family other than your own.
104
In al- Muwa Mlik b. Anas adds the gloss, .
105
Some writers say that the verse was lost accidentally when some domesticated animal
200 mu

al-Nsbr, al-abar, p. 1821, al-Zamakhshar, al-Nasaf, al-Suy, al-Itqn
(a), and al-Muttaq, Kanz al-umml, vol. 1, no. 4751, all state that, accord-
ing to isha and Ubayy, the verse allegedly was part of sra 33. This, how-
ever, is impossible, as the verse rhymes with n but the sra thoughout
with . According to another tradition, this verse was originally part of sra
24.
106
This sra is more likely to fit, as it not only agrees with the rhyme of
the fragment but also treats only the adultery (zin) of men and women.
Moreover, verse two, which stipulates the penalty of flogging without any
qualification, is in contradiction to the Verse of Stoning.
107
One ought to
assume that this verse was abrogated by this verse two. This, however, is
neither documented nor can it be harmonized with the development of
Islamic criminal law. If it is true that during his Medinan period Muam-
mad repeatedly condemned unchaste persons to death by stoning,
108
it is
incomprehensible that such a revelation should have become abrogated,
or that it could have been lost. As the credibility of this tradition cannot
be verified, it must be taken for granted that the penalty of stoning was
practised under the first caliphs; it is taught in Islamic legal texts down to
( , al-Nsbr, al-Zamakhshar, al-Nasaf) devoured the spot where this verse had been
written. The same is also said of the verse of nursing (see below, p. 202sq.) by (Ibn Mjah,
Sunan, cap. ; al-Mabn li-nam al-man, vol. 4, fol. 40
r
; al-Damr, K. al-ayawn,
s.v. ). All these traditions allegedly go back to isha, but in most cases (e.g. al-Nsbr,
al-Zamakhshar, and al-Nasaf) are considered a fabrication of sectarians
.
106
al-Bukhr, K. al-Muribn 7; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 14.
107
By the stretch of imagination verse 2 might be interpreted not to be in contradiction to
the precepts given in this instance; but such harmonizing is hardly admissible. Subsequent
law recognizes both the penalties mentioned, namely requiring flogging for lighter offences
but stoning for serious cases of adultery. This is explained in more detail below, p. 201
n. 109.
108
al-Bukhr, K. al-Muribn 7, 8, 10, 11, 14 and 15. Ibn Sad (al-abaqt): Biographie
der Genossen (vol. 4, part 2), p. 51, al-Diyrbakr, Tarkh al-khams, vol. 1, p. 467, vol. 2, p. 139.
Another tradition, ibid., p. 52, l 7sqq., relates that on one occasion Muammad did not have
a repentant adulteress stoned, but pardoned her.
On the other hand, the words allegedly said after the conquest of Mecca
(al-Bukhr, K. al-Buy 100, Way 4, Maghz 54, Fari 18, Muribn 9, Khumt
5, al-Wqid by Wellhausen, p. 338; for additional sources see Goldziher, Muslim studies,
vol. 1, Germanoriginal pagination, Seite 188, note, andhis TurbundHaar, p. 589)cannot
be applied here. Al-Qasalln on al-Bukhr, Fari 18 (vol. 9, p. 438sq.) explains as follows

, i.e. the person who fornicates

forfeits the right of kinship (nasab) with the begotten child. Only this interpretation of
those words together with the protasis, , constitutes a meaningful entity. The other
interpretation which has the stoning ( ) in mind is rejected on good grounds by
al-Qasalln as well as others.
mu

this day.
109
According to most of the reports, it was Umar who urged the
Medinans to observe this cruel law: I have seen the Prophet have someone
flogged, andwe thenalsoflogged. If people hadnot accusedme of attempted
innovation, I would have included the Verse of Stoning in the Koran. We
actually recitedit.
110
The words we quote leave the impressionthat the claim
to the divine origin of this verse merely serves as a means to establish the
law in practice. As the stoning of unchaste men and women was foreign to
ancient Arabian common law, which did not know criminal punishment,
the example of the Jewish law in Deuteronomy 22:2124
111
would have pre-
dominated.
The beginning of the verse has no apparent internal relation to the part [i/251]
concerned with stoning and, furthermore, as far as the content and form
is concerned, shows Koranic characteristics. Since both parts are regularly
connected with each other by the name Verse of Stoning, however, what
was judged probable for its second part applies also to its first part, namely
that it was never part of the Holy Book.
112
The beginning of the verse appears a few times
113
in the following form

as a mere saying of Muammad, without

supplying particular circumstances.
(8) al-Muttaq, Kanz al-umml, vol. 1, no. 4752, from Ubayy b. Kab:

109
According to lawnot only the personwho is married at the time of committing adultery
is to be stoned but also the person who had previously been living in a legal marriage. Who-
ever had never been married before (
and
respectively) is flogged with one

hundred lashes. Cf. Ed. Sachau, Mohammedanisches Recht, p. 809sqq.; Snouck Hurgronjes
review of Sachaus Mohammedanisches Recht, p. 161 sqq.; Th.W. Juynboll, Handleiding tot de
kennis van de Mohammedaansche wet, p. 302sqq.
110
Ibn Hishm, p. 1014sq.; al-abar, vol. 1, p. 1821; Ibn Sad (al-abaqt): Biographien der
mekkanischen Kmpfer (= vol. 3, part 1), p. 262; al-Yaqb, ed. M.Th. Houtsma, vol. 2, 184;
al-Tirmidh, Abwb al-udd, cap. 8; al-Diyrbakr, Tarkh al-khams, vol. 1, p. 14; al-Mabn
li-namal-man, part iv, fol. 32
r
sq.
111
When sentencing a Jewish couple who had been brought before Muammad for sus-
pected adultery reference is made to this Biblical passage; al-Bukhr, K. al-Mubirn, 10;
al-Diyrbakr, Tarkh al-khams, vol. 1, p. 467.
112
This is also Caetanis opinion in his Annali, vol. 2, part 1, p. 305. From a lexical point of
view it might be noted that the words and
do not occur in the Koran.

113
Muslim, K. al-mn27; al-Khabal-Tibrz, Mishct, bbal-lian, fal 1 at the endwhere
there is instead ; of course in this case it cannot be read either
or but must be
.
202 mu

This text is composed entirely of Koranic passages. From the beginning [i/252]
to it is found in sra 17:34, from to
in sra 4:26; from

in sra 25:70, and the final phrase to
in sra 4:128. This, of course, is

no decisive argument against its authenticity, because we find in our Koran
quite a few verses that look like a combination from other parts. Still, such
a text cannot claim any reliance unless it can be firmly established that it
belongs to the Koran.
(9) al-Muttaq, Kanz al-umml, vol. 1, no. 4753, fromAbIdrs AL-KHAWL-
N;
114
al-Mabn, part iv, fol. 37
r
:

115

116
The beginning of this text up to
, as well as the end starting with

, follow one another immediately in sra 48:26. In the middle passage,

however, from to , only isolated idioms are from the Koran.
(10) al-Muttaq, Kanz al-umml, vol. 1, no. 4754, from Bajla (Ibn Abda):

.
Nearly the entire text up to
is identical with sra 33:6. The three

following words are now and then listed by the exegetic tradition among
the canonical readings.
117
The designation of the Prophet as father of the
believers is nowhere else to be found in the Koran. On the contrary, he
rejects this outright in sra 33:40. By the same token, he never addresses the
believers as my children.
(11) isha said:
118
114
Died 80/699; cf. al-Dhahab, Tadhkirat al-uff, vol. 1, p. 49.
115
Kanz al-umml inserts after an additional . However, this reading variant is
extremely suspect since it does not really make sense, and since it can readily be explained
by an inadvertent double .
116
All that follows is missing from al-Mabn li-namal-man.
117
For example, al-abar, Tafsr, s.v., vol. 21, p. 70 l 15, ; al-Nsbr, Tafsr, s.v.;
in al-abar, Tafsr, vol. 21, p. 84 in the margin, .
118
Muwaa, p. 224; Muslim, K. al-Ra 1 in two forms (a, b); al-Tirmidh, Abwb al-ra
3; al-Nas, K. al-Nik 49; al-Khab al-Tibrz, Mishct, Bb al-muarrimat, fal 1 6;
al-Mabn li-nam al-man, part iv, fol. 40; al-Suy, al-Itqn, p. 517; al-Diyrbakr, Tarikh
al-khams, vol. 1, p. 14.
mu

119

120

121

122

123

124

: ]
125
[ ;
126

127

: ]
128
[

]
129
[ ;
or
130
finally:

.
This concerns the number of breast-feedings that put a child in the same
relationship to its wet-nurse and her relativesinsofar as the legality of
marriage is conceivableas tothe natural mother andher relatives. Accord-
ing to the earlier version of this alleged Koranic passage, ten breast-feeding
sessionsand, according to a later version, fiveresult in a relationship
that excludes marriage. The text of the respective revelation is best viewed
from the third of the above-mentioned recensions: (scil. )
. It is not easy to arrive at a sound opinion regarding the reliability
of this transmission. However, given that the controversy about the number
of breast-feedings required to constitute an impediment to marriage had
created lively discussions even in the early legal schools,
131
we have to realize
the possibility that this saying is a fabrication, serving to support a particu-
lar scholasticism of the basic law (sra 4:27), either a priori as a verse of the
Koran or first put into the mouth of the Prophet as a adth.
(12) al-Wqid in Wellhausens Muhammed in Medina, p. 187:
132
[i/254]

. This alleged verse fromthe Koran, those who slander decent women,
God shall curse in the present world and the world to come, seems to be
nothing but a free translation of sra 24:23.
119
Missing from al-Nas.
120
Missing fromMuslimb also (
), al-Itqn, Tarkhal-khams; Muslim(a) adds
.
121
[yuarrimna] vocalization fromal-Khab al-Tibrz, al-Mishct, and al-Mabn li-nam
al-man, fol. 40
v
; the word is missing from al-Itqn.
122
Muslim; al-Nas,
.
123
al-Diyrbakr, Tarikh al-khams, omits all the following words.
124
Tarikh al-khams; al-Nas, ; al-Mabn li-namal-man, .
125
al-Tirmidh.
126
, al-Khab al-Tibrz, Mishct; Muslim a.
127
al-Muwa, however, according to the gloss, another manuscript has .
128
Different version in al-Mabn, part iv, fol. 35
r
.
129
Ibn Mjah, Sunan, cap.
.
130
Muslim b.
131
Cf. al-Sharn, al-Mzn, Cairoed. (1317/1899), vol. 2, p. 131; Juynboll, Encyclopedia, p. 725,
col. 1.
132
The Arabic text has been kindly supplied by August Fischer of Leipzig.
204 mu

Our study has thus led to a great variety of results. In no single case was
it possible to procure evidence of the reliability of the transmission. Against
that we may put the fact that in the case of three fragments (nos. 1, 2, 4, 7,
and 11) their reliability can be challenged on good grounds, and in the case
of two of them (nos. 3 and 12) there are at least doubts. The fragments from
8 to 10 differ from the rest in so far as the text can be found literally in the
Koran, either completely, as in the case of no. 8, or to a large extent, as in the
case of nos. 9 and 10. Tradition quite rightly points out that numbers 9 and
10 represent nothing but different recensions of sras 48:26,
133
and 33:6 (cf.
above, foot-note 116). In this case, we may also consider number 8 to be, at
best, a variant of sra 17:34.
134
Non-Extant Revelations [i/255]
In addition, we have other bits of information about lost parts of the Koran,
although no fragments have survived.
Sra 33 is supposed to have been considerably larger. As it nowcomprises
only seventy-six verses, some traditions ascribe to it some two hundred
verses; according to still others, it was once as long as the second sra, or
evenlonger.
135
The truthcannownolonger be established. If the information
does not rest on false conjectures, this reference is probably to an old copy
in which our sra thirty-three comprised several other pieces. Additionally,
sra nine
136
and, as indicated above on pp. 172 and 174, sra ninety-eight
are both supposed to have been considerably larger. The origin of this
information we shall investigate in the following chapters.
The two alleged sras, which are called , will also have to be [i/256]
treated in this context.
133
Kanz al-umml, vol. 1, no. 4753. The text to be found in the Errata.
134
The words , Ibn Hishm, 810, l 5sq., = al-abar, vol. 1, 1627, l 11 sq.,
are not a locus Korani deperditus (GlossariumTabari, s.v. ), but present a revelation only
hypothetically ( ).
135
al-Mabn, part iv, fol. 33
v
, 35
r
; al-Zamakhshar on sra 33; al-Qurubi, on sra 2:100 and
on sra 33; al-Naysabr and al-Nasaf on sra 33; al-Suy, al-Itqn, 524; al-Muttaq al-Hind,
Kanz al-umml, vol. 1, no. 4751.
136
Fakhr al-Dn al-Rz and Al al-Dn on sra 9:65:

.
mu

The Uncanonical Sayings of the Prophet
To the third type of promulgations belong Muammads utterances, which,
although recognized as divine,
137
are not explicitly claimed to be an integral
part of the Koran. There are manuscript texts where these dicta are col-
lected.
138
We must here restrict ourselves to six examples.
(1) al-Bukhr, K. al-Tawd, cap. 50, 1:

) (
.
Whenthe manapproaches me by the spanof the hand, I approach himby an
ell, and when he approaches me by an ell I approach him by a fathom,
139
and
when he walks towards me, I run towards him.
140
(2) al-Bukhr, Kitb al-awm, cap. 9, Muslim, Kal-iym, cap. 22 (al-Qasal-
ln, VI, p. 135):

141

142
Every activity concerns man except fasting, which concerns me and which I [i/257 ]
repay; fasting is a fence, and when one of you has a day of fasting, he should
not use obscene language nor yell, and when someone wants to exchange
insults with him or quarrel, let him say I am fasting, and by Him in Whose
hand is the life of Muammad, the smell from the mouth of a fasting man
is for Allh sweeter than the scent of musk; allotted to the fasting man are
two pleasures, the joy whenever he is allowed to eat again, and the joy of his
fasting when some day he returns to his Lord.
(3) al-Bukhr, K. al-Tawd, 50:

137
Several different formulas were in use for this purpose, e.g., ) ( ) (
or
, etc.
138
Cf., for example, , in Brockelmann, GAL, vol. 1, p. 446, no. 83.
139
Cf. al-Qasalln, s.v., vol. 10, p. 464: .
140
In a third recension in al-Bukhr, loc. cit., the words are merely indicated as a saying
of Muammad.
141
On the expression ibn dam, man see above, p. 194 n. 62.
142
In al-Bukhr, K. al-Tawd, cap. 50, and al-Mabn li-namal-man, cap. iv, fol. 36
r
, l 9,
there is a still shorter recension; several can be found in Muslim, loc. cit.
206 mu

No one can say that he is better than Jonah, the son of Amittai, although his
descent goes back to his father.
143
(4) Ibn Sad, Biographien der medinischen Kmpfer Muhammeds, p. 123:
.

Assured of my compassion are those who in me love one another, in me
144
sit
together, and in me help and visit one another.
[No number (5), the count jumps from no. 4 to no. 6.]
(6) al-Ji, K. al-Masinwa-l-add = Le Livre des beauts, editedby Gerlof [i/258]
van Vloten, p. 168, l 2sq.; al-Bayhaq, K. al-Masin wa-l-masw, edited by
Friedrich Schwally, p. 310, l 4sq.:

O, man,
145
if you make me a journey, then I will provide you with sustenance.
Islamic encyclopaedic science
146
unites uncanonical promulgations under
the heading
or
. There is the proper,

clear differentiationbetweenthis category andthe Koran,
147
onthe
one hand, and
, the ordinary sayings of the Prophet on the other

hand. But Islamic science makes a mistake by considering the so-called holy
adths outright revelations,
148
since inno single instance is it apriori certain
that we are even dealing with an authentic utterance of Muammad.
143
These words are transmitted as simple words of the Prophet and to be found in al-
Bukhr, K. al-Tafsr on sra 4:161, and sra 6:86; al-Nawaw, Tahdhb, ed. Wstenfeld, p. 641,
etc., partly omitting the last phrase, the sense of which is not clear. Al-Qasalln does not
supply an explanation. The Prophet Jonah ( ) is mentioned four times in the Koran (4:161,
6:86, 10:98, 37:139) however never with the kunya.
144
This is a typical Christian expression and frequently found in the New Testament but
became quite common in Islamic literature. There are numerous examples that have been
collected by Ignaz Goldziher, Muslimstudies, vol. 2, p. 355sqq. Cf. also M.J. de Goeje in glosses
in al-abar, s.v., . The expression is foreign to Koranic diction because in the phrases
, sra 22:77, and the addition of is likely to be intended. This expression
in the last line seems to originate fromthe mystic cultures of antiquity; cf. Alb. Dietrich, Eine
Mithrasliturgie, p. 109sqq.
145
On the expression ibn dam see above, p. 194, n. 62.
146
al-Tahnaw, Dictionary of the sciences of the Musulmans (

) critical edition by Muammad Vajh and Maulavi Abdul

Haqq and Maulavi Ghulam Qadir (Calcutta, Bibliotheca Indica, 1854), p. 280sqq., s.v. .
Al-Jurjn (d. 816/1413), Definitiones, ed. G. Flgel, p. 88, is able to trace back this terminology
yet two more centuries.
147
Thus the Koran is outright called , for example, al-Masd, al-Tanbh, p. 191, l 5.
148
Cf. al-Mabn li-nam al-man, cap. iv, fol. 36
r
:

.
mu

Miscellaneous Revelations Preserved in the Traditions
Finally, we must remember that there is an exceedingly large number of [i/259]
traditions stating that Muammad received revelations on the most diverse
occasions, consisting of warnings or commands as well as disclosures of
contemporary or future events. Of the abundant material, some specimens
must suffice:
After a revelation, Muammad suddenly shunned Amr b. Jish, who
wanted to kill him with a stone (al-Wqid, p. 355sq.).
He learns from divine information where a stray camel went (al-Wqid
in Wellhausen, Muhammed in Medina, p. 183).
If it is a command from Heaven, Usayd [Juynboll, Encyclopedia] said to
Muammad, comply! (al-Wqid in Wellhausen, Muhammed, p. 204).
The Prophet is foretold by a revelation that God graciously received Ab
Lubba (loc. cit., p. 214), and is informed of the circumstances of the new
prayer place of the Ban Slim (loc. cit., p. 410).
Near al-Jirna
149
he was approached by a man dressed in a jubbah (cloak),
with perfume sprinkled all over himself and his clothes, who asked a ques-
tion, upon which Muammad succumbed to a fit, waited for a divine reve-
lation, and then replied (al-Bukhr, K. al-Fail al-Qurn 2).
This type of revelation seems to have a very high degree of authenticity.
That no revealed texts are extant alone adds to their credit. Yet even in the
case that all these episodes belong to the realm of fancy, it remains a fact
that we have here an appropriate image of Muammads moods and frame
of mind. As the history of religiondemonstrates, it is the mark of prophets to
remaininnearly constant connectionwiththe deity, and this not only inthe
case of great and momentous actions but also being receptive to inspiration
in innumerable minor daily affairs.
It can thus be taken for granted that, apart from the qurns, Muam-
mad experienced the influence of other revelations.
150
If, in addition, we
remember his many independent manifestations, the question must be
raised as to how it was possible for him to see his way through this confu-
sion. The Koranic revelations, according totheir owninterpretation, goback
to a divine book in heaven. Therefore, Muammad is likely to have consid-
ered only such revelations as qurns which, according to his belief, origi-
nated from that heavenly archetype.
151
We can still add to this pure formal
149
An oasis between al-if and Mecca; Juynboll, Encyclopedia, p. 140, col. 1.
150
Cf. already above pp. 1011, and 14.
151
Cf. the fitting remark regarding the so-called divine adths in al-Mabn li-nam
al-man, vol. 4, fol. 36
r
:
.
208 mu

principle the material aspect, namely that it concerned generally applicable
ordinances and important affairs of religion.
On such unstable ground the Prophet himself might occasionally have [i/260]
been confronted with doubt; the epigoni, who were tasked with the collec-
tion of the literary bequest, must have been so much more likely to fall into
error. For this reason, common utterances could easily acquire the reputa-
tion of holy adths, or even penetrate the Koran as revelations of prime
importance, just as, vice versa, genuine qurns that, for whatever reason,
did not gain entry in the canonical collection might have found their way
into the collections of adths.
Still, up to this point, we have not been able to consider any adth to
be Koranic with any degree of certainty. On the other hand, it seemsas
shall be shown later in the following sectionthat there is no passage in
the Koran that should be banished to the adth collections for good reason.
This negative result might be related in part to the inherent difficulty, and to
our insufficient research aids, as well as to the self-reliance and competence
prevailing during the collection of the Koran.
The main credit for the reliability of traditions belongs to the Prophet [i/261]
himself. Presumably from the first moment he was convinced to receive
communications from a divine book he laid down that, vis--vis the Bible
of the Jews and Christians, these revelations were to serve as a true and
unadulterated testimony of the divine will. Therefore, it must have been in
his interest to save these revelations from oblivion and misrepresentation
by means of fixing themin writing. Tradition, indeed, not only mentions the
secretaries to whom he dictated the qurns,
152
but also supplies important
information about the form of these copies. On the other hand, all other
communications from Allh were not officially recorded, our knowledge of
them depending instead on the accidents of oral transmission.
152
Cf. above, p. 23sqq.; Caetani, Annali, vol. 2, part 1, p. 706sq.
WRITTEN COLLECTIONS
IN MUAMMADS LIFETIME
The Preservation of the Manuscripts of the
Revelation during Muammads Lifetime,
Based on References in the Koran and
on the Literary Formof the Sras
The numerous individual revelations that constitute the Holy Scripture of [ii/1]
Islam are based on Koranic references to a book preserved in heaven in an
accurate form. Althoughthe Bible of the Christians andJews originates from
the same archetype, it has beensubject toserious falsifications. Additionally,
the different names for revelation, like qurn,
1
kitb, andway,
2
are allusions
to its written origin. In such circumstances it would be difficult to under-
stand that Muammad had not endeavoured at a very early stage to estab-
lish a new document of revelation as well as its written fixation.
3
Already
the Meccan sra 29:47 contains an allusion to writing down the revelations.
Tradition is quite explicit in this matter and even records the names of the
men to whom the Prophet dictated the revelations.
4
Yet we lack reliable
information
5
regarding the particulars of procedure and the preservation
and arrangement of the material. According to Lammens,
6
in sra 75:1617
Allh summoned Muammad not to hasten the edition of the Koran as a
special collection so as to have a free hand to change the text leisurely. This
interpretation, however, is wrong; moreover, in this context, hastening can
refer only to the Prophets arbitrary interpretation, from which he should
refrainuntil a complete, appropriate revelationhadbeenreceived. Similarly,
sra 20:113 commands: And hasten not with the qurn ere its revelation is
1
Qurn is both the infinitive of qaraa, to recite, and loan word from Aramaic eryna
lector.
2
Goldziher, Muslimstudies, vol. 2, p. 20, n. 2.
3
Sprenger, Mohammad, vol. 3, p. xxxiii; Hirschfeld, New researches into the composition
and exegesis of the Qoran, pp. 136 and 141 as well as many others deny this for no valid reason.
Cf. sra 25:34 and above, p. 208.
4
Cf. above, p. 36sqq.
5
6
H. Lammens, Ftima et les filles de Mahomet; notes critiques pour l tude de la Sra, p. 113.
210 written collections in mu
hammads lifetime
accomplishedunto thee. Aliterary analysis of the extant sras reveals, how-
ever, that Muammad himself occasionally combined individual qurns
into a larger unit, or wanted to present markedly artistic literary composi-
tions as emanating from singular and uniform occasions of promulgation.
The homilitic arrangement of most of the sras makes it exceedingly [ii/2]
difficult to penetrate the secret of their composition and express an opinion
as to what extent the combination of individual revelations of different
provenance within one and the same sra is to be attributed to the Prophet
himself or only to subsequent editors. The literary unity in the case of the
larger sras can be maintained with some degree of certainty only in cases
when there is identity or homogeneity of content, as it is the case of sras
12 and 18, or when a refrain threads the entire sra, as in sras 26, 56, 70,
and 77, or when style, rhyme, and rhythm display such a great harmony, as
in sra 37. Much more doubtful is the matter in the case of sras 17, 41, and
7. No conclusion at all can be reached as far as sras 2, 8, 63, 4, and 9 are
concerned. None of the respective sras can be claimed to have come about
without the aid of notes.
I like to assume the same in instances when Muammad at Medina [ii/3]
enlarged earlier revelations by means of small additions or interpolations,
7
or even replaced or abrogated them by a new text with different content.
8
This was intended to loosen somewhat the chain of revelations, which he
had carelessly put around the neck of his prophetic liberty when he fixed
them in writing.
On the other hand, the numerous dangling verses, as well as the frag-
mentary groups of verses that are either embedded in sras or now placed
together in the final part of the canonical edition, require a separate expla-
nation. No matter how much importance Muammad attached to writing
down, we cannot expect too much completeness and archival perfection,
least of all at Mecca, where he had to fight for life or death as the recognized
Messenger of God. Under the pressure of external circumstances, the keep-
ing of records, even if intended, will have been neglected more than once.
7
E.g. sras 74:3134; 80:1733 and 4860; 95:6; 85:811; 78:37sqq.; 19:3541.
8
Sra 2:100. Its Koranic expressionis nasakha, whichlater entered scholarly usage. Origi-
nally this meant either tointroduce a variant reading denominatedafter the Judeo-Aramaic
loanwordnuskhah, codex, copy, or it originates fromthe Aramaic verbwiththe meaning of
to delete. As I emphasized earlierabove, p. 41 sqq.the theory of the abrogated passages
of the Koran is unlikely to be Muammads free invention; rather it is related to a concept
transmitted to him, possibly the New Testament concept of the abrogation of the law by the
Gospel ( .)
written collections in mu
hammads lifetime 211

In the earliest period everything is very likely to have been left to memory,
which, at times, must have failed the Prophet. Accordingly, he consoles the
believers in sra 2:100 with the promise that for whatever verse is cast into
oblivion Allh will give them a similar or better one.
Apart from records made on Muammads own demand, there were [ii/4]
probably also such of smaller or larger size produced or commissioned by
eager followers of his teachings. Concurrently there was the preservation
by memory, which must have been of greatest importance at a time when
reading and writing was still a rare art. In addition to the not insignificant
number of companions who knew shorter Koranic passages by heart as far
as this was required
9
for the prayer liturgy, there were individuals who had
memorized considerable sections that they could repeat correctly, as from
a book, and thus save from oblivion many a revelation that either had not
been recorded or had been lost.
As long as the Prophet was alive revelation was like a flowing stream. But
when after his death this sourcebecause of its vital importance for the
new religionwas suddenly exhausted,
10
the congregation was sooner or
later bound to feel the need to have access to the entire corpus in a reliable
form.
The credit for this collectionof the Korantraditionascribes inremarkable
unison to the three first caliphs.
11
We have at our disposal a considerable
number of old and more recent traditions on this subject. Even if most of
them are agreed in principle, they still differ markedly in important details.
Since sources dealing with such important affairs of religion are a priori
suspect of tendentious interpretation, we must approach confirming as well
as contradicting information with equal scepticism. An investigation based
on such principles and in connection with careful consideration of the
extant form of the Koran as the final result of developmentand, in the
final analysis, the only unconditional safe cluemight still get close to the
truth of the matter.
9
Cf. e.g. al-Bukhr, Adhn, 94sqq.; Ab Isq al-Shrz, al-Tanbh f fur al-Shfiiyya,
p. 21 sqq. [Sezgin, GAS, vol. 9, p. 202, last line].
10
Inqaaa l-way, al-Khab al-Tibrz, Mishct, Manqib Ab Bakr, at the end.
11
al-Khzin al-Baghdd, Tafsr, vol. 1, p. 6, says that in this entreprise the legitimate
caliphs had the support (waffaqa) of Allh. al-Suy, al-Itqn, p. 133, even uses the strong
expression (alhama) inspired. Cf. also below, p. 219sqq., regarding Als alleged activity in
the collecting of the Koran.
THE INDIRECT COLLECTORS
OF THE KORAN OR THE CUSTODIANS
OF THE ORAL REVELATION
Since the Messenger of God left the earthly stage suddenly and unexpect- [ii/5]
edly, it is obvious that the Korancannot have beencompletely collecteddur-
ing the Prophets lifetime. If a tradition going back to Zayd b. Thbit
1
main-
tains that at this time the Koran had not been collected at all, then this
is based on a different concept regarding the accounts of how Ab Bakrs
edition
2
came about. According to this account, this caliph found the tra-
ditions dispersed and on different pieces oras al-Suy adds by way of
explanationneither in one place nor arranged according to sras. This
interpretation, however, is not in complete agreement with the results of
the previous chapter, according to which even at that time there had been
not only sras, which from the outset were conceived as literary units, but
rather also sras that Muammad himself later put together from pieces of
different provenance.
Although the answer to this question must be postponed until the study
of Ab Bakrs recension of the Koran has been completed, yet another
strange contradiction to the prevalent opinion can be cleared up right here.
There are not a few traditions that quite harmlessly and without a trace
of polemics against divergent opinions mention a large number of persons
who allegedly collected the Koran during the Prophets lifetime. Ibn Sad
devotes a separate chapter
3
to this subject, even though in another part of
his work he considers the first caliphs to be the first compilers of editions
of the Koran. In such circumstances there is hardly any doubt that there is
something peculiar about it. Indeed, the idiom used, jamaa l-Qurn, does
not refer to collecting the dispersed revelations in one book but rather,
as it was already known to Muslim authorities of the interpretation of
1
al-Suy, al-Itqn, p. 133, l 6sqq.
2
Cf. below, p. 220sqq.
3
Dhikr man jamaa l-Qurn al ahd rasl Allh, Ibn Sad (al-abaqt, vol. 2, part 2):
Letzte Krankheit, Tod und Bestattung Muhammads [Muammads last illness, death and
funeral], ed. Fr. Schwally, pp. 112115.
214 the indirect collectors of the koran
adth, to learning by heart.
4
In the case of this interpretation it remains
uncertain whether the individual collectors really had memorized the
entire revelation or only fairly large portions. As we shall see later, knowing
the holy texts by heart has always been of prime importance; the written
preservation of the revelation has always been considered the means to an
end.
The opinions of the different traditions differ not only with respect to the [ii/6]
number of these so-called collectors, but also as to their names. Most fre-
quently the following four are mentioned together:
5
Ubayy b. Kab, Mudh
b. Jabal,
6
Zayd b. Thbit, and Ab Zayd al-Anr [Sad b. Aws b. Thbit,
d. 215/830]. In the numerous variants of this tradition many new names
appear: e.g., Ab l-Dard, Uthmn, Tamm al-Dr,
7
Abd Allh IBN
4
al-Nawaw, Tahdhb, ed. Wstenfeld, p. 516, l 4, paraphrases
by they
had completely remembered it by heart. al-Qasalln, vol. 6, p. 162, on al-Bukhr manqib
Zayd b. Thbit explains
by
. Al-Suy, al-Itqn, bottom, it reads:

[namely
] . Regarding other synonymous expressions cf. below, p. 223 n. 4.

In the tradition Ibn Sad, loc. cit, p. 112, l 16,
, common to all the other recensions of

this chapter, is replaced by
, of course also with the meaning to remember by heart.

In other passages, e.g., Ibn Sad (al-abaqt, vol. 3, part 1): Biographien der mekkanischen
Kmpfer [biographies of the Meccan combatants], p. 53, l 15 and 18, the phrase means to
recite all of the Koran.
5
al-Bukhr, Bad al-khalq, 149, manqib Zayd; Muslim, Fail, cap. 58; al-Tirmidh,
manqib Mudh; al-Khab al-Tibrz, Mishct, Jmi al-manqib, fal 1, 9; al-Mabn li-nam
al-man, part 4; al-Qurub, vol. 1, fol. 22
r
; al-Suy, al-Itqn, p. 166, l 5sqq.; al-Shshw,
iii; Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung Muhammads,
p. 113, ll 7, 1417, etc. These names are spread over different traditions as follows: al-Itqn,
p 166, mentions Mudh, Zayd, Ab Zayd, Ab l-Dard (Uwaymir al-Khazraj, d. 32/652)
[EI
2
; G. Juynboll, Encyclopedia of canonical adth, p. 550, n. 1] (four).Ibn Sad, Letzte
Krankheit, p. 113, l 5sq.: Ubayy, Mudh, Uthmn, and Tamm al-Dr (four).al-Bukhr,
Bad al-khalq, 123, 147, and 148; Muslim ibn al-ajjj, Fail al-Qurn, 59; al-Suy, al-
Itqn, p. 165; al-Nawaw (Tahdhb al-asm), edited by Wstenfeld, p. 267: Ubayy, Mudh,
Abd Allh b. Masd, and Slim b. Maqil, clients of AB HUDHAYFAH (Ms b. Masd,
[d. 240/854; F. Sezgin, Geschichte des arabischen Schrifttums, vol. 1, p. 41.]) ibid.Ibn Sad,
ibid., p. 113, l 11 sq.: Ubayy, Mudh, Zayd, Ab Zayd, and Tamm (five men)Ibn Sad, ibid.,
p. 113, l 20sq., and l 114, l 1 sqq.: Ubayy, Mudh, Ubda b. al-mit, Ab Ayyb, Ab l-
Dard (thus, also five.)Ibn Sad, ibid., p. 113, l 1 sqq.: Ubayy, Mudh, Zayd, Ab Zayd
al-Anr Sad b. Aws b. Thbit [F. Sezgin, Geschichte, vol. 9, pp. 6768], Ab l-Dard,
Sad b. Ubayd (six men).Ibn Sad, ibid., p. 113, l 20sqq.: the same men, and Mujammi b.
Jriya.Ibn Sad, ibid., p. 113, l 23sqq.: Ubayy, Mudh, Zayd, Ab Zayd, Uthmn, Tamm
al-Dr.al-Suy, al-Itqn, p. 169, l 13sqq.: Ubayy, Mudh, Zayd, Ab Zayd, Ab l-Dard,
Mujamm.Ibn al-Nadm, al-Fihrist, ed. Flgel, p. 27: Ubayy, Mudh, Ab Zayd, Ab al-
Dard, Sad b. Ubayd, Al b. Ab lib, Ubayd b. Muwiya b. Zayd b. Thbit b. al-ak
(seven men).
6
EI
2
; EQ; Juynboll, Encyclopedia, p. 22, col. 2, p. 40, col. 1.
7
EI
2
; EQ; Juynboll, Encyclopedia, p. 35sqq.
the indirect collectors of the koran 215
MASD, Slim b. Maqil,
8
Ubda b. al-mit,
9
Ab Ayyb, Sad b. Ubayd,
Mujammi b. Jriya,
10
Ubayd b. Muwiya (Ibn Zayd b. Thbit b. al-ak),
and Al b. Ab lib.
From among these men we shall later meet again Al b. Ab lib, Slim,
Zayd b. Thbit, Ubayy b. Kab, and Abd Allh IBNMASDas alleged or true
editors of written collections of the Koran.
8
EQ: died 12/634.
9
EI
2
; Juynboll, Encyclopedia, p. 38, col. 2, p. 419, col. 2.
10
EQ; Juynboll, ibid.
POPULAR ACQUAINTANCE
WITH THE KORAN UNDER THE FIRST CALIPHS
As far as the Koran is concerned, the ignorance of the average believer in [ii/7]
the early years of Islam was beyond imagination. Literature contains many
drastic pieces of evidence. After the Battle of al-Qdisiyya, not far from
al-ra (637), Umar b. al-Khab ordered the commander-in-chief, Sad b.
Ab Waqq,
1
to distribute the spoils among the Koran reciters (amalat
al-Qurn). When the famous warrior, Amr b. Madikarib,
2
appeared before
him and was questioned about his acquaintance with the revelation he
apologized with the words: I embraced Islam in the Yemen, and spent the
rest of my life as a soldier andtherefore hadno time to memorize the Koran.
Bishr b. Raba of al-if, whomSad b. Ab Waqqs asked the same question,
replied with the well-known phrase, bismillh al-Ramn al-Ram.
3
When
in the Battle of Yamma the Anr were addressed by their leader with the
honorific name, People of the Sra of the Cow, a ayyi warrior asserted
that he had not even memorized a single verse of this sra.
4
Aws b. Khlid,
a respected Bedouin of the Ban ayyi, was once so maltreated by the
commissioner of Caliph Umar for not remembering even a single passage
from the Koran that he died.
5
Indeed, even under the Umayyads a khab at
Kufa allegedly went up to the pulpit and recited a part of the dwn of AD b.
Zayd (IBN AL-RIQ),
6
thinking it was a verse from the Koran.
7
Even if these
stories are nothing but anecdotes they certainly give an accurate picture of
how versed in the Koran the Bedouin soldiery of the nascent Islam really
were. And strange characters like this preacher might have still appeared at
much later times.
1
EI
2
; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 8, p. 171.
2
EI
2
3
Ibn ajar al-Asqaln, vol. 1, no. 764; Ab l-Faraj al-Ifahn, al-Aghn, vol. 14, p. 40.
4
Abd al-Ramn b. Muammad IBN UBAYSH; Sezgin, GAS, vol. 1, p. 295, last line;
[al-Ghazawt al-damina al-kfila, Ms. Berlin, cod. Wetzstein, no. 173 = Wilh. Ahlwardt, Verze-
ichnis der arabischenHandschriften, no. 9689], fol. 13; [Beirut editionby Suhayl Zakkr, 1992].
5
Ab Tammmabb b. Aws, al-amsa, vol. 1, p. 389; Ab l-Faraj al-Ifahn, al-Aghn,
vol. 16, p. 58.
6
7
al-Fihrist, Flgels edition, p. 91.
COLLECTING AND EDITING
Al as a Collector of the Koran
A variety of traditions mention Al b. Ab lib, a cousin and son-in-law [ii/8]
of Muammad, as the author of a collection of qurns. According to one
tradition, he started his collection while the Prophet was still alive, and
on the latters explicit order. It is said that Al collected the Koran from
leaves, silk rags, and scraps of paper that he discovered under the Prophets
pillow, immediately vowing not to leave the house before he completed the
task.
1
Others move the event to the time immediately after Muammads
death, and suspect Al to have used this vow as an excuse to delay homage
toAbBakr.
2
It is alsosaidthat inthe face of Muammads deathAl realized
the fickleness of man and was determined to complete the writing in three
days.
3
The author of the Fihrist even claims to have seen a fragment of the
original. The fact of the matter is that there is absolutely no truth to this
claim.
Even the sources of these accountsShite commentaries on the Koran, [ii/9]
and Sunnite historical works with Shite influenceare suspect, since
everything that Shites say about the most saintly man of their sect must be
considered a priori a tendentious fabrication. The content of these reports
contradicts all sound facts of history. Neither the traditions regarding Zayd
b. Thbits collection of the Koran nor those about other pre-Uthmnic col-
lections knowanything of an analogous work by Al. He himself never refers
1
Cf. the Shiite commentaries, Tafsr Al b. Ibrhm in: W. Ahlwardt, Verzeichnis der
arabischen Handschriften no. 929 = cod. Sprenger, no. 406; Muammad b. Murtad, K. al-f
(f tafsr al-Qurn) in: W. Ahlwardt, ibid., no. 899 =cod. 1 Petermann, no. 553; Mirza Alexander
Kazem-Beg, Observations sur le Chapitre inconnu du Coran, publi et traduit par Garcin de
Tassy, p. 386. All these traditions go back to Als family, thus making them for us so much
more suspectable.
2
Ibn Sad (al-abaqt), vol. 2, part 2: Letzte Krankheit, Tod und Bestattung Muhammeds,
p. 101, ll 1620; al-Suy, al-Itqn, p. 134sq. Both sources, however, challenge the credibility
of the information. In Ibn Sads work Ikrima when questioned replies that he does not
know anything about it. al-Suy quotes Ibn ajar al-Asqalns remark that jamaa in that
tradition means to remember.
3
al-Fihrist, edited by Flgel, p. 28.
220 collecting and editing
to his own collection, neither during his caliphate nor before, and it is
certain that the Shites were never in possession of such a document.
4
According to al-Yaqb,
5
the arrangement of the sras in Als collection [ii/10]
of the Koran established shortly after the death of Muammad is said to be
as follows: 2, 12, 29, 30, 31, 41, 51, 76, 32, 79, 81, 82, 84, 87, 98 (first section.)3,
11, 22, 15, 33, 44, 55, 69, 70, 80, 91, 97, 99, 104, 105, 106 (second section.)4, 16,
23, 36, 42, 56, 67, 74, 107, 111, 112, 103, 101, 85, 95, 27 (third section.)5, 10, 19,
26, 43, 49, 50, 54, 60, 86, 90, 94, 100, 108, 109 (fourth section.)6, 17, 21, 25,
28, 40, 58, 59, 62, 63, 68, 71, 72, 77, 93, 102 (fifth section.)7, 14, 18, 24, 38, 39,
45, 98, 57, 73, 75, 78, 88, 89, 92, 110 (sixth section.)8, 9, 20, 35, 37, 46, 48, 52,
53, 61, 64, 65, 83, 113, 114 (seventh section).
Although some sras are accidentally omitted in the manuscript trans-
mission (sras 1, 13, 34, 47, and 107), the method of arrangement is quite
clear. It is based on a particular combination of the canonical edition with
the sections or reading parts (ajz, sing. juz). Whereas in other cases these
sections represent incisions in the text according to the transmitted order,
here, in each of the seven sections a fixed number (1617) of select sras is
united. Yet this selection is not entirely arbitrary, since every section regu-
larly begins with a lower numbered sra (27), according to the authorized
order of sras, and then on through the different decimalswith minor
exceptions that are themselves probably subject to textual corruptions
andthencontinues to the highnumbers, so that every sectionoffers insome
measure a cross section of the entire Koran.
If thus the arrangement of the sras testifies to a dependence on the
Uthmnic recension, so muchmore does this apply to the later divisioninto
reading portions, which did not appear until the Umayyad period.
According to yet another equally untenable account,
6
the arrangement of [ii/11]
the six first sras of the Alid Koran was as follows: 96, 74, 68, 73, 111, 81.
The Collection of Slimb. Maqil
Another collection that, as it seems, allegedly appeared immediately after
the death of Muammad is ascribed to Slim b. Maqil,
7
a client of Ab
4
For details see below, Sectarian Reproach to the Uthmnic Text, pp. 246266.
5
al-Yaqb, Historiae, ed. by M.Th. Houtsma, vol. 2, pp. 152154. In the known manu-
scripts of al-Fihrist the list of sras of the Alid Koran is omitted.
6
See below, p. 251, Sprenger, Leben und Lehre des Mohammad, vol. 3, p. xliv.
7
EQ.
collecting and editing 221
udhayfa.
8
When he set out on the project he vowed, like Al, not to leave
the house before the task had been accomplished. Afterwards there was a
discussion as to the name of the collection. Some men suggested sifr but
Slim b. Maqil rejected this proposal because this was the basmala of the
Jews; preference should be given to muaf, which he had come to know
in Abyssinia in a similar meaning. Thus the matter was decided. Al-Suy
records yet another tradition in the same context, according to which Slim
was among those who, at the command of Ab Bakr, undertook the col-
lection of the Koran. This tradition contradictsas shall be demonstrated
laterall sound facts of the history of the Koran. Al-Suy, therefore, right-
fully classifies the matter among curiosities (gharb).
8
al-Suy, al-Itqn, p. 135; Sprenger, Leben und die Lehre des Mohammad, vol. 3, p. xliv.
THE (FIRST) COLLECTION OF ZAYD B. THBIT
The Prevailing Tradition
There is a long tradition
1
about this collection going back to Zayd himself. In [ii/11]
spite of its wide dissemination, it has undergone relatively few changes.
2
Its
content is as follows: during the war against the prophet Maslama, particu-
larly in the decisive Battle of Aqrab in al-Yammain 11/632 or 12/633
3
the death toll was particularly high among the reciters of the Koran.
4
It
was for this reason that Umar b. al-Khab feared that a large part of the
Koran could be lost should all these men die in battle.
5
He then advised the
1
Muammad b. Muslim AL-ZUHR (d. 124/741) from Ubayd b. al-Sabbq [EQ; Juynboll,
Encyclopedia, p. 421 sqq.] (Fihrist wrong, Salaf) from Zayd.
2
Izz al-Dn IBNAL-ATHR, Chronicon, vol. 2, p. 279, vol. 3, p. 86; Ibn al-Nadm, al-Fihrist,
p. 24; Ab l-Fid, Annales moslemici, ed. J. Reiske, vol. 1, p. 212; al-Yaqb, Historiae, ed.
M.Th. Houtsma, vol. 2, p. 154; al-Bukhr and al-Tirmidh in Tafsr on sra 9:129sq.; al-abar,
Tafsr; al-Bukhr, Fail al-Qurn 3, Akm 37; al-Khab al-Tibrz, Mishct, Fail
al-Qurn, fal 3; al-Mabn li-nam al-man, part ii; al-Dn, al-Muqni f marifat rasm
maif al-amr, Ms. Sprenger, fol. 2
r
sq.; Silvestre de Sacy, Commentaire sur le pome
nomm Rayya, p. 343sq.; commentary on the Aqla in the Mmoires de l Acadmie des
inscriptions, vol. 50 (1808), p. 421; al-Qurub, Jmi akm al-Qurn, Ms. fol. 19; al-Suy,
al-Itqn, pp. 133sq., and 138.
3
Fighting took place probably during the last months of 11/632, and the first months of
12/633. Cf. Caetani, Annali, vol. 2, p. 724, andhis ChronographiaIslamica, fasc. 1, pp. 112 and121.
Conversely, the majority of the sources do not supply any particular year for the collecting.
4
Most of the sources listed in note 2 above call the men who knew by heart large parts
of the Koran qurr reader. Some, like al-Yaqb (otherwise al-Aghn, vol. 14, p. 40, l 18, and
al-abari, vol. 1, p. 1940, l 2) use the expression amalat al-Qurn, which is normally trans-
lated by bearer of the Koran. The actual meaning is obscure since there is nothing to lead
over from Arabic amala to carry, bear either to the meaning to commit to memory or
to the meaning to transmit, derived from expressions like amalat al-adth (al-Nawaw,
p. 63, according to M.J. de Goeje in the gloss, al-abar) or amala adthan an (al-Mizz,
cod. Landberg, no. 40, [sic] according to Eduard Sachau onIbnSad (al-abaqt): Biographien
der mekkanischen Kmpfer, vol. 3, part 1, p. 453, l 3), and ilman an (al-Dhahab, Tadhkirat al-
uff, Hyderabad ed., vol. 1, p. 37, l 5sq.) respectively. For this reason we are probably dealing
with the mechanical transfer of a foreign idiom. Since there is nothing corresponding avail-
able in either the Judeo-Aramaic or the South Arabian-Abyssinian lexicon, there remains
only Middle Persian. But which meaning is intended when rendering harbidhat, namely
the Arabic plural of Persian hrbu [ ] priest (Avestic athrupatai head of school) by
amalat al-dn, is also questionable.
5
224 the (first) collection of zayd b. thbit
caliph to issue an order that the Koranic fragments should be collected. At
first, Caliph Ab Bakr was hesitant to embark upon a task that the Prophet
had not authorized, but he eventually consented and commissioned Zayd
b. Thbit, an intelligent young man who had already been called upon by
the Prophet to write down certain passages of the revelations.
6
After some
hesitation Zayd accepted, although he thought it would be easier to move
a mountain. He collected the Koran from pieces of scrap papyrus or parch-
ment,
7
thin whitish stones,
8
palm branches,
9
shoulder-blades,
10
ribs,
11
pieces
of tanned skin,
12
and small boards.
13
Tradition lists as the final source the
6
His vita will be presentedlater inthe chapter onthe members of the KoranCommission [ii/13]
appointed by Uthmn. [Cf. A. Jeffery, Materials for the history of the Qurn, pp. 223224.]
7
, Izz al-Dn IBN AL-ATHR, Chronicon, vol. 3, p. 86; Ibn al-Nadm, al-Fihrist, ed. by
G. Flgel, p. 24; Ibn Khaldn, al-Ibar wa-l-khabar f ayym al-Arab, vol. 2, appendix ( ),
p. 136; al-Bukhr, Akm, 37, and Tafsr on sra 9:129; al-Mabn li-nam al-man, Ms.
fol. 6; al-Dn, al-Muqni, Ms. fol. 2 b; al-Qurub, Ms. fol. 18 b; Ab Muammad MAKK IBN
AB LIB, Kashf an wujh, p. 502; al-Naysbr in al-abar, Tafsr, vol. 1, p. 23; Al al-Dn,
vol. 1, p. 6. These scraps, according to al-Suy, al-Itqn, p. 137, were made of papyrus or
parchment. L. Caetani, Annali, vol. 2, p. 711, deals with the question, and thinks that at the
time the latter material was more common in Arabia. Ab l-Fid, vol. 1, 212, outright uses the
expression .
8
Fihrist, p. 24; al-Bukhr, Akm, 37, Fail 3; al-Tirmidh on sra 9:129; al-
Khab al-Tibrz, Mishct, Fail, fal 3; al-Itqn, pp. 134 and 137; Muqni, fol. 2 b; Ibn Aiyya,
fol. 25; al-Nsbr; in Al al-Dn, vol. 1, p. 6, the curious scholiumhas
, cf. also Ibn Aiyya, al-Jmi al-muarrar, fol. 25 .
9
; cf. the sources mentioned in notes 6 and 7; Ab l-Fid, vol. 1, p. 212, reads
; Ab Muammad MAKK IBN AB LIB b. Mukhtr al-Qays, Kashf, [Ahlwardt,
Verzeichnis, Ms. no. 578,] p. 502, . The use of this writing material in pre-Islamic Arabia
is documented in Imru al-Qays, Wilh. Ahlwardt, The Divans of the six ancient Arabic poets,
no. 63, 1; Labd b. Raba, Der Dwn des Lebd, ed. J. Chlid, p. 61; al-Sukkar, Poems of the
Huzailis, ed. Kosegarten, no. 3, l 7, and al-Fihrist, p. 21. Muammad used this writing material
for a letter to the [Ban] Udhr, in Wellhausen, Skizzen und Vorarbeiten, Heft 4, no. 60;
al-Wqid, ed. by Wellhausen, p. 388. Cf. also G. Jacob, Studien in arabischen Dichtern, vol. 3,
p. 162.
10

, al-Bukhr in the Tafsr; al-Suy, al-Itqn, p. 137; al-abar, Tafsr, vol. 1, p. 20; Ab
Muammad MAKK IBN AB LIB, al-Kashf, p. 503. Cf. Fihrist, p. 21; al-abar, vol. 1, p. 1806,
l 15sqq. = Ibn Sad, Letzte Krankheit, Tod und Bestattung Muhammads [Muammads last
illness, death, and funeral], p. 37, l 6sqq.; Musnad Amad IBNANBAL, vol. 1, p. 355; Goldzi-
hers review, C.H. Becker, Papyri Schott-Reinhardt, col. 2250. Shoulder blades of camels were
still in use in East Africa by the Suahelis until most recent times, particularly in elementary
schools, cf. E. Ruete, Memoiren, 4th ed., 1886, vol. 1, p. 90.
11
, al-Itqn, p. 137; al-Dn, Muqni, fol. 2 b; Carl G. Bttner, Suaheli-Schriftstcke in
arabischer Schrift, p. 189, refers to thigh-bones of camels as being used still in East Africa.
12
, al-Suy, al-Itqn, p. 137; Ab Muammad MAKK IBN AB LIB al-Qays,
loc. cit., p. 17. Muammad made use of this writing material in his missives, cf. Wellhausen,
Skizzen und Vorarbeiten, Heft 4, nos. 48 and 52; al-Waqid, in Wellhausen, Muhammed in
Medina, p. 388; cf. also G. Jacob, Das Leben der vorislamischen Beduinen, p. 162.
13
the (first) collection of zayd b. thbit 225
hearts of men,
14
namely that Zayd b. Thbit supplemented his archival
researches by interviews with men who had memorized Koranic passages.
Finally, it is reported that he chanced uponsra 9:129sq. withKhuzayma
15
or
with Ab Khuzayma (Ibn Aws b. Zayd)
16
fromMedina. The individual pieces
Zayd b. Thbit wrote down on uniformleaves
17
and handed themover to the
caliph. After his death the copy descended to his successor Umar, who in
turn bequeathed it to afa, Muammads widow.
14
( ) : the previously cited sources, and Ab l-Fid, Annales, ed. Reiske,
vol. 1, p. 212, .
15
al-Tirmidh inthe tafsr onsra 9:129; al-Mabn li-namal-man, fol. 6; Ab Muam-
mad MAKK BN AB LIB b. Mukhtr al-Qays, loc. cit., vol. 2, p. 302; al-Muttaq, Kanz
al-umml, vol. 1, nos. 4759 and 4767; al-Tirmidh knows the full name of this man, Khuza-
yma b. Thbit, but although the name is found in biographical dictionaries, he is not brought
in connection with the collection of the Koran. [Gautier Juynboll, Encyclopedia, pp. 185, col. 1,
and 696, col. 1.]
16
Fihrist, p. 24; al-Suy, al-Itqn, pp. 134 and 136; al-Bukhr, Fail al-Qurn, 3; al-
Khab al-Tibrz, Mishct, Fail al-Qurn, fal 3. The man is commonly identified as an
Anri and therefore allegedly identical with Ab Khuzayma b. Aws b. Zayd (Ibn Sad (al-
abaqt) Biographien der medinischen Kmpfer (vol. 3, part 2), p. ; Ibn al-Athr, Usd al-
ghba, vol. 5, p. 180). The designation as Ibn Thbit (al-Itqn, p. 136, end) is based on a
confusion with the above-mentioned Khuzayma [EI
2
; Juynboll, Encyclopedia, p. 185, col. 1,
p. 696, col. 1.]
The variants, Khuzayma and Ab Khuzayma, appear side by side in al-Bukhr, tafsr on
sra 9:129, Akm, 37, al-Dn, Muqni, fol. 6, al-Qurub, fol. 18 b, and Al al-Dn, vol. 1, p. 6.
People attemptedto explainthe variants indifferent ways: e.g., that inthe first collectionsra
9:129 was found at Ab Khuzaymas quarters, but later in the Uthmnic recension, sra 33:23
was foundat Khuzaymas quarters (al-Qurub, fol. 20
v
, andal-Bukhr, Fail al-Qurn, etc.).
Still other combinations are offered by al-Qurub, in al-Muqni, and Ibn Aiyya al-abar in
his Tafsr, vol. 1, p. 21, has those two Koranic passages found on two different men named
Khuzayma, andhe dates the event inthe rule of Uthmn. According toIbnajar al-Asqaln,
vol. 1, no. 1395, Ibn al-Athr, Usd al-ghba, vol. 1, p. 326, and Ibn Aiyya, fol. 26
v
, sra 9:129sq.
was discoveredunder the rule of AbBakr onrithb. Khazma. InIbnal-Athr, Usdal-ghba,
vol. 5, p. 180, it is merely stated that the names indicate different men, with no more in
common than their Anr origin.
Less frequently we find the statement that the missing verse was sra 33:23, e.g., al-Dn,
al-Muqni, fol. 2 b, and al-Qurub, fol. 18 b. In al-Mabn li-nam al-manwhere nearly
everything that al-abar, Tafsr, vol. 1, p. 20, said about Uthmans recension is applied to the
one by Ab Bakrsra 33:23 is missing at first examination, and sra 9:129 at the second
examination. According to al-Suy, al-Itqn, p. 143, Umar finds the end of sra 9 (verses 129
and 130) on rith b. Khuzayma. The opinion that these verses are the last of the Koran to
have been revealed is naturally somehow connected with the fact that they were discovered
rather late as mentioned earlier. However, we cannot say for certain what was the causal
relation of the two opinions. As I emphasized above, p. 182sq., in any case, this chronological
arrangement must be considered unfounded. Conversely, William Muir (Life of Mahomet,
vol. 1, p. xxvi) contests that the verses hadbeendiscoveredsolate; as the last of the revelations
they must have been familiar to all men.
17
uuf; cf. on this below, p. 232sq.
Divergent Traditions
Whereas Umar b. al-Khab appears in the standard transmission only as [ii/15]
the intellectual author of the first collection, it is Ab Bakr in his capac-
ity as the ruling caliph who is responsible for the realization of the plan,
who appoints the technical director, and who takes charge of the project.
However, there is yet another tradition that, in so far as the brief text per-
mits, completely disregards the first caliph and attributes all the above-
mentioned functions to the initiative of his energetic successor. The words
of the tradition,
18
Umar was the first tocollect the Koranonleaves, possibly
include the wider sense that not only did the completion of the enterprise
fall into the reign of this caliph, but also its beginning. On the other hand,
the remark that Umar died even before he collected the Koran
19
is a refer-
ence to the final canonical recension that he is supposed to have already
contemplated.
20
At other places we learn various details about the way Umar proceeded
with his first collection. A later source cites as a reason for its collection
that Umar, inquiring about a certain Koranic verse, was told that the man
who knew it by heart had been killed in the Battle of Yamma.
21
Further-
more, it is said, e.g., that he included only such passages that had been
attested by two witnesses.
22
Also traditions relating to the Verse of Ston-
ing
23
seemto assume that Umar was involved in the collection of the Koran.
As some reports point out,
24
he fearedthat one day believers wouldpainfully
miss the verse if they did not find it in the Book of God.
25
According to
18
IbnSad(al-abaqt): Biographiender mekkanischenKmpfer (vol. 3, part 1), p. 252, l 8sq.
When al-Suy in al-Itqn, p. 135, interprets the verb jamaa collected as ashra bi-jamih
he advised him to collect, this, I think, is harmonistic fancy.
19
Ibid., p. 212, l 4.
20
al-Suy, al-Itqn, p. 430; cf. below, p. 255sq.
21
al-Itqn, p. 135.
22
al-Itqn, p. 136, appendix (from Yay b. Abd al-Ramn b. ib (d. 104/722723)),
[Juynboll, Encyclopedia, p. 14, col. 2]. The Shiite commentary (by Muammad b. Murta,
K. al-f), Ahlwardt, Verzeichnis, no. 899, is trying to follow from this the imperfection of
Umars Koran because among the passages that could not be endorsed by two witnesses
there must have certainly been authentic passages.
23
24
al-abar, vol. 1, p. 1821; Ibn Hishm, p. 1015; al-Tirmidh, udd, 6; al-Khab al-Tibrz,
Mishct, udd, beginning; al-Mabn li-namal-man, sections 2 and 5.
25
Ibn hir al-Baghdd, al-Nsikh (Ahlwardt, Verzeichnis, no. 478) quotes Umar: If I did
not fear to be suspected of interpolation, I would have written the Verse of Stoning in the
margin (shiyat al-muaf) of the codex.
others,
26
he freely admits that he did not include the verse, not wanting to
be reproached for an addition to the revelations. According to al-Suy,
al-Itqn, p. 137, he made the decision because he could not find the two
required witnesses.
27
All these traditions are based on the opinion that
this verse is part of the revelation. Even if this is a mistake, as I tried to
prove earlier,
28
it is still difficult to believe that a man like Umar stubbornly
defended its authenticity.
A third group of traditions
29
attempts to harmonize the first and second [ii/17]
groups. At the behest of Ab Bakr, Zayd b. Thbit then wrote down the
revelations on pieces of leather, shoulder blades, and palm branches. After
the death of the Caliph, namely under Umar, he copied these texts on a
single sheet,
30
the size of which unfortunately remains a mystery.
Finally, a strange story
31
must be recalled. According to this, Ab Bakr
refused to collect the Koran, since Muammad did not also do this. There-
upon Umar took over and had it copied on leaves. He then commissioned
twenty-five Quraysh and fifty Anrs to copy the Koran and submitted the
work to Sad b. al-.
32
It is obvious that in this case the traditions regard-
ing the first collection of the Koran and the canonical edition are mixed up.
Such a large number of collaborators is nowhere else mentioned for the first
collection. Sad was only a child of eleven when Umar became caliph. Nei-
ther al-Yaqb nor any of his sources can realistically be held accountable
for this hopeless confusion; this must be due to a lacuna in the manuscript
used by the editor.
33
When emphasizing Zayd b. Thbits youth and intelligence, as well as
his former activity as special amanuensis to Muammad,
34
all our sources
35
26
Ibn Sad (al-abaqt): Biographien der mekkanischen Kmpfer, p. 242; al-Yaqb, Histo-
riae, vol. 2, p. 184; al-Mabn li-namal-man, vol. 2; al-Muwaa, p. 349 (udd, 1, end).
27
There is the unique report in al-Suy, al-Itqn, p. 528, that the controversy over the
inclusion of the Verse of Stoning arose during the canonical edition.
28
Above, pp. 198202.
29
al-abar, Tafsr, vol. 1, p. 90; al-Suy, al-Itqn, p. 138; Ab Muammad MAKK IBN
AB LIB b. Mukhtr al-Qays, al-Kashf an wujh al-qirat wa-l-ilalh wa-l-hajajh in Wilh.
Ahlwardt, Verzeichnis der arabischen Handschriften, 578 (= cod. 1 Petermann, no. 17, p. 503).
30
f afa wida.
31
al-Yaqb, Historiae, vol. 2, p. 152. Cf. G. Juynboll, Encyclopedia, p. 232, col. 1.
32
Juynboll, Encyclopedia, p. 232; Sezgin, GAS, vol. 1, p. 3, and vol. 9, p. 6.
33
This lacuna would have to be assumed before wa-ajlasa, p. 152, l 15.
34
The chapter on Muammads correspondence and the embassies to him in the Sra
of Ibn Sad lists fourteen men who served as the Prophets secretaries, but Zayd is not
mentioned.
35
Cf. above, p. 223 n. 2.
agree that Zaydwas particularly suitedtocollecting the revelations. Scholars
usually give as the first reason the greater willingness of a young man to
comply with the orders of the caliph than would have been the case with
an old, obstinate servant. On the other hand, the sources cited say nothing
about Zayds skill to learnthe Koranby heart, although this ability is inother
cases frequently mentioned.
36
The reports regarding Zayd b. Thbits working procedure tacitly presup- [ii/18]
pose that he generally followed the originals at his disposal. Still, his treat-
ment of the last verses of sra nine, [Repentance,] nos. 129 and 130, which he
simply appended to a large sra, demonstrates that he occasionally did not
shun arbitrary decisions. Zayd or Umar allegedly said on this occasion that
if this part had consisted of three instead of only two verses, he would have
created a separate sra.
37
Criticismof the Traditions
As we can see, Muslims hold three different views regarding the develop-
ment of the first collection of the Koran. According to the first point of view,
the standard tradition, this took place under the rule of Ab Bakr; accord-
ing to the second point of view, this happened during the rule of Umar; and
according to the third point of view, the actual work commenced under Ab
Bakr but was completedonly under his successor. As there is noclear answer
to the question as to which one of them is to be preferred, a complicated
investigation is necessary.
The standard tradition is made up of different elements that either con-
tradict one another or other historical reports:
1. Ab Bakr, indeed, organized the initial collection, but the intellectual
author, and the actual driving force, was Umar.
2. The occasion of the Battle of Yamma, the solemn motive to save the [ii/19]
word of God from decline, the participation of the ruling caliph as
well as the most powerful man in the theocracy of the timeall these
circumstances together give the collection the character of a basic
36
A.I. Silvestre de Sacy, Notice du manuscrit arabe al-Muqni f l-marifa , p. 305; Ibn
al-Athr, Usd al-ghba; al-Nawaw (akhadha l-Qurn); al-Dhahab, Tadhkirat al-uff, vol. 1,
p. 26 (afia l-Qurn). It is not clear what Ibn Sad, [al-abaqt al-kabr], vol. 2, part 2: (Letzte
Krankheit, Tod und Bestattung Muammads p. 116, l 9) has in mind when he is praising Zayds
qira.
37
In al-Itqn, p. 143, l 3, it should read instead .
fundamental work for religion and state. It was therefore only to be
expected that after the death of Ab Bakr the project was not handed
over to his family but to Umar, his successor in office.
3. Umars bequest of the collection to his daughter afa, however,
obliges one to conclude that it did not constitute the property of either
the community or the state but was private ownership. Adocument of
official or public character couldnot have beenbequeathed, least of all
to a woman, even if she was the widowof the Prophet, but it belonged
to the succeeding caliph.
In favour of the private character of the collection speaks the fact that after
the great conquests it was not usedinany foreignprovince, while, as we shall
see later, the editions of Abd Allh IBNMASDand Ubayy b. Kab attained
this success, even without such a high patronage.
4. Ab Bakrs brief term of office of two years and two months
38
is, in the
eyes of tradition, rather short for the difficult process of collecting the
dispersedtexts. Somuchmore sosince the project was not starteduntil
after the Battle of Yamma,
39
so that there remained only the time of
some fifteen-odd months.
5. The connectionof the collectionand the campaignof al-Yamma rests [ii/20]
on a very weak foundation. L. Caetani
40
points out that in the registers
of Muslim casualties in the Battle of Aqrab there are but few men
with profound knowledge of Koranic matters; nearly all of them were
recent converts. The contention that many Koran reciters were killed
in the encounter, and that Ab Bakr feared that a great part of the
Koran might get lost, therefore cannot possibly be true.
41
Little can be
said against this opinion, provided that Caetanis list
42
of 151 casualties
is correct and that we are in the possession of nearly comprehensive
information of the Koran reciters at that time.
In the reports on casualties accessible to me there are actually only two
men whose knowledge of the Koran is explicitly attested.
43
Abd Allh b.
38
From 13th Rab II 11/8 June 632 to 21st Jumd 13/22 August 634.
39
The campaign occurred probably in the three final months of 11/632 and the first three
months of 12/633. Cf. above, p. 217.
40
Caetani, Annali, vol. 2, 2, n. 1.
41
42
Annali, vol. 2, pp. 739754.
43
According to a legendary note in al-Muttaq, Kanz al-umml, vol. 1, no. 4770, even four
hundred Koran reciters are supposed to have been killed. Cf. Juynboll, Encyclopedia, p. 695,
col. 1.
af b. Ghnim
44
and Slim(Ibn Maqil), client of Ab udhayfa,
45
who after
him carried the banner of the muhjirs. If the words of Ab udhayfa: O,
people of the Koran, decorate the Koran with deeds,
46
are indeed authentic,
they would suggest a considerable number of such men among the Muslim
combatants.
But even if the contradictions pointed out by Caetani were to disappear,
the traditional connection of the collection with this military campaign still
cannot be maintained because, as the additional report states in cryptic lan-
guage, the collection is based almost exclusively on written sources. There
can be no doubt in this matter since we knowthat Muammad himself had
arranged for a written copy of the revelations.
47
In such circumstances, the
death of any odd number of reciters of the Koran did not jeopardize the sur-
vival of the Prophets revelations.
The subject of the traditions offers no additional means to discover the [ii/21]
historical truth in this chaos of contradictions and errors. We must conse-
quently try to find clues from the form of the tradition and extract an older
core by way of literary analysis. The overwhelming majority of the individ-
ual pieces of evidence suggests recognizing the collection of the Koran as an
affair of the state. The only aspect touching private law, namely the passing
of Umars leaves to the property of his daughter afa, can easily be elimi-
nated from the text. It thus seems that there cannot be any doubt that the
other view is older and more relevant.
Nevertheless, this solution, which in itself is simple and obvious, must be
considered wrong. It is precisely the fact that after the death of Umar the
collection was the property of afa bt. Umar that is the most reliable of
the entire report, as it is confirmed by reports about the canonical edition of
the Koran. There it is said about Uthmn that he had the leaves brought
from afas and used them as the basis of his recension. This is the most
solid point from which we must turn our gaze backwards. Although the
reports of the two versions of the Koran are nowmostly connected with one
another, nevertheless, in the older sources, each one has its own isnd and
consequently its own literary position.
44
In al-abar, vol. 1, p. 1940, 1945, and Ibn al-Athr (al-Kmil), vol. 2, p. 276, he is called
mil al-Qurn.
45
Loc. cit. he is called mil al-Qurn or ib Qurn, but without mention of his death.
His death in the battle, however, is in other sources frequently attested: al-Baldhur, Liber
expugnationis, p. 90; Ibn Qutayba, p. 139; al-Nawaw; Ibn al-Athr, Usd al-ghba. Cf. also
Caetani, loc. cit., p. 750, no. 113. We have already come across this man, above, p. 209.
46
al-abar, vol. 1, p. 1945, l 2; Ibn al-Athr (al-Kmil), vol. 2, p. 277.
47
See above, p. 217.
For the moment we must postpone the discussion as to how far we can [ii/22]
trust what has beensaidabout the early story of those leaves. Initself, there
is nothing more natural than their passing to afa by way of a bequest.
The affair, however, might possibly be different. If afa was literate,
48
she
might have acquired or commissioned the collection of the Koran for her
own use. If this were not the case, there is more than one reason that might
have prompted one of the most respected women of contemporary Medina
to such an action. If Umar was not the previous owner, his authorship is
also in doubt. It is quite obvious how the error attributing authorship to
Umar could have come about. After the believers had resigned themselves
to the bitter reality that an unable and obnoxious ruler like Uthmn had
become the father of the canonical recension, it might have occurred to
them as a demand of retributive justice to give at least partial credit to his
overshadowing predecessor.
In no circumstances does a path lead from Umar back to Ab Bakr,
so that if any caliph at all can be considered to be the author, it must be
Umar. He is also referred to in the explicit wording of one of the divergent
traditions, and of the principle tradition in so far as it presents this caliph as
the driving force behind the project.
The assumption of Ab Bakrs cooperation is dependent on the premise
that Umars predecessor had in fact been the actual or the alleged author
of the collection. If Umar represented the intellectual greatness of the first
caliphs, Ab Bakr had the advantage of having been one of the first believ-
ers and the closest friend of Muammad. In such circumstances many men
might have been surprised that such a man did not also pursue the collec-
tionof the Koran, andthat this pious wisheventually sublimatedtoa histori-
cal expression. Inthese endeavours perhaps evenisha, Muammads well-
known widowand daughter of Ab Bakr, who was inclined in family politics
to sacrifice truth and honour to her ambitions, might have played a part.
The last of these three Muslim views, which attributes the collection to [ii/23]
the terms of office of the two caliphs, represents an artificial harmonization
of the first and second options, and furthermore makes the enterprise again
more a matter of state whenit obviously hadtobe considereda private affair.
Zayds editorial activity is consistent with all the forms of tradition men-
tioned, and has the additional advantage of not being easily suspected of
tendentious fabrication. It is true that his appointment by Uthmnlacks the
explicit reference that he, after all, is the editor or writer of afas leaves.
48
al-Baldhur, Liber expugnationis, ed. de Goeje, p. 472.
The causal connection of the first collection with the campaign of Ya-
mma has been shown above to be ahistorical. It is futile to look for a dif-
ferent kind of cause, as it is only natural that after the death of Muammad
sooner or later the need would have arisen to have a reliable copy of the
revelations, the most valuable legacy the Messenger of God had left his com-
munity of the faithful. Least of all, one shouldthink, woulda competent man
such as Zayd need inducement for a project whose usefulness and expedi-
ency was so obvious.
Formand Content of the First Collection
The state of the written documents of the Koran after the death of Muam-
mad is fairly depressing. They are not only supposed to have been scattered
and in disorder but also written on at least half a dozen different materials.
On the other hand, there is suspicion that tradition is greatly exaggerating
this either toenhance the merit of the collectors or toillustrate emphatically
the touching simplicity of the past.
49
Harmless passages in Ibn Sads
50
biog-
raphy of the Prophet make it certain that at that time letters were written on
palmbranches and pieces of leather. It is not far-fetched to assume that they
attempted to use uniform material for higher literary purposes. This would
apply particularly in this case, since it concerned texts of divine origin and,
as we have shown above,
51
not only smaller individual revelations had to be
taken down but also long sras.
The word uuf
52
leaves, by which Zayds collection is identified, very [ii/24]
likely suggests that it refers to uniform material and size. Of the various
49
Cf. also A. Sprenger, Leben und die Lehre des Mohammad, vol. 3, p. xxxix.
50
J. Wellhausen, Seine Schreiben und die Gesandtschaften [Muammads correspondence
and embassies], 48, 52, 60.
51
See above, p. 210sq.
52
The singular of uuf, afa, is an Arabic neologismrespectively formed after Abyssin-
ian and South Arabian aafa to write, meaning written upon. The word has been used
already in pre-Islamic poetry, The Poems of the Huzailis, edited by Kosegarten, no. 3, l 6, al-
Mutalammis, Gedichte, nos. 2, l 2, and 9, l 6, Labd b. Raba, Die Gedichte, no. 47, l 1, al-Aghn,
vol. 20, p. 24, l 30, Aws b. ajar, p. 29, l 9. As already the Arabs themselves have noted(al-Suy,
al-Itqn, pp. 120 and 135) maaf, or more frequently muaf, is a loan word also according
to its form which, however, is rarely found in pre-Islamic poetry (Ahlwardt, The Divans of
the six ancient Arabic poets, no. 65, 2). In Ethiopic, maaf serves as the most common des-
ignation for book, codex. Cf. also S. Fraenkel, Die aramischen Fremdwrter im Arabischen,
p. 248, I. Goldziher, Muslimstudies, vol. 1, pp. 106107; Nldeke, Lehnwrter in und aus dem
thiopischen [Borrowings Words in and from Ethiopic], p. 49sq.
writing materials of the allegedliterary bequest of the Prophet this canapply
only to leather. I cannot ascertain that parchment had already been in use
in Arabia of the time. On the other hand, the designation leaves seems to
indicate that the individual parts of the collectionhadnot yet beenas solidly
arranged,
53
as the later canonical editions are named muaf
54
book.
This interpretation, however, is untenable. Even if those uuf had been [ii/25]
loose sheets, the text of each single sheet must have had a defined arrange-
ment. This alone constituted a not insignificant limitation to individual
treatment. After all, such a sheet consisted of at least two pages, and if
folded even four. The fixed layout of the text might amount to several dou-
ble sheets, namely as many as required to make a signature. While one
signature of old Greek Bible codices is commonly made up of three or
four
55
double sheets, the large Kfic parchment codices I examined con-
sisted of three to five double sheets, namely twelve to twenty sheets. Other
methods for the proper collation of the text offered themselves when a
verse broke off at the end of a sheet or signature or when a sheet begin-
ning with a new verse, could be linked with certainty by its content to the
previous sheet. Serious doubts were actually conceivable only when a sig-
nature began with a new verse. But this case was the great exception and
happened even less frequently the larger the signature was; given five signa-
tures, it probably did not even happen once.
56
As can be seen, a (relatively)
53
As evidence of the loose order of the first collection Nldeke refers in the first edition of
this book, Seite 195, to the following tradition in Ibn Aiyya, fol. 25
r
, and al-Qurub, fol. 18
v
:
fa-jamaahghayr murattabal-suwar badtaabshadd, whichhe translates andZaydcollected
the Koran after much labour but without orderly arrangement of the sras. H. Grimme,
Mohammed, vol. 2, p. 13, wants to apply the disarray to the condition of the copies of the
revelation before the first collection. As both opinions are legitimate, I made no use of them
at all, particularly as they are totally unimportant for the present study. Incidentally, it says in
al-Itqn, Cairo ed., vol. 1, p. 60, l 22 (Calcutta, p. 133) = al-Qasalln onal-Bukhr, vol. 7, p. 446,
that the Koran was not collected (majm) during the Prophets lifetime, and was without
definite order of the sras (wa-l murattab al-suwar).
54
Occasionally also the first collectionis called muaf, e.g., IbnSad (al-abaqt): Biogra-
phiender mekkanischenKmpfer, p. 242; al-abar, Tafsr, vol. 1, p. 20; al-Suy, al-Itqn, p. 138.
According to a very strange note in al-Itqn, p. 135, it is Slim (Ibn Maqil), the client of Ab
udhayfacf. about him above, pp. 190 and 201 n. 45who was the first person to collect
the Koran in a muaf.
55
Cf. V. Gardthausen, Griechische Palographie, vol. 1, part 2, p. 158; Th. Birt, Kritik und
Hermeneutik, p. 356. Sections belonging together were either fastened to a common spine or,
if lying loose, one on top of the other, and kept in a case or leather bag; see Gardthausen, loc.
cit., p. 176, and Birt, loc. cit., p. 357.
56
In order to express an opinion I compared the situation in Flgels 1858 printed edition
of the Koran. There it happens thirty-one times that a page begins with a sra. Seventeen
extraordinary accuracy of the arrangement can be achieved even without a
modern mnemonic method like signatures, pagination or catchwords.
57
In such circumstances, the accuracy of the order of the sras on the [ii/26]
leaves of the first collection of the Koran is unlikely to be much different
from the later editions. It is therefore difficult to understand why, as a rule,
this collection is not granted the designation codex or muaf. Whether the
leaves were joined together or not cannot be the criterion, since we do not
even know if the Uthmnic model manuscripts were sewn. As a matter of
fact, it is quite common in the Islamic Orient still to this day to keep even
printed works merely as loose sheets.
58
We cannot approach the question of content and completeness of the [ii/27]
first collection, nor its form, division of sras, their possible separation
by basmala, logograms or other marks, until after an investigation of the
genesis of the other pre-Uthmnic redactions as well as the canonical
recensions.
cases must be excluded as being due to the arbitrary decision never to separate a three-line
sra heading from the beginning of the verse. If from the remaining fourteen cases a sra
heading occurred more than once at the beginning of a section of sixteen pages, it would be
astonishing.
57
In old Greek manuscripts of the Bible not the pages are counted but the signatures. In
the Kfic manuscripts of the Koran which I examined I did not find anything like this, nor
catch-words.
58
To prevent them from falling out of their covers or jackets they are furnished with
claps. In addition they are frequently put into a capsule (mafaa [or mifaa]). Unlike our
books, which stand in shelves, those are lying flat. It may be noted in passing that because
of reverence, a copy of the Koran must not be kept together with other books but put on a
pedestal (kurs).
THE OTHER PRE-UTHMNIC COLLECTIONS
The Personalities of the Editors
From the short period of twenty years between Muammads death and [ii/27]
Uthmns recension we knowof no less than four famous collections of the
Koran, apart from the leaves belonging to afa bt. Umar, for the origin
of which no one else can be held responsible but the person by whose
name they are known. There might have been other editions as well, but
they did not attain the same reputation and therefore disappeared without
leaving a trace. The editors of those four renowned collections are Ubayy
b. Kab,
1
Abd Allh IBN MASD,
2
Ab Ms AL-ASHAR,
3
and Miqdd b.
al-Aswad.
4
As long as nothing definite is known about the literary procedure of
these men, it remains to be seen whether we are dealing with independent
collections of dispersed texts of revelation or whether they were borrowed
from extant collections. We might be best advised generally to call their
works editions.
Ubayy b. Kab,
5
a Medinan of the Khazrajite Najjr clan, was an early [ii/28]
Muslim who fought against the unbelievers at Badr and Uud. He distin-
guished himself in pre-Islamic times by his competence in writing, and he
served as Muammads amanuensis, not only for correspondence
6
but also
for revelations.
7
No wonder that he also made a name for himself as a reciter
1
A. Jeffery, Materials for the history of the text of the Qurn, pp. 114116.
2
Ibid., pp. 2024.
3
Ibid., pp. 209211; EI
2
; EQ.
4
Juynboll, Encyclopedia, p. 714, and his lemma in EI
2
, s.v.
5
Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung Muhammads
[Muammads last illness, death and funeral], p. 103, and ibid., vol. 3, part 2: Biographien der
medinischen Kmpfer Muhammeds, p. 59sqq.; Ibn Qutayba, p. 133sq.; Ibn ajar al-Asqaln,
al-Isba f tamyz al-aba, Ibn al-Athr, Usd al-ghba f marifat al-aba, s.v.; al-Dhahab,
Tadhkirat al-uff, vol. 1, p. 15; al-Nawaw, Tahdhb al-asm. Cf. above, p. 217 n. 5, where he
is reckoned among those who had memorized the entire Koran. EI
2
; Juynboll, Encyclopedia,
s.v.; Sezgin, GAS, vol. 1, s.v.
6
J. Wellhausen, Seine [Muammads] Schreiben und die Gesandtschaften, nos. 17, 18, 25,
42 to 70 = Ibn Sad (al-abaqt): Biographie Muhammeds; Ereignisse seiner medinischen Zeit,
p. 21, l 25 and 27; p. 23, l 27; p. 28, l 2 and 6; p. 35, l 11.
7
Ibn Sad (al-abaqt): Biographien der medinischen Kmpfer Muhammeds, p. 59.
236 the other pre-uthmnic collections
of the Koran. His date of death is variously given as 19/640, 20/641, 22/642
643, 30/650651, or 32/652653.
8
AbdAllhIBNMASD,
9
a Hudhal of lowparentage, was anearly Muslim
who fought at Badr. He was Muammads servant and nearly always near
him, thereby acquiring great familiarity with the revelations. He claimed
to have already known seventy sras when Zayd b. Thbit was still a boy
playing in the street. Umar sent him as a q and treasurer to Kufa, where
he died in 32/652 or 33/653. According to others he died in Medina.
10
Ab Ms Abd Allh b. Qays AL-ASHAR
11
was in 7/628 a member of
the Yemenite Ban Ashar embassy that appeared before Muammad while
he was besieging the Jewish stronghold Khaybar.
12
He embraced Islam and
held administrative and military offices under the caliphs Umar and Uth-
mn, being particularly suited on account of his personal valour. In 17/638
he became governor of Bara and in 34/654 even took the place of Sad b.
alat Kufa. Concurrently he was an active teacher of the Koran and a
reciter, a position for which he was particularly suited because of his fine
and mighty voice. As a traditionist he rigorously insisted that his transmis-
sions not be written down but passed on only orally. He died in 42/662 or
52/672.
13
Miqdd b. Amr,
14
of the Yemenite Ban Bahr, became involved in a [ii/29]
blood feud and had to flee and eventually ended up at Mecca, where he
became a client of al-Aswad b. Abd Yaghth, apparently a Yemenite com-
patriot. There he became one of the first to embrace Islamand participated
8
Cf. in this connection, below, p. 254.
9
Juynboll, Encyclopedia, pp. 78; Sezgin, GAS, vol. 1, s.v.
10
Ibn Qutayba, p. 128; Ibn Sad (abaqt): Biographien der medinischen Kmpfer, p. 104,
sq., and Biographien der mekkanischen Kmpfer, p. 106sq.; al-Nawaw, Tahdhb al-asm, Ibn
ajar al-Asqaln, and Ibn al-Athr, Usd al-ghba, s.v.; al-Qurub, fol. 20
r
; E. Sachau in his
introduction to Ibn Sad (al-abaqt): Biographien der mekkanischen Kmpfer Muhammeds,
p. xvsq. Cf. above, p. 217 n. 5, where he is mentioned among those men who had memorized
the entire Koran.
11
Jeffery, Materials for the history of the text of the Qurn, pp. 209211.
12
Ibn Sad (al-abaqt): Biographie Muhammeds; Ereignisse seiner medinischen Zeit p. 79
= Julius Wellhausen, Seine [Muammads] Schreiben und die Gesandtschaften an ihn, no. 132.
13
Ibn Sad (al-abaqt): Letzte Krankheit, p. 105sq., ibid. (al-abaqt): Biographien der
Genossen, pp. 7886 (the main source), ibid., Biographien der Kufier, p. 9; Ibn Qutayba,
p. 135; al-Bukhr, Fail al-Qurn 31; al-Nawaw, Tahdhb al-asm, Ibn ajar al-Asqaln,
al-Iba f tamyz al-aba, and Ibn al-Athr, Usd al-ghba, s.v.
14
Ibn Sad (al-abaqt): Biographien der mekkanischenKmpfer, pp. 114116; Ibn Qutayba,
p. 134; al-abar, vol. 3, p. 2544; al-Nawaw, p. 575; Ibn ajar al-Asqaln, al-Iba, and Ibn
al-Athr, Usd al-ghba, s.v.; Juynboll, Encyclopedia, s.v.
the other pre-uthmnic collections 237
as a horsemanindicating noble extractionin nearly all the campaigns
against the unbelievers. During the conquest of Egypt
15
he held a command,
and under Muwiya he participated in the Cypriot campaign.
16
The sources
are silent as far as his religiosity is concerned; the same applies to his
knowledge of the Koran. When he died in 33/653, Uthmn said the prayer
for the dead.
Dissemination and Preservation of the Editions
As far as the dissemination of the Koran editions of these men is concerned,
the Damascenes and Syrians
17
respectively followed the reading of Ubayy b.
Kab, the Kfans the one of Ibn Masd, the Basrans the one of Ab Ms
(AL-ASHAR) and the inhabitants of im the one of Miqdd b. Amr.
18
It is
not surprising that the editions of Ibn Masd and Ab Ms (AL-ASHAR)
in Kfa and Bara respectively attained such a reputation, considering the
influential positions that these men held in the respective places. On the
other hand, nothing is known about the outward relations of either Miqdd
to im or Ubayy to Syria.
Not a single edition of these men has come down to us, so we are depen- [ii/30]
dent on indirect sources regarding their outer form and text. There is not
even a trace of Miqdd b. Amrs edition in these indirect sources. I know
of only one reference to Ab Ms AL-ASHAR in al-Suy, al-Itqn, p. 154,
which states that he incorporated in his Koran the sras of Ubayy b. Kab
as well as the traditions regarding two verses peculiar to his canon.
19
Con-
versely, from the texts of Ubayy and Ibn Masd we not only have a certain
number of readings, which are collected below in the chapter The Read-
ers and Readings, but also lists regarding the number and order of the
sras.
15
Ibn Taghrbird, Annales, ed. Theodorus G.J. Juynboll, vol. 1, pp. 9, 21, 53, 76, and 102.
16
al-Tabar, vol. 1, p. 2820; al-Baldhur, Fut al-buldn, ed. de Goeje, p. 154.
17
Ibn al-Athr, al-Kmil f l-tarkh, vol. 3, p. 86, merely states that the people of Damascus
considered their reading to be the best. On the other hand, it reads in al-abar, Tafsr, vol. 1,
p. 20, that the people of al-Sham followed the reading of Ubayy.
18
Ibn al-Athr, al-Kmil, vol. 3, p. 86; Ibn Aiyya, fol. 25
v
; al-Qasalln, vol. 7, p. 448, on
al-Bukhr, Fail al-Qurn 3.
19
Cf. above, pp. 191 and 196f., and below, p. 248.
The Koran of Ubayy b. Kab
According to the Transmission of al-Fihrist
According to the Fihrist
20
Ubayys Koran was arranged as follows:
21
sras 1, 2,
4, 3, 6, 7, 5, 10, 8, 9, 11, 19, 26, 22, 12, 18, 16, 33, 17, 39, 45, 20, 21, 24, 23, 40, 13, 28,
37, 38, 36, 15, 42, 30, 43, 41, 32, 14, 35, 48, 47, 57, 58,
22
25, 32, 71, 46, 50, 55, 56,
72, 53, 68, 69, 59, 60, 77, 78, 76, 81, 79, 80,
23
83, 84, 95,
24
96, 49, 63, 62, 65,
25
89,
67, 92, 82, 91, 85, 86, 87, 88, 64,
26
98,
27
61, 93, 94, 101, 102,
,
28

,
29
104, 99, 100, 105, 113, , 108, 97, 109, 110, 111, 106, 112, 113, 114.
20
Ibn al-Nadm, al-Fihrist, ed. G. Flgel, p. 27. The authority of the Fihrist is the moderate
Shite writer Ab Muammad al-Fal IBN SHDN (Fihrist, p. 231; al-s, Tusys List of
Shyahbooks [Fihrist kutubal-Sha], editedby A. Sprenger (Calcutta, 1858), p. 254, [EI
2
; Sezgin,
GAS, vol. 1, pp. 537538]). His authority refers to a codex of Ubayy, which he claims to have
seen at a place not far from Bara on a certain Muammad b. Abd Allh al-Anr.
21
Although traditional indices list sras by their name, herefor reason of simplicity
the arrangement of the sras follow the respective number in our editions.
22
The text has , which Flgel considers a spelling mistake of , the name of the
fifty-second sra. Correct is , which according to al-Itqn, p. 127, as well as in Ubayy, is
the name of the fifty-eighth sra, commonly called al-Mujdala. This is also as it reads in
al-Itqn, p. 150, in the catalogue of the sras of Ubayy.
23
The text has , the common name of the eightieth sra. That this interpretation is
correct is confirmed in al-Itqn, p. 150. For this reason the appearing once more in the
index to the Fihrist after sra 80 must be an error.
24
of the text is the common name of the ninety-fifth sra. This identity is confirmed
by al-Itqn, p. 150. The name appears once more later in the Fihrist between sras 113 and
108. It is difficult to determine whether this is an accidental duplicateit is preceded by
or some other mistake because in this case al-Itqn has a much different sequence.
25
Text , which in al-Itqn, p. 151, reads
.
26
Text , which according to al-Itqn ought to read . It is strange that Itqn, p. 150,
has before sra 80 ( ), which is obviously a spelling mistake for , the name of the
sixty-fifth sra.
27
Text:
. Itqn, p. 151: .
In the Fihrist it then reads: . Conversely, the words
are
useless. An attempt at interpretation is made in the next note 29.
28
. This is the name of a sra not contained in our text, which actually consists of
three verses; it shall be treated in detail, below, p. 241 sqq. Flgel did not understand this any
more than August Mller, the editor of the foot-note, since both of them did not trouble to
consult either al-Itqn or Nldekes Geschichte des Korans, which at that time had long been
published.
29
this is how it ought to be read according to al-Itqn, pp. 151 and 527, instead of
the text. This is the name of the second sra which only Ubayy knows. In the Fihrist this is
followed by the additional words
, which constitute beginning

and end of this sra. The
, missing at the beginning, slippedas it seemsinto the name

of the ninety-eighth sra in the form of
. Everything that Flgel suggests in this respect

is nonsense, which could have been avoided if he only had looked at Nldekes Geschichte des
Korans.
The Koran of Ubayy b. Kab
According to the Transmission of al-Suys
al-Itqn, and Its Relation to al-Fihrist
Al-Itqn, p. 150sq., supplies the following list of the Koran of Ubayy: Sras
1, 2, 4, 3, 6, 7, 5, 10, 8, 9, 11, 19, 26, 22, 12, 18, 16, 33, 17, 39, 41 or 43,
30
20, 21, 24,
23, 34, 29, 40, 13, 28, 27, 37, 38, 36, 15, 42, 30, 57, 48, 47, 58, 67, 61, 46, 50, 55,
56, 72, 53, 70, 73, 74, 54, 41 or 43,
31
44, 31, 45, 52, 51, 69, 59, 60, 77, 78, 75, 81, 65,
79,
32
80, 83, 84, 95, 96, 49, 63, 62, 66, 89, 90, 92, 82, 91, 86, 87, 88, 61, 64, 98, 93,
94, 101, 102, 103, ,
33
,
34
104, 99, 100, 105, 106, 107, 108, 97, 109, 110, 112, 113,
114.
Missing are the sras 14, 25, 32, 35, 68, 76, 85, and 111, whereas the number [ii/33]
of sras missing in the Fihrist amounts to twice this number, namely four-
teen, but entirely different ones. The overall number of Ubayys sras also on
this list makes one hundred and sixteen; admittedly, the explicit statement
is missing. Al-Suy, on the other hand, mentions two traditions in a differ-
ent passage,
35
according to which this recension consisted only of 115 sras
because either sras 105 and 106
36
or sras 93 and 94 were combined. There
is no way of further establishing the reliability of these facts. The fact that
30
Text:
. As the logogram appears in front of each of the sras 40, 41, and 4345,
and since the sras 40, 44, and 45 are undoubtedly mentioned at other parts of the list, here
reference can be only to either sra 41 or 43, whereas on page 39 on the list of the Fihrist
sra 45 follows. The reading
must be challenged, regardless whether or not this word is

connected with the preceding or put behind because it is not not customary in the list
to mention the opening words of a sra. The deterioration must be very deep as neither the
names of the two -sras nor that of the other missing sras showany similarity with . In
addition, the word , which is regularly used to separate the names of the sras, is missing
once or twice respectively.
31
Text: . According to foot-note 30 this can refer only to either sra 41 or 43.
32
Between 79 and 80 the text has . If this word were to be changed to , sra 64
would be listed twice, appearing once more below, between the sras 91 and 98. For such
reason we must be dealing with a different kind of deterioration, be it a spelling mistake
from the name of a missing sra or more likelyas also in this passage al-Fihrist offers the
sequence 79, 80, 83, 84a duplicate to the preceding or to the following , about
which compare above, p. 238 n. 26.
33
Cf. above, p. 238 n. 28.
34
Cf. above, p. 238 n. 29.
35
al-Itqn, p. 154.
36
This is frequently also said of these two sras; cf. the commentators on sra 106;
Takprlzade, Mift al-sada, cod. Wien, neue Folge 12 = G.L. Flgel, Die arabischen,
persischen Handschriften, vol. 1, p. 25sq. In the Koran of the Shiite sect of the Immiyya,
apart from the sras 8 and 9, also those two pairs were combined to one chapter each. Cf.
below, p. 289 n. 51.
these four sras are enumerated individually in the transmitted lists need
not speak against this, since sra 94 follows directly upon sra 93, and sra
106 stands immediately after 105.
The Sras Peculiar to the Koran of Ubayy b. Kab
Far more weighty is the fact that Ubayys collection contains two sras that
are missing from the canonical recension. These are called at times Srat
al-Khal and Srat al-afd
37
They are also known under the collective title
of Srat l-qunt
38
or even Srat al-Qunt.
39
The other designations, Du
al-Qunt,
40
Du al-Fajr
41
or al-Du
42
are based on the assumption that this
concerns mere prayers and not sras. It is extremely rare to find the text
itself. al-Mabn li-namal-man, part 3, supplies the beginning of the first
sra, al-Fihrist, the opening verse of the first sra as well as the twolast words
of the second sra, al-Jawhar, al-i, and Lisn al-Arab, the third verse
of the latter, and al-Zamakhshar, the two first verses. It appears that the
complete text of the sras is supplied only by much later writers, al-Suy
(d. ad1510 [sic]),
43
Takprlzade (d. 1560), and al-Birgl (AL-BIRKAW)
(d. 1562 [sic]). Al-Suys isnds, or chain of authorities, are traced back to
the first century ah, in so far as they could be established chronologically.
44
The two sras were first published by J. von Hammer-Purgstall
45
in 1850. [ii/34]
Apart from the fact that he did not scrupulously reproduce his manuscript,
a better text can now be established with the aid of parallel versions.
37
al-Fihrist, p. 27; al-Suy, al-Itqn, pp. 151 and 527; Takprlzade, loc. cit.
38
39
Umar b. Muammad (IBN ABD AL-KF), Ms., fol. 3
r
.
40
Frequently, e.g., al-Zamakhshar on sra 10:10; the catechism of Taq l-Dn Muammad
b. Pr Al AL-BIRKAW (Birgil d. 970/1562 [Brockelmann, Geschichte, vol. 2, pp. 440444,
suppl. vol. 2, p. 654]); Handschriften in Gttingen, cod. Asch., 88, [ or ,
translated by Garcin de Tassy entitled L islamisme d aprs le Coran,] 90, 87, 97.
41
Lisn [al-Arab], vol. 4, p. 130.
42
al-Jawhar, al-i, vol. 1, p. 223.
43
In al-Suy, al-Itqn, p. 153sq., conversely, he transmits the text in three variants, in a
small monograph on these two sras (W. Ahlwardt, Verzeichnis, no. 438), however, according
to six traditions.
44
AbdAllhb. Zurayr al-Ghfiq (al-Itqn, p. 153), d. 81/700; Ubaydb. Umayr (Itqn, p. 153,
W. Ahlwardt, Verzeichnis, 438, no. 4) d. 64/683; Umayya b. Abd Allh b. Khlid b. Usayd [EI
2
]
(al-Itqn, p. 154), d. 84/703 or 87/705; Maymn b. Mihrn (Ahlwardt, Verzeichnis, 438, no. 2),
d. 117/735; [EI
2
; EQ; Sezgin, GAS, vol. 7, p. 11]. These dates of death have been taken from
al-Khazraj, Khulsat tadhhb al-kaml.
45
Literaturgeschichte der Araber, vol. 1, p. 576.
Text [ii/34]
46

47

* *
48

49
*
50

51

52

* * *
53

*

54

*
55

Translation
Sra of the Abandonment.
In the Name of God, the Merciful, the Compassionate;
(1) O God, we ask You for help and forgiveness;
(2) We praise You, and are not ungrateful to You;
(3) We renounce and leave anyone who sins against You.
Sra of the Serving with Alacrity.
In the Name of God, the Merciful, the Compassionate;
(1) You we serve;
(2) And to You we pray and worship;
(3) And to You we speed and strive after;
(4) We anticipate Your mercy;
(5) And dread Your punishment;
(6) Truly, Your punishment reaches
56
the unbelievers.
46
Cf. above, p. 240 n. 37. [Although inadvertently all the numbers of the foot-notes follow
the direction of the Latin script this does not effect the content; vocalization omitted.]
47
The Basmala is missing from al-Zamakhshar on sra 10:10, al-Itqn, p. 153 (a), cod.
Landberg, 343 (= Ahlwardt, Verzeichnis der arabischen Hands., no. 438), nos. 1, 3, 4, 5.
48
Birgl inserts .
49
al-Itqn, p. 154 (c) inserts ; Birgl inserts .
50
Ahlwardt, Verzeichnis, cod. 438, no. 5, inserts ; Birgl inserts
.
51
Birgl, cod. 87 and 97 inserts .
52
Cf. above, p. 240 n. 37.
53
This verse is missing in Takprlzade.
54
Thus read al-Fihrist, al-Zamakhshar, al-Suy, al-Itqn (a, c), Cod. Landberg, 343,
nos. 2, 4, 5, 6; al-Birkaw; conversely, al-Itqn, b (W. Ahlwardt, Verzeichnis der arabischen
Handschriften, 438 =) cod. Landberg, 343, nos. 1 and 3, Takprlzade .
55
al-Suy, al-Itqn, 154 (Ahlwardt, Verzeichnis der arabischen Hands., 438) = cod. Land-
berg, 343, nos. 1, 3, position verse five before verse four.
56
is in the Koran always causative: to cause someone to reach someone (used only
of persons); in other cases it is also simply to reach; for this reason it can be read as either
an active or passive participle.
The Question of Authenticity
Since these texts are prayers in content and form, their character as revela- [ii/36]
tion can be upheld only if the word qul say! has been omitted, which, as it
is well-known, is used in the Koran to identify prayersfor example, sras
113 and 114or subjective remarks of Muammad as the Word of God. Such
qul is indeed missing from the opening sra of our editions of the Koran,
but this fact is precisely one of the reasons that the Ftia is suspected of
not being part of the revelation. The other reasonsas elaborated above
(p. 79sq.)are based first on the strong dependence on the language of
Jewish and Christian liturgies, which prompted the usage of idioms not oth-
erwise foundinthe Koran,
57
secondonthe use of grammatical constructions
otherwise alien to the Koran,
58
and, third, on the awkward construction of
the last verse, possibly caused by problems of translation. In contrast, the
style of Ubayy b. Kabs sras is much smoother, following generally the line
of Koranic idioms. Considering the short text, a not insignificant quantity
of linguistic exceptions can be documented. As just mentioned, the con-
struction of istana with the person in the accusative can be documented
inthe Koranonly inone passage of the Ftia. The verb to praise is alto-
gether missing fromthe Koran,
59
although other verbs of the same meaning
are found rather frequently in the Koran.
60
Also missing is , to hurry.
, to run, although proper Koranic Arabic, does not occur in connection
with , to be heading for.
61
, to sin is here followed by the person
in the accusative, whereas it is used in the Koran with a direct object only
(sras 75:5, and 91:8). occurs only once in the entire Koran (sra 20:12),
but notlike heremetaphorically.
62
57
The name of Allh as King of the Day of Judgement malik yawmal-dn.
58
Istana to ask for assistance with the accusative, whereas usually everywhere in the
Koran with of the person.
59
Nevertheless, it is likely to have been used in this meaning already at the time of the
Prophet; cf. amsa, 777, in a poemof Umayya b. Ab l-alt (= Dwn, ed. Schulthess, no. 9, 5;
Muallqat Antara (Th. Arnold), v. 35 = Ahlwardt, Divans, p. 82, no. 21, l 40; al-Sukkar, Poems of
the Huzailis, no. 91, l 3; [Divans of the six ancient Arabic poets, ed. W. Ahlwardt, p. 28], Zuhayr b.
Ab Sulm, no. 4, l 20; Labdb. Raba, Gedichte, editedby AntonHuber andCarl Brockelmann,
no. 53, l 18.
60
The Koran has instead kabbara, sabbaa, amida.
61
The idiom, , sra 62:9, is no proper parallel.
62
Doubtful is . With the meaning of to renege ones faith in the Koran is always
constructed with of the person, whereas the meaning of to be ungrateful, which in this
case is also permissible, it is commonly followed by the person in the accusative., also sra
11:63.
For such reasons it is unlikely that these sras constitute a genuine part [ii/37]
of the Koran, or that they are even from the Prophet. Yet to all appearance
they are old prayers, current already during the Prophets lifetime. Tradi-
tion, as we have been able to see, frequently calls them du, and Umar
as well as Ubayy b. Kab are supposed to have used them in the prayer of
qunt.
63
Fromhere on it was an easy step towards the opinion of their heav-
enly origin.
64
Other interpreters might be tempted to agree for the sole rea-
sonthat these sras were introducedby the Basmala.
65
Still others pretended
to know even more and date their revelationas well as the words

sra 3:123to the time when the clans of Muar were cursed
by Muammad.
66
But this information is based on the combination of du
al-qunt, the name coined for Ubayy b. Kabs sras, and the tradition that
Muammad said a qunt prayer
67
after that curse. According to one tradi-
tion Ab Ms AL-ASHAR also had the two sras in his canon,
68
whereas
Ibn Abbs was guided by the common reading of Ab Ms AL-ASHAR
andUbayy b. Kab. Al allegedly presentedthese sras toAbdAllhb. Zurayr
al-Ghfiq (d. 81/700) as being part of the Koran. That an excellent authority
like Ubayy b. Kab could have been deceived must not come as a surprise;
no less a person than the expert Ibn Masd cast away the Ftia, whereas
Zayd (Ibn Thbit) incorporated it in his canon.
The Relation of the Transmitted Lists of Ubayy b. Kabs
Sras with One Another and with the Canonical Edition
Despite the many andconsiderable deviations, the arrangement of the sras [ii/38]
in Ubayys Koran generally follows the principle of the canonical recension,
namely the progression fromthe longer to the shorter chapters. In both lists
this is most obvious at the beginning and the end, less so in the middle
section. The list of the Fihrist is identical with the Uthmanic arrangement
in the following sixteen places: (1) sras 6 and 7; (2) 8 and 9; (3) 20 and 21;
63
al-Itqn, p. 153.
64
Muslimscholars who deny the authenticity of Ubayys sras approach the subject from
quite a different angle. They seem to fear for the sanctity of the Uthmanic text if these sras
are recognized as revelation.
65
al-Itqn, p. 153: .
66
al-Itqn, p. 154; cf. Juynboll, Encyclopedia, p. 660, col. 2.
67
Cf. above, p. 156 n. 96.
68
al-Itqn, p. 154, beginning. Ibn Abbs, too, allegedly had the two sras in his copy. Cf.
above, p. 237.
(4) 37 and 38; (5) 57 and 58; (6) 55 and 56; (7) 68 and 69; (8) 59 and 60; (9)
77 and 78; (10) 79 and 80; (11) 83 and 84; (12) 95 and 96; (13) 85, 86, 87 and 88;
(14) 99 and 100; (15) 109 and 110 and 111; (16) 112, 113 and 114.
According to the list in al-Suy, al-Itqn, the numbers 5, 6, 7, 10, and 13
should be dropped, for which five other identical sequences appear, namely
the sras 1, 2, 73, 74, 93, 94, 101, 102, 103, 105, 106, 107 and108. Bothlists agree on
the position of the qunt prayers, which are both placed between the sras
103 and 104.
The Koran of Ibn Masd According to the Transmission of al-Fihrist
As in the case of Ubayy b. Kabs collection of the Koran, we also have here [ii/39]
two detailed reports. According to the list in al-Fihrist, p. 26, the sras were
arranged as follows: 2, 4, 3, 7, 6, 5, 10,
69
9, 16, 11, 12, 17, 21, 23, 26, 77, 38, 28, 24,
8, 19, 29, 30, 36, 25, 22, 13, 34, 35, 14, 38, 47, 31,
70
39, 40, 43, 41, 46, 45, 44, 48, 57,
,
71
32, 50, 65, 49, 67, 64, 63, 62, 61, 72, 71, 58, 60, 66, 55, 53, 51,
72
52, 54, 69,
56, 68, 79, 70, 74, 73, 83, 80, 76, 55, 79, 78, 81, 82, 88, 87, 92, 89, 84, 85, 96, 90,
93, 94, 86, 100, 107, 101, 98, 91, 95, 104, 105, 106, 102, 97, 103, 110, 108, 109, 111, and
112.
The Koran of Ibn Masd According to
the Transmission of al-Suys al-Itqn
The list of al-Itqn, p. 151 sq., is arranged as follows: 2, 4, 3,
73
7, 6, 5, 10, 9, 16, 11, [ii/40]
12, 18, 17, 21, 20, 23, 26, 37, 33, 22, 28, 27, 24, 8, 19, 29, 30, 36, 25, 15, 13, 34, 35, 14,
38, 47, 31, 39, 40, 43, 41, 42, 46, 45, 44, ,
74
48, 59, 32, 65, 68, 49, 67, 64,
69
It is very strange that according to al-abar, Tafsr, vol. 1, p. 2163, l 9, still in 35/655 the
tenth sra (Jonah [Ynus]) of the Medinan Koran was designated as the seventh sra, as this
enumeration corresponded to Ibn Masds arrangement. It is more likely that in this passage
ought to be replaced by , the ninth.
70
Text instead of as it is correctly given in the index of al-Itqn.
71
cannot refer to the sras 61 or 87, beginning with the word
, as these are
clearly indicated at other places of the list. But also among the sras 15, 18, 20, 27, 42, and
49, which are missing from the list, there is no known nameapart from the forty-second
sra ( )which would easily lend itself to a perversion to . In this case the list of
al-Itqn is of no help as its arrangement differs considerably.
72
Another transmission of the Fihrist has the reverse order 52, 51 as in al-Itqn.
73
Also al-Itqn, p. 145, indicates the sras 2, 4, and 3 as the beginning of the codex of Ibn
Masd.
74
The interpretation of this name poses problems. This cannot refer to the sixtieth sra
63, 62, 61, 72, 71, 58, 60, 66, 55, 53, 52, 51, 54, 56, 79, 70, 74, 73, 83, 80, 76, 77, 75,
78, 81, 82, 88, 87, 92, 89, 85, 84, 96, 90, 93, 86, 100, 107, 101, 98, 91, 95, 104, 105,
106, 102, 97, 99, 103, 110, 108, 109, 111, 112, and 94.
The Relation of the Two Lists to One
Another and to the Uthmanic Recension
According to al-Suys al-Itqn, the sequence of the sras in the recension
of Ibn Masd is congruent with the standard text of Uthmn in the follow-
ing places: (1) sras 11 and 12; (2) 29 and 30; (3) 34 and 35; (4) 39 and 40; (5) 41
and 42; (6) 81 and 82; (7) 104 and 105. According to the Fihrist, there are four
additional places: sras 77 and 78, 84 and 85, 93 and 94, 111 and 112. With this,
the sequence of the Fihrist comes close to the arrangement of the canonical
arrangement.
The sras missing from al-Fihrist (16, 18, 20, 27, 42, and 49) are all con-
tained in al-Itqn, and conversely, those missing from al-Itqn (50, 57, and
64) are found in al-Fihrist. Therefore, all these omissions are purely acci-
dental. If they are inserted in the respective indices, it becomes evident
that both of them contain an identical number of sras, namely all the
sras of the standard text of Uthmn, with the exception of sras 1, 113, and
114. The accuracy of the result is confirmed by remarks at the end of both
lists.
75
Although the number of sras in the codex of Ibn Masd is not explicitly [ii/41]
stated in al-Itqn, in al-Fihrist it is calculated to consist of only one hundred
and ten sras. This is very strange. As this codex is short three sras, that
total ought to have been one hundred and eleven, unless two of them
as its name, , appears later, and particularly at the same place as in the Fihrist. By the
same token, cannot easily be a distortionof another name of the sras, as inthe index
to the Fihrist, the sras 44 and 48 follow directly one another, and as the names of the sras
50, 57, and 69, omitted from the Itqn, provide not a trace of similarity with that word. For
this reason must be a duplicate of the name of the sixtieth sra, appearing in the
following line.
75
Fihrist, p. 26, l 27sq.
. Shorter, al-Itqn, p. 152, . Cf. also Umar b. Muammad,
fol. 3
v
; al-Mabn li-nam al-man, parts 2 and 4; al-Qurub, i, fol. 20
r
, and 22
v
; al-Suy,
al-Itqn, p. 186 end and p. 187 top; al-Shshw according to Ibn Qutayba; Takprlzade,
loc. cit. Most traditions simply state that these sras had never been part of Ibn Masds
codex; it is rare to find the statement that he erased them (
, al-Mabn li-nam al-man,

part 4; al-Itqn, p. 187) or eliminated them ( , al-Itqn, p. 186, end).
were combined to form a single sra. The sras that come first to mind
when thinking of a combination, namely sras eight and ninewhere sra
nine, at least in the Uthmnic recension, lacks the basmalacannot be
considered in this connection, as they do not appear one after the other in
either of the two lists.
76
Since our sources offer no other combinations of
this kind, we must suspect that the strange numerical count is nothing but
a mistake in the text.
In another tradition
77
it says that the codex of Ibn Masd consisted of 112
sras, where the two sras of invocation are wanting. According to this, the
Ftia is again ascribed to him, and this opinion is widespread. Al-Suy
mentions in a later passage of his Itqn
78
yet three other traditions in the
same vein. The author of al-Fihrist reports
79
to have seen a two-hundred-
year-old codex of Ibn Masd in which this sra was written down, adding
that in general no two manuscripts of this recension were truly identical.
Even the opposite statement in the Fihrist about the three sras missing
in the codex of Ibn Masd
80
leaves the impression that the words
were inserted later in the text.
In each case, Ibn Masds reservation regarding the three sras is not [ii/42]
without foundation, since they are so different from all the other sras
in content as well as form that their authenticity is open to much criti-
cism. Whereas the Ftia is conspicuous for its dependence upon Jewish
and Christian liturgies,
81
the sras of conjuration are full of gross pagan-
ism,
82
although their initial word, qul, ostensibly gives themthe appearance
of a revelation. Whoever is responsible for the current place of these prayers
in the Koran evidently intended them as a kind of religious shield: just as
prayers of praise of the Ftia were likely intended to invoke the protection
of Allh, so the two prayers of conjuration should serve to ward off the influ-
ence of evil spirits.
76
Anote inal-Itqn, p. 152, states explicitly that inthe codex of IbnMasdsra 9 included
the basmala.
77
78
al-Itqn, p. 187.
79
al-Fihrist, p. 26, bottom.
80
Cf. Seite 40, Anm. 3: This reference is wrong.
81
Cf. above, p. 242.
82
The Relation of the Korans of Ubayy, Ibn Masd, and
Ab Ms to One Another and to the Canonical Version
Ubayys Koranconsistedof all the sras of the canonical versionplus the two
prayers of qunt. IbnMasd, onthe other hand, hadtwoor three sras (1, 113,
and 114) less than the subsequent authorized recension. The arrangement
of the sras in these two codices is so different that a succession of sras
coincides only twice in the lists, namely at the beginning (sras 2, 4, 3)
and at the end (sras 105 and 106) as well as in the middle (sras 43 and
41), according to al-Itqn and al-Fihrist respectively. The relation of both of
them to the canonical version is much more favourable, since Ibn Masds
sequence is identical tothis versionineight (al-Itqn) andtwelve (al-Fihrist)
places respectively, while Ubayys sequence is in sixteen places. Conversely,
the agreement of the sras arranged exactly by length is less favourable, as
there, according to my table below, pp. 264266, the overlap of sequence
occurs only in four or five places, namely sras 2, 4, and 3; 9, and 11; 63 and
62; 110 and 111; 112 and 113 (Ubayy according to al-Fihrist) and sras 2, 4, 3, 7, 6,
and 5; 48, and 57; 77, and 78; 111, and 112 (Ibn Masd according to al-Fihrist)
respectively.
Despite all the differences and inaccuracies, both codices are evidently [ii/43]
basedonthe principle of anarrangement according tothe decreasing length
of the sras. This principle is so peculiar that its double application cannot
be accidental but must be the result of textual relationships, even if there is
noother evidence. As this principle is alsoadheredtointhe canonical recen-
sion, which, as we shall see presently, is based on the first collection of Zayd
b. Thbit, the textual relationships alluded to must also include the latter.
The same conclusion emerges from another observation. Since throughout
Ubayy b. Kabs and Ibn Masds lists of sras nearly the same headings
appear as in the subsequent canonical recension, the underlying presup-
position is undoubtedly that behind identical names there must basically
be the identical content. This arrangement of the sras must therefore have
already existed in the codex of afa bt. Umar, becauseas tradition has
itit served as a model for the Uthmnic recension. Naturally, this can-
not be accidental but rather the result of a textual link. Unfortunately, it is
impossible to be specific. The former codex is allegedly based on the very
first collection of the revelations. But this is not necessarily reliable, since
the tradition regarding Zayds collection has been shown to contain other
errors as well. One must also not forget that Ibn Masd and Ubayy b. Kab
were Zayd b. Thbits senior contemporaries and had served Muammad
longer than he did.
In addition, there are some isolated traditions regarding verses that were [ii/44]
allegedly contained in these two recensions but are missing in our editions.
With regard to the lost verse that Ibn Masd allegedly had in his codex,
we have a tradition traced back to him in Hibat Allh b. Salma, al-Nsikh
wa-l-manskh, Cairo edition, p. 10, saying: The Messenger of God had me
recite a verse which I then memorized and entered in my codex. When I
returned to my sleeping quarters, I could not remember it, and the place
in my codex was blank. Thereupon I consulted the Prophet who replied
that the verse had already been abrogated the day before. Such general
statements are of course totally useless, even if they can be trusted.
More useful are the reports about the three lost verses that Ubayy b. Kab
had in his codex.
The first one says: If man had a vale of treasures, he would covet yet
a second one; and if he had two, he would covet yet a third one; neither
shall the belly of man be filled, but with dust. Allah will turn unto him who
shall repent.
83
This verse is allegedly from sra 10:25 or somewhere in sra
98,
84
which is impossible, if only for the difference of rhyme. Yet no other
place is possible, since the word for man used here in the original, ibn
dam, is an expression found nowhere in the Koran.
85
The second verse reads: Religion for God is the moderate Hanfiyya, nei-
ther Judaism, nor Christianity; whoever does good shall not remain without
reward.
86
This verse is also said to have belonged to sra 98, but its content
and rhyme do not allow it. It is altogether not genuine, as the expressions
used for the three different religions are foreign to the Koran.
The third verse reads: If an elderly man and woman commit adultery, [ii/45]
then stone themdefinitely as penalty fromGod All-mighty and wise.
87
This
so-called Verse of Stoning cannot have been either part of sra 33, given
the non-matching rhyme, or of the Koran in general, as this cruel, criminal
ordinance can only have been instituted after the death of Muammad, as I
explained previously.
88
It is also said of the first verse of Ubayy b. Kab that Ab Ms (AL-
ASHAR) read it in a sra similar to the ninth. According to one tradi-
83
The Arabic text can be found above, p. 190, the English translation in Sales Koran,
Preliminary discourse, section 3, p. 52, ll 1821.
84
Loc. cit., above, p. 193.
85
Loc. cit., above, p. 195.
86
Cf. loc. cit.; the Arabic text, above, p. 195.
87
Cf. the Arabic text, above, p. 198.
88
Cf. above, pp. 199201.
tion,
89
he presented the verse as he still remembered it to a meeting of three
hundred readers of the Koran at Bara.
It is reported that on this occasion Ab Ms (AL-ASHAR) presented
yet another verse, belonging to the sras similar to the so-called musab-
bit.
90
Whether this sra has been lost or constitutes a part of our codices,
but had escaped Ab Ms (AL-ASHAR), is not clear from the text of the
tradition. The verse reads: O, you who believe, wherefore do you say what
you do not do? Atestimonial will be written on your neck, and on the Day of
Resurrection you will have to explain.
91
The first half of the text is identical
with sra 61:2. Because of the rhyme, the second part cannot have belonged
to either sra sixty-one or another of the musabbit. Even if this part does
not offer an argument raising doubt on its authenticity, this is still not likely,
as all the other traditions about lost verses of the Koran turned out to be
unreliable.
92
We have no clue as to how the sras in the pre-canonical codices were [ii/46]
separated: Was this done merely by empty space, or by some other sign or
word? According to al-Zamakhshar on sra 106, sras 105 and 106 consti-
tuted one single sra without separation in the codex of Ubayy b. Kab,
93
whereas Al al-Dn,
94
s.v., says that between the two of them the sign of
division, basmala, was missing.
95
This formulation presupposes that all the
other sras were preceded by the basmala. If this were correct, the basmala
ought to have been used already in afas codex, on whichas has been
demonstrated abovethe other pre-canonical codices are dependent.
Obscure and Dubious Codices of the Koran
It is not inconceivable, and possibly even probable, that apart from the
renownedcollections of the Korandiscussedabove, there were still others of
less repute, which therefore left no mark on literature. If it is claimed, how-
ever, that some of the Companions of the Prophet, such as Al, had the sras
89
Muslim, K. al-Zakt 26 (al-Qasalln on al-Bukhr, Cairo edition, 1303/1885, vol. 4,
p. 444sq., in the margin).
90
Under this name the sras 57, 59, 61, 62, 63, and 64 are subsumed.
91
Cf. the Arabic text above, p. 196.
92
For details see loc. cit., above, p. 196sq.
93
bi-l fal.
94
The German original does not distinguish between the two Al al-Dn: al-Khzin
al-Baghdd and al-Muttaq al-Hind. There is no reference to part 2 in the index, which in
any case is limited to the first five occurences.
95
wa-lamyufal baynahum f muafih bi-bismillh al-Ramn al-Ram.
in a chronological arrangement, this tradition
96
lacks credibility; for such an
arrangement presupposes somewhat lengthy, learned, exegetic activity and
would have been difficult to realize, since even Muammad himself com-
bined earlier and later revelations in the records he commissioned.
97
For the
same reason, the chronological lists described above, p. 48sq., also belong
to a much later period than the isnds would suggest. For internal reasons,
however, they must be older than the chronologically arranged codices of
the Koran, if indeed such works had ever existed. The alleged sequence of
the first six [sic] sras in Als chronological Koran (sras 96, 74, 68, 73, and
111, according to al-Suy, al-Itqn, p. 145) incidentally does not agree with
any of the above-mentioned chronological lists, in which generally sras 68
and73 are placedbefore sra 74. Inany case, it is certainthat Al cannot have
produced such a codex. By the same token, such a scholarly and historically
orientated type of arrangement can neither be expected of a contemporary
of the Prophet nor can it escape our notice that all traditions regarding Als
activity as compiler or editor of the Koran are from the outset suspect of
Shite fabrication.
98
This, however, is not to say that Al did not have his own codex of the [ii/47]
Koran. Onthe contrary, it is extremely probable not only that he did, but also
several of his contemporaries from the theocratic elite possessed records of
Muammads revelations. Yet these codices, as far as they were reasonably
complete, were basically modelled on those renowned collections. This
group would also have to include the codex of isha, which is said to have
been arranged differently, even though she attached little importance to it.
99
96
al-Qurub, fol. 22
v
; al-Suy, al-Itqn, p. 145.
97
Cf. below, next chapter, The arrangement of the sras in the Uthmnic Koran, p.
263sqq.
98
Cf. above, p. 219sq., regarding the collection of the Koran which Al allegedly produced
immediately after the death of Muammad; in addition, below, p. 293, about the so-called
Sra of the Light.
99
al-Mabn li-namal-man, part 2.
THE GENESIS OF THE AUTHORIZED REDACTION
OF THE KORAN UNDER THE CALIPH UTHMN
The Established Tradition
1
[ii/47]
During the campaigns in Armenia and Azerbaijan the warriors from Iraq
and Syria quarrelled about the genuine form of the text of the Koran. The
soldiers from im (Emesa) considered their text traced back to Miqdd b.
al-Aswad to be the best. The Damascenes and the Syrians respectively pre-
ferred their own text.
2
The Kfans recognized as the authority only the read-
ing of Abd Allh IBN MASD, and the Barans only that of Ab Ms AL-
ASHAR.
3
When the renowned commander udhayfa [Ibn al-Yamn]
4
was
back at Kfa after his campaign in Armenia and Azerbaijan, he complained
to the governor, Sad b. al-, about this situation which, according to
his view, seriously threatened the future of Islam. Many members of the
theocratic elite agreed with him; only the followers of Ibn Masd stub-
bornly insisted on the authority of their master. Upon udhayfa [Ibn al-
Yamns] return to Medina, he reported to Caliph Uthmn on what he had
observed. After he had consulted with the old Companions of the Prophet,
there was unanimous agreement with the commanders view of the sit-
uation. Thereupon the Caliph appointed a commission consisting of the
Medinan Zayd b. Thbit and three respected Quraysh, Abd Allh b. al-
Zubayr, Sad b. al-, and Abd al-Ramn b. al-rith,
5
and ordered them
1
al-Bukhr, Fail al-Qurn 3; al-Tirmidh in the Tafsr on sra 9 at the end; al-Khab
al-Tibrz, Mishct, Fail al-Qurn, fal 3 5; Fihrist, ed. Flgel, p. 24sq.; Izz al-Dn IBN AL-
ATHR, al-Kmil f l-tarkh, vol. 3, p. 85sq.; Ibn Khaldn, al-Ibar, Cairo ed., vol. 2, p. 135sq.;
al-Naysbr, Gharib al-Qurn in al-abar, Tafsr, 1st edition, vol. 1, p. 23; Al al-Dn, vol. 1,
p. 6sq.; al-Dn, Muqni; K. al-Mabn li-nam al-man, fol. 6sqq.; al-Qurub, fol. 20
r
; al-
Suy, al-Itqn, p. 138sq. Cf. Silvestre de Sacy, Mmoire sur l origine et les anciens monu-
ments de la littrature parmi les Arabes, p. 426sqq.
2
These are the general terms used by Ibn al-Athr, Usd al-ghba, vol. 3, p. 86, to describe
it. al-abar, Tafsr, vol. 1, p. 20, says that the Syrians followed the reading of Ubayy b. Kab. Cf.
also, above, p. 237 n. 17.
3
Cf. above, p. 235sq.; A. Jeffery, Materials for the history of the text, pp. 209211.
4
He was appalled by the different ways in which his warriors and those from Syria
recited the Qurn, Juynboll, Encyclopedia, p. 231, col. 2.
5
Juynboll, Encyclopedia, p. 232, col. 1; Sezgin, GAS, vol. 9, p. 6.
252 the koran under the caliph uthmn
to procure copies of the codex of the Koran in the possession of afa
bt. Umar.
6
After the work was completed, afas original was returned,
and the copies were dispatched to various foreign destinations to serve
as the authorized, model recension. The older collections, however, were
destroyed. It seems that the populace everywhere obediently accepted this
decision. Only the ever-unruly Kfans under the leadership of Ibn Masd
offered resistance.
When exactly this happened can only be guessed. The respective cam- [ii/49]
paigns are usually dated 30/650.
7
But their connection with other battles
reported by the chroniclers in the same region with identical actors
8
is by no
means explicit. If Ibn Masd indeed lived to see the completion of the Uth-
mnic recension, this must have happened before 32/652 or 33/653,
9
dates
which are given as the years of his death. The latest date is the death of Uth-
mn, which occurred at the end of 35/655 (18 Dh l-ijja).
According to the unanimous tradition, the initiative for the project came
not from the Caliph but from one of his most renowned commanders who,
in the wake of disagreements about the correct recitation of the Holy Text,
feared for the unity of Islam and for the Islamic theocratic government.
10
In
any case, it remains the undeniable merit of Uthmn to have followed upon
the advice and sped up the realization of the plan. He thus accomplished
his most reasonable act of government, and the only one through which he
won fame. The insurgents, of course, later reproached him for this benef-
icent decision (al-abar, Tafsr, vol. 1, p. 2952). On the other hand, outstand-
ing persons like Abd Allh b. Umar and Al, although personal and political
opponents, allegedly agreed with him in this particular case.
11
6
Cf. above, pp. 225sq., and 228sqq. In Ibn Aiyya, and al-Qurub, fol. 20
r
it reads after
al-abar: The leaves in the possession of afa served as a model (juilat imnan) for this
second collection.
7
al-abar, vol. 1, p. 2856; Ibn al-Athr, Chronicon, ed. Tornberg, vol. 3, p. 85; Ibn Khaldn,
ed. Cairo, vol. 2, append. 135; al-Dhahab, Tarkh al-Islm, vol. 1, cod. Paris, fol. 151 [sic]
(according to Caetani). Cf. J. Wellhausen, Prolegomena zur ltesten Geschichte des Islams,
p. 110.
8
Cf. L. Caetani, Chronographia Islamica, 32/652 4; Ibn Taghrbird, Annales, ed. Juyn-
boll, vol. 1, p. 97; Abulfedae Annales moslemici, ed. Reiske, vol. 1, p. 204; and al-Nuwayr
(Nihyat al-arab) give the year 29/649. When al-Suy, al-Itqn, p. 139 (according to al-
Qasalln, vol. 4, p. 438) mentions 26/646, this is confused with an earlier campaign. Cf.
Leone Caetani, loc. cit., 25/645, 3.
9
This would correspond to Eutychios, Annales, ed. Cheikho, vol. 2, p. 341.
10
11
Ibn al-Athr, al-Kmil f l-tarkh, vol. 3, p. 87.
the koran under the caliph uthmn 253
The tradition just paraphrased has been called established in the chap- [ii/50]
ter heading because it is the most widespread in literature, adth, Koranic
commentators, and works of history. The verification is not as sound as in
the case of the tradition of the first collection, where the isnd ends with
the renowned traditionist Anas b. Mlik
12
and does not go back to an eye-
witness. The criticism of that other tradition has particularly demonstrated
how little reliable such formalities are.
Deviating Traditions and Their Value
Deviating traditions are outwardly no less attested than the so-called estab-
lished traditions, since the authorities to which they are traced back, like
Abd Allh b. al-Zubayr, Kathr b. Afla, and [Muammad b. Muslim] AL-
ZUHR, d. 124/742,
13
are among the most respected traditionists. For this
reason, in the case of each of the following traditions the question of their
intrinsic reliability will immediately pose itself. A tradition in al-Dns al-
Muqni omits Sad(Ibnal-) andreplaces himwithbothAbdAllhb. Amr
b. al-
14
and Abd Allh IBN ABBS.
15
The former is known for asceticism,
great competence in adth and literacy, and he is the alleged author of a
collection of adth.
16
It is nevertheless unlikely that he was a member of
the Commission, since his father, whom Uthmn had deposed as governor
of Egypt in 28/648,
17
joined the forces of the opposition to the Caliph. Abd
Allh IBN ABBS, on account of his considerable theological and exegetic
erudition,
18
would have been well suited for this task. The fact that his name
is mentioned evidently relates to the tendentious goal of having at least one
man of the Prophets family be associated with the establishment of the
canonical text.
12
Muammad b. Muslim AL-ZUHR (d. 124/742) from Anas b. Mlik (d. ca. 90/708).
13
EI
2
; EQ; G. Juynboll, Encyclopedia of canonical adth, pp. 690730; F. Sezgin, Geschichte
des arabischen Schrifttums, vol. 1, pp. 280283.
14
Died 65/684; EI
2
; EQ; Juynboll, Encyclopedia, pp. 23; Sezgin, GAS, vol. 1, p. 84.
15
EI
2
; G. Juynboll, Encyclopedia, p. 1; F. Sezgin, Geschichte, vols. 1, and 8.
16
Cf. Ibn Qutayba, p. 146; al-Nawaw, p. 361 sq.; Ibn Sad (al-abaqt): Biographien der
Genossen, die sich noch vor der Eroberung Mekkas bekehrten (vol. 4, part 2), p. sqq.; Ibn
al-Athr, Usd al-ghba, vol. 3, p. 33; Juynboll, Encyclopedia, pp. 23.
17
Cf. J. Wellhausen, Prolegomena zur ltesten Geschichte des Islams, p. 127.
18
For details see the literary-historical appendix, p. 272.
Another source
19
adds Ubayy b. Kab
20
to those four men; he was one of [ii/51]
the most renowned experts on the Koran and the editor of a special col-
lection.
21
This must be rejected, however, because according to the reliable
report of al-Wqid, who had made inquiries among his family, he died
in 22/642 or, according to other informants, even two or three years ear-
lier.
22
The statement that he died in 30/650 or 32/652 is strongly suspect of
being a forgery for the sake of making it plausible that he collaborated with
the Uthmnic recension.
As is claimed in two traditions, Zayd (Ibn Thbit) was given only a single
assistant from the Quraysh. One of these two traditions mentions Sad
b. al-,
23
whose name was already on the above-mentioned list of four.
Uthmn, it is said, put the question to the Companions of the Prophet of
who knew Arabic best and who had the most beautiful hand. They then
decided on Sad as linguist, and Zayd as calligrapher. Thus, the former
dictated, and the latter wrote. If I amnot mistaken, we are here dealing with
a simplification of the established tradition. Abd al-Ramn b. al-rith
was probably left out, as no appropriate job could be found for him. If this
were the case, then the respective tradition would be dependent on the
established tradition and therefore of later origin.
The second of those two traditions
24
mentions Zayd besides one Abn b. [ii/52]
Sad b. al-, probably an uncle of the frequently mentioned Sad. Abn,
who had served as the Prophets amanuensis,
25
although according to al-
19
Silvestre de Sacy, Antoine I. Commentaire sur le pome nomm Rayya intitul
Akila, p. 427.Ibn Sad (al-abaqt, vol. 3, part 2). Biographien der medinischen Kmpfer,
p. 62, merely mentions that Uthmn ordered him to collect the Koran, then follows a tradi-
tion which reckons him among the members of the commission of twelve.al-Itqn, p. 430
at the top, mentions a tradition, in which Hni al-Barbar, a slave of Uthmn, relates that his
master once sent him to Ubayy b. Kab with several passages of the Koran (sras 2:261, 30:29,
and 86:17) written on the shoulder bone of a sheep, to have them corrected, which he did.
20
EI
2
; A. Jeffery, Materials for the history of the text of the Qurn, pp. 114116; Juynboll,
Encyclopedia, p. 93, col. 2, p. 265, col. 2, p. 479, col. 2.
21
22
Cf. Ibn Qutayba, p. 134; Ibn Sad (al-abaqt, vol. 3 part 2): Biographien der medinis-
chen Kmpfer, p. 62; Ibn Taghrbird, Annales, ed. Juynboll, vol. 1, pp. 58 and 97; Ibn ajar
al-Asqaln; Ibn al-Athr, Usd al-ghba f marifat al-aba, vol. 1, p. 50; al-Dhahab, Tadhki-
rat al-uff, vol. 1, p. 15.
23
Umar b. Muammad (IBN ABD AL-KF); Mabn li-nam al-man, part 3; G. Weil,
Geschichte der Chalifen, vol. 1, p. 167, note 3, according to al-Dhahab, Geschichte des Islams
[Tarkh al-Islm], cod. Gothan., p. 171 [sic].
24
Ibn Aiyya, fol. 25; al-Qurub, fol. 20
r
; al-abar, Tafsr, vol. 1, p. 20; Ibn al-Athr, Usd
al-ghba, vol. 1, p. 37.
25
al-Baldhur, Fut al-buldn, p. 473; al-abar, vol. 1, p. 1782; Ibn al-Athr, Usd al-ghba
f marifat al-aba, vol. 1, p. 50, etc.
abar, Tafsr (vol. 1, p. 2349), he died previously in 14/637 in the Battle of
Yarmk. The latest date of his death that I can find, 29/649, is likely to have
been calculated later with consideration of his alleged collaboration with
the Uthmnic recension, although it is still dated two years too early. Ibn
Aiyya and al-Qurub are thus certainly right when they consider the entire
tradition weak (af).
Yet another genre is represented by a transmission for which we are
indebted to the erudition of al-Suy:
26
Ibn Ab Dd al-Sijistn
27
from Muammad IBN SRN al-Anr
28
from
Kathr b. Afla.
29
When Uthmn was ready to have copies of the Koran pro-
duced, he gathered around him twelve men from the Quraysh and the Anr.
He then had the container with the Koran
30
brought fromUmars house, and
administered the oath to the group: whenever they were at variance with
themselves, they waited with the decision until it was determined who was
authentically the last person to have heard the passage.
The truth of the report that Uthmn did not hesitate to consult men, even
though they were living three days from Medina,
31
and whether this can be
connected with this tradition, is doubtful. Al-Qasalln,
32
using the same
source as al-Suy, includes among the twelve men both Ubayy b. Kab and
Muab b. Sad,
33
whereas Ibn Sad ((al-abaqt, vol. 3, part 2): Biographien
der medinischen Kmpfer, p. 62, l 19sq.) mentions Zayd b. Thbit in addition
to Ubayy, while al-Muttaq, Kanz al-umml, vol. 1, no. 4763, still adds to both
of them Sad b. al-, but reckons all to be Quraysh.
Not a word of this tradition is true. For all intents and purposes, the [ii/53]
story of this large commission simply aims at a better representation of
the Medinan community in establishing the Koran. The number twelve is
conspicuous, as it reminds us of the twelve chieftains (naqb) of the Children
of Israel (sra 5:15). The silence regarding most of the names also arouses
suspicion. Finally, as we shall see, the description of the procedure for the
establishment of the text starts from false assumptions.
26
27
EI
2
28
29
According to al-Khazrajs Khulat tadhhb al-kaml he transmitted fromUthmn and
Zayd, and was killed in the battle of al-arra.
30
raba.
31
al-Suy, al-Itqn, p. 139; al-Qurub; al-Muqni.
32
Vol. 7, p. 449, draws from Kitb al-maif of Ibn Ab Dd al-Sijistn.
33
Goldziher, Schools of Koranic commentators, p. 167; Juynboll, Encyclopedia, p. 506.
The greatest deviation fromthe established tradition is definitely the tra-
dition that al-Suy
34
adopted from the muaf work of AB BAKR (Mu-
ammad b. Abd Allh) IBN ASHTAH AL-IBAHN.
35
Abd Allh b. al-Zubayr said: One day a man went to Umar and told himthat
people were at variance with the Koran. Umar therefore decided to produce
the Koran in a unique reading, but he was murdered before he could embark
uponhis project. The manthenapproachedthe CaliphUthmnandrepeated
this offer. Thereupon Uthmn collected the codices, and ordered me to fetch
the codex belonging to isha. After we had checked and corrected it, he had
the other codices torn up .
36
This is likely also the reference in another source
37
that Umar was mur-
dered before he had collected the Koran. The tradition evidently endeav-
ours to belittle Uthmn in preference to his great predecessor, not unlike
the Biblical story of the Temple at Jerusalem that plays David off against
Solomon.
38
That the model codex originated from the private property of
isha is also tendentious, as this woman was the aunt of the Zubayr
through her sister Asm.
Critique of the Established Tradition:
The Biographies of the Members of the Commission,
and the Qualification of the Members for Their Task
Given that our investigation led us to reject all the divergent traditions as
well as their details about the composition of the Koran Commission, the
reliability of the established tradition must now be determined.
As far as the biography of the four members of the Commission is con- [ii/54]
cerned, Zayd b. Thbit was a Medinan of the Ban Najjr, a sub-clan of the
Khazraj. As a youngster he oftenacted as anamanuensis to the Prophet, par-
ticularly for his revelations,
39
and later produced the codex of afa.
40
Under
Uthmn, he held office as a q,
41
while according to others he was in
34
al-Itqn, p. 430.
35
Died 360/970; cf. G. Flgel, Die grammatischen Schulen [schools of grammarians],
p. 299.
36
shuqqiqat.
37
Ibn Sad (al-abaqt): Biographien der mekkanischen Kmpfer [Biographies of the Mec-
can combatants], ed. by E. Sachau, p. 212, l 14.
38
Wellhausen, Prolegomena zur Geschichte Israels, 2nd ed., p. 187sq.
39
al-abar, vol. 2, p. 836; above, p. 37.
40
41
al-abar, vol. 1, p. 3058; Ibn al-Athr, [al-Kmil,] vol. 3, p. 150.
charge of the treasury,
42
or perhaps the chancellery.
43
As an intrepid fol-
lower of the Caliph,
44
he remained loyal to the cause of the Umayyads, and
refused to pay homage to Al.
45
The year of his death is usually given as
45/665.
46
Sad b. al- was born shortly after the hijra. He was an Umayyad and a [ii/55]
favourite of the Caliph Uthmn. Among the numerous women he married
throughout his life, two daughters of this caliph are also mentioned.
47
After
Wald b. Uqba was deposed in 29/649, Sad b. al- became governor of
Kfa, a post he held until the end of 34/654.
Abd al-Ramn b. al-rith was of about the same age, and belonged to
the prominent family of the Makhzm. After his father diedby the notorious
plague of Emmaus, Umar married the latters widow Fima. Among Abd
al-Ramns women, apart from a daughter each of Ab Bakr and Zubayr,
a daughter of Uthmn is also mentioned, namely the same (Maryam) who
is also named among the women of Sad. According to the sources, he did
not play a part in politics. His connection with the Umayyads seems to have
been permanent and good since two of his daughters entered the harem of
respected members of these families, of Muwiya and Sad b. al-.
48
Abd Allh b. al-Zubayr, who was of about the same age as Abd al-Ra-
mn, also belonged to a prominent Meccan family. Through his mother
Asm bt. Ab Bakrs
49
side he was not only the grandson of Ab Bakr and the
nephew of isha, but subsequently became a stepson of the Caliph Umar.
He is said to have distinguished himself not only as a soldier but also by his
great religious fervour, and his assiduous praying and fasting. As the son of
42
al-Nuwayr, p. 259; al-Yaqb, vol. 2, p. 195; Ibn al-Athr, Usd al-ghba, vol. 2, p. 222.
43
Ibn al-Athr, vol. 3, p. 154.
44
al-abar, vol. 1, p. 2937; Ibn al-Athr, vol. 3, p. 119.
45
al-abar, vol. 1, p. 3070, 3072; Ibn al-Athr, [al-Kmil,] vol. 3, p. 154; Ibn al-Athr, Usd
al-ghba, vol. 2, p. 222.
46
Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod [Final illness and death of
Muammad], ed. F. Schwally, p. 116; Ibn al-Athr, Usd al-ghba, p. 222 [sic]; Ibn al-Athr
(al-Kmil), vol. 3, p. 378; Ibn Qutayba, [al-Marif,] p. 133. If it is true (Ibn Hishm, p. 561)
that in the Battle of the Trench (at the end of 5/626627) he was fifteen years old, he can
have been only twenty years old when Muammad died.
47
Ibn Sad (al-abaqt, vol. 5): Biographien der Nachfolger in Medina [biographies of the
successors at Medina], ed. by K.V. Zettersten p. 19sq.
48
IbnSad(al-abaqt): Biographiender Nachfolger inMedina [the successors at Medina],
edited by K.V. Zettersten (vol. 5), p. 1 sqq.; Ibn al-Athr, Usd al-ghba, vol. 3, p. 283sq.
49
Juynboll, Encyclopedia, s.v.
a man who later played an ambiguous role, to say the least, in the rebellion
against Uthmn, and then himself reached for the caliphate, he cannot be
counted upon to have been exactly a follower of the caliph.
50
The Procedure of the Establishment of the Text of the Koran, and the
Qualification of the Members of the Commission for Their Task
Judging by Zayd b. Thbits former activity, his qualification for Uthmns [ii/56]
Koran Commission is beyond doubt. Unlike anyone else, he belonged here,
and is therefore also the only person on whose collaboration with the Com-
mission there is unanimous agreement.
It is extremely rare that Muslim scholars express surprise that the first
choice did not fall upon Ibn Masd, who had embraced Islam at a time
when Zayd had not yet been born and who was endowed with still other
merits.
51
Yet they finally took solace in the thought that Zayd b. Thbit knew
the entire Koran by heart, whereas Ibn Masd knew only seventy sras.
This view is totally untenable and is based on the misinterpretation of a
tradition that states that the Prophet had already personally recited seventy
sras before Ibn Masd
52
when Zayd was still a youngster. This, however,
neglects the fact that Ibn Masd himself was the author of his own Koran
recension. On the other hand, this tradition fails to recognize the fact that
the Uthmnic Koran is nothing but a copy of the codex of afa, and that
therefore there was no one better qualified to head the copy work than Zayd
b. Thbit, the former copyist and editor of the model codex.
Conversely, it is extremely difficult to pass judgement regarding the rea-
son that led to the election of the three Qurayshites. Sad (Ibn al-) had
beengovernor of Kfa since 29/649. We donot knowwhether he was at Med-
ina when the Commission was constituted, nor whether he was summoned
by the Caliph for precisely this project, nor the reason for the appointment.
The fact that Sad was well acquainted with the conditions in Iraq, and
that he had earlier received udhayfa [Ibn al-Yamns] complaints
53
right
then and there, can hardly have been the reason, since these advantages
were irrelevant to a collaboration with the Koran Commission. Since the
50
Ibn Qutayba, p. 116; al-Nawaw, p. 34sq.; Ibn al-Athr, Usd al-ghba f marifat al-aba,
vol. 3, p. 161 sqq.; Wellhausen, Prolegomena zur ltesten Geschichte des Islams, p. 131 sq.
51
al-Qurub, fol. 19
r
.
52
Cf. above, p. 235.
53
biographies of the other two Quraysh supply no facts as to why they were
included in the Commission, we must look and see whether the tradition
itself might supply a clue.
This is indeed the case. Uthmn issued the following guideline for the [ii/57]
Koran Commission: If you disagree, write in the dialect of the Quraysh,
which was used in the revelations.
54
These words apparently justify the
conclusion that the Quraysh majority of the Commission should vouch for
the dialectical accuracy of the text. Another tradition also recognizes these
three menas the greatest authorities onthe Qurayshidiom, althoughincase
of disagreement leaving the final decision to the Caliph. For example, when
Zayd b. Thbit once wanted to write (with ) whereas the rest of them
preferred with (sras 2:249, 20:39), Uthmn declared the latter form
to be the true Quraysh form.
55
This interpretation is wrong, however. The
example alone is an unfortunate choice, since tbt is not at all genuine
Arabic but anAbyssinianloan-word. Tbhunis a gruesome deformity. Even
the controversy about such a word formation is totally against the spirit
of that early period. Neither the Prophet nor his closest successors and
followers had the least idea of scrupulous, philological exactitude.
56
The viewof Muslimscholars is closely connected with the frequently dis-
cussed question of the relation between the canonical recension and the
first collection of Zayd b. Thbit. Since both redactions were theoretically
identical,
57
according to their dogmatic prejudice founded on the divine ori-
gin of the Koran, Muslim doctors devised the theory of the seven aruf, or
the variant readings within the limits of the seven sets of readings, to jus-
tify the preparation of the second edition.
58
Accordingly, the first collection
contained the variants in seven different Arabic dialects,
59
while the Uth-
mnic recension constitutes only one dialect, the Quraysh idiom,
60
which
54
Ibn al-Athr, al-Kmil f marifat al-tarkh, vol. 3, p. 86, l 24sq.
55
al-Tirmidh in the Tafsr on sra 9 at the end; Muqni; Ibn Aiyya, fol. 25
v
. A tradition in
the Mabn li-namal-man, fol. 7
r
(cap. 2) reports a controversy inthe first editionbetween
Abn b. Sad and Zayd regarding this word.
56
Nldeke, Das klassische Arabisch und die arabischen Dialekte, p. 4.
57
Ibn Aiyya, cod. Sprenger 408; al-Qurub, fol. 22sq.; Abd al-Aad b. Muammad Abd
al-Aad al-arrn; Nuzhat; Leiden, cod. 653 Warner; al-Itqn, p. 145.
58
Jeffery, Materials for the history of the text of the Qurn, pp. 12.
59
al-Muqni; A. Silvestre de Sacy, Commentaire sur le pome nomm Rayya intitul
Akila, p. 425; al-Suy, al-Itqn, p. 133; al-Shshw, cap. 2. As regards other interpretations
of aruf cf. above, p. 252sqq.
60
al-Suy, al-Itqn, p. 140. rith (Ibn Asad) AL-MUSIB al-Anaz (d. 243/857) [EI
2
;
EQ; Sezgin in his GAS (vol. 1, pp. 639642) says: According to established opinion Uthmn is
the collector of the Koran, but this is not the case, rather he only initiated the acceptance
ultimately was used by the angel Gabriel when communicating the revela-
tions to the Prophet.
Generally, any tradition connecting the Uthmnic text in any way with [ii/58]
dialectal questions must be rejected, since the Koran is not written in a local
dialect at all but rather has a language identical to that of the pre-Islamic
poems. These, however, cannot possibly have beenwrittenindialectal form,
as their authors belonged to quite different tribes, living so far apart that the
texts would have to show strong idiomatic differences. Admittedly, when
fixing a text in such a defective script as Arabics, where vowels are gener-
ally not indicated and many consonants are expressed by the same sign,
some idiosyncrasies of the verbal presentation were simply not recogniz-
able at all. Still, the lexical and grammatical agreement is suchthat anactual
uniform language must be assumed. After all, given what we know about
linguo-geographical conditions in other parts of the world, it would be a
total contradiction if such a drastic disappearance of dialects were to have
occurred in large areas of the Arabian Peninsula. We are, thus, obliged to
conclude that the ancient poems, as well as the Koran, were composed in a
generally intelligible standard language,
61
the difference of which from the
local dialects of cultural centres like Mecca and Medina was naturally less
than from that in the more distant areas of the Peninsula.
Ifregardless of objectionsUthmn indeed intended to gather the [ii/59]
best authorities of the Quraysh dialect, he should have addressed himself
to other men and not to people who, although from Quraysh families, had
actually grown up in Medina.
Another seemingly simple solution to the problemseems to be indicated
by the remark of tradition that Zayd b. Thbit and his associates copied
the codex of afa.
62
To me it still seems extremely doubtful that the noble
of one particular variant reading which he, together with authorities from the Emigrants
(Muhjirn) andHelpers (Anr) agreedupon, for he fearedthat the Iraqi andSyriancommu-
nities might get embroiled in disagreement because of their variant readings. Cf. Juynboll,
Encyclopedia, p. 231, col. 2.
61
Julius Wellhausen, Arabisches Heidentum, 2nd ed., p. 216, speaks of a language above
the dialects existing in the illiterate literature of pre-Islamic Arabia. Nldeke comes to
a different conclusion in his Beitrge zur semitischen Sprachwissenschaft, p. 2, and Neue
Beitrge zur semitischenSprachwissenschaft, p. 4. Otherwise I aminagreement withNldekes
critical reviewof K. Vollers book, Volkssprache und Schriftsprache imalten Arabien [common
parlance and literary language in ancient Arabia] (1906).
62
Fa-nasakhh f l-maif. For references see above, p. 251 n. 1. Frequent in the biogra-
phies are expressions like fa-katab l-maif, e.g., Ibn al-Athr, Usd al-ghba f marifat al-
aba, vol. 3, p. 281 and 284; al-Nawaw, p. 281; al-Khazraj, Khula, s.v.; Abd al-Ramn b.
al-rith.
Quraysh would have lent themselves to such a troublesome and exacting
writing task, even if they were qualified. But since Zayd b. Thbit could not
have accomplished the enormous work single-handedlyit was a task of
three or four copiesI suspect that the actual copy work was done by a
staff of professional calligraphers, with Zayd b. Thbits activity limited to
the overall charge of the project. Whether the Quraysh mentioned were
indeed sufficiently qualified to assist Zayd in this activity is also shrouded
in obscurity. In any case, this combination is more plausible than the literal
interpretation of tradition.
Another conjecture can be connected with the report that Uthmn [ii/60]
attempted to collect as many revelations as possible.
63
Nearly every ver-
sion
64
of the established tradition agrees that sra 33:23 was originally lack-
ing but was later found at Khuzayma b. Thbit al-Anrs quarters, where-
upon the verse was put in its present place. According to al-abar (Tafsr,
vol. 1, p. 20), this verse was missing during a first revision of the new text,
whereas, during a second revision, the end of sra 9 was discovered on
the authority of another Khuzayma.
65
In the Tafsr, only al-Tirmidh men-
tions this ending on the latters authority. For such textual investigations,
the three Qurayshites could have rendered great service, particularly since,
through their connections with the most prosperous and recognized fam-
ilies, they must have been best informed as to copies of the revelations
in their possession. Those traditions, however, are undoubtedly based on
an actual or alleged incident that occurred in the course of Ab Bakrs
collection,
66
althoughlike everything else reported about the Uthmnic
recension with regard to variant readings or dialectal formsthey are in
stark contradiction to the clearly stated fact of the established tradition
that the Uthmnic Koran was merely a copy of the codex of afa. In
such circumstances, the preconditioned familiarity with codices, that
63
In this case one might actually refer to the passage quoted on p. 225 n. 31, but this is not
part of the established tradition. For this reasonalso the following traditioncannot be used,
which is transmitted by al-Tirmidh and al-Bukhr in Kitb al-Tafsr, and al-Suy, al-Itqn,
p. 142: Ibn al-Zubayr said, I spoke to Uthmn: Verse 241 of The Sra of the Cow is abrogated
by verse 234. Why do you still write it? Thereupon Uthmn replied: You ought to leave it,
cousin, for I do not intend to remove anything from its place. Incidentally, these verses do
not contradict one another in any way. Verse 234 permits the widow to remarry, when four
months and ten days have passed after the death of her husband. The right of the widow to
collect aliments from the estate of the man for one year applies onlyas it is clearly stated
in verse 241if she remains unmarried this long.
64
Only Flgels Fihrist, p. 24, and Ibn al-Athr (al-Kmil f l-tarkh), vol. 3, p. 86, do not
mention missing verses of the Koran.
65
Juynboll, Encyclopedia, p. 696, col. 1, n. 1.
66
Cf above, p. 223sqq.
was expected of the three Qurayshites was of no practical importance and,
thus, could not have been the reason for their selection for the Koran Com-
mission.
I can now offer only one other alternative for consideration, namely that [ii/61]
on account of their high social esteem those Qurayshites were expected to
add prestige to the enterprise inthe eyes of the public. Yet there was no need
for such a measure, as the Caliph had made the decision in mutual agree-
ment with the old Companions of the Prophet, and it was the best possible
recommendation. If it were done without their assent, one would at least
have expected that older and more mature men would have been selected.
Whoever does not find this argument convincing has no choice but to
consider the inclusionof the three Qurayshites to be ahistorical. Inthis case,
he is under the obligationtodemonstrate the forgery of the historical fact, be
it for the interests of the Qurayshparty or for other ambitions. Whoever ven-
tures to make this attempt will soon run into the greatest difficulties, as the
conflicting interests of the Umayya and Zubayr [Ibn al-Awwm]
67
families
certainly cannot be reconciled. The appointment of the three Qurayshites
is therefore likely to be true, even if we we knownothing about the purpose
and manner of their employment.
Furthermore, the questionis of little importance, since, inviewof the task
of simply producing several copies of a model text, the Commission could
only have played a subordinate role. Far more important was the basic deci-
siontocreate a uniformtext of the Koran. Uthmnassembledthe Compan-
ions and informed themof the situation. They attached great importance to
it, and followed the opinion of udhayfa [Ibn al-Yamn.]
68
Unfortunately,
we learn nothing of the members of this council. udhayfa, who fathered
the idea, would have certainly deserved a seat. Sad b. al-, too, would have
here found a more appropriate place than in the technical commission.
After having ascertained that Uthmns role in the establishment of the [ii/62]
uniform text of the Koran consisted of no more than commissioning a copy
of the most respected codex at the time in Medina, from now on we can no
longer speak of Uthmns collectionbut only of his recension. As a matter of
fact, the expression to collect is never used in the established tradition but
rather only inisolatedother traditions,
69
andthereafter frequently inthe his-
67
EI
2
; G. Juynboll, Encyclopedia, pp. 273, col. 1, 462; col. 2, 510, col. 1.
68
Ibn al-Athr (al-Kmil f l-tarkh), vol. 3, p. 86. When Al came to Kfa in 36/656 he
repliedaccording toIbnal-Athr, ibid., vol. 3, p. 87, al-abar, vol. 2, p. 747; al-Suy, al-Itqn,
p. 139 endto those who criticized unfavourably the Uthmnic recension of the Koran that
the Caliph had acted in accordance with the Companions.
69
al-Suy, al-Itqn, p. 430 according to Abd Allh b. al-Zubayr, translated above, p. 255.
torical
70
and Koranic
71
literature. It is also not appropriate to perpetuate an
expressionappearing insuchlate sources whenthe mainpurpose of the col-
lection is considered to be its destruction rather than its textual use.
72
This
conjecture is supported in al-Yaqb (vol. 2, p. 196) but otherwiseas far as
I can seetradition uniformly and exclusively considers the sole purpose
of the collection to be the establishment of the text. Moreover, that combi-
nation betrays a conception of the fate of the pre-Uthmnic redactions of
the Koran which, as we shall see in a later chapter, is by no means beyond
dispute.
The Arrangement of the Sras in the Uthmnic Koran
The purpose of this arrangement is by no means obvious. Among the points [ii/63]
of view that might possibly be considered, the one of content must be
excluded from the outset. It is known that not only the sras but also
many individual revelations are concerned with the most diverse subjects.
The chronological principle must also be excluded. It would contradict
not only the spirit of the periodas mentioned above when discussing
pre-Uthmnic redactions of the Koran
73
but would also be impractica-
ble for archival considerations because, apart from passages Muammad
himself added to earlier ones, the fragments probably became hopelessly
mixed up already in the first collection of Zayd b. Thbit. For such reason
Ikrima could rightly reply to Muammad IBN SRNs (d. 110/729) question
of whether the Koran had been arranged in chronological order that this
would have been impossible even if men and demons had joined efforts.
74
Later generations prohibit outright the paying attention to the chronology
of the Koran, condemning this as heretical.
Insuchcircumstances nothing else remains but to consider a mechanical
arrangement according to the length of the sras, a principle already sug-
gested by early Muslim scholars. Uthmn collected the Koran, compiled it
(allafa), and put together the long sras with the long ones, and the short
sras with the short ones.
75
Disregarding the first sra, the canonical edi-
70
al-Yaqb, ed. Houtsma, vol. 2, p. 196; Eutychius, Annales, vol. 2, p. 341.
71
al-abar, Tafsr, vol. 1, p. 20; Ibn Aiyya; Muqni; al-Suy, al-Itqn, pp. 138, 140, 430;
al-Qasalln, vol. 7, p. 449.
72
Th. Nldeke in the first edition of this work, p. 212.
73
Cf. above, p. 249.
74
75
al-Yaqb, ed. M.Th. Houtsma, vol. 2, p. 196 [
]; similarly Eutychius, vol. 2,

tion, indeed, begins with the longest chapters, followed by shorter ones, and
ends with the smallest ones. The system is not as strange as it might at first
seem, for when arranging chapters according to size it is equally rational to
start with the largest as well as with the shortest part. The observant reader
will likely find an equal number of examples of both methods in world liter-
ature. I would merely mention that the paragraphs within the order of the
Jewish Mishna are arranged by the descending numbers of the chapters.
76
As measurement of length, the early Muslims cannot have resorted to [ii/64]
the verse count; instead, it must have been the obvious number of pages
of an evenly written copy. After all, the length of the verses is so different
that, for example, the seventh sra has thirty more verses than the fourth
sra, even though the latter one is one page longer, the twentieth sra has
even five more verses than the ninth sra, despite being less than half as
long, and the twenty-sixth sra takes up approximately only one fourth of
the pages of the second sra, while still consisting of only sixty-nine verses
less than the latter. The general guideline of the overall proportion has been
followed only very roughly and incompletely. In order to supply a tangible
illustration of these facts, I produced the following table, which lists the
sras of the canonical Koran, including the number of verses as well as the
outer dimensions in pages and lines of Gustav Flgels 1858 Arabic edition of
the Koran,
77
adding the perfect arrangement that corresponds to the exactly
descending proportion.
Canonical edition Sequence of
the sras of
No. No. Volume the canonical
of of edition in strict
sra verses Pages Lines arrangement
a b c d e
1 7 - 5 2
2 286 22 11 4
3 200 13 11 3
4 175 14 4 7
5 120 10 18 6
the sras of
a b c d e
6 165 11 16 5
7 205 13 3 9
8 76 4 20 11
9 130 9 21 16
10 109 7 1 17
p. 341, and al-Itqn, p. 140: Uthmn collected the leaves in a codex arranged by its sras
(murattaban li-suwarih;) al-Suy, al-Itqn, p. 145: Uthmn ordered them to have the long
sras follow one another.
76
H.L. Strack, Einleitung in den Talmud, p. 25; from Abraham Geiger, Einiges ber Plan
und Anordnung der Mischnah, p. 489sqq.
77
Not counted are the headings of the sras (name, place of revelation, basmala); the
incomplete lines are reckoned to be complete. When there is no heading the page of Flgels
Koran consists of twenty-two lines.
the sras of
a b c d e
11 123 7 7 10
12 111 6 20 12
13 43 3 7 18
14 52 3 8 26
15 99 3 18 28
16 128 7 6 20
17 111 6 4 33
18 110 6 6 24
19 98 3 18 22
20 135 5 7 8
21 112 4 9 21
22 178 5 - 40
23 118 4 7 39
24 64 5 6 27
25 17 3 14 23
26 227 5 15 29
27 95 4 18 37
28 88 5 4 19
29 69 4 - 25
30 60 3 6 43
31 34 2 2 34
32 30 1 1 14
33 73 5 7 13
34 54 3 9 42
35 45 3 2 30
36 83 3 - 41
37 182 4 - 35
38 88 3 1 36
39 75 4 5 38
40 85 4 18 15
41 54 3 3 46
42 53 3 7 48
43 89 5 11 57
44 59 1 1 47
45 36 1 21 31
46 35 2 1 45
47 40 2 6 58
48 29 2 7 56
49 18 1 10 59
50 45 1 12 55
51 60 1 12 51
52 49 1 7 50
53 62 1 11 32
54 55 1 11 44
the sras of
a b c d e
55 78 1 15 53
56 96 1 18 54
57 29 2 7 49
58 22 1 20 60
59 24 1 18 48
60 13 1 9 52
61 14 - 20 67
62 11 1 - 68
63 11 - 17 69
64 18 1 - 65
65 12 1 4 72
66 12 1 - 74
67 30 1 7 76
68 52 1 7 62
69 52 1 4 64
70 44 1 - 66
71 29 - 21 70
72 28 1 3 71
73 20 - 18 61
74 55 1 2 77
75 40 - 16 73
76 31 1 1 78
77 50 - 20 79
78 41 - 18 63
79 46 - 18 75
80 42 - 14 80
81 29 - 11 89
82 19 - 8 81
83 36 - 17 85
84 25 - 10 84
85 22 - 11 88
86 16 - 10 98
87 19 - 7 82
88 26 - 10 90
89 30 - 13 92
90 20 - 8 87
91 15 - 6 96
92 21 - 8 86
93 11 - 4 91
94 8 - 3 1
95 8 - 4 93
96 19 - 7 95
97 5 - 3 99
98 8 - 9 100
the sras of
a b c d e
99 8 - 4 101
100 11 - 4 104
101 9 - 4 94
102 8 - 3 97
103 3 - 2 102
104 9 - 4 105
105 5 - 3 107
106 4 - 2 109
the sras of
a b c d e
107 7 - 3 103
108 3 - 1 106
109 6 - 3 110
110 29 1 11 89
111 5 - 2 112
112 4 - 2 113
113 5 - 2 114
114 6 - 2 108
This table shows that the two arrangements differ so markedly from one [ii/66]
another that actually only six sras, namely 3, 12, 21, 51, 80, and 84, are at the
proper place; one wonders why the system has been carried out in such an
imperfect way when there were no difficulties whatsoever in its consistent
application.
Many an inaccuracy could be explained by the fact that the editor was [ii/67]
dealing with many sras in copies of a variety of leaf sizes and hands, thus
disguising the true scope. This, however, can hardly explainthe most glaring
and obvious violations of this principle, such as, for example, how sras
13, 14, and 15, with a size ranging from 3 to 3 pages, ended up between
sras of seven pages, and why sra 8 (five pages) is placed before sra 9
(ten pages), or why sra 32 (1 page) stands before sra 33 (5 pages).
On the other hand, it is also difficult to believe that afas copy of the
Koran should not have been of the uniform shape of a codex. One is thus
tempted to assume that the current order of the sras must go back to
the incomplete textual condition of afas copy, which Zayd b. Thbit,
either because of personal qualms or being under the spell of contemporary
prejudices, did not dare to change, or at least not drastically. It cannot even
be excluded that the hands of the compiler of this copy had already been
tied. However, when we realize that the redactions of Ubayy b. Kab and Ibn
Masd, although differing from one another as well as from the Uthmnic
Koran, reveal its general arrangement without, however, getting any closer,
it would consequently seem that the logical conclusion was conveniently
avoided. The rationale behind this strange procedure might have been the
reluctance to produce something perfect, thereby provoking the jealousy
of strange demonic powers, a superstition particularly widely-held among
primitive people.
There are only two exceptions to the rigid arrangement, both of which we
can appreciate reasonably well. One of them is the most serious departure
from the principle found in the canon. It concerns the position of the
five-line Ftia, immediately before the longest chapter, the so-called Sra
of the Cow. The other exception is insignificant and concerns the fact that
the shortest srathe one-line hundred and eighthis not the last one,
but rather two of the distich type. The fact that these two sras (113 and 114),
also like the first sra, are prayers in terms of content would support the
assumption that they are purposely placed where they now are. Although
we do not precisely knowwhat religious or superstitious idea motivated the
editor, it does not seem unduly strange that he considered it appropriate to
open such a holy book with a prayer of thanks, and finish it with protecting
spells. The possibility that the respective sras were originally not a part of
revelation has been previously discussed in detail.
78
As far as the total number of sras in the Uthmnic Koran is concerned, [ii/68]
a system is no more apparent than in the canons of Ubayy b. Kab and Ibn
Masd. It is purely accidental.
The Mysterious Letters Preceding Certain Sras
Twenty-nine sras of our Koranare precededby single letters of the alphabet
or by combinations of letters (logograms), which tradition considers to be
part of the Koran. They read as follows:
preceding sras 10, 11, 12, 14, 15
preceding sras 2, 3, 29, 30, 31, 32
preceding sra 13
preceding sra 7
preceding sras 40, 41, 43, 44, 45, 46
79
preceding sra 42
preceding sra 38
preceding sra 27
preceding sras 26, 28
preceding sra 20
preceding sra 50
preceding sra 19
preceding sra 68
preceding sra 36
78
Cf. above, pp. 89sq, 231 sq., and 245sq.
79
These sras were thus collectively called .
Muslims took great pains to lift the veil of mystery off of these letters. [ii/69]
Many explanations refer to Ibn Abbs and other celebrities of the first cen-
tury, or even to all the Companions of the Prophet, who ought to have been
well informed. However, their comments, like the exegetic tradition in gen-
eral,
80
are strongly suspect of having been forged by later Muslims for the
sake of justifying better their own opinions, so that this criticism must con-
form to internal arguments. In recent times it has become fashionable to
ignore the traditional attempts of interpretation. But this is unwarranted.
As will become evident later, Christian scholarship of the Occidenteither
accidentally or by borrowingfrequently created identical or similar con-
cepts, but evenwhere they pursued their ownway they did not always arrive
at a better argument. The Muslim explanations, of which I can naturally
present only a selection from among the remarkable examples,
81
can be
divided into two major groups.
The first group recognizes in the logograms abbreviations of words or
phrases:
(al-Itqn, p. 486) ; :
(al-Itqn, p. 490) (al-Itqn, p. 486) ; :
(al-Bayw, on sra 13:1) :
(Itqn, p. 486) ;
; ; :
(al-Itqn, p. 493)
(al-Itqn, p. 487) :
(al-Itqn, p. 487) :
;
; ; :
(al-Itqn, p. 493)
(al-Itqn, p. 487) :
(al-Itqn, p. 487) :
(al-Itqn, p. 487) :

(Itqn, p. 487) ; :
(al-Itqn, p. 493)
;
; ; :
; ;
80
Cf. in this connection the remarks in the literary-historical appendix.
81
Other material can be found in O. Loth, Tabaris Korankommentar, pp. 603610.
(al-Itqn, p. 486sq.)
(Itqn, p. 486) (Itqn, p. 487) ; (Itqn, p. 493) :
(al-Itqn, p. 493)
:
It is evident that all these interpretations belong to the realm of unlimited [ii/70]
possibilities. Since every abbreviated word can be replaced at convenience
by one or several letters, the interpretation of such abbreviations is con-
versely subject to the same arbitrariness. The only interpretation that can
be substantiated is the one of (sra 68) as al-t, fish. Because the North
Semitic nn came to mean fish when assimilated to Arabic,
82
and since
Jonah is otherwise also called Dh l-Nn,
83
and in sra 68:48 named ib
al-t, it is conceivable that might be a kind of name or heading of the
sixty-eighth sra.
Although in the second group there is agreement that the letters do not [ii/71]
represent abbreviations, in other respects the approach is quite different.
(a) The letters are mysterious names for the Prophet, which defy fur-
ther interpretation ( ,
84
al-Suy, al-Itqn, p. 492; , al-Itqn, p. 493;
,
85
al-Itqn, p. 492); for particular sras ( , Ibn al-Khab al-Da-
hsha, Tufa, p. 29; , Tufa, p. 29,
86
al-Itqn, p. 488, Tufa, p. 29); for
a mountain ( , al-Itqn; p. 493; a mountain surrounding the
earth, al-Itqn, p. 493); for an ocean (the ocean on which there is
the Throne of the All-merciful, or where the dead become alive, al-
Itqn, p. 493); or finally for writing table
87
or ink-well (, al-Itqn,
p. 493).
(b) The letters are signsderived, by the way, from the numerical value
order of the North Semitic alphabetwhich are here read symboli-
cally or apocalyptically; for example, = 71 years (al-Itqn, p. 489sq.),
= 271 years (al-Itqn, p. 489), = 14 = moon, on account of the cor-
responding number of the stations of the moon (al-Itqn, p. 493), etc.
(c) The letters are auxiliaries to attract attention, either to lead the busy
Prophet to the voice of Gabriel, or to astonish the Prophets listeners
82
Mufaaliyyt, ed. H. Thorbecke, no. 16, l 39.
83
al-Bayhaq, al-Masin wa-l-masw, ed. Schwally, p. 32, l 2.
84
As a consequence, h has become for Muslims a common mans name. According
to Ibn Jubayr and ak in al-Bukhr, Tafsr on the twentieth sra, is said to mean in
Nabataean o, man, which of course is nonsense.
85
Also Ysn has become a common Muslim mans name.
86
al-Baldhur, ed. de Goeje in the glosses s.v.; Ibn al-Athr, Usd al-ghba, vol. 4, p. 322.
87
This is inferred from the conjuration opening the sixty-eighth sra: By the pen, and
what they inscribe.
by this unusual method in such a way that they pay more attention to
the revelations (al-Itqn, p. 491 sq.)
(d) The letters attest that the revelations were put inwriting inthe familiar
and generally intelligible Arabic alphabet. They are very intelligibly
chosen, together representing exactly half (14) of the alphabet, and
contain also half of every phonetic symbol (al-Itqn, p. 492).
(e) The letters are dividers (fawil [ ]) of the sras (al-Itqn, p. 494). [ii/72]
Amere first glance reveals that the fantastic ideas, the numerical acrobatics,
and the other theories of the second group are as far-fetched as the arbitrary
interpretations of the alleged abbreviations. Moreover, the important ques-
tion why only twenty-nine sras are preceded by such mysterious letters is
not even touched upon.
Among the Western works on the subject, only those works that help our
understanding of the problem deserve consideration. Theodor Nldeke, in
the first edition of the present work,
88
regrets that it has not been possible
to find definite facts about the meaning of the logograms, particularly as
this would have undoubtedly led to important conclusions regarding the
composition of the Koran. They do not originate from Muammad at all
because it would indeed be strange if he had put such unintelligible signs
in front of his revelations which, after all, were intended for everyone;
but rather, they represent letters and clusters of letters, probably marks of
possession, originating from the owners of the Koranic copies which were
used in the first collection of Zayd b. Thbit, and which found their way
into the final version of the Koran by mere carelessness. This is supported
by the whole string of successive sras of different periods furnished with
the sign , suggesting the idea that we are dealing here with a copy of the
original that contained these sras in the identical order. Further, it would
not be impossible that these letters were no more than monograms of the
owner. The following abbreviations are conceivable: = = al-Zubayr,
= al-Mughra, = = ala or ala b. Ubayd Allh,
89
and =
Abd al-Ramn. In the middle letter might indicate , the
two final letters, , etc. But even the possibility of variant readings make
everything uncertain.
This view was well received. It is supported by the fact that the mono- [ii/73]
grams are exclusively found at the head of sras whichoriginally do not con-
stitute a unit. On the other hand, the individual explanation of the names
88
p. 215sq.
89
Juynboll, Encyclopedia, s.v.
is as arbitrary as that of the early Muslim authorities. The explanation of
the long logograms, and , poses insurmountable difficulties. The
assumption that the letters entered the canonical edition by mere careless-
ness is totally out of the question, since anauthority like Zayd b. Thbit, who
twice had to pass judgement on the form of the text, cannot conceivably
have been capable of such a lapse.
Encouraged by the survey of the interpretation of the mysterious letters
which al-abar supplies in the introduction to his great commentary on
the Koran, but particularly by Ikrima al-Barbars alleged explanation that
the combination of the three monograms, , and , produce the word
al-Ramn, the Most Gracious, Otto Loth
90
recognizes also in the other
monograms indications of certain catchwords of the Koran. When he then
recalls Aloys Sprengers conjecture
91
that the letters might partly be read
also inreverse orderroughly like those onsealshe reckons, for example,
to be an abbreviation of and of , and of
; and
he then puts , , , and possibly also in relation to the familiar
words known from sra 56:78, l yamassuhu ill l-muahharn, and ,
sra 42:1 to the words laalla l-sata qarb in 42:16. These combinations
are an honour to the authors ingenuity but are too arbitrary to be taken
seriously. Particularly suspicious is the reckless transposition of letters. I
have never encountered anything like this except in Arabic calligraphy,
when an empty space had to be decorated artistically. More valuable are
Loths general views that serve as an introduction to his arguments. First of
all, he turns against Nldeke. It remains incomprehensible how the editors
of the Koran could include the private notes of the former owners in the
Holy Book. On the other hand, the argument that Muammads inclination
for the wonderful and the obscure led him personally to devise such signs
would not seem to be strange. Since all the relevant sras belong to the
late Meccan or early Medinan period, when Muammad was approaching
Judaism, the letters might be Kabbalistic figures. Not all of these objections
are of identical importance. For all appearances, the question of whether or
not we can imagine Muammad capable of such mysterious ciphers can be
answered in the affirmative as well as in the negative. Although we know
nothing definite regarding the date of the Jewish Kabbalah, most probably
it is several centuries later than the Koran.
As Loth continues in the same context on p. 603 of his article, the impar- [ii/74]
tial reflection on those sras reveals that their beginning contains mostly
90
Otto Loth, Tabaris Korancommentar, pp. 588610.
91
Das Leben und die Lehre des Mohammad, 2nd ed., vol. 2, p. 182sq.
an allusion to the preceding letters. He is likely to be thinking mainly of the
frequent introductory formula, those are the yt of the Book (sras 10,
12, 13, 15, 28, and 31, and similarly sra 27). It is certainly quite possible to
translate yt with symbols,
92
and consider parts of the alphabet symbols
of the revelation. This, however, is in contradiction to the common Koranic
meaning of yt miraculous sign or verse (sras 11:1; and 41:2 and 44), and
that in the similar introduction to sra 2:1, that is the revelation, wherein
is no doubt, the demonstrative does not refer to the preceding logogram
A L M but unconditionally to what follows.
93
It is more likely that in sra
3:1 there might be a reference to the logogram A L M, provided these let-
ters can be interpreted as abbreviation of the words, Allhu l ilha ill
huwa l-ayyu l-qayym. But more likely, the first verse, which is identical
with the so-called Verse of the Throne of sra 2:256, is but an old inter-
pretation of that logogram and the original beginning of the sra is verse
two.
94
On the other hand, Loth is quite correct in his observation that in the [ii/75]
initial verses of the coded sras their content is nearly always identified
as the revealed Word of God.
95
There are, of course, considerably more
sras with such beginnings that lack logograms (sras 18, 24, 25, 39, 52, 55,
97), whereas other sras are also preceded by letters but have an entirely
different beginning (sras 29 and 30). However, the passages upon which
Loth bases his argument might possibly be too numerous to consider this a
mere accident.
Based on this and other considerations, Nldeke subsequently aban-
doned his earlier opinion. I thinkas he saysMuammad seems to have
wanted these letters to be a mystical reference to the archetypal text in
92
Downright characters, letters like Late Hebrewt, andSyriac tt, is never the Arabic
equivalent of yt. Essentially different is the writing on the heavenly gold plates, from
which Joseph Smith translated the Book of Mormon with the help of the stones of the seer,
because this alphabet consists of a wild fantastical and unsystematic sequence of all possible
characters (caractors), from which no alphabet at all can be enucleated. Cf. Eduard Meyer,
Ursprung und Geschichte der Mormonen, pp. 3383.
93
Likewise, the demonstrative pronoun in the formula, tilka yt Allhi (sras 2:253 and
3:104) refers to what follows. As far as I see, in this case as in all the other passages mentioned
above, this is also the view of the entire exegetic tradition.
94
Although in this case, the verbal predicates had better be put into the passive voice.
95
More particulars emerge from the following compilation: sra 2:1, dhlika l-kitbu; 3:2,
nazzala alayka l-kitba wa-anzala l-furqna; 7:1, kitbun unzila ilayka; 10:1, 12:1, 13:1, 15:1,
26:1, 28:1, 31:1, tilka ytu l-kitbi; 11:1, kitbun ukimat ytuhu; 14:1, kitbun anzalnhu ilayka;
20:1, manzalnalaykal-furqna; 27:1, tilkaytul-Qurni wa-kitbinmubnin; 32:1, 40:1, 45:1,
46:1, tanzlu l-kitbi; 36:1, 38:1, 50:1, wa-l-Qurni; 41:1, tanzlun min al-Ramni l-Rami; 42:1,
ka-dhlika y ilayka; 43:1, 44:1, wa-l-kitbi l-mubni; 68:1, wa-l-qalami wa-m yasurn.
heaven. Toa manwhoregardedthe art of writing, of whichat the best he had
but a slight knowledge, as something supernatural, and who lived amongst
illiterate people, an A B C may well have seemed more significant than they
do to us who have been initiated into the mysteries of this art from child-
hood.
96
This point of view has the advantage of better relating the logograms to [ii/76]
the opening verses of the respective sras. On the other hand, this presup-
poses a measure of illiteracy of the Prophet that is incompatible with my
previous remarks.
97
This objection does not apply ifaccording to Nldekes additional re-
marksthe mysterious, solemn impression which Muammad attempted
to make is related to the mass of his listeners. If this had been the Prophets
only intention, it would be difficult to comprehend why the logograms are
only foundat the beginning of chapters but not evenonce infront of individ-
ual revelations in the middle of sras. This fact cannot possibly be acciden-
tal, regardless of whether normally the inconsistent use of the logograms
is based on old distortions,
98
or on the imperfect condition of the texts of
the revelation at their first collection. In these conditions, Nldekes more
recent opinionwhich, by the way, is connected most closely with some of
the Muslim theories set forth above, p. 269 (c and d)again raises doubts
and enforces the conjecture that the logograms are somehow related to the
editorial work of the sras.
The real existence of the logograms leads back to a very early period.
Because of the connection of the Uthmnic Koran with its original, the
logograms must already have been part of afas copy. Apparently, Ibn
Masd also had them in his recension, since it is reported that he read the
logograms of sra forty-two without the letter ayn.
99
When even Loth, and
nowadays Nldeke, plead for Muammads authorship, they are in agree-
ment with tradition, which considers the logograms to be part of the reve-
lation. The knowledgeable Zayd b. Thbit would have hardly included the
strange scribble in the final redaction if he had not been convinced of
the authority of the Prophet.
100
If Muammad is indeed the author of the
logograms then he must also be the editor of the ciphered sras. Although
96
His article Koran in the Encyclopdia Britannica, 11th ed., p. 904; and Orientalische
Skizzen, p. 50sq.
97
Above, p. 209sq.; and before, p. 36sqq.
98
Nldeke, Orientalische Skizzen, p. 51.
99
al-Zamakhshar and al-Bayw, s.v. This is also reported of Ibn Abbs.
100
Cf. above, p. 271.
this contradicts earlier established opinion, it would conform to our previ-
ous observation that the Prophet kept an amanuensis to whom he dictated
his revelations,
101
that he early set out to create his own book of revela-
tion,
102
and that the way of combining pieces of different provenance but
similar content in certain sras leaves the impression of originating from
the Prophet himself.
103
Unfortunately, this point of viewis in no way helpful
for the question of the importance of the individual case.
H. Hirschfeld
104
still persists totally in Nldekes early point of view, with [ii/77]
the exceptionthat he associates every individual letter of the logograms with
an explicit name. He thus arrives at the following equations, which, as he
himself admits, are purely hypothetical:
AL the definite article
M Mughra
S afa
R (Z) Zubayr
K Ab Bakr
H Ab Hurayra
N Uthmn
ala (Ibn Ubayd Allh)
S Sad (Ibn Ab Waqq)
udhayfa [Ibn al-Yamn]
Umar or Al, Ibn Abbs, isha
Q Qsim b. Raba
Consequently, a single letter should indicate that the sra following goes [ii/78]
back to the copy of this owner, whereas sras consisting of several letters
were found to be in partial or total possession of several persons. This must
have been governed by the principle of uniting the monograms, which
actually belonged in front of the fragments of the current chapters, at the
beginning together with the others. Whether the marks of possession go
back to copies of the respective owners or editors can be decided no more
than the question of why Zayd b. Thbit retained or added them. That sras
2 and 3 are nowso far apart fromthe four other, equally ciphered sras 29 to
32 must be explained simply by the systemof arranging according to length
that governed the collection. However, the hypothesis regarding the marks
of possession can be maintained only if the logograms do not go back to the
101
Cf. above, p. 209sq.; and previously p. 36sqq.
102
Cf. above, p. 80, 81, 106, 117, 129, and 175.
103
Cf. also the conjecture mentioned below, p. 274 n. 105.
104
Newresearches into the composition and exegesis of the Qoran, pp. 141143.
Prophet.
105
This is for Hirschfeld self-evident, since after all that we know,
Muammad cannot have collaborated in the composition of the sras. On
the contrary, I have explainedthe error of this assumptionmore thanonce.
106
105
Conversely, it would indeed be compatible with Muammads authorship if the names
of his secretaries were hiddenbehind the logograms. However, not a single transmitted name
of the writers of the revelations can be identified.
106
Cf. above, p. 209sq. and p. 273sq.
THE BASMALA
Whereas the afore-mentioned logograms, which are encountered in mani- [ii/79]
fold forms, can only be found in front of certain sras and are considered
part of the text of the revelationand are therefore reckoned to be the
first verse of the respective srathere is yet another unchangeable phrase
placed at the beginning of all the sras in the Koran, excepting one, albeit
without usually being considered a part of the actual text.
1
This is the for-
mula bismillh al-Ramnal-Ram, which is shortened and called basmala
or tasmiya. Given that it is not reported anywhere that it was introduced
only by the Caliph Uthmn, it must have existed already in the copies of
afa and other pre-Uthmnic recensions.
2
Muammad was undoubtedly
familiar with the formula, as, after all, he had it placed at the head of the
Pact of udaybiyya in 6/627628.
3
Many letters and epistles to the pagans,
Jews, and Christians of Arabia also open with it.
4
The basmala occurs even
once in the very text of the Koran (27:30) at the beginning of the epistle of
Solomon to the Queen of Sheba. Since the basmala otherwise occurs only
at the beginning of sras, suggesting its editorial origin, the Prophet can be
accountable for it only in instances where a particular sra received its cur-
rent form from him. On the other hand, an earlier origin seems again to be
1
The MeccanandKfanreaders recognizedthe basmala as a separate verse, whereas the
readers of Medina and Syria hold that it is only placed there to separate the sras (kutibat lil-
fal wa-l-tabarruk bi-l-ibtid). The difference is also of practical importance to the respective
school. The schools following the first opinion, like the Shfites, pronounce the basmala in
a loud voice in the liturgy, whereas, for example, the followers of the second group of readers
utter it in a lowvoice. Cf. al-Zamakhshars commentary on the Koran, Cairo, 1308, vol. 1, p. 21,
and above, p. 94sq., on sra 1.
2
Cf. above, p. 249.
3
Cf. above, p. 132; Ibn Hishm, p. 747; al-abar, vol. 1, p. 1546.
4
Ibn Sad (al-abaqt (vol. 1, part 2): Biographie Muhammeds; Ereignisse seiner medinis-
chen Zeit [Biography of the Prophet; events of his Medinan period], p. 28sq.; J. Wellhausen,
Seine [Muammads] Schreiben und die Gesandtschaften, nos. 24, 30, 35, 47, 75. According
to one tradition (Sendschreiben Nr. 10) Muammad first wrote bismik Allhumma like the
Qurayshcf. also Sendschreiben Nr. 61until the revelation of sra 11:43, and subsequently
bismillh until the revelation of sra 17:110, bismillhi l-Ramn until the revelation of sra
27:30, and from then on adding also al-Ram. According to a tradition in al-Wid, Asbb
al-nuzl, Cairo ed., pp. 6 and 10, this complete form of the basmala is the earliest revela-
tion.
278 the basmala
indicated by the fact that the basmala predates the logograms which, what-
ever their meaning might be, nevertheless, in one way or another, also seem
to be connected with the redaction.
5
Of all the sras of our Koran it is the ninth sra alone that lacks the [ii/80]
basmala. Muslims attribute this to intentional omission. Among the diverse
reasons they advance there is only one worth mentioning. According to it,
Muammads Companions couldnot agree whether or not tocombine sras
eight and nine into a single one and therefore reached a compromise and
left a free space between the two sections, although without placing the
sign of division, the basmala.
6
This alleged lack of resolution among the
Companions is however incomprehensible, as not only is the entire content
of the two sras considerably different and chronologically far apart, but
also the first verse of the ninth sra stands out prominently as the beginning
of a new section. On the other hand, it appears to me that the device of the
editors to help them out of the dilemma is too ambiguous and trivial. In
this instance, it is far more natural and simple to consider an accident and
assume that in the canonical recension, or the original text, the basmala
between the two sras was either omitted because of a writing mistake or
disappeared because of external damage and that people later did not dare
to make any alterations in the state of the transmitted form. It is known that
the development of many peculiarities in the textual form of the Hebrew
Bible is due to similar conditions.
5
Cf. above, p. 273sq.
6
Cf. al-Tirmidh in the chapter tafsr on sra 9:1; al-Bayw, and generally the Commen-
tators; al-Farr al-Baghaw; al-Khab al-Tibrz, Mishct, Fail al-Qurn at the end; Ab
l-QsimUmar b. Muammad[IBNABDAL-KF, titulus operis me latet], cod. Catalogus cod-
icumorientaliumBibliothecae Academiae Lugduno Batavae, vol. 4 (1864), p. 5, Ms. MDCXXXIV
= 1634 (= cod. 674 Warner); Sezgin, GAS, vol. 1, p. 16, no. 24.
SECTARIAN REPROACH AGAINST THE UTHMNIC TEXT
The Alleged Corruption of the Text of
the Koran by Ab Bakr and Uthmn:
Reproach of Christian Scholars of the West
Some Christian scholars in the West suspected the text of the Koran, both [ii/81]
the Uthmnic recension and its original version, to have been the subject
of deliberate forgeries. The first among European scholars to suspect the
genuineness of certain verses in the Koran was Silvestre de Sacy,
1
who ques-
tioned the authenticity of sra 3:138. G. Weil added to this both verse 182
and sra 39:31 sq.,
2
later extending this to the related sras 21:35sq. and
29:57
3
by blaming for these interpolations no less a person than the Caliph
Ab Bakr, who allegedly initiated the first collection. The main argument is
the tradition that Umar did not want to believe in Muammads death, and
loudly proclaimed this conviction in front of all Muslims until persuaded by
Ab Bakr by reciting sra 3:138 or 39:31 sq., or both passages, which refer to
Muammads death. But it had occurred to Umar, or, as other versions state,
to the Muslims, that they had never heard this revelation.
4
This, however,
might be nothing but a harmless reference to the fact that at the moment
of dismay over the unexpected death of the Prophet Umar and his friends
did not recall the respective verse,
5
a viewwhich seems to correspond to the
tacit consent of tradition. Conversely, it is difficult to believe that a forged
quotation from the Koranparticularly something that Ab Bakr would
1
Silvestre de Sacy, Ettaberi Annales, [review,], Journal des savans, 1832, p. 536.
2
Weil, Mohammed der Prophet, p. 350, and his Historisch-kritische Einleitung in den
Koran., 1st ed., p. 43.
3
Weil, Historisch-kritische Einleitung in den Koran, 2nd ed., p. 52sqq. This enlarged revi-
sion is evidently occasioned by Nldekes objections raised in the first edition of the present
work, particularly p. 199, bottom.
4
IbnHishm, p. 1012sq.; al-abar, vol. 1, p. 1815sqq.; IbnSad(abaqt vol. 2, part 2): Letzte
Krankheit, p. 52sqq.; al-Yaqb, ed. Houtsma, vol. 2, p. 127; al-Shahrastn, ed. Cureton, vol. 1,
p. 11; al-Bukhr, al-Maghz, cap. 85, Bb al-khalq, cap. 101 (Fail Ab Bakr) 9 and the other
relevant parallels in my [Schwallys] notes on Ibn Sad.
5
A similar case is reported by umayd b. Ziyd in al-Farr al-Baghaw: He asked Kab
al-Qura [EQ: b. Quraya; Juynboll, Encyclopedia, p. 343, col. 2] about Muammads Com-
panions and was told that they were all in Paradise. When in addition sra 9:103 was cited as
proof, he said: It seems to me that I never read this verse.
280 sectarian reproach against the uthmnic text
have invented on the spur of the momentcould easily be imposed upon a
man like Umar. The evidence of Ab Hurayra, who is one of the authorities
and allegedly also did not knowthat verse, does not say much. After all, this
traditionist is neither one of Muammads earliest Companions, as he did
not become a Muslim until 7/628, nor can his words claim credibility, since
later research has exposed him more and more as a liar.
6
Weils interpretation also contradicts Muammads philosophy as we [ii/82]
know it. It is without a shadow of doubt that certainly during his last years
Muammad did not want to leave the believers in doubt about his own
mortality. Onthe contrary, his intent was touse every chance todemonstrate
by way of revelation (sras 17:95, 18:110, and 41:5) that he was only a mortal.
After all, not only verses 3:138 and 39:31, which consider Muammads death
inevitable, but also verses 3:182, 29:57, and 31:35sq., express the truism that
all men must die and thus fit perfectly into this context.
This proves the authenticity of the respective verses in every way. Yet it [ii/83]
is necessary to go further and criticize the basis and the point of depar-
ture of Weils error, namely the tradition itself. The whole controversy over
Muammads corpse is strongly suspect of having been invented to defend
his human nature against groups who, with reference to certain Jewish and
Christianexamples,
7
consider it self-evident that a prophet sent by God can-
not die a natural death but must rather disappear in a mysterious way. If,
then, Muammads death had really been such a stumbling block, the belief
in his reappearance ought to have left more traces in tradition. However, it
was not until the reign of Uthmn that the man appeared who related this
concept to the person of Muammad, namely Abd Allh b. Sab.
8
Hartwig Hirschfeld
9
is unable either toput Weils lame arguments backon
their feet or to refute Nldekes objections. In spite of this, Hirschfeld sticks
to the interpolation of sra 3:138 by insisting on the new evidence that all
Koranic passages containing the name Muammad (3:138, 33:40, 47:2, and
6
Goldziher, Muslimstudies, vol. 2, p. 56; Caetani, Annali, vol. 1, pp. 5156.
7
Let us recall Biblical and apocryphal stories of the sunset of life of Enoch, Moses, Elijah,
Isaiah, Jesus. A reference to Moses is the Islamic legend: By God, the Messenger of God is
not dead, but only returned to his Lord like Ms b.Imrn. Verily, he will return and cut off
the hands and feet of those who believed in his death. (Ibn Hishm, p. 1012; al-abar, vol. 1,
p. 1815). Al-Shahrastn, ed. Cureton, vol. 1, p. 11, mentions s b. Maryam (Jesus) instead of
Moses.
8
T. Andrae, Die Person Muhammeds, p. 23, [who has it from I. Friedlnder, Abdallh b.
Sab, der Begrnder der a in Zeitschrift fr Assyriologie, 23 (1909), p. 299; EI
2
; Juynboll,
Encyclopedia, p. 503, col. 1].
9
Newresearches into the composition and exegesis of the Qoran, pp. 138141.
sectarian reproach against the uthmnic text 281
48:29) are spurious. Following A. Sprenger
10
and Fr. Bethge,
11
he apparently
is of the opinion that Muammad is no actual name but a Messianic term.
However, the reasons supporting himand his predecessors, as well as Leone
Caetani later on,
12
are invalid. In particular, there can be not a shadow
of doubt that Muammad had been a common mans first name even in
pre-Islamic Arabia, a fact that has already beendiscussedabove onpage 6sq.
No less suspect is Gustav Weils
13
interpretation of sra 46:14: [ii/84]
We have charged man, that he be kind to his parents; his mother bore him
painfully, and painfully she gave birth to him; his bearing and his weaning are
thirty months. Until, when he is fully grown, and reaches forty years,
14
he says,
O my Lord, dispose me that I may be thankful for Thy blessing wherewith
Thou hast blessed me and my father and mother, and that I may do righteous-
ness well-pleasing to Thee; and make me righteous also in my seed. Behold I
repent to Thee, and I am among those that surrender.
Tradition attaches this verse to Ab Bakr, for among the early Companions
of the Prophet there was no one so privileged as to see not only his parents
but also his children embrace Islam. By following this interpretation, Weils
challenging the authenticity of the verse means no more and no less than
that the first caliphinterpolatedthe entire verse, or at least its secondhalf, in
the codex of revelations in order to enhance the reputation of his family and
to serve base and selfish motives. This serious charge, however, cannot be
maintained. If it were justified, the impressionof the CaliphAbBakr would
be drastically contrary to what we know from the historical sources. On the
other hand, it would be incomprehensible that Ab Bakr, if indeed he once
wanted to emphasize his excellence, would have chosen such obscure and
ambiguous expressions. These difficulties lead one to challenge the accu-
racy of the exegetic tradition on which Weil bases his argument. Whoever
follows indigenous interpreters in the case of this verse will necessarily be
forced to apply verse 16, closely related with the same authorities, to Ab
Bakrs son, Abd al-Ramn (Ibn Ab Bakr),
15
who remained a pagan longer
than his father and rejected his first invitation to accept Islam with con-
temptuous words. This interpretationis of course impossible. What onearth
would have prompted Ab Bakr to fabricate a tradition to rebuke his son
10
Das Leben und die Lehre, vol. 1, p. 155sqq.
11
Ramn et Amad, thesis, Bonn, p. 53sq.
12
Annali, vol. 1, p. 151.
13
Weil, Historisch-kritische Einleitung in den Koran, 1st ed., p. 67, 2nd ed., p. 76sqq.
14
The words and reaches forty years seemto me to have been added later as an exegetic
gloss.
15
Died 130/747 or 135/752. Sezgin, GAS, vol. 1, p. 284, no. 14.
who at that time had longsince 6/627been converted, and by including
the forgery in the Koran forever, stigmatizing him in the eyes of believers?
Reference to the well-knownintegrity of this caliph is ill suited to explaining
this strange behaviour, as the virtue that indeed distinguished the historical
Ab Bakr hardly agrees with the presumed activity as a forger. Actually, the
verse does not go back to either Ab Bakrs son or any other historical figure,
a fact that is also concededby some interpreters.
16
Inthis case, all precise ref-
erences in 46:14 must also be dispensed with and it must be assumed that
the words, as frequently in the Koran, merely purport to express a general
truth.
In the end, Weil denies the authenticity of sra 17:1: Glory be to Him, [ii/85]
who carried His servant by night from the Holy Mosque to the Further
Mosque the precincts of which We have blessed, that We might show him
some of Our signs .
17
This is to say that the verse was rhymed only after
the death of Muammad, and possibly incorporated in the Koran in the
time of Ab Bakr. Muammad cannot possibly have claimed to have made
the mysterious journey to Jerusalem, since he always maintains throughout
the Koran that he is a messenger and warner but not a miracle worker.
The objection is quite legitimatecompare only sras 13:8 and 27, 17:95,
25:8sqq., and 29:44yet it becomes untenable when the night excursion
is regarded as a dream. Traces of this opinion are to be found even in
Muslim tradition, which in other cases clings to the miracle.
18
The text of
the Koran does not supply a hint that it was a dream but speaks of the
Night Journey as a fact. In order to escape from these contradictions one
may assume that the Prophets excited fantasy, which here touches upon
the thinking of primitive man, experienced the dream as reality, the same
way as Muammads visions (sras 53:6sqq., and 81:23sq.) are depicted
as true events. Given that we learn nothing else from this episode of the
Koranand sra 17:62 cannot be related to thisand that the traditions
referring to verse 1 are inconclusive, one might want to consider the night
excursion to represent merely another hero of the past. Here, unfortunately,
newdifficulties arise, since nosuchmiracle is reportedfromBiblical persons
who, as far as we know, legend associates with the Kaba, such as Adamand
Abraham, whereas Ezekielof whom it is said that a spirit once took him
by a lock of his hair and lifted him up between the earth and the heaven,
16
al-Zamakhshar, and above, p. 130 n. 101.
17
Weil, Historisch-kritische Einleitung in den Koran, 1st ed., p. 65sq., 2nd ed., pp. 7476.
18
and brought himto Jerusalem
19
has nothing to do with the Kaba in any
legend we know of.
Weils observation
20
that the verse does not continue into the following [ii/86]
verse is true but irrelevant in terms of its authenticity, as this applies to
many other verses of the Koran that so far have not been objected to. The
circumstances can be explained by the fact that the verse lacks its original
continuation. The different rhyme with r, in contrast to all the other 110
verses that rhyme with a without exception, would suggest that the entire
section was previously placed somewhere else.
The fact that the verse is part of the revelation cannot be contested. Its [ii/87]
alleged linguistic inaccuracy exists only in Weils mind. Whether the phrase
asr laylan can be considered a pleonasm seems extremely doubtful, as
laylan can equally be translated one night. In this case laylan is equally
dispensable, as is al-layl, laylihimor laylahumin passages such as sras 11:83
and 15:65, Ab Tammm, al-amsa, 744, v. 5, al-Mubarrad, al-Kmil, ed.
Wright, p. 62, l 9, al-amsa, 384, verse 3. In any case, the passage to which
Weil is objecting is alsofoundinsra 44:25; evenif his stylistic interpretation
of the phrase were correct, all this would not argue against the authenticity,
as pleonasms are common to all the languages of the world. Furthermore,
the usage of the fourth verbal form asr, with or without a preposition, is
quite common. Finally, the transition from the third person singular to the
first person plural, when Allh is talking about himself, can be documented
in the Koran with hundreds of examples.
21
Within a single verse this seldom
occurs, but from sras 30 to 50 I identified two instances (35:25 and 40:77;
in reverse order, 39:2), while in two other cases (48:1 sq., and 8sq.) this
change of person spreads over two verses, forming a single period. Whoever
seeks to avoid recognizing the weight of these arguments by imagining an
extraordinary imitator of the Koranic style would immediately encounter
new problems, as such an ingenious forger would be expected to supply a
more appropriate connection to what follows and a more suitable rhyme.
Most importantly of all, the motive for the interpolation would have to be
discovered, which no one has succeeded in doing.
19
Ezekiel 8:3; cf. also above, p. 110 n. 104, at the end.
20
In the first edition of his Historisch-kritische Einleitung in den Koran, Weil supplies no
evidence; this is only found in his review Nldeke ber Mohammed und den Koran in
Heidelberger Jahrbcher der [sic] Literatur, 1862, p. 7, which constitutes a reviewof Nldekes
1860 edition of the Geschichte des Qorns.
21
It would be useful to collect all the material. It would tell us much about the composi-
tion of the sras.
Like AbBakr, Uthmnis alsoaccusedof forgery. He allegedly omittedall [ii/88]
the passages in which Muammad earlier came up against the Umayyads.
Weil,
22
however, neither produces evidence for this assertion nor even ex-
plains it properly, so that we do not even know whether he meant the dele-
tion of entire passages or only of the names of individual persons. The elim-
ination of anonymous polemics naturally would have been futile, because
the respective addressee could later no longer be identified with certainty;
this also applies to the commentaries that identify many passages of the cur-
rent text
23
withmembers of the Umayyadfamily, whichUthmnwouldhave
hadtohave overlookedat the time. The deletionof individual names is quite
conceivable, althoughBanUmayyacertainly inthe earlier periodwere
no worse opponents to Muammads sermons than other eminent Meccan
families, with the result that there was no cause to attack them more fre-
quently or violently than other families of the town. We must thus assume
that the names of many other enemies of Islam have also been suppressed,
including, for instance, some of the Jews and the munfiqn, whom the
Prophet hated from the bottom of his heart. However, no reason at all can
be supplied for this. There still remains the fact that it is completely con-
trary to Muammads habit to mention names of his surroundings, be they
personal or geographical. This can hardly be an accident but must rather be
the deliberate intention of the document of revelation, which was destined
for all of humanity, to minimize as far as possible the particular in favour of
the general. If occasional revelations, in which names are likely to have been
found more frequently, were later incorporated into the Koran, such names
are likely to have been omitted at this occasion by the Prophet himself. This
system, however, is not rigorously applied.
Place names in the Koran are mentioned five times: Mecca twice (sras [ii/89]
48:24, 3:90), andonce eachBadr, unayn, andYathrib,
24
(sras 3:119, 9:25, and
33:13) respectively. The names of contemporariesother than Muammad
himself (sras 3:138, 33:40, 47:2, and 48:29)
25
occur twice, i.e. Muammads
mawl and adopted son, Zayd b. ritha,
26
(sra 33:37) and his uncle Ab
Lahab IBNABDAL-MUALIB (sra 111), whereas not even a single man of
22
Geschichte der Chalifen, vol. 1, p. 168.
23
E.g., sra 49:6 to Uthmns cousin Wald b. Uqba (Ibn Ab Muayth). As I explained
above, p. 178sqq., this interpretation is very doubtful.
24
al-Madna (sras 9:102, and 63:8), is not yet a proper name; the same applies to umm
al-qur (6:92) or al-qaryatniMecca and al-if(sra 43:30).
25
Sra 61:6 has instead Amad.
26
EI
2
; EQ; Juynboll, Enclyclopedia, p. 452, col. 1.
the most trusted of Muammads friends or the most steadfast supporters
of the young community is mentioned by name.
This fact can be explained in various ways. It could have been on account
of the incomplete textual condition in which Muammad left at least parts
of the Koran, or some special motivation which prompted him to deviate
fromthe rule, or, finally, the penetration of old exegetic glosses into the text
proper. Which of these possibilities deserves first priority must be deter-
mined in each particular case. In conformity with the start of this discus-
sion, I can here limit myself to the personal names. The mention of Zayd
(Ibn ritha, d. 8/630)
27
in sra 33:37, some people
28
consider an honour
that was awarded since he had let the Prophet have his wife Zaynab bt.
Jash.
29
Conversely, the mention of Muammads uncle, Ab Lahab (IBN
ABDAL-MUALIB), aims permanently to stigmatize himfor his disbelief.
The purpose in each of the cases, however, eludes me. As far as we can judge
the conditions of the time, neither did the compliant adopted son deserve
such mild consideration, nor did the disbelieving uncle merit such a sharp
denouncement. In these circumstances one ought to take into considera-
tionwhether the name Zayd[Ibnritha] inthis instance (sra 33:37) might
not be an old exegetic gloss, particularly as the cumbersome reference to
this personby the relative clause at the beginning of the verse
30
gives no hint
that shortly thereafter his name will be mentioned. Furthermore, whenAbd
al-Uzz IBNABDAL-MUALIBreceived the nickname Ab Lahab merely
on the basis of sra 111, we are dealing in this passage not with a personal
name. On the contrary, it is doubtful if the interpretation of that designa-
tion of Muammads uncle is correct at all, no matter how unambiguous
this tradition may be.
31
Weils other attempts at finding Uthmn guilty of intentional suppres- [ii/90]
sionof larger portions of the Koran
32
alsofailed. Wheninal-Dhahabs histor-
ical work
33
the rebels accuse Uthmn of having combined the Koran, which
originally consistedof books (kutub), intoone single book (kitb), they prob-
ably intended to say no more than that he replaced previously common
27
Ibid.
28
Hirschfeld, Newresearches, p. 121.
29
Cf. above, p. 168; Hirschfeld, Newresearches, p. 139.
30
When thou saidst to him whom God has blessed and thou hadst favoured, Keep thy
wife to thyself, and fear God, and
31
32
Geschichte der Chalifen, vol. 1, p. 168.
33
Tarkh al-Islm, cod. Paris, 1880, fol. 164. On the author cf. Brockelmann, GAL, vol. 2,
p. 46sq.; Sezgin, GAS, vol. 1, p. 504, l 7.
different recensions with a uniformedition of one reading, a scenario which
corresponds to historical truth.
Weils other contention
34
does not deserve a separate refutation, i.e., that
from the various Koranic versions on the same subject existing in the doc-
uments collected by Zayd (Ibn Thbit) under Ab Bakr, Uthmn included
only a single one of them, paying little or no attention to other collections
or fragments found in the hands of Muammads old Companions. We have
earlier
35
furnished evidence that both of Zayds codices are identical, and
that Uthmns recension is nothing but a copy of the codex of afa.
In the final analysis, there are several general historical considerations [ii/91]
that speak in favour of Uthmn. Although the old Caliph remained a tool in
the hands of his family, he was still an upright, pious, and religious man who
cannot possibly have been suspect of falsifying the word of God. Further-
more, in the Koran Commission
36
there was only one Umayyad representa-
tive. Of the other members, Abd Allh b. al-Zubayr, a member of a family
jealous of the Ban Umayya, and Zayd b. Thbit, the former amanuensis of
the Prophet, are above suspicion of illegally favouring Uthmn.
Even if the character of these individuals were somewhat less favourable,
any attempt at a tendentious change of the text on their part would have
failedfor other reasons. During the nearly twenty years since Zaydb. Thbits
first redaction, the number of circulating codices of the Koranhadincreased
markedly, and we have identified no less than five famous collections from
the period before Uthmn. From one of them, the codex of afa, the
Uthmnic editionwas copied. The original was returnedtothe owner. Thus,
there were so many references to the original text available that any serious
change in the text would have been noted immediately and, particularly
when malicious tendencies were suspected, a storm of indignation would
have followed.
In addition to the written means of control there were the oral ones. [ii/92]
Even after all the copies of the pre-Uthmnic collections had either been
destroyedor disappeared,
37
there must have remaineda sufficient number of
people to reconstruct any suppressed passages from memory.
38
This would
have been no problem, as it was possible to have recourse to the collabo-
rators in that redactionin so far as they were still aliveparticularly to
34
Weil, Historisch-kritische Einleitung in den Koran, 2nd ed., p. 56sq.
35
Cf. above, pp. 251 sq. and 262sq.
36
37
Cf. above, pp. 252 and 256257.
38
Cf. below, p. 235sq.
Ibn Masd, who was so proud of his Koranic competence and was deeply
hurt by the preference given to Zayd (Ibn Thbit).
39
Although Ibn Masd
had ample reason to be angry with Uthmn for having rejected his collec-
tion of the Koran, he never accused Uthmn of forgery.
40
How much more
would the innumerable other enemies of this generally unpopular ruler
have exploited the slightest suspicion and spread it throughout the Islamic
world. Nevertheless, the older sects and opposition parties, though they
were largely recruited from among the circles of reciters of the Koran, were
apparently unable to charge the Caliph with anything more serious than
being a man who dismembered the Koran,
41
and a man who burned the
Koran,
42
epithets that refer to the destruction of the pre-Uthmnic codices.
For this reason, all attempts at justification put into the mouth of the Caliph
lead in this direction.
43
Thus everything seems to indicate that the Uthmnic text was as com- [ii/93]
plete and reliable as could be expected. It was primarily these merits that
facilitated its quick and easy acceptance in the Muslimcommunity. Official
force alone would have never succeeded.
39
40
The following words are commonly put into his mouth: People of Iraq (variant: of
Kfa)! Hide the copies of the Koran in your possession and defraud them, for Allh, the
Exalted says: Whoso defrauds shall bring the fruits of his fraud on the Day of Resurrection
(sra 3:155)and then approaches Allh with these copies. Cf. Ibn al-Athr, Chronicon, ed.
Tornberg, vol. 3, p. 87; IbnSad(al-abaqt, vol. 2, part 2): Letzte Krankheit, TodundBestattung,
p. 105; al-Tirmidh, Tafsr; al-Mabn li-nam al-man, fol. 6
v
; al-Qurub, vol. 1, fol. 20
r
. This
interpretationof the Koranic passage departs far fromits proper sense. Although this address
by Ibn Masd is unhistorical, it corresponds more or less to what can be expected, given
our knowledge of the situation. Conversely, there is a very derogatory remark of this man
concerning Uthmns editorial activity in the following tradition in Mliks al-Muwaa,
p. 62: Ibn Masd said to a man: You are living at a time when there are many jurisconsults
(fuqah) but few readers of the Koran; though the laws of the Holy Book are observed, its
very letters however being neglected. Its continuation with its reference to the future when
inversely the laws of the Koran are violated, but its letters being observed, clearly indicates
that the entire tradition is fabricated from the point of view of a much later period.
41
Shaqqq al-maif, al-abar, vol. 2, p. 747.
42
arrq al-maif, al-Qurub, fol. 20
r
.
43
This becomes obvious fromthe following passage of the Persiantranslationof al-abar,
cod. Leiden[no number supplied:]
. They say that I burned the Koran;

(this I did) because people only had fragments in their hands, and everybody considered his
own to be the best; thereupon I collected them all, placing a long sra in front, another of
medium length in the middle, and a short one at the end, properly arranged them all, and
handed them over to the people; whatever they had in their possession I took and burned.
Reproach of MuslimSectarians,
Particularly the Shites, against Uthmn
The doubts voiced within Islam about the integrity of the Koran are an
entirely different matter. They are not based on scholarly facts of historical
criticism but on dogmatic or ethnic prejudices. Pious Mutazilites consider
as spurious all passages that curse the enemies of Muammad, since this
cannot possibly be, as they say, a lofty revelation from the Well-Guarded
Table.
44
The Khrijite sect of the Maymniyya takes exception to including the [ii/94]
Story of Joseph in the Holy Book, since a love story is ill suited to it.
45
Far more numerous and diverse are the objections to the canonical text
by the party of Al, the so-called Sha, the Shites. These exceptions refer
not only to the insertion or omission of entire sras but also to verses and
single words.
46
Whereas other sects apparently considered passages that
they contest to have entered the Koran by accident or mistake, the Shites
suspected everywhere nothing but bias and malice. Since nowhere in the
Koran did they find expressed the sanctity to which Al and his family were
entitled to in their own view, they accused Ab Bakr and Uthmn of having
changedor totally suppressedthese passages, nomatter hownumerous they
might have been.
47
Additionally, all the passages that were considered lost
by Sunnite tradition the Shites claim dealt with Al. By the same token,
verses in which the anr, Muammads old Medinan followers, and the
muhjirn, i.e. the Companions who emigrated with the Prophet to Yathrib,
were accused of disgraceful treatment, were allegedly deleted. But since the
crime of these men consisted of refusing to vote for Al in the election of
the first caliph, the Prophet would have reproved his most trusted followers
44
Cf. Fakhr al-Dn al-Rz, Maft al-ghayb, Blq edition, 1289, vol. 1, p. 268, according
to Goldziher, Vorlesungen ber den Islam, pp. 207 and 260, no. 11; translation: Introduction to
Islamic theology and law (1981).
45
al-Shahrastn, Religionsparteien, translated by Haarbrcker, vol. 1, pp. 143 and 145
(Curetons edition, vol. 1, p. 95sq.) Ibn azm, Milal wa-nial, in: Israel Friedlaender, Het-
erodoxies of the Shiites , vol. 1, p. 33.
46
The technical term is tabdl, cf. I. Friedlaender, loc. cit., vol. 2, p. 61.
47
Al b. Ibrhm. Tafsr, Ahlwardt, Verzeichnis, Berlin, cod. 929; Muammad b. Mur-
ta, al-f; ibid., cod. 899. Journal asiatique, 4e srie (1843), p. 406sqq. [This reference is
to Extraits du Modjmel al-tewarikh, relatifs l histoire de la Perse, traduits par J. Mohl,
pp. 385432. As Mohl is not even mentioned by Nldeke-Schwally, it does not seem to be
relevant at all.] Still more ridiculous are the fables which are mentioned in this connection,
for example, when it is said that Al offered Ab Bakr his complete Koran in order to deprive
him of any excuse at the Final Judgement; or that Ab Bakr allegedly wanted to kill him, etc.
for an action that did not occur until after his death, and could not possibly
have entered the ken of the parties concerned. What an accumulation of
impossibilities!
It is quite evident that the other contentions are untenable. The notion [ii/95]
of Als and his descendants sole title to the caliphate, which is supported
neither by the religion of Islam nor its national origin, crystallized only
quite some time after Als death. Furthermore, the Shite deification of Al
sprung up on Iranian soil. If Al had even only once been suggested as suc-
cessor in the Koran, this would have likely been binding on the electoral
college. A departure from such a guideline would have created complica-
tions that would have left clear marks on tradition. Evennumerous passages
of the Holy Book allegedly pleaded for Al as the most exalted of men with-
out even one single member of the electoral college or some other Compan-
ion interceding for the candidate of the Prophet! Let him believe who is so
inclined. By the same token, Al never inhis life referredtosuchKoranic pas-
sages,
48
althoughhe was later twice passedby, and, having finally securedthe
caliphate, he was obliged to defend his right against Muwiya, the governor
of Syria, with sword and word. Yet even the Shites make use of the Uth-
mnic recension to this very day, irrespective of all insinuations. According
to their faith, however, this is nothing but a temporary solution until the
coming of the Messianic kingdom. The genuine and unadulterated text is
in the possession of Als mysterious successors, the Twelve Imms, who
keep it hidden
49
until the last imm, the Shite Messiah, or as he is called by
the Shites, al-Mahd l-qim, brings it forth fromoccultation.
50
Some Shite
sects, like the Immiyya, accept this and patiently await the expected reap-
pearance in the distant future.
51
Others, pretending foresight, are obliged to
48
This argument the author of al-Mabn li-nam al-man uses against the Shites. Ibn
azm in Isr. Friedlaender, loc. cit., vol. 2, p. 62, says that otherwise it would have been Als
duty to fight the interpolators.
49
Many questions in this context are explained in Muammad b. Murta, K. al-f
(tafsr al-Qurn), in Ahlwardt, Verzeichnis der arabischen Handschriften, cod. 929 (= cod. 1
Petermann, no. 553), but after listing the contradicting opinions of scholars, in the end he
himself does not know what to say.
50
Muammad b. Murta, K. al-f, Ahlwardt, cod. 929; Journal asiatique, 4e srie (1843),
pp. 399 and 402sq.; cf. Goldziher, Schools of Koranic commentators, p. 178.
51
Mirza A. Kazem-Beg, Observations , p. 403. According to Ibn azm in I. Fried-
laender, loc. cit., vol. 1, p. 51 sq., there are few authorities of the respective sect who repu-
diate interpolations in the Koran. According to a confession in Mirza A. Kazem-Beg, loc. cit.,
p. 401 sq., the Immiyya generally follow the Uthmnic Koran, with the exception that they
combine to one chapter each the sras 93 and 94 as well as 105 and 106. In this case they are
approaching the Koran of Ubayy b. Kab where, according to al-Itqn, p. 154, either the one
or the other of these Koranic pairs formed one chapter. Cf. also above, p. 239.
obtain evidence of the forgery by dangerous exegetic conjectures or free
inventions, unless they even have the audacity to attribute their knowledge
of the true reading to wonderful encounters. One of these narrators main-
tains that he receiveda codex of the Koranfromone of the above-mentioned
imms, and, though he was forbidden to look at it, he did nevertheless and
thereupon found his own reading.
52
According to fourth-century writers, altogether some five hundred pas- [ii/96]
sages of our Koran are allegedly forgeries.
53
Whether the material available
comes even close to this figure I do not know. In any case, a complete list-
ing of this would serve no useful purpose for this investigation, and thus
we limit ourselves to pointing out the various types and documenting them
with typical examples.
There are first of all reports of lacunae in the Uthmnic redaction, the [ii/97]
text of which is not known or, in any case, not supplied. Among the sras
which originally were much longer, sra 24 allegedly had over one hundred
verses, and sra 15 even one hundred and ninety.
54
As far as the original
length of sra 33 is concerned, the Sunnite sources supply fantastic particu-
lars.
55
Insra 25:30, where there is a reference tofulnan, a somebody, a par-
ticular name is alleged to have been supplied originally.
56
In sra 98, which,
according to some Sunnite sources, was also originally much longer,
57
the
names of seventy Quraysh men and their fathers are alleged to have been
purposely omitted.
58
It goes without saying that there is notruthtothis at all.
Muhmmad, who displayed such a strong aversion to the mention of names
in the Koran,
59
cannot possibly have decided to list seventy all at once, and
evenincludedthe names of their fathers. Onthe other hand, if AbBakr had
the audacity to omit this many names, he certainly wouldnot have hesitated
to include his own name once.
According toa conspicuously similar tradition,
60
sra 9:65 originally listed
the names of seventy munfiqn (hypocrites) together with the names of
their fathers. This would suggest that we not interpret the word in its usual,
52
Muammad b. Murta K. al-f (Tafsr al-Qurn), Ahlwardt, Verzeichnis, no. 899.
53
al-Mabn li-namal-man, part 4, fol. 32
r
; Goldziher, Muslimstudies, vol. 2, p. 109.
54
Muammad b. Murta, K. al-f, Ahlwardt, Verzeichnis, cod. 929.
55
Cf. above, p. 204.
56
Fakhr al-Dn al-Rz, Maft al-ghayb, vol. 4, p. 470, according to Goldziher, Muslim
studies, vol. 2, p. 109, and his Schools of Koranic commentators, p. 270sq.
57
Cf. above, p. 193 and 195.
58
al-Mabn li-namal-man, part 4.
59
60
Ibn Abbs says: God Most High revealed the mention of seventy men fromamong the
general meaning but rather consider it a particular reference to the Quraysh
of the other tradition, who, according to Shite opinion, were the leaders of
the Muslims who did not permit Al to succeed to the leadership. This is
particularly appropriate because the first two caliphs, who also belonged to
the circle of Als enemies, were frequently referred to by the Shites with
the abusive word, munfiq.
61
This equation, however, is untenable, since the
second part of the tradition,
62
which presents the children of the munfiqn
as believers, cannot be Shiite. Furthermore, the alternative argument, that
a Sunnite tradition was grafted onto a Shite one, is too far-fetched.
The individual readings which the Shites fabricated and presented [ii/98]
against the alleged forgeries of Ab Bakr and Uthmn all deal with the sub-
ject of Al and the imms. This corresponds to the familiar bias that the
Imm Ab Abd Allh [JAFAR AL-DIQ] (d. 148/765)
63
allegedly once put
like this: If you had read the Koran in its original version you would have
found us (i.e. the imms) mentioned by name.
64
This saying must originate
from before the beginning of the fourth century, since it occurs already in
Al b. Ibrhm AL-QUMMs commentary on the Koran
65
and, according to
Ibn al-Anbr (d. 327/939 or 328/940
66
), such readings were already in circu-
lation during his time.
67
If we knew more about Shite literature we would
most likely arrive at much earlier dates and could possibly trace back the
beginnings of this exegesis to the second century.
The majority of the readings consist of the words Al and l Muam- [ii/99]
mad (Muammads family) being inserted without consideration for the
munfiqnby name andthat of their fathers; He thenabrogated(nasakha) the mentionof the
name for compassion for the believers so that they would not slander one another because
their children had become believers. Cf. al-Baghaw, and al-Khzin al-Baghdd, Tafsr on
sra 9:65. The Arabic text canbe foundabove, p. 204. It is difficult tosay whether the reference
to the tradition in al-Itqn, p. 527, that sra 9 was originally four times the size, has also these
names in mind.
61
Thus already Al b. Ibrhm AL-QUMM, Tafsr. (W. Ahlwardt, Verzeichnis, cod. 929).
62
It is obvious that the entire tradition is a fabrication. The mention of so many names is
as unthinkable as their omission once they had been included, regardless of whether this is
blamed on the Prophet or on one of the first caliphs.
63
I.e., JAFAR [AL-DIQ] b. Muammad b. Al b. al-usayn. In Shite adth sources
often called only by his kunya, Ab Abd Allh, the sixth imm of the Twelver Sha; EI
2
; EQ;
Juynboll, Encyclopedia, pp. 260266; Sezgin, GAS, vol. 1, pp. 528531.
64
Law qaraa l-Qurn ka-m unzila la-alfaytan fhi musammayn. [Juynboll, Encyclope-
dia, pp. 260262.]
65
Ahlwardt, Verzeichnis, cod. 929, Ab l-asan Al b. Ibrhm b. Hshim AL-QUMM,
Tafsr. About the author cf. Sezgin, GAS, vol. 1, pp. 4546.
66
Cf. EI
2
67
Stated by al-Qurub, fol. 31
r
.
respective meaning.
68
Where the text has hdhirunmustaqmun
69
(sras
3:44, 19:37, 36:61, 43:61 and 64), we now read for no rhyme or reason iru
Aliyyin. In the case of sra 3:119, the phrase bi-sayfi Aliyyin, by the sword of
Al, is added after words and Allh most surely helped you at Badr.
70
Like-
wise, insra 4:67 if, whenthey wrongedthemselves, they hadcome tothee,
is augmented by the address O Al.
71
In sra 4:164, f Aliyyin is inserted
after anzalahu, just as it is in sra 5:169 after fa-inna. In sra 4:166 after wa-
alam, as well as in sra 26 at the end after alam, la Muammadin
aqqahum is inserted as an object. Instead of (kuntum khayra) ummatin
(sra 3:106) it is readaimmatin; insteadwajalnlil-muttaqnaimman(sra
25:74) wajal lanminal-muttaqnaimman. Insra 11:20 the words imman
wa-ramantan are placed behind shhidun min-hu.
72
Where in the Koran
the phrase, verily, Al is guidance, is to be inserted, the source does not
say.
73
Complaints about large parts or entire sras of the Uthmnic redac-
tion do not seem to have been voiced in older Shite literature. What little
has become known in more recent times lacks precise dating, since the
sources have not yet been examined. However, a systematic survey of Shite
literature is likely to reveal many strange details.
According to a work of the Turkish mufti Asad Efendi against the Shites, [ii/100]
quoted in a seventeenth-century Occidental work,
74
we read: You deny the
verse called the covering in the Alchoran [sra 88] to be authentick; you
reject the eighteen Verses, which are revealed to us for the sake of the holy
Aische. These verses apparently refer to the beginning of sra 24, where
this wife of the Prophet was defended for her dubious behaviour during the
campaign against Ban Mualiq.
75
It is understandable that Shites are less
than happy to see the vindication of the mortal enemy of their saint, Al,
in the Koran. On the other hand, I have no idea what might have been the
reason for their lack of confidence in the case of sra eighty-eight.
68
Thus the description in Muammad b. Murta, al-f, adding, however, that in other
such passages the names of the doubters [munfiqn] have now disappeared.
69
In sra 15:41 there is the variant reading, hdh irun alayya mustaqmun.
70
From al-Qurub.
71
al-Qumm, Tafsr, W. Ahlwardt, Verzeichnis, cod. 929, s.v. Here, as in many of these
passages, there is an allusion to events from the period after the death of Muammad.
72
The last seven examples have been taken from Kazem-Begs article, p. 407sqq.
73
Goldziher, Beitrge zur Literaturgeschichte der a und der sunnitischen Polemik,
p. 14; Muslimstudies, vol. 2, p. 109.
74
Sir Paul Rycaut, Histoire de l tat prsent de l empire ottoman, translated from the
English (Paris, 1670), from Goldziher, Muslim studies, vol. 2, p. 109. [Here quoted from the
original, The Present State of the Ottoman Empire (London, 1668), pp. 119 and 121].
75
Juynboll, Encyclopedia, p. 720, cols. 12.
The Shite Sra of the Two Lights ( )
According to some Shite authorities, several sras originally in the Koran
were later eliminated from the book.
76
Only one of them has so far become
known. This is the so-called Sra of the Two Lights. The first person to
bring news of the textfrom the Persian Dabistn-i mahib of the mid-
seventeenth-century Musin FNto Christian Europe was the French
Orientalist Garcin de Tassy.
77
The edition of the Kazan professor Kazem-
Beg
78
offers the same text, with corrections of obvious textual and printing
mistakes, including the orthography, vocalization, and division of verses as
is common in more recent manuscripts of the Koran. In the introduction
Kazem-Beg expresses his pleasure finally, after eighteen years, at being in
possession of the complete chapter, whereas up to then only fragments had
been known.
79
It thus appears that Kazem-Beg did not learn of the complete
text until its French publication. Conversely, Garcin de Tassy maintains in
his epilogue
80
that Kazem-Beg succeeded after eighteen years of research in
acquiring a copy of the complete text. It is difficult to tell whether this claim
is caused by a misunderstanding of the above-mentioned words of Kazem-
Beg, or possibly by a private letter that is more precise than the introduction
to his article. In the latter case, one ought to be extremely sceptical, since
Kazem-Beg keeps the provenance of his discovery a secret and, secondly,
does not supply a single variant. Additionally, it is unlikely that two such
discoveries are made shortly one after another.
Although I cannot offer a better text, for a better understanding of the [ii/101]
researchit seemedtome useful topresent the Arabic original of the sra
81
to-
gether with a translation.
76
Kazem-Beg, loc. cit., p. 424; Goldziher, Schools of Koranic commentators, p. 172.
77
His Chapitre inconnu du Coran, publi et traduit pour la premire fois.
78
Journal asiatique, 4e srie, 2 (1843), p. 414sqq. (Kazem-Beg apparently found the sra.
He first handed it over to Garcin de Tassy for publication, but later edited it once more him-
self more accurately; from A Fischers additions and corrections on p. 220 of the German
edition.)
79
P. 373sq.: Je suis enfinassez heureux pour possder dans ce moment, aprs dix-huit ans
couls, tout le chapitre inconnu du Coran, dont je n avais lu prcdemment que quelques
fragments, et de communiquer mes ides sur cette dcouverte. M. Garcin de Tassy, auquel
nous sommes redevables de la publication de ce chapitre, dit dans son introduction, etc.
80
Loc. cit., p. 428: le chapitre que j ai publi est si peu rpandu dans le monde
musulman, que ce n est quaprs dix-huit ans de recherches que le savant professeur de
Kazan a pu sen procurer une copie exacte.
81
Kazem-Begs text is reproduced without changes. Critical remarks and suggestions for
correctionare tobe foundeither inthe foot-notes tothe Germantranslationor inthe append-
ed lexical and stylistic juxtapositions. I supply vowels only where the pronunciation is not
self-evident or where it differs from previous forms. Kazem-Begs division of verses has been
retained. Conversely, the numeration of the verses is newin order to facilitate their location.
The Sra of the Two Lights
82
In the name of God, the Merciful, the Compassionate
83
(1) O you who believe, believe in the two lights which We have sent down
to recite to you My signs, and to warn you of the penalty of a mighty day.
(2) It is two lights, one of them emanating from the other. Verily, We hear
and know. (3) Those who keep the pact of God and his Messenger,
84
they
will share delightful gardens. (4) And those who renounce their belief after
having become believers by breaking their pact and the agreement with the
Messenger will be thrown into hell-fire. (5) They have brought outrage upon
themselves and opposed the agent of the Messenger, these will be made to
drink from the bottomless pit. (6) It is God who illuminates heaven and
earth with what He wanted,
85
made a selection from among the angels and
messengers and ranked them with the believers. (7) These are part of His
creatures, God does what He wills, there is no god but He, the Merciful, the
Compassionate. (8) Already earlier peoples were perfidious vis--vis their
messengers, so I seized them for their perfidy, verily, my seizure is strong and
hurts. (9) God destroyed the dites and Thamdites because of what they
(howthey) deserved(it), and made themanexample for you; will younot
86
be
god-fearing? (10) And Pharaoh I drowned with all his followers, because he
rebelled against Moses and his brother Aaron so that he (it) may be a sign
for you, most of you are truly wicked. (11) Verily, God shall assemble them
on the Day of Judgement, then they do not have an answer when they are
asked. (12) Verily, the bottomless pit is their abode, andGodknows andis wise.
(13) OMessenger, let My warning reach them, they will soon know.
87
(14) Lost
are those who turned away from My signs and My judgement.
88
(15) They are
unlike
89
those who keep their pact, and whomI recompensed with gardens of
bliss. (16) Verily, God offers forgiveness and large recompense. (17) Al is truly
one of the godfearing. (18) Onthe Day of Judgement We shall give himhis due.
(19) We shall not ignore that he was wronged. (20) We also distinguished him
and bestowed honour on him before his entire family. (21) He and his family
are waiting confidently. (22) But their enemy is the imm of the sinners. (23)
Speak to those who renounce their belief after having become believers: You
desired the splendour of this life, you were in a hurry and forgot the promises
of God and his messengers, and broke the contract after you had concluded
82
In this translation I tried to retained the abruptness of the style as well as the ideas.
83
For the most likely meaning of ram cf. above, p. 92.
84
The words f ytin do not make sense.
85
Bi-m sha cannot mean according to his will as translated by both, Kazem-Beg, and
G. Weil, Historisch-kritische Einleitung in den Koran, 2nd ed., p. 93.
86
I read like Garcin de Tassy and Gustav Weil a-fa-l instead fa-l of the text; cf., e.g., sra
7:33.
87
Instead yamaln I read yalamn like sra 15:3 and 96 as well as Gustav Weil.
88
Murin is wrong for murin.
89
As also others have noted before, the beginning of the verse ought to start with some-
thing like mataluhuml ka-(mathali).
it, so We made parables that you might be guided. (24) O messenger, We sent
you down conspicuous signs in which is a
90
who accepts him (it) or who
after your death
91
turns away from it, they all shall see the light. (25) So turn
away fromthem, they are apostates. (26) Verily, they shall be summoned on a
day when nothing shall help them, and they shall not find mercy. (27) In hell
they will be assigned a place from which there is no escape. (28) So praise
the name of your Lord and be one of the adorers. (29) We sent Moses and
Aaron with that which
92
they used violence on Aaron,
93
but steadfastness
is good. Afterwards We made them apes and swine, and cursed them to the
Day of Resurrection. (30) So be patient, one day they shall be afflicted.
94
(31)
We gave you power like those of the messengers that were before you. (32)
From among them
95
We gave you one as guardian, perchance they rescind.
(33) Who turns away from my command, I shall bring him back. May they
enjoy the brief period of their unbelief! Do not ask the perfidious! (34) O,
messenger, onthe neckof those whobelieve we have put a pact for you.
96
Keep
to it and follow the thankful ones! (35) During his vigil Al is afraid of the
hereafter and is looking forward to the recompense of his Lord. Say! Are those
indeed equal who and those who committed outrage? But they will come
to know my punishment.
97
(36) Chains will be put around their neck, and
they shall repent their deeds. (37) We announced to you pious
98
descendants.
(38) They shall not disobey our command.
99
(39) Upon them from Me prayer
and compassion, may they be dead or alive like on the Day of Resurrection.
(40) Upon those which they violate after your death,
100
be My wrath, they
are wicked people who perish. (41) But those who follow the right path upon
them is My compassion, and they are terraced gardens
101
[lofty chambers] of
Paradise. (42) Praise be to God, the Lord of the World; Amen.
90
At least one word is missing here. Gustav Weil thinks, ahd or mthq. This and similar
expressions would of course be un-Koranic. The Koran, though, frequently speaks of happen-
ings which might be a hint, but never of hints which contain something.
91
I read innahumla-muarn perhaps to be added f l-adhbi following sras 30:15 and
34:37. As an absolute the word stands in all the other passages, but always with reference to
the Final Judgement.
92
The words bi-m istukhlifa by no means make any sense to me.
93
Bagh with the apparently intended meaning to use violence is usually constructed
in the Koran and elsewhere with al, which in this case should be added.
94
If the graphic outline is correct, it can read only yublawna. In this case the word cannot
represent the end of the verse.
95
This naturally does not refer to the earlier messengers of God mentioned previously as
waiyy undoubtedly refers to Al. Possibly something is missing between the verses 31 and 32.
96
The idiom to put a pact on someones neck is nowhere found in the Koran.
97
The verse is tailored to fit sra 39:12. Something seems to be missing between alladhna
andalam. The last little phrase might alsohave a concessive meaning: althoughthey know
my penalty.
98
It is a faulty text: either it is lin without article or bi-l-dhurriyyati.
99
Read yukhlifna.
100
A totally un-Koranic notion, cf. above on verse 24.
101
This meaning of ghuruft arises from sra 39:21 [but not according to Arberry].

.1

.2

.3

.4

.5

.6
] [

.7

.8

.9

.10

.11

.12

.13

.14

.15

.16

.17

.18

.19
.20

.21

.22

.23

.24

.25
.26
.27

.28

.29

.30

.31

.32

.33

.34

.35

.36

.37

.38

.39

.40
.41

.42

It cannot be deniedthat a first glance at this sra leaves one withthe impres- [ii/107]
sion of good Koranic Arabic; most of the sentences and idioms are found
literally or with minor differences in our Koran. However, it is particularly
this fact that Kazem-Beg cites as proof of the forgery.
102
One may say by way
of objection that the Koran, as already shown by Garcin de Tassy in his epi-
logue,
103
also abounds with repetitions, and that it contains passages that
look almost as though they have been made up of scattered phrases of other
passages. These arguments are thus not unambiguous, and do not allow a
safe conclusion. They can be seen in their proper light only after we have
traced back the relation of the Sra of the Two Lights to the Koran according
to other points of view. Nevertheless, the substantial congruency is in fact
facedwitha considerable number of lexical, stylistic, andfactual exceptions.
Lexical cases: anzala to send down is connected in the Koran with
objects only, whereas nrayn in v. 1 indicates persons, Muammad and
Al.Nr light, v. 1, is actually a common and harmless word (in the
Koran religious illumination), but the application to persons, as in this
102
Loc. cit., p. 425.
103
Loc. cit., p. 429.
case, seems to have been used first among Shite circles.
104
In the Koran
only Allh is once called the Light of the heavens and the earth (sra
24:35.)Nawwara to illuminate (v. 6) as well as the verbum finitum, na-
dima to repent (v. 36) are foreign to the Koran.For tawaff, v. 24, the
context demands the meaning to comply with an agreement or obey a per-
son, whereas in the Koran it applies only to Gods acceptance of man after
death.Waiyy (vv. 5 and 32) testator, mandatory does not occur in the
Koran.
105
Imm is here unlikely to have the general meaning used already
in the Koran of model or leader; rather it refers specifically to the head of
a religious community who is empowered by birth and divine providence,
except that in this case the word is not applied, as it usually is, to the pope of
the Shite community but rather, with sarcastic connotation, to the ruling
caliph, who is the master of the secularized, ungodly, state church only by
arbitrary human action.
106
A to renounce ones obedience, which here (v. 5) is constructed
with the dative, is in the Koran regularly followed by the accusative. The
construction of the words, wa-l a laka amran (sra 18:68) is not quite
clear.Khalafa with the dative of the person, and with the meaning re-
quired by the text (v. 38) to disobey is neither Koranic nor Arabic at all;
correctly it ought to be the III. Form, khlafa, with the accusative.The
phrase yawm al-ashr (v. 11) is never used in the Koran for the Final Judge-
ment, although the verb ashara to assemble the people in Allh is very
frequently used.That the plural uhd (v. 23) cannot be documented in
the Koran is surprising, particularly as the singular ahd occurs so often. In
the parallel passage, sra 16:93, which the writer has in mind, we find aymn
instead uhd.Bagh to violate (v. 39) is connected in the Koran and
everywhere in Arabic with al of the person, whereas with the person in
the accusative it means to seek.Maslak path (v. 41) does not occur in
the Koran, although the corresponding verb is found quite frequently.
Also stylistically the text leaves something to be desired. If the words [ii/109]
bi-m sha (v. 6) are indeed intended to mean as he wanted, it would be
a poor substitute for m sha.The phrase atayn bi-ka l-ukma we gave
you power is definitely not Arabic; correctly it ought to read atayn-ka bi-
l-ukm.The connection balligh indhr be warned (v. 13) does not quite
seem like Koranic Arabic, although ballagha as well as andhara are com-
mon; still, insteadof the infinitive one wouldrather expect mandhartu(n)
104
Cf. below, p. 301 sq.
105
For details see below, p. 301.
106
For this interpretation of imm cf. Goldziher, Introduction to Islamic theology, p. 183.
bi-hi. If the text of the Two Lights had been better transmitted, perhaps
some of these linguistic problems would not be present.
Not affected by this limitation is the mixture of style which runs through
the entire chapter of the Two Lights. This consists of short verses, reminis-
cent of the late Meccan sras, whereas the addresses O, you who believe
(v. 1), and O, Messenger (vv. 13, 24, and 34) are peculiar to the Medinan
sras.
The least harmless among the violations of Koranic thought is probably
the equationof the worshippers of idols, who are made apes andswine, with
the enemies of Moses andAaron(v. 29), whereas the corresponding sra 5:65
does not supply a historical connection. Far more important is the thematic
double character corresponding tothe above-mentionedstylistic mixture of
the chapter. Thus, the admonition to the Prophet to bear calmly the insults
(v. 30) as well as the strong emphasis on the Final Judgement (vv. 1, 11, 18,
26, and 39) and the early peoples with their messengers
107
(vv. 8, 9, 10, 29),
are some of the pet ideas of the Meccan sras. On the other hand, ignor-
ing the infidels completely andto the authors minddividing humanity
exclusively into believers and those who renounced their belief (vv. 4 and
23), does not make sense, even on the basis of Muammads late Medinan
period. On the contrary, this seems to refer to the conflict within Islam,
which did not arise until long after the death of the Prophet (vv. 24 and 39).
This conjecture is confirmed by several comments in this chapter, which [ii/110]
all culminate in the person of Al, the saint of the Sha,
108
addressing him
sometimes by his actual name (vv. 17 and 35), and sometimes by the com-
mon Shite by-word waiyy.
109
This way the fate of Al and his house is
predicted (vv. 5, 17sqq., 24, and 40). The favourite Shite name of honour,
imm, for Al and his descendants does not occur in this sra, yet the hos-
tile caliph is once mockingly inserted as the imm of the sinners (v. 22
110
).
The name nr light,
111
added to the names of Muammad and Al (vv. 1
and 2), is related to a well-known Shite theory under somewhat gnostic
influence.
112
According to this, since creation a divine, luminous substance
107
The chapter always uses the designation rasl for Muammad as well as for earlier
prophets. Nab is not even used once.
108
Cf. above, p. 291.
109
Cf. Th. Nldeke, Zur tendenzisen Gestaltung der Urgeschichte des Islms, p. 29;
Goldziher, Muslimstudies, vol. 2, p. 114; Goldziher, Introduction to Islamic theology, p. 175.
110
Cf. above, p. 300.
111
The Koranic meaning of nr, cf. above, p. 300.
112
Goldziher, Neuplatonische und gnostische Elemente, pp. 328336; Tor Andrae, Die
Person Mohammeds in Lehre und Glauben, p. 319sqq.
has been passing from one chosen descendant of Adam into the next until
reaching the loins of Muammads and Als common grandfather; here,
this divine light split in two; one part entered Abd Allh, the father of
Muammad, and another the formers brother, Ab lib (IBN ABD AL-
MUALIB), the father of Al. From him this divine light passed on from
one generation to the next into the particular imm.
113
From this easily fol-
lows the idea of the miraculous union of Muammad and Al, which finds
expression in the words two lights, one from the other, similarly to what
Al says about himself in al-Shahrastn:
114
I amfromAmad like light from
light. The first verse of the Sra of the Two Lights also says that it is intended
to recite to man the Signs of God, and threaten torment and affliction. The
functions here ascribedtoAl the Koranreserves exclusively toMuammad,
the greatest and last of the prophets. Equally audacious is the invitation to
belief in the two lights (v. 1). Muammad, of course, is met several times
in the Koran as the subject of belief, but every time only after Allh (sras
7:158, 24:62, 48:9, 49:15, 57:7 and 28
115
).
This ought to be the overwhelming evidence that the so-called Sra of [ii/111]
the Two Lights is a Shite falsification, just as Kazem-Beg also recognizes it
to be.
116
For the time being the exact date of its origin cannot be determined
with precision, since little research has been done in Shite apocryphal lit-
erature. The Shite exegetes (Ab l-asan) Al b. IbrhmAL-QUMM
117
and
Muammad b. Murta [AL-KSH
118
] (d. 911/1505 or 6) do not seemto have
known the sra, or else they would have mentioned it in the introduction
to their commentaries to the Koran.
119
According to Kazem-Beg, there is no
authentic work onthe Immite traditionthat mentions this sra; andbefore
the sixteenth century there is no writer who knows nrayn as its title; after
all, nrni as the name of the twin-constellation Muammad-Al does not
appear until the fourteenth century.
120
113
Goldziher, Introduction to Islamic theology, p. 183; Sprenger, Leben und Lehre des Mo-
hammad, vol. 1, p. 294sq. Somehow connected with this is likely to be the light (nr) which
according to Sunnite tradition (Ibn Hishm, p. 101; al-abar, vol. 1, p. 1078; Ibn Sad, Biogra-
phie Muammads bis zur Flucht, p. 58sq.) became visible in the face of Abd Allh b. Abd
al-Mualib, and another which appeared when Muammad was born, and radiated a long
way. (Ibn Sad, Biographie Muammads bis zur Flucht, p. 36, with three variants.)
114
Translated by Haarbrcker, vol. 1, p. 128. Cf. also Kazem-Beg, loc. cit., p. 411.
115
Believe then in Allh, and in His Messenger.
116
Loc. cit., p. 428.
117
A student of al-Kuln (d. 328/939); Sezgin, GAS, vol. 1, p. 45, no. 29.
118
Brockelmann, GAL, vol. 2, p. 200.
119
Cf. the literary-historical appendix, p. 310sq.
120
Loc. cit, p. 424.
Just as obscure as the dating of the Sra of the Two Lights are the name [ii/112]
and person of its author. In any case, he masters his Koran as well as any
theologically trained Muslim. Nevertheless, as we have seen, he confuses
the literary periods of Muammad and occasionally fails to keep with the
linguistic usage of the Koran and even, at timesassuming the accuracy of
the transmitted textagainst the rules of Arabic grammar in general, even
in cases where it was not demanded by the formulation of new ideas or
concepts. The overwhelming congruence with the language of the Koran
is, thus, not natural and accidental but rather artificially created with the
intentional aim to disguise the falsification.
ADMINISTRATIVE MEASURES
FOR THE REALIZATION OF THE
UTHMNIC TEXT
This is a case where we are again faced with a dearth of information. Most [ii/112]
of the sources for Uthmns undertaking
1
utilized up to this point contain
nothing but insignificant remarks about the number of copies produced
or the places to which they were distributed, except the statement that
a codex was sent to every region under the sun.
2
More precise informa-
tion can be obtained almost exclusively from Muslim works related to the
Koran. According to the most widespread view, one copy was retained at
Medina and the three others were dispatched to Kfa, Bara, and Damas-
cus.
3
Other writers add Mecca, considering this to be in accordance with the
general opinion.
4
Still others mention seven places, adding Yemen and Ba-
rayn.
5
Ibn Wi [al-Yaqb] mentions even Egypt and Mesopotamia in his
1
Cf. above, p. 251 n. 1.
2
Fa-arsala il kull ufq bi-muaf, from al-Fihrist, Ibn al-Athr, al-Bukhr, al-Tirmidh,
al-Suy, al-Itqn; and Al al-Dn AL-KHZIN al-Baghdd [Juynboll, Encyclopedia, p. 232,
col. 1, n. 3]. The scholiast of the Riyya (Silvestre de Sacy, Commentaire sur le pome nomm
Rayya) puts it somewhat differently, but is equally general, arsalaUthmn il kull jund min
ajnd al-Muslimn muafan.
3
al-Dn settles for these four places in al-Muqni, Ahlwardt, Verzeichnis, no. 419, fol. 5
r
(cf. Silvestre de Sacy, Commentaire Akila, p. 344) cited by al-Qasalln on al-Bukhr,
Bulaq edition, 1303, vol. 7, p. 449. This edition is considered the best or most common
edition by al-Nuwayr (cod. Lugdun., 273), and (Ab l-Qsimb. Firrukh b. Khalaf AL-SHIB
[Sezgin, Geschichte, vol. 9, p. 41 ii]); cf. Brockelmann, GAL, vol. 1, p. 400; in the Riyya (cf.
Silvestre de Sacy, loc. cit., vol. 5 = p. 431), Ibn Aiyya, al-Qurub, vol. 1, fol. 21
r
, the scholia of
Muammad IBNAL-JAZAR (d. 833/14291430, cf. Brockelmann, Geschichte, vol. 2, p. 201) on
the Muqaddima of Ab Amr Uthmn b. Sad AL-DN (d. 444/10531054, cf. Brockelmann,
loc. cit., p. 407) and Shar al-Muqaddima al-Jazariyya, MS Wien, no. 1630, fol. 309 b, in Flgel,
Die arabischen, persischen und trkischen Handschriften zu Wien, vol. 3 (1867), pp. 6566.
4
Umar b. MuammadIBNABDAL-KF, cod. Lugd., 674Warner, no. MDCXXXIV(1634),
fol. 2, versus al-Dn, Muqni; Silvestre de Sacy, Mmoire , p. 432. According to al-Suy,
al-Itqn, p. 141, and al-Qasalln, loc. cit., this was the general opinion. [Brockelmann, GAL,
suppl. vol. 2, p. 212 (sic)] it was already found in al-Makk b. Ab lib (d. 437/10441045,
Sezgin, GAS, vol. 8, p. 251).
5
Umar b. Muammad IBN ABD AL-KF, fol. 2
v
; al-Dn, Muqni; al-Nuwayr, Ibn
Aiyya; al-Qurub. According to al-Tibyn f db amalat al-Qurn (according to cod.
Sprenger 403, W. Ahlwardt, Verzeichnis der arabischen Handschriften) by al-Nawaw (d. 676/
12771278, cf. Brockelmann, loc. cit., vol. 1, p. 394) and according to al-Itqn, p. 141, this view
306 administrative measures for the uthmnic text
historical work.
6
(Ibn) al-Jazars statement that there were originally eight
copies seems altogether to be a misunderstanding, as he makes no com-
ments whatsoever.
7
Among these different views preference should be given to the one that [ii/113]
corresponds best to the most reliable tradition on the genesis of the Uth-
mnic redaction. This tradition is known to connect with a disagreement
between Iraqi and Syrian contingents during a military campaign over the
different ways of reciting the Koran.
8
Most appropriate for our purpose is
the very first of the afore-mentioned views, which mentions among the for-
eign places only Kfa, Bara, and Damascus. They were the most important
cities and garrisons of the period in the provinces Iraq and Syria. It would
thus seem that the Caliph merely had in mind the settling of a dispute
among his troops. There was apparently no need at all for the more distant
goal of favouring his entire dominion with a uniform text of the Holy Book,
although the idea was attributed very early to Uthmn that since there was
in Islam only one God and one Prophet there ought to be also only one
Koran. Such dogmatic considerations are likely to be primarily responsible
for the growth of the oldest demographic statistics, even if all the minutiae
cannot be explained this way. The mention of Mecca evidently owes to its
importance as the birthplace of the Prophet and seat of the ancient sanctu-
aries, eventhoughthe people there likely always followed the first recension
of Zayd b. Thbit, just as in Yemen. The province of Barayn
9
probably fol-
lowed the custom of Iraq, just as Egypt followed Syria, from where it had
been conquered. The mention of seven localities might reflect the goal of
making the number of standard texts equal to the number of aruf, or vari-
ant readings of the Koran, and the later Koranic recitations.
10
It is not recorded whether, according to the unanimous tradition, the [ii/114]
copy retainedat Medina was the codex of afa or one of its newly produced
copies. Incidentally, this codex was allegedly destroyed by the Umayyad
concurred with the renowned grammarian AB TIM Sahl b. Muammad AL-SIJISTN
(d. 255/869 or 250), [EI
2
; Sezgin, GAS, vol. 8, pp. 9396; vol. 9, pp. 7677]; (cf. Fihrist, p. 58sq.;
G. Flgel, Die grammatischen Schulen der Araber, p. 87sqq.).
6
Ibn Wadhih qui dictur al-Jaqubi Historiae, ed. Houtsma, vol. 2, p. 197, where al-Jazra
takes the last place, and Mir (sic!) between Mecca and Syria.
7
Ibn al-Jazar, K. al-Nashr f l-qirt al-ashr, Ahlwardt, Verzeichnis, 657, fol. 3
r
, bottom.
8
Cf. above, p. 251 sq.
9
The scholiast of the Riyya thinks that tradition does not report that Yemen and
Barayn where places to which copies were dispatched. Cf. Silvestre de Sacy, Mmoire sur
l origine et les anciens monuments de la littrature parmi les Arabes, p. 432.
10
Goldziher, Schools of Koranic commentators, pp. 2628.
administrative measures for the uthmnic text 307
Marwn b. akam while governor of Medina (45/665 or 47/667), since
he suspected non-Uthmnic variant readings.
11
This account, however, is
doubtful, for the stated motive cannot possibly be reconciled with the fact
that the Uthmnic redaction was a copy of that codex.
The destruction of the codices of divergent recensions of the Koran, [ii/115]
which Uthmn ordered, according to the same account in traditions, is
likely to have been also limited to Iraq and Syria, given what was argued
previously. As far as this concerned public property, the governors certainly
had the means to enforce such a measure, although copies in private hands
were practically beyond their reach. Some traditions state that the method
employed was tearing up
12
the codices. This cannot possibly have been the
case, as this would not have protected the individual pieces and shreds from
further profanation. This interpretation could possibly be designed to load
the obnoxious caliph with yet another sacrilege. The superstitious reserve
and reverence displayed by Islam vis--vis the word of God demanded
total destruction, best accomplished by burning. This is indeed what
most authorities report.
13
If, according to the commentary on the Koran by
Muammad b. Murta [AL-KSH] (d. 911/1505), Uthmn had those
codices first torn up (mazaqa) and then burned, this taleas the Shite
character of the work would suggestevidently aims to make the sacrilege
appear even worse, although the burning might also have served as com-
pensation for the mischievous tearing it up.
It would seemthat the general public recognized the utility of the admin- [ii/116]
istrative measures. Difficulties allegedly arose only at Kfa. The old Com-
panions living there were glad when the newmodel copy arrived,
14
although
11
So Ab Muammad Makk, al-Kashf an wujh al-qirt wa-ilalih wa-ujajh, Ahl-
wardt, Verzeichnis, no. 578; al-Qasalln, vol. 7, p. 419, from IBN AB DWDprobably
from K. al-Maif of Ab Bakr Abd Allh b. Sulaymn AL-SIJISTN (Sezgin, GAS, vol. 1,
pp. 174175, d. 316/971972)says Marwn: I have done this only because I was afraid that
in the course of time someone might doubt it.
12
Ibn al-Athr, Chronicon, ed. Tornberg, vol. 3, p. 87, and Ab Muammad Makk, loc.
cit., p. 503 express this with kharaqa, al-Suy, Itqn, p. 430, and al-abar, vol. 2, p. 747,
unmistakably with shaqqa, and Muammad b. Murta [AL-KSH], K. al-f, Wilhelm
Ahlwardt, Verzeichnis der arabischen Handschriften, no. 899, with mazaqa.
13
al-Bukhr, al-Tirmidh, al-Khab al-Tibrz, al-Khzin, Mishct, al-Suy, al-Itqn,
p. 138, al-Fihrist, Ibn Wadhih qui dicitur al-Jaqubi Historiae, vol. 2, p. 196, Ibn Khaldn, al-
Muqaddima, vol. 2, p. 135. For this reason this viewis defended quite rightly in al-Muqni, Ibn
Aiyya, fol. 25
v
, and al-Qurub, vol. 1, fol. 20
r
. As the word for to burn, araqa is in Arabic
writing ( ) distinguished from kharaq, to tear up ( ), only by the omission of one dia-
critical point, the written transmission is somewhat uncertain. It is commendable that some
transmissions have instead of kharaqa unequivocal synomyms like shaqqa and mazaqa.
14
Ibn al-Athr, Chronicon, ed. Tornberg, vol. 3, p. 87; al-Tirmidh, Tafsr on sra 9, end.
in other respects they were hostile to the Caliph. Ibn Masd, however,
requested his supporters to resist and furtively hide their Korans.
15
In
retaliationaccording to one sourcehe was called to Medina and there
subjected to heavy corporal punishment on the order of Uthmn.
16
Not much credibility ought to be attached to this account, as the same
source contains another conspicuous note. Accordingly, the governor who
demanded the copy of the Koran from Ibn Masd was Abd Allh IBN
MIR
17
who, however, according to most accounts, was since 29/649 gover-
nor of Bara, whereas at that time Sad b. al- ruled at Kfa, being recalled
at the end of 34/654 and replaced by Ab Ms AL-ASHAR.
18
Among the
other great Koranic authorities of the age who produced their own recen-
sions Ubayy b. Kab was no longer alive.
19
A reflection of Miqdd b. Amrs
attitude is the fact that, when he died in 33/653, Uthmn said the prayer
for the dead.
20
If only we knew the year the canonical Koran was intro-
duced.
21
It is certain that the third of the renowned collectors, Ab Ms
AL-ASHAR,
22
saw the introduction of the Uthmnic Koran, as he did not
die until 41/661 or 42/662.
23
However, we do not know whether this hap-
penedbefore his appointment as governor of Kfa. Inany case, preparations
for a newredaction must have been well under way at that time. The Caliph
would hardly have entrusted Ab Ms AL-ASHAR with such a high office,
particularly in permanently unruly Iraq, if he had not been sure that the lat-
ter would carry out the anticipated innovation.
The accounts of the disposal of the pre-Uthmnic codices of the Koran [ii/117]
reflect such certainty and unanimity, and are provisioned with so many
details that could not easily have been fabricated, that it is difficult to doubt
their historicity. For Christian scholars, this fact is so firmly established that
they cannot visualize the enforcement of the new redaction without the
support of the police. Conversely, I cannot quite see either the necessity of
the measure or its purpose.
15
Cf. above, p. 287 n. 41.
16
Ibn Wadhih qui dictur al-Jaqubi Historiae, vol. 2, p. 197.
17
Abd Allh IBN MIR b. Yazd al-Yaub, d. 118/736; Sezgin, GAS, vol. 1, pp. 67.
18
L. Caetani, Chronographia Islamica on the respective years.
19
Cf. above, pp. 235236 and pp. 253254.
20
21
Cf. above, p. 251252.
22
A. Jeffery, Materials for the history of the text of the Qurn, pp. 209211; Juynboll, Ency-
clopedia, s.v.
23
administrative measures for the uthmnic text 309
Most importantly, however, the objective, the realizationof the canonical
redaction, was not at all achieved by the destruction of the earlier recen-
sions. In order to understand this one must begin with the conditions of the
present. Our present reciters of the Koran present the revelation by heart,
even if during the service they keep a copy in front of them in order to pre-
serve its solemnform.
24
Inteaching, too, writtenor lithographedcopies serve
only as an auxiliary; the important aspect is and remains the free verbal
presentation of the teacher.
25
If this is the case still today, when there are
numerous manuscripts and innumerable lithographs
26
available, howmuch
more importance must have been attached to recitation from memory in
the time of Uthmn, when complete codices of the Koran were a great rar-
ity. Thus, it must apply to the past what can be observed everywhere in the
Islamic Orient today, i.e. that the reciter of the Koran who once memorized
his text according to a certain reading (qira) is unable to learn anew. In
such circumstances, the newrecension could not prevail until a newgener-
ation of reciters had grown up. But in order to encourage this it would have
sufficed to prescribe the use of the canonical recensioninthe public schools
of the Koran. The older recensions would then gradually disappear of them-
selves, without the necessity to destroy them.
Another reason that would have made the disposal appear inexpedient [ii/118]
is the consideration for the rarity and price of leather and parchment, the
only contemporary writing materials for books, particularly in the case
manuscripts of high quality and oversize. In viewof such circumstances one
could have simply corrected individual textual variants, rearranged pages or
signatures, and, at worst, obliterated all writing and rewritten the page, a
method that was common throughout the Middle Ages in the Orient as well
as in the Occident.
27
24
In Egypt many of the reciters of the Koran are totally blind.
25
Quite similar, possibly even more so is the system of transmission among the Indians.
In the History of Indian literature (vol. 1, p. 29) Maurice Winternitz says: It is an interesting
phenomenon that in India from time immemorial up to the present day what mattered for
the whole literary and the scientific activity was the spoken word and not the script. Even
today when Indians have known the art of writing for centuries, when there are innumerable
manuscripts and these manuscripts enjoy even a certain degree of sanctity and veneration,
whenthe most important texts are accessible eveninIndia incheapprintthe entire literary
and scientific activity in India is based on the spoken word. Not from manuscripts or books
do they learn the texts, but from the mouth of the teachertoday as millennia ago.
26
Ordinary printing of the Koran is prohibited.
27
This is probably behind the idea in Ibn Wadhih qui dictur al-Jaqubi Historiae (vol. 2,
p. 196, bottom) when he reports that Uthmn had the old manuscripts of the Koran cleansed
However that may be, after the introduction of the Uthmnic redaction [ii/119]
all forms of the older recensions, regardless of their reputation, disappeared,
save some uncertain traces. This was undoubtedly a great blessing for the
unity of the Islamic Church, but an irreparable loss for our knowledge of the
beginnings of Islam and the genesis of its Holy Book.
with hot water and vinegar. When once under the Abbsid ruler Amn the chancellery was
ransacked the people of Baghdad used the parchment documents for writing purposes after
they had washed them; cf. Ibn al-Nadm, al-Fihrist, ed. Flgel, p. 21, l 18sqq.
THE ISLAMIC CANON
AND ITS RELATION TO ITS CHRISTIAN
AND JEWISH COUNTERPARTS
The Founders of Judaism
The Jewishreligionwas not foundedby a single personbut rather developed [ii/119]
gradually over the course of several centuries from an older first stage, the
religion of the Israelites. The mutation of important religious documents of
various periods of this development toward an authentic ritual prestige also
proceededgradually over a periodof some five hundredyears. The historical
development, however, remained so much alive that the different parts of
the canon, namely law, prophets, and hagiographies, retained a sequence
corresponding to their chronological origin and were never amalgamated
into one entity within Judaism.
The Position of Jesus in Christianity
Although the foundation of Christianity emanated from one person, Jesus
cannot be considered its founder. In the Messianic community that sprang
up after his death, Christ immediately became the object of religion.
1
Since
Jesus left neither revelations nor other writings, nascent Christianity had
no holy document of its own and instead had to content itself with the
canon of its origin, the Synagogue. The NewTestament, consisting of a great
variety of Christian writings from different periods, did not reach a kind
of literary completion in the Occident until the end of the fourth century,
whereas the process in the Oriental Church took even longer. Afterwards, it
also became customary for Christianity to recognize the triple Jewish canon
as a unit and, unlike the New Testament, combine it under the name of the
Old Testament.
1
Valuable related ideas can be found in Eduard Meyer, Ursprung und Geschichte der
Mormonen, p. 279.
312 the islamic canon
The Development of Canonical Islam
The genesis of the Islamic canon is entirely different; one might even say [ii/120]
that it was the product of the opposite development. It is not the work of
several authors but of only one man, and was therefore accomplished in the
span of a lifetime. The form of the Koran as we now have it was essentially
complete two to three years after the death of Muammad. The Uthmnic
redaction is only a copy of the collection of afa and was completed under
Ab Bakr or in the reign of Umar at the latest. This redaction was probably
limited to the composition of the sras and their arrangement. As far as the
individual revelations are concerned, we can be certain that their text is
transmitted generally in the form in which it was found in Muammads
literary bequest.
Added to this considerably different development is a difference of liter-
ary form. The writings of the Jewish as well as the Christian canon are the
work of man, although very early the conception prevailed that the Bibli-
cal writers were moved by the Holy Ghost (2Peter 1:21) in an extraordinary
way. The direct word of God is found only where He speaks to prophets or
other chosen, pious men. In the Koran, the situation is entirely different.
Muammad is, of course, objectively and actually the writer of the revela-
tions recorded in this Book, although he does not consider himself to be
the author but rather only the mouth-piece of Allh and mediator of His
word and will. In the Koran, therefore, only God is speaking, and God alone.
Although the historian of religion will recognize this as mere fiction, for the
Prophet it was utterly true, given his enthusiasm for the divine origin of the
revelations, and his congregation believed it.
Muammad was as familiar with Judaism and Christianity as one could [ii/121]
be during his time, becoming so dependent on these religions that hardly
a single religious idea in the Koran is not derived from them. He was also
aware that both religions were in possession of sacred writings and for this
reason called their followers the People of the Book. In other respects,
he had the strangest notions about historical relations. He imagined, for
example, that Jews and Christians had received the same revelation, which
in each case was falsified. For this reason, he, the Arab Prophet, was chosen
by Allh once again to recite the text of the ancient revelation from the
Celestial Tablets. As soon as he was certain of his divine mission, he had
those revelations recorded the way he received them.
The creation of a particular holy document was consciously envisaged
at the birth of Islam. This is an artificial and unoriginal trait of the reli-
gion that would indicate close connections with certain gnostic sects. Islam
the islamic canon 313
is also connected with themin sharp contradistinction to Judaism and
Christianityinsofar as Islam goes back to a particular person as its
founder.
The peculiar theory of the relation of the Koran to the earlier revelation
apparently stems from Muammads accurate realization that all his reli-
gious and ethical ideas are borrowed fromthe religions of the Book. As far
as we know, this theory is certainly original. We would probably arrive at a
different conclusion if the sectarian original literature of the first Christian
centuries had been better preserved.
THE ISLAMIC SOURCES
ON THE ORIGIN OF THE REVELATIONS AND THE
GENESIS OF THE BOOK OF THE KORAN
The MuslimSources: The Foundations of the Systemof Transmission
Islamic literature, insofar as it is relevant for our study, is almost exclusively [ii/122]
in Arabic. It comprises biographies of Muammad and his Companions,
canonical adth, history of the first caliphs, contemporary poetry, commen-
taries onthe Koran, andprolegomena. It goes without saying that this survey
must be limited to the most important works used in this History of the
Qurn. The form and content of these works must be more thoroughly dis-
cussedthanis possible ina general history of literature. Awelcome aidinthe
field of canonical adth comes fromIgnaz Goldziher,
1
and for the history of
the Prophet the excellent studies of EduardSachau,
2
Carl Brockelmann,
3
and
Leone Caetani are of great assistance.
4
In spite of this, the number of works
devoidof details about concepts andsources still remains enormous. Insuch
circumstances, and in order not to reduce this appendix to a scanty list of
names and titles, I had no choice but to attempt myself, even without the
support of comprehensive monographs, to establish the bare minimum in
the hopeful expectation of the indulgence of the experts.
It is in the nature of the subject that modern Christian studies take up [ii/123]
only one fourth of the appendix. Works still important for the present [1918]
generally do not date from before the middle of the nineteenth century. I
shall consider older works only occasionally, provided they are of perma-
nent influence on subsequent development, and endeavour to present all
results and their characteristics, including their merits and shortcomings,
objectively and impartially, so as to enable general historians and scholars
of religion also to orientate themselves in the subject, even if they are not
Arabists.
1
Goldziher, Muslimstudies, vol. 2 (1977).
2
Sachau in the introduction to his edition of Ibn Sad (al-abaqt, vol. 3, part 1): Biogra-
phien der mekkanischen Kmpfer Muhammeds in der Schlacht bei Bedr [Biographies of Mu-
ammads Meccan combatants in the Battle of Badr], pp. vixl.
3
C. Brockelmann, Das Verhltnis von Ibn el-Ars Kmil fit-Tar zu Tabaris Abr
er-rusul wal-mulk [sic,] Dr.phil. thesis, Straburg, 1890.
4
Annali, vol. 1 (1905), pp. 2858.
316 the islamic sources
The Foundations of the Systemof Transmission
The most reliable part of the extant sources on Muammads life and work
are undoubtedly archival documents such as treaties, letters, and official
lists.
Arab chroniclers write with a confidence unknown in similar world liter-
ature, even when not supported by documents. As a sign of authenticity, the
individual transmissions are normally headed by no more than the isnd, or
chain of authorities. This consists of the strand of intermediaries between
the author of a work and the eyewitness of the reported event. An illustra-
tion can be found in the section concerning Muammads final illness in
FerdinandWstenfelds editionof IbnHishm, Das LebenMuhammeds (
), p. 1005, l 1619:
Said(qla) IbnIsq(d. 151/768): I was told(addathan) fromYaqbb. Utba
al-Thaqaf (d. 128/745) from Muammad b. Muslim al-Zuhr (d. 124/741) from
Ubayd Allh b. Abd Allh b. Utba (d. 94/712) fromisha, Muammads wife,
she said: The Messenger of God went out between two men of his family, Fal
b. Abbs and another man, with his head bandaged and his legs trailing on
the ground, until he entered my room.
Such an account attested by a chain of authorities is called adth. These
adths are connected with one another corresponding to the disposition
followed by the authors, and the chronology of the events.
The chain of transmitters is not always as complete as in this example. [ii/124]
Not infrequently one link or another is missing, less because of negligence
than because of literary principles, as shall be demonstrated in detail in the
following two chapters. Still more typical of the isnd is that no distinc-
tion is made between oral and written transmission, with the result that a
literary dependence upon an earlier work is considered a verbal account
of its author. This is connected with the fact that the content (matn) of
the tradition originally goes back to oral reports. It also has to do with the
fact that later on, when transmitting had become a literary profession, a
pupils oral instruction by the teacher was of utmost importance, whereas
the accompanying written notes were considered as no more than an aid
to memory.
5
Consequently, no distinction is made between that part of the
isndderivedfromwrittensources andthat part fromtheir authorities, even
though it is obvious that the former are a far more reliable guaranty, as they
can be referred to at any time.
5
For a different point of view see F. Sezgin, Goldziher and hadith, pp. xivsqq.
the islamic sources 317
By no means did Muslim scholars accept an isnd uncritically,
6
although [ii/125]
they appliedformal criteria only andwere generally satisfiedwhenthe chain
of authorities was sound, when the individual reporters were known to have
been in contact with one another, and when, at the end of the chain, the
name of a Companion appeared. When these formalities were met, even
the most glaring logical or historical absurdities of the text, called (matn),
was quietly accepted. Christian research in the Occident only gradually
freed itself from this spell during the last two decades of the nineteenth
century. In reality even the most perfect isnd from the Arab point of view
represents no more than the history of transmission of a given event and is
thus of only textual-historical importance without any value judgement. It
can be demonstrated that the names of eyewitnesses are not infrequently
falsified and that, even more frequently, they are likely the fabrication of
traditionists, who often recognize in this a legitimate means of stamping
their accounts with the seal of absolute reliability. It is rather suspicious that
those Companions who most frequently serve as authorities belong to the
younger generation, whereas the earliest and most respected followers of
the Prophet appear only rarely in such a role. For example, in the isnds
of Ibn Isq according to Wstenfelds edition, Ibn Abbs is mentioned
thirty-eight times,
7
Ab Hurayra eight times,
8
Anas b. Mlik six times,
9
while
Umar, the caliph, only twice.
10
In al-abars Annals the testimony of Ibn
Abbs is referred to 286 times, Ab Hurayras fifty-two times, and Anas
b. Mliks forty-seven times; the first four caliphs, however, are not even
mentioned once.
11
Muslims, who are quite aware of these facts, think that
the early Companions were preoccupied with spreading Islam, religious
wars, and the salvation of their own souls.
12
This is certainly a pertinent
observation.
The first generation of believers was too involved in the events to reflect
on themhistorically. In spite of this, Muslims hold in no less esteemthe reli-
ability of the younger generation. Objections were raised only occasionally
6
Cf., e.g., Muslim in the introduction to his ah; al-Suy, Tadrb al-rw, Cairo, 1307/
1889; Goldziher, Muslimstudies, pp. 135144.
7
Pages 131, 138, 204, 207, 227, 302, 323, 368, 371, 376, 395bis, 428, 446, 449, 450, 470, 484,
551, 585, 604, 642, 749, 750, 789, 790bis, 796, 810bis, 927, 943, 960, 965, 1010, 1013, 1017, 1019.
8
Pages 368, 400, 468, 579, 765, 964, 996, 1012.
9
Pages 261, 571, 574, 757, 849, 903.
10
Pages 64 and 463.
11
Goldziher, Muslimstudies, vol. 2, p. 139sq.; Caetani, Annali, vol. 1, p. 43.
12
Ibn al-Athr, Usd al-ghba f marifat al-aba, vol. 1, p. 3, top.
against individuals such as Anas b. Mlik and Ab Hurayra. The criticism,
however, was not directed against the subject of their traditions but rather
against their lower social standing, for both of themhad been servants; after
all, the same Muslim criticism accepts without hesitation the most obvious
falsehoods and farces of other authorities. To the younger generation also
belongs isha, who is frequently referredto, withmore than1,200 traditions
to her credit. Although she had been married to Muammad for eight years,
she was a young girl of eighteen years at his death. She became a mature
woman and influential person only as a widow. What we know of her later
life and career as an unscrupulous, political, scheming woman certainly
raises serious doubts about the reliability of her statements. For Muslims,
however, as the mother of the believers, andthe apparent favourite wife
13
of
the Prophet, she has always been held in high, almost sacred esteem. Thus,
so many fabricated traditions were placed in her mouth that she cannot
possibly be held responsible for everything that is currently attributed to
her.
For obvious reason, andincontradistinctiontothe procedure of Muslims,
the isnd can only be considered the secondary or final criterion when
judging the historicity of the matn; the main criterion, of course, remains
the criticism of the content of the tradition.
The reliability of the Arab chroniclers is generally no greater or less than [ii/127]
other early historical sources that cover related matters from the same dis-
tant remove. Accordingly, critical research is inevery case governed by iden-
tical principles. For instance, the accounts fromthe time when Muammad
was the recognized head of the Medinan ecclesiastic state are more reliable
than those concerning his childhood and the beginning of his career, since
interest in his vicissitudes during the Meccan period can have developed
only much later. With respect to both periods it must not be forgotten that
outstanding personalities, particularly founders of religions, are, for per-
sonal, political, or dogmatic motives, affected especially early and easily by
the tendentious twists of transmitted material. Since the driving forces are
rarely evident or extant, it is always difficult todetermine the formanddirec-
tion of the transformation. It is likely to require decades of work until the
most glaring distortions of the Prophets biography have all beenrecognized.
The early development of proper historical writing in Arabic is nearly
inconceivable without the mighty impetus of Islam, although, on the other
13
This wide-spread opinion, too, must be tantamount to a colossal swindle of the enter-
prising widow.
hand, this presupposes the existence of an Arabic national literature. In the
first instance there is poetry, which was quite advanced, even in pre-Islamic
Arabia. Without this background, the Holy Book of the new religion would
have rather appeared in a Syriac or Ethiopic form. The transmission of
poetic products from one place to another, and from one generation to the
next, was in the hands of professional men whom we might even consider
rhapsodists. The indigenous name is rw, which originally meant water
carrier andlater became transmitter; their professionthus became known
as riwya. Since these technical terms later became common expressions in
the historical field of tradition as welleven though they did not penetrate
the isnd schemefrom a practical point of view a dependence on the
rhapsodist guild cannot be excluded.
But all these factsprovidedthey are correctare at best nothing but an [ii/128]
explanation of certain peculiarities of historical prose, like the Arabic dic-
tion and the isnd, the poetic accessories and some technical terms, but not
at all conducive to the rise of historiography. Given the cultural situation in
Arabia of the time, suchanactivity cannot have appearedspontaneously but
must have arisen froma related literary genre. The question that poses itself
is: where should one look for it? In this regard, there is nothing available
on Arab soil except an ancient narrative prose that, being an explanation
of the songs, was likely also recited by the rhapsodists. It generally dealt
with the armed quarrels of individual heroes and the feuds between fam-
ilies and tribes. A dependence on foreign literature, which one finds among
contemporary Byzantine chroniclers of foreign lands, seems to be absent in
the Orient, unless one considers the Middle Persian chronicles, of which we
know hardly anything at all.
Joseph Horovitz
14
attempted to trace the isnd back to a Jewish origin.
Although he offers some surprising parallels, the evidence leaves something
to be desired. On the one hand, the chain of authorities never played the
same role in Jewish literature as it did the Arabic adth, even at the end
of the first century ah. On the other hand, the Jewish usage has no history
either within Jewry proper or in Israelite thinking, facts which are likely to
point to a foreigninfluence. And finally, the questionof the Arabic isnd can
hardly be separated from the older historical literature of the Arabs, which
no one would want to blame the rabbis for.
14
Alter und Ursprung des Isnd in der Islam, vol. 8, pp. 3947.
The Biography of the Prophet
The interest in the life of the Prophet first centred on the military cam- [ii/129]
paigns. Among the earliest authors of such maghz books there is men-
tion of Abn, a son of Caliph Uthmn (d. 105/723724), Urwa,
15
the son
of the familiar Companion al-Zubayr b. al-Awwm (d. between 91/709710
and 101/719720) as well as the two freed slaves, Shurabl b. Sad [Sad
16
]
(d. 123/740741) andMs b. Uqba
17
(d. 141/758759). Their books, apart from
a small fragment of the latter,
18
have been lost, although they have partially
survived in later works and are now their most precious parts.
The predominance of that interest becomes evident immediately in the
earliest extant work of Muammad IBN ISQ, a Medinan client, who,
however, was writing at the court of the second Abbsid caliph and died in
151/768769. Since more than half of the work is devoted to the campaigns,
though it is usually entitled Sra, it is occasionally also called Book of the
Maghz.
19
The book is no longer extant in its original version but only
15
EI
2
; EQ; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 1, pp. 278279.
16
17
EI
2
; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 1, pp. 286287.
18
See E. Sachau, Das Berliner Fragment des Ms b. Uba.
19
al-Dhahab, Tadhkirat al-uff, Hyderabad edition, vol. 1, p. 155; al-Dhahab, Tajrd
asm al-aba, vol. 1, p. 4; Izz al-Dn IBN AL-ATHR, Usd al-ghba f marifat al-aba,
vol. 1, p. 11; according to J. Horovitz, Aus den Bibliotheken von Kairo, Damaskus und Kon-
stantinopel, (p. 14sq.) in a manuscript of the Kprl Ktphanesi, Constantinople; al-
Masd, Les Prairies d or (vol. 4, p. 116) it reads Kitb al-Maghz wa-l-sayr. However, I am
sceptical of the reading and think that ought to be changed to .
(August Fischer adds the following remarks: Ibn Isqs work is not occasionally, but
commonly called Book of the Maghz. As far as I know, Sra it is called merely in Yqt,
Irshd, vol. 6, p. 399, l 4 [ ]. Conversely, Ibn Hishms excerpt regularly has this
title.) Cf. M. Hartmann, Die angebliche sra des Ibn Isq; further, Ibn Sad in Ibn Hishm,
Das Leben Muammads, vol. 2, p. vii, bottom; Ibn Qutayba, Handbuch der Geschichte, p. 247,
M. [sic]; Yqt, Irshd, vol. 6, p. 399, ll 10 and 16, 400, l 9, 401, l 9; Ibn al-Athr, Chronicon, vol. 5,
p. 454; Ibn Khallikn, Blq ed. 1299/1881, vol. 1, p. 612, l 15; Ab l-Fid, Annales moslemici,
vol. 2, p. 26; Ibn Sayyid al-Ns [EI
2
], Ibn Hishm, loc. cit., vol. 2, p. 19, l 16; al-Suy, abaqt
al-uff Kl. V. 12 [sic]; IbnTaghrbird, Annales, vol. 1, p. 388, l 4; jj Khalfa, vol. 5, p. 646, etc.,
[and the following additions: Loc. cit., same note: The form of the title, is
found not only at the passage indicated by Schwally but also in Ibn Khallikn, vol. 1, p. 611, 1st
line, p. 365, l 13; Ab l-Fid, Annales, vol. 2, p. 150; al-Dhahab, Tadhkirat al-uff, vol. 1, p. 156,
l 1, and Khula, p. 326, l 2 from bottom (cf. also
, jj Khalfa, vol. 5, p. 646)].

Yqt, Irshd, p. 401, l 9, reads instead
. alone in jj Khalfa, vol. 3, pp. 629

and 634; al-Fihrist, p. 92, l 1, has (in Ibn Hishm, vol. 2, p. x, Ibn Sayyid al-Ns biography
of the Prophet is entitled
). This establishes the accuracy of .

But this is not, as Schwally wants, to be read al-sayr but al-siyar (= way of life, biography;
see Brnnow-Fischer, Chrestomathie aus arabischen Prosaschriftstellern (1913), the available
translations of the passages cited, as well as Martin Hartmann, loc. cit.).
in the edition of Abd al-Malik IBN HISHM,
20
a South Arabian scholar
later living in Egypt (d. 218/833834), from a copy made by Ziyd b. Abd
AllhAL-BAKK,
21
a private pupil of the master. Unfortunately, IbnHishm
did not limit himself to comments and additions, instead considerably
abridging the text as well. As he himself stated in his preface,
22
he omitted
all accounts that did not mention Muammad, which are unrelated to
Koranic verses, and which can be considered neither cause, explanation
nor evidence of other events mentioned in the book; furthermore, he left
out poems which no other scholar knew, and, finally, passages which to him
seemedobjectionable or likely tooffendothers or were not authenticatedby
al-Bakk. Still, it might be possible to establish Ibn Isqs original version,
since other copies circulatedthat were usedby later historians like al-abar,
Ibn Sad, and Ibn al-Athr.
23
The abundant literary use of Ibn Isq by outstanding writers is a reflec- [ii/130]
tion of posteritys gratitude. Nevertheless, derogatory remarks about the
author were also current among Muslims. In the science of tradition he is
allegedly weak or unsound, frequently merging several traditions with
one another without supplying details of the deviation, citing useless ac-
counts of unknown persons, attributing false names, and being called an
outright liar.
24
Since these verdicts aim at the form of the isnd, they are
based on the pertinent observation that Ibn Isq does not conform to the
requirements that were instituted a century later. Whereas a model isnd
of al-Bukhr or al-abar consists of an uninterrupted chain of transmit-
ters from a reported event down to these two authors, Ibn Isq follows no
fixed system, sometimes omitting links, sometimes supplying an allusion
20
Das Leben Muhammeds, Arabic text edited by F. Wstenfeld, Gttingen, 18581860,
p. 1026 [followed by an illegible rest of a numeral, possibly 8, referring to line?]. G. Weils
1864 translation is awkward, and philologically leaves much to be desired. The importance of
the work would warrant a new translation. In English there is now the translation by Alfred
Guillaume, The Life of Muammad (1955), a work which has been repeatedly reprinted.
21
EI
2
; Sezgin, GAS, vol. 8, p. 119: al-Buk.
22
Wstenfelds edition, p. 4, l 6sqq.
23
Cf. Sachaus lucid exposition in his edition of Ibn Sad (abaqt, vol. 3, part 1): Biogra-
phien der mekkanischen Kmpfer Muhammeds, p. xxivsq. This desideratum has now been
realized by Gordon D. Newby with his The Making of the last prophet; a reconstruction of the
earliest biography of Muammad, the Kitb al-Mubtada (1989). This is a work of great impor-
tance. Prof. Newby has managed to reconstruct much, perhaps nearly all, of the lost first book
of Ibn Isqs Sra. This reconstruction should encourage scholars to look again in detail at
the early development of Muslim attitudes to the older religions.
24
Wstenfelds edition, ibid., pp. xxxxiii.
instead of the actual name
25
or even omitting everything.
26
These irregu-
larities and inconsistencies cannot be attributed to Ibn Hishms irrespon-
sibility. Instead, they are connected with the mutation of the isnd from
the undefined to the particular form, so that Ibn Isq is the intermedi-
ary in the historical development. This had already been known not only to
his principle predecessors (Muammad b. Muslim) Ibn Shihb AL-ZUHR
and Urwa b. al-Zubayr,
27
but most likely can be traced back to the very
beginning of the adth. No matter how complete or incomplete the isnd
might be, the historical value of the respective account will never depend
on it.
The other objection of Muslim criticism, which concerns the form of [ii/131]
the account, also appears to be immaterial. Although it is correct that Ibn
Isq frequently created a coherent account, consisting of a variety of tra-
ditions,
28
this is not serious since the variants are frequently very minor and
do not affect the meaning. But even where greater deviations are glossed
over by harmonizing methods, it is hardly appropriate to criticize outright
a method that would be applauded as artistic expression in other types of
literature.
Ibn Hishms importance is not only based on the chronological prece- [ii/132]
dence by being the author of the earliest extant maghz history, but also
in his unrivalled abundance of good information, which is far from being
sufficiently explored by research. I am thinking first of all of the numer-
ous genuine contemporary poems. The fact that he permits the opposi-
tion to give vent to their partially malignant abuses of the Prophet is evi-
dence of an astounding impartiality. These and his other merits are best
reflected in his recognition by his successors. He demonstrated his sense of
historicity by transmitting the most important document of early Islam in
its entirety, the so-called constitution of Medina,
29
whereas later writers
25
For example, a man, men, a shaykh, a client, some one, some (fifty-three times),
someone who I trust (four times), someone who I do not distrust (thirty-three times),
according to the lists compiled by Wstenfeld in his edition, ibid., pp. lviiilxix.
26
Most of the time it merely says: Ibn Isq says.
27
Goldziher, Neue Materialien zur Literatur des berlieferungswesens, p. 474.
28
For example, in Wstenfelds edition, p. 263, Muammads Night Journey; p. 428, the
Battle of Badr; p. 555, the Battle of Uud; p. 699, the siege of Medina; p. 725, the cam-
paign of al-Murays; p. 894, the campaign of Tabk; I. Goldziher, Neue Materialien zur
Literatur des berlieferungswesens, p. 474, points out that already al-Zuhr (d. 124/741) did
not list the individual authorities separately, instead he combined them whenever he saw
fit.
29
Edition by F. Wstenfeld, ibid., pp. 341344.
because of dogmatic prejudices, it wouldseemeither ignoredit altogether
or have only poor references tothis document.
30
The only work that contains
the unabridged text after Ibn Isq appears in the Prophets biography by
Muammad b. Ab Bakr IBN SAYYID AL-NS (d. 734/13331334).
31
The second oldest extant historical work on the Prophet was written in [ii/133]
Baghdad by the Medinan Muammad b. Umar AL-WQID
32
(d. 207/822
823) and contains nothing but the campaigns. Alfred von Kremers edi-
tion
33
has the text proper only up to page 369, line 16; the rest is nothing
but a worthless addition of a later period.
34
Thanks to two newly discov-
ered London manuscripts, Julius Wellhausen then published a brilliant Ger-
man version.
35
This, however, cannot be a substitute for the Arabic original,
particularly since the Calcutta edition is philologically insufficient and rep-
resents only a third of the actual work. As is the case with Ibn Isq, we
also do not have al-Wqid in its original form but only in the recension
of (Muammad b. al-Abbs) IBN AYYUWAYH,
36
a scholar of the fourth
century [295/907381/991]. The omission of several poems promised in the
text must be ascribed to him or to one of his predecessors. Wherever there
are deviations from Ibn Isq, in most cases Ibn ayyuwayh offers the bet-
ter or more original alternatives. In general, numerous passages look like
shortened versions of Ibn Isq, although he is never mentioned as author-
ity.
37
Al-Wqids main merit is the complete collection of the material but
not its chronological parts, where his comments are not infrequently con-
tradicted by odd remarks in the accounts. Although the material added is
largely legendary and anecdotic, it still offers much local flavour and natu-
ral colour. The isnd is a bit more regular and complete than in Ibn Isq.
The general indication of sources, which are so dear to the latter, are totally
wanting. Where the majority of the sources agree, he refers to them at the
30
The passages from biographers have been collected by Caetani, Annali, vol. 1, p. 376sq.
A.J. Wensinck, Muammad and the Jews of Medina, p. 59sqq. completed them, particularly
with parallels from the collections of adths.
31
The book is entitled Uyn al-thr (f funn) al-maghz wa-l-shamil wa-l-siyar Cod.
Leiden, 340, fol. 62
v
; W. Ahlwardt, Verzeichnis, nos. 9577 and 9578. Cf. Brockelmann, GAL,
vol. 2, p. 71; F. Sezgin, Geschichte des arabischen Schriftums, vol. 1, p. 301; EI
2
.
32
EI
2
33
Calcutta, Bibliotheca Indica, 1856.
34
In addition, also p. 7, l 9 (ql al-Wqid) to p. 9, 2 (al-ya) is an unknown substitute.
35
Muhammed in Medina; das ist Vakidis Kitab al Maghazi in verkrzter deutscher Wie-
dergabe. The following description I took partly literally from Wellhausens preface.
36
Sezgin, GAS, vol. 1, p. 206, no. 247.
37
The only passage of Kremers edition where he is mentioned is not genuine; cf. above,
foot-note 29.
beginning of the section, followed by the common text.
38
Variant readings
he indicates in between, supplying their origin; and he introduces the con-
tinuation of the main text by ql they said.
The Bara-born Muammad IBN SAD (d. 230/844845), a pupil and sec- [ii/134]
retary of al-Wqid, is the author of a biography of the Prophet, which has
been transmitted as manuscript together with a book of classes (abaqt),
but originally probably did not constitute an independent work. In the
Fihrist (on page. 99) it is called Akhbr al-nab, a name that fits the sub-
title, which in the London Ms. appears in the colophon;
39
but it is commonly
called Sra. The Berlin [i.e. Leiden] edition (al-abaqt al-kabr) consists of
vol. 1, part 1 (161 p.), vol. 1, part 2 (186 p.), vol. 2, part 1 (137 p.), vol. 2, part 2
(pp. 198). As far as the volume is concerned, the work ranks even consider-
ably behind Wstenfelds abridged Ibn Isq. The missing portion, accord-
ing to my estimate, amounts to 213 and 152 pages respectively, depending
on the page size of Ibn Hishm or Ibn Sad. The abridgements apply to the
first partearly history,
40
Muammads childhood,
41
first events after the
hijra, where Ibn Hishm has nearly twice as muchas well as the cam-
paigns, where Ibn Sad has three times as much. On the other hand, Ibn
Sad is again more comprehensive in early Biblical history (vol. 1, part 1,
pp. 5 to 26), Muammads genealogy (pp. 28 and 2736), and the signs of
prophethood (pp. 95126). He treats Muammads character in particular
detail [ifa] (vol. 1, part 2, pp. 87 to 186,
42
appearance, dress, habits, etc.) to
which Ibn Hishm (pp. 149sq. and 266sq.) devotes but two pages, the last
illness and Muammads death (vol. 2, part 2, pp. 1 to 98, which get five
times less coverage in Ibn Hishm, pp. 9991027), and Muammads cor-
respondence, including the messages
43
to him (vol. 2, part 2 (pp. 15 to 86),
which receive twice as many pages as in Ibn Hishm). The great value of
the Sra is its last mentioned section, consisting almost exclusively of doc-
uments. Although the content is by no means complete, and some parts
38
Caetani, Annali, vol. 1, p. 34sq.
39
khir khabar al-nab, Ibn Sad (al-abaqt vol. 2, part 2): Letzte Krankheit, Tod und
Bestattung [final illness, death and funeral], edited by Fr. Schwally, p. 98.
40
Totally omitted is nearly everything related to pre-Islamic Arabia, and which, in Ibn
Hishm up to p. 100, takes up so much space.
41
Still, also in in this part there are complete legends, or legendary traits that are missing
in Ibn Hishm. All these problems await monographic research.
42
Inthis sectionto page 166 al-Wqid is not evenonce mentionedas a source, conversely,
repeatedly on the last twenty pages.
43
The most recent study of Muammads letters is Jakob Sperbers thesis Die Schreiben
Muammads an die Stmme Arabiens (1916).
are wanting, which have been preserved in Ibn Hishm,
44
hereas Well-
hausen
45
saysyou get a much better idea of the course of conversionof the
Arabs. This is partially the result of the systematic presentation of the mate-
rial, for which Ibn Sad displays an undeniable inclination. Let me recall the
list of persons named Muammad before the rise of Islam,
46
then the chap-
ter on the signs (almt) of prophethood, which later became the impetus
to the extensive literature of dalil al-nubuwwa, and, finally, the chapter on
Muammads character (ifa), whichlikely servedal-Tirmidh as a model for
his al-Shamil.
Ibn Sads main source for the campaigns is of course al-Wqid, his lord [ii/135]
and master. How he made use of him cannot be determined until this work
is completely available in the original Arabic version. Not only in the cam-
paigns
47
but also in other parts of the sra
48
the pupil vindicates Ibn Isqs
work, which al-Wqid made use of without advertising so. In this case, it is
of great importance that he had at his disposal two different versions,
49
nei-
ther of which was identical with the one available to Ibn Hishm, for in
this case we can avail ourselves of the important help for the reconstruc-
tion of Ibn Isqs original text. As far as the composition of the isnd is
concerned, Ibn Sad is roughly equal with al-Wqid, although the latters
compressed narratives are made still more uniformby his consistently plac-
ing the variants at the end, which Ibn Sad spreads over the entire article.
In this respect, the comparison of larger sections is extremely informative,
especially the presentation of the battles of Badr and Uud in both writers.
44
Ibn Hishm, pp. 963sq., 940sq., 971 sq., and 944.
45
Skizzen und Vorarbeiten, 4th Heft (1898), p. 88.
46
Ibn Sad (abaqt, vol. 1, part 1): Biographie Muammads bis zur Flucht [biography of
Muammad to his flight], p. 111 sq.
47
For example, Ibn Sad (abaqt, vol. 2, part 1): Die Feldzge Muhammeds [Muammads
campaigns], p. 1, l 11; p. 3, l 2; p. 39, l 7; p. 40, l 13; p. 57, l 9; p. 134, l 16.
48
For example, Ibn Sad (abaqt, vol. 1, part 1): Biographie Muammads, p. 25, 4; p. 29, l 11;
p. 108, l 4; p. 122, l 21; (abaqt, vol. 2, part 2): Letzte Krankheit, TodundBestattungMuammads
[Muammads last illness and death], p. 19, l 19; p. 48, l 6; p. 78, l 20; p. 105, l 10; p. 171, l 5;
(abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung Muammads [Muammads
last illness and death], p. 3, l 12; p. 35, l 4; p. 44, l 26; p. 79, l 18.
49
These are the copies of IBRHM IBN SAD b. Ab Waqq al-Zuhr al-Madan (d. 183/
798799) [Sezgin, Geschichte des arabischen Schrifttums, vol. 1, p. 95, no. 14], and Hrn b.
Ab s al-Sham. Cf. the passages listed in the preceding notes 43 and 44; Ibn Sad (abaqt,
vol. 3, part 2): Biographien der medinischen Kmpfer Muhammeds in der Schlacht bei Bedr
[Medinancombatants at Badr], pp. 1719; and E. Sachau inthe introductionto IbnSad, vol. 3,
part 1, Biographien der mekkanischen Kmpfer Muhammeds in der Schlacht bei Bedr [Meccan
combatants at Badr], p. xxv.
In that case also the structure of the composition is most solid in the
chapters on the early history as well as in the letters and missives, whereas
the Prophets characteristics (ifa) and his death constitute something more
like a loose conglomeration of traditions.
50
Amost useful complement to the sra are the biographies of the Compan- [ii/136]
ions in the abaqt. The relevant sections of Ibn Sads Berlin (Leiden) edi-
tion, al-abaqt al-kabr, are as follows: vol. 2, part 2, pp. 98136; vol. 3, parts 1
and 2; vol. 4, parts 1 and 2; vol. 5, pages 328341, 369379, 382390, 400403,
406412; vol. 6, pp. 643; vol. 7, part 1, pp. 163, part 2, pp. 6465, 99101,
111151, 176177, 188199; andvol. 8, whichdeals exclusively withwomen. The
total of these accounts (1213 pages) constitute nearly three times the size of
the entire sra (445 pages).
The entry for the Prophet in Ibn Qutaybas
51
Kitb al-Marif (IbnCoteibas
Handbuch der Geschichte) is too short (pp. 5683) to be of any help. The
author does not present a cohesive account of the events and limits himself
to compressed statistical surveys, devoting a disproportionally large space,
two thirds of the pages (pp. 5674), to Muammads characterhis descent,
family relations, women, children, slaves, and riding animals. The sources
are rarely supplied: al-Wqid (p. 59), and Ibn Isq (p. 75) once each, Ab
l-Yaqn
52
three times (p. 69, ll 1 an 6, p. 76), Zayd b. Akhzam (d. 257/870)
with following isnd (pp. 80 and 83). Abd Allh b. [al-]Mubrak
53
(d. 181/797,
p. 77, l 4) and, unknown to me, one Jafar of Ibn Ab Rfi (p. 83). More useful
are the chapters on Muammads Companions (pp. 83174), which contain
many important details.
The compendium of world-history, Ibn Wadhih qui dicitur al-Jaqubi His- [ii/137]
toriae, by al-Yaqb (d. 278/891892), edited by M.Th. Houtsma (1883), is
remarkable not only because it is the only Shite work of history from the
earlier period but also because it is based on sound, early sources. In its
interest in cultural-historical matters it reminds us of Ibn Qutayba and the
50
This connection becomes evident in some similarities to the linguistic usage. For
example, the phrases rajaa l-adth il and dakhala l-adth baduh f ba never once occur
in Ibn Sad (al-abaqt) Letzte Krankheit, Tod und Bestattungen (vol. 2, part 2), pp. 198, and
only once each in (al-abaqt): Biographie Muhammeds; Ereignisse seiner medinischen Zeit
(vol. 1, part 2), pp. 87187. Conversely, in Ibn Sad (al-abaqt), vol. 1, part 2, pp. 186, the
first phrase is found nine times and the last twice in (al-abaqt): Biographie Muhammeds;
Ereignisse seiner medinischen Zeit (vol. 1, part 2), whereas both phrases together occur twelve
times in (al-abaqt): Biographie Muammads bis zur Flucht (vol. 1, part 1).
51
Sezgin, GAS, vol. 9. pp. 154158.
52
Sezgin, ibid., vol. 1. pp. 266267, no. 3.
53
118/736181/797; Juynboll, Encyclopedia, p. xxxii, col. 2, etc.; Sezgin, GAS, vol. 1, p. 95.
later al-Masd.
54
The section on the life of the Prophet (vol. 2, pp. 1141)
contains nothing new. Still, it does contain some compilations that were
not taken into account in earlier biographies, such as, for example, a list
of the Meccan and Medinan sras (pp. 3234), of Muammads amanuen-
sis (pp. 8792), and a rich anthology of his alleged sermons (pp. 98121). As
regards astrologywhich here probably makes its first appearance in the
biography of Muammadthere is the regular reference to one Muam-
mad b. Ms al-Khwrizm
55
(pp. 5, 21, and 126). This is an invented nar-
rative and refers only now and then to the sources listed systematically
at the beginning of the second volume (p. 31). Following his Shite lean-
ing, the author most frequently quotes the sixth imm of the Twelver Sha
(d. 148/765), Ab Abd Allh JAFAR b. Muammad AL-DIQ (d. 148/765)
56
pp. 7, 8, 21, 34, 44conversely, he cites Ibn Isqwhom, according
to the preface, he used in the recension of Ibn Hishm from [Ziyd] al-
Bakkas well as al-Wqid but only twice each (pp. 20, 45; 43, 121) and
four other traditionists
57
twice each. He otherwise refers to his sources; in
the case of a difference of opinion he resorts to general expressions.
58
On the history of the caliphs the reports on the recension of the Koran [ii/138]
by Ab BakrUmar (p. 152) and Uthmn (p. 196sq.) deserve particular
attention as well ashis presentation is again typically Shitea detailed
description of Als collection of the Koran (pp. 153154).
The famous work, the conquest of the countries (Fut al-buldn),
of Amad b. Yay AL-BALDHUR
59
(d. 279/892893)a Persian-born
resident at the court of the Abbsid caliphs Mutawakkil and Mustan
refers to the Prophet (pp. 194), although not exclusively, since the con-
quests are dealt with beyond the Umayyad period. Also relevant are some
sections of the final chapter, penmanship, about literate men and wom-
en among Muammads acquaintances (p. 472sqq.). The work abounds in
54
Sezgin, ibid., vol. 1, pp. 332336.
55
In the introduction to the second volume, p. 4 l 6, called al-munajjim, the astrologer.
This of course cannot refer to a namesake of his who died in 428/10361037, to whom
C. Brockelmann refers in GAL, vol. 1, p. 225.
56
Juynboll, Encyclopedia, pp. 260262.
57
Ab Abd Allh Fal b. Abd al-Ramn, Muammad IBNKATHR, Muammad b. Sib,
and Ab l-Bakhtar.
58
For example, wa-raw bauhum, wa-qad ruwiya, wa-qla, wa-yuqlu, pp. 7, 8, 15, 1823,
33, 34, 37, 4046, 49, 52, 58, 59, 64, 73, 79, 92, 97, 98, 121, 125, 127.
59
Liber expugnationis regionum(Leiden, 1866). There is a cheap Cairo reprint, and a Ger-
man translation by Oskar Rescher, El-Beloris kitb fut el-buldn (Buch der Eroberung
der Lnder), based on de Goejes edition.
valuable additions to the accounts of older works. The form of the chain
of authorities is very precise. The numerous sources used would warrant a
separate investigation. Among the parts that are relevant for us, he refers
six times to Ibn Isq,
60
seven times to Ibn Sad, and twenty-three times to
al-Wqid.
In al-abars (d. 310/923 in Baghdad) world-history, the era of Muam- [ii/139]
mad is covered in vol. 1, pp. 10731836, of the Leiden edition.
61
What Ibn
Isq relates in his early history naturally appears in al-abar spread over
different places. For a comparison of the two works it is best to limit one-
self to the sections dealing with the Medinan period, namely al-abar,
vol. 1, pp. 12271836 (609 pages), and Ibn Hishm, pp. 3141026 (712 pages).
But since al-abars 609 pageswithout the foot-notesare equivalent
to only 430 pages of the layout of the Wstenfeld edition, al-abars pre-
sentation is 282 pages shorter. Of his predecessors he uses primarily Ibn
Isqwho, with two hundred references, far outdistances al-Wqid (47)
and Ibn Sad (15)and it is to him that the strand of the narrative always
returns.
62
Included in his numerous references are many that are wanting in
Wstenfeld,
63
since al-abar made use of a different recension.
64
Because of
60
On p. 10, Ibn Isq refers to two transmissions that are not found in either Ibn Hishm
or al-abar.
61
Annales [Akhbr] quos scripsit cumaliis edidit M.J. de Goeje, 13 volumes totalling 8054
pages in Arabic. The period relevant for us has been edited by Pieter de Jong. The Arabic title
of the Leiden edition is given as Akhbr al-rusul wa-l-mulk, or simply Tarkh (Fihrist, p. 234;
al-Masd, vol. 4, p. 145.)
62
Raja al-adth il adth Ibn Isq: p. 1299 l 5; p. 1301 l 6; p. 1308 l 9; p. 1315 l 3; p. 1389 l 18;
p. 1392 l 6; p. 1398 l 1; p. 1465 l 13; p. 1487 l 13; p. 1492 l 14; p. 1514 l 17; p. 1532 l 6; p. 1540 l 3; p. 1620
l 12; p. 1770 l 18.
63
al-abar, vol. 1, p. 1126 l 111127 l 8; 1142 ll 1619 and 1143 l 3 respectively; 1162 l 81163 l 2;
1171 l 11173, l 1; 1192 l 41191 l 10; 1253 ll 716; 1318 ll 26; 1321 ll 1315; 1340 l 101341 l 15; 1344 l
91345 l 6; 1357 ll 1014; 1365 l 151366 l 9; 1369 ll 815; 1398 ll 1416; 1400 ll 914; 1416 l 91417 l
6; 1419 ll 812; 1441 l 511; 1454 l 91455 l 2; 1496 ll 914; 1560 ll 36 and 1719; 1561 l 81568 l 2;
1569 l 11570 l 7; 1572 ll 1013; 1574 l 41575 l 5; 1576 ll 2 and 3; 1578 ll 59, 131579 l 1; 1617 ll 47;
1640 l 17; 1641 l 7; 1642 l 171644 l 13; 1657 ll 11 and 12; 1683 ll 312; 1705 ll 14, 15; 1809 l 171810 l 1;
1834 ll 1316. Best known of these passages is p. 1190sqq. on the temporary acceptance of the
Meccan idols in the Islamic cult, and p. 1441 on the capture of Abbs [IBN AL-MUALIB],
Muammads uncle, in the Battle of Badr. The historicity of this eventfor which Sprenger
pleaded at the time in his Leben und die Lehre, vol. 3, p. 131Leone Caetani tried to challenge
in his Annali dellIslam, vol. 1, p. 517. Cf. against this, Nldekes review of Annali dellIslam in
WZKM, 21 (1907), p. 309sq.
64
The isnd consistently runs as follows: Ibn umayd from Salma b. al-Fal from
Muammad b. Isq. The formers full name is Muammad b. umayd b. ayyn Ab
Abd Allh al-Rz (d. 248/862.) Salma b. al-Fal Ab Abd Allh AL-RZ died after 170/786
(Khula).
such circumstances, and because of the numerous variants in the two texts,
al-abar is the most important aid in the re-establishment of the original
Ibn Isq.
65
Al-abars next merit is the communication of new material as well [ii/140]
as the conscientious separation of divergent reports, either by listing all
traditions known to him directly one after another or only a single tradi-
tion but including all the divergent readings.
66
Before a group of related
transmissionsor occasionally at the endhe readily supplies a compar-
ative or statistical table of contents,
67
whereas a direct personal comment
on the respective question is rare.
68
It happens occasionally that general
references like some people think, others say, it is claimed, etc.
69
are
not followed up with a reference to a specific source. On the other hand,
al-abar avoids altogether the favourite habit of previous writers, namely
incorporating various traditions into a uniformnarrative, unless such a har-
monization is already part of his original, in which case he makes an appro-
priate note.
70
As can be seen, al-abars traditionalist formalismtook a shape that had [ii/141]
never existed before. Although this is a step back artistically, it enhances the
usefulness of his work for the historian.
Other parts of his world-history that deal with the first caliphs contain
many important reports concerning Muammad and his time. Of prime
importance are, for example, the accounts of the appearance of the prophet
Maslama, which, despite their fragmentary condition, are unique in their
65
66
Cf., for example, al-abar, vol. 1, p. 1565 l 12sq.
67
For example, fromthe text pp. 1245 l 71247 l 3, the following scheme can be uncovered:
qla Ab Jafar wa-ikhtalaf al-salaf min ahl al-ilmf dhikr man ql dhlik fa-ql bauhum
wa-qad wfaq qawl man ql dhikr man ql dhlik wa-ql kharn ql Ab Jafar
wa-qad wfaq qawl man ql. This system is changed in the most variant ways as becomes
obvious from the following passages: p. 1227 l 16sqq.; p. 1242 l 10sqq.; 1249 l 16sqq.; 1250 l
12sqq.; 1256 l 12sqq.; 1259 l 10sqq.; 1281 l 8sqq. [sic;] 1262 l 12sqq.; 1263 l 4sqq.; 1270 l 12sqq.;
1273 l 6sqq.; 1276 l 15sqq.; 1279 l 9sqq.; 1281 l 4sqq.; 1296 l 13sqq.; 1357 l 15sqq.; 1362 l 1 sqq. and
16sqq.; 1367 l 9sqq.; 1375 l 8sqq.; 1502 l 9. sqq.; 1767 l 14sqq.
68
For example, al-abar, vol. 1, p. 1259 l 14, wa-al-ah indan f dhlik. In other parts of
the annals still other expressions are to be found, for example, ana ashukk, I have doubts
(vol. 1, p. 522 ll 3 and 13).
69
For example, p. 1297 l 12, wa-qad zaama bauhum, ql kharn, ql, yuql, p. 1233 and
1245 l 5, p. 1248 l 9.
70
For example, al-abar, vol. 1, p. 1291 l 171292, l 1 = IbnHishm, p. 428 ll 2 and3; al-abar,
vol. 1384 ll 36 = Ibn Hishm, p. 555 l 12sqq., hence follows the account of the battles of Badr
and Uud.
form.
71
Al-abar everywhere shows himself to be an industrious and consci-
entious collector, although nowhere does he display a particular historical
aptitude.
We have a Persian adaption from352/963 of al-abars annals by Ab Al [ii/142]
Muammad AL-BALAM,
72
vizier of the Smnid ruler, Ab li Manr b.
N, or possibly by an unknown author on the order of this vizier. Although
this translation is considerably abridged and omits the chain of authorities,
it adapted much from other sources. If we had a critical edition of the
work it would be of no small usefulness for the Arabic original. Its French
translation by H. Zotenberg
73
does not serve the purpose, particularly as it
is based solely on Paris manuscripts,
74
whereas other manuscripts seem to
contain significant variants. Hence, a text edition on a broad manuscript
base still remains a desideratum. The Turkish edition, which is said to have
been printed in 1260/1844 in Constantinople,
75
remained inaccessible to me.
The famous work Murj al-dhahab (Meadows of gold), by the widely trav-
elled, learned and intelligent Ab l-asan Al b. al-usayn AL-MASD
(d. 345/956), is a mine of political, cultural, and literary history. The space
allotted to the Prophet in the French edition, including a translation, pp.
114175,
76
is, however, far too limited to permit the author to present new
material, not to mention his literary qualities. He narrates freely and rarely
makes use of the cumbersome technicality of the isnd. Of the earlier biog-
raphers he mentions Ibn Hishm, p. 116, Ibn Isq, p. 144 ll 6 and 11, p. 145 l 4,
al-Wqid, p. 144 ll 6 and 10, p. 145 ll 1 and 8, Ibn Sad, p. 145 l 8, and al-abar,
p. 145, l 8. The details of the first caliphs are also inadequate. The authors
interest is not aroused until he comes to Al, whose brief rule takes up three
times as many pages as that of his three predecessors in office.
In the great world-history (al-Kmil f l-tarkh) of (Izz al-Dn) Ab l- [ii/143]
asan Al IBN AL-ATHR (d. 630/12321233), which consists of twelve vol-
umes in the Leiden edition,
77
the life of Muammad takes up only a little
space, namely vol. 2, pp. 1 to 252. As the author states in the preface (vol. 1,
71
al-abar, vol. 1, pp. 1738, 1916sq., 1933sq., and 1951.
72
EI
2
; Sezgin, GAS, vol. 1, p. 327.
73
Chronique de Abou-Djafar MohammedbenDjerir benYezidTabari, traduite sur laversion
persane d Abou Ali Mohhammed Belami par Herm. Zotenberg, t. IIV, Paris, 18671874.
74
Here, Schwally overlooked the subtitle, traduite sur la version persane d Abou-Ali
Mohammed Bel Ami d aprs les manuscrits de Paris, de Gotha, de Londres et de Canterbury.
75
Brockelmann, GAL, vol. 1, p. 143; Sezgin, GAS, vol. 1, p. 327, Auszge.
76
Maoudi, Les prairies d or, texte et tradution par Barbier de Meynard et Pavet de
Courteille (18611877).
77
Ibn-el-Athiri Chronicon quod perfectissimum.
p. 4), he complements and continues al-abar, whose annals he uses as the
basis of his own work. Unlike the latter, he does not attempt to list as many
traditions as possible on one event, instead opting for a coherent account.
To this end, he fuses the traditions and tries to express himself as simply
and briefly as possible, omitting unimportant secondary details in order
to streamline the narrative. In the interest of this presentation he likewise
omits the chain of authorities and citations. If, for exceptional reasons, he
does want to supply a reference, he adds only the author of the work or one
of his authorities. For example, regarding the life of the Prophet, he refers
only ten times to Ibn Isq,
78
eight times to al-Wqid,
79
and only once to his
main source, al-abar.
80
Even the general indications of sources like qla,
it is said, are rare.
81
Disproportionately many authorities are cited in the
chapters on the prophetic mission, the early converts, and Muammads
biography.
82
Brockelmann, to whom we are indebted for an excellent [doc-
toral] thesis onthe relationof Ibnal-Athrs al-Kmil toal-abars Akhbr al-
rusul wa-l-mulk, documents numerous passages that have been borrowed
directly and without credit fromIbnHishmand al-Wqid.
83
All things con-
sidered, Ibnal-Athr is a great advancement, althoughhe cannot claiminde-
pendent recognition vis--vis older works. On the other hand, the sections
on the first caliphs contain important material not found in his predeces-
sors, such as, for example, the detailed account of Ab Bakrs and Uthmns
collection of the Koran.
84
The Canonical adth
85
[ii/144]
The biographic adth, including their origin and development, which we
traced through the centuries in the preceding chapter, are quite different
fromthe canonical adith.
86
The former lists the sayings and deeds ascribed
78
Vol. 2, pp. 29, 42, 43, 44, 86, 107, 111, 112, 144, and 155.
79
Pp. 28, 36, 44, 86, 107, 111, 174, and 131 [sic].
80
P. 144.
81
For example, pp. 33, 34, 36, 234, 236, 237, 238, and 239.
82
Pp. 3236, 4144, and 231 sq.
83
Straburg thesis, 1890, Das Verhltnis von Ibn el-Ars Kmil fit-tar zu al-abars
Abr , p. 31 sq. (see above, p. 315).
84
Vol. 3, pp. 8587.
85
There is nowthe most useful Encyclopediaof canonical adthby G.H.A. Juynboll (2007).
86
The canonical adth is occasionally called in Western Islamology adth proper or
in the proper sense. Whenever there is mention of the literature of adth in general, this is
meant.
to the Prophet Muammad, which are the binding models for Muslims
religious exercises and ritual behaviour, and which also serve as the basis
for the canonical and ethical handbooks.
87
The public events, which dom-
inate the sra, yield far less for such purposes than the Prophets private
affairs. Although this is not altogether without interest for the sra as well,
it remains much in the background. The canonical adth, however, are
in their proper element here and do not stop at either private matters,
like the secrets of intercourse, or relieving oneself.
88
If in both adth and
sra the same material is being used, it is not arranged chronologically but
rather according to dogmatic, ethical, or ritual aspects, with the result that
what is grouped together in the sra is widely dispersed in the adth. The
public events in the life of Muammad, which earlier works of adth still
consider, gradually recede and have totally disappeared when we come
to sunna-books. As we can see, there is a basic difference when dealing
with the two types of adthnot in the subject itself but in the way it is
treated.
Although reports on the behaviour of historical persons at home and in [ii/145]
the family are generally quite suspicious because of problems of evidence,
Western scholars have been particularly impressed by this aspect of the
canonical adith. What struck themmost was the unbelievably broad spec-
trum of the traditions, the long chains of authorities, the intimacy of the
content as well as the oftentouching, nave, andcandidstyle.
89
Evena distin-
guishedscholar like R.P.A. Dozy, the historianof the Moors of Spain, without
hesitation used half of al-Bukhr as a historical source.
90
But as had been
shown by Goldziherto whom we are indebted for the basic criticism of
the canonical adth
91
all movements and counter-movements in the life
of Islamhave beenprecipitated by the formof the adth, including not only
the dogmatic and canonical party rivalries of the first centuries, but also
the political confrontations. This is to say that theological defenders of doc-
trines or customs traced their opinions back to freely fabricated traditions,
and to alleged sayings and deeds of the Prophet, all for the purpose of more
87
Goldziher, Muslimstudies, vol. 2, German pagination in the margin, p. 5.
88
Juynboll, Encyclopedia, s.v. (46 and 4 references respectively).
89
Cf. for example, the frequently used phrases wa-ka-ann ar l-nabiyya, it seems to me
as if I were seeing the Prophet, wa-ka-ann wqif bayn yadayh, it seems to me as if I were
standing before him.
90
Dozy, Essai sur l histoire de l islamisme, Leiden, 1879, p. 124.
91
Muslim studies, vol. 2, On the development of the adth (1977), pp. 17251. See also
Fuat Sezgin, Goldziher and adth, translated from his GAS, vol. 1 (1976), pp. 5384, in
Goldziher, Schools of Koranic commentators (2006).
effective popularization of their view. The public in general did not per-
ceive this as a pious trick but merely as a legitimate literary device. The rise
of Muslim criticism of the traditionwhich has almost always been con-
cernedwiththe formonly, swallowing the most absurdcontent providedthe
isnd was soundwas unable in this case, as in the case of the biographical
adth, to effect a change for the better. This, however, is not to deny that
underneath the learned veneer of errors and lies there might still lie hidden
some credible traditions. Still, until mention is made to the contrary, every
canonical adth must a priori be considered a falsification.
This adth takes an exceptional position also as far as the treatment of [ii/146]
the isnd is concerned. Despite all the changes undergone by the judgement
of authorities inthe course of historywhichwill be shownwhenreviewing
individual worksthe canonical adth remained all along the domain of
the complete and uninterrupted chain of authorities. In contrast, within the
biography of the Prophet, as well as historiography in general, this form
reached its pinnacle in al-abar, later declining to the point that, in the
end, all reference to the sources disappears. On the other hand, the isnd
of the canonical adth is even far more conspicuous, since here the text
(matn) consists of many small sections, each required to be preceded by its
verification. This is why, in collections of adth, the isnd takes up at least
the same space as the actual text.
The Literature of adth
92
The high esteem accorded to isnd in the canonical adth, as just shown,
led early on to the development of the so-called musnad literature, where
the traditions are arranged according to isnds, more particularly to the
Companion in its final link. The earliest of the musnad works, and the
only one up to now that is available in a printed edition, is by Amad b.
Muammad IBN ANBAL (164/780241/855).
93
Within the general frame-
work of this arrangement some individual categories are listed in additional
subsections according to either relationship or sex, such as, for example,
92
The following presentation is based mostly on Ignaz Goldzihers Muslimstudies, vol. 2,
The hadth literature, pp. 189251, as well as his article, Neue Materialien zur Literatur des
berlieferungswesens [new facts on the literature of adth]. Cf. also F. Sezgin, Goldziher
and hadith, in: Goldziher, Schools of Koranic commentators (2006), pp. xiiixxxviii.
93
Cairo, 1311/18931894, four volumes in large quarto amounting to 2,888 pages. F. Sezgin,
GAS, vol. 1, pp. 504505.
the traditions of the Companions of Muammads family, of the anr, of
women, or according to place of residence or birth, such the Companions
fromIraq, Bara, Kfa, Mecca, andMedina, etc. As we might guess, this struc-
ture is not an invention of the author but rather a remainder of an earlier
period when the literature of adth consisted of such small, independent
collections. This is the easiest way to explain that not only the complete
work of Ibn anbal, but also each one of the above-mentioned categories
of tradition, is called musnad. The content extends to all matters that have
ever been the subject of adth information, to ritual laws and regulations,
legal norms, moral sayings, legends, and fables. Also, historical transmis-
sion, maghz, takes upa large space. Criticismnever exceeds the bounds of
moderationcommonamong Muslims, clinging tothe mere external aspects,
andusually stops far short of the limits. Yet whenIbnanbal goes so far as to
include even sayings of the Prophet in praise of the generosity of al-Saff,
the first Abbsid caliph, or which predict the conquest of India or proclaim
the fame of his hometownMerv, one is temptedtothink of a joke rather than
carelessness.
The concept of the musnad books complicates matters greatly, especially [ii/148]
since the general system of arranging by Companions
94
name, as pointed
out above, is then further subdivided into special groups of people, so that
a tradition might appear in different places or, if it appears only once, in
a place that no one would expect it. These difficulties were conducive to
the appearance of another type of adth literature, the so-called muannaf
books, in which the transmissions are arranged by content and brought
together in chapters.
95
Of course, this system also has its drawbacks, for
when a traditions content suits two different subject categories it is usually
assigned to different places, either partially or wholly, whereas if secondary
considerations determined its classification, it ends up in a place where one
would never expect to find it.
The most esteemed muannaf work is al-a
96
of Muammad b. Isml
AL-BUKHR (194/810256/870). The title of the book, meaning sound,
might be an indication that the author endeavoured to present his material
in a better formthan his teacher Ibn anbal. In any case, he made it a point
to eliminate traditions fromsuspicious authorities or of doubtful content, to
present both the text of the tradition as well as the isnd with conscientious
94
al l-rijl so the Arabic technical term.
95
al l-abwb.
96
It is also called al-Jmi al-a, cf. al-Qasalln inthe preface to his Irshd, Bulaq, 1303.
fidelity, to identify explanations and additions, particularly if they were his
own, and to list at the end, and under his name, his personal opinions
regarding various parallel transmissions. Yet all these improvements are
nothing but external features and are not in the least related to what we
call historical criticism. Even if we credit him with having considerably
reduced the learned jumble of tradition-related materialhis book makes
up only one third of Ibn anbals Musnadone must not believe that the
remainder is worth more than the discarded material. His teachers pet
authorities, Anas b. Mlik, the mendacious isha, and the evil-disposed
Ab Hurayra, are also preferred by al-Bukhr. The second part of the book,
Bad al-khalq,
97
and, most of all, the great maghz book, which has no
equal among the adth works, contain considerable historical material.
This is followed by commentary on the Koran. All the restin the edition I
used,
98
670 and 856 pagesis made up of the canonical adth, intermixed
with historical material here and there.
Another of Ibnanbals pupils, Abl-usaynMUSLIMIBNAL-AJJJ al- [ii/149]
Naysbr (d. 261/875
99
), alsoleft us a highly esteemedcollectionentitled[al-
Jmi] al-a.
100
The order of chapters differs fromthat of al-Bukhr. Addi-
tionally, the latters characteristic headings are missing, although [Yay
b. Sharaf] AL-NAWAW (d. 676/1277)
101
later added these to his own com-
mentary. Whereas al-Bukhr oftensprinkles identical traditions withdiffer-
ing isnd throughout the respective chapters,
102
Muslim b. al-ajjj lists all
97
Starting withthe chapter manqib ab al-nab. The chapter headings (bb) are from
al-Bukhr himself and constitute an integral part of the work, only the text differs according
to the respective recension. Conversely, the overall appellation, book (kitb), seems to be a
later addition. Cf. Goldziher, Muslimstudies, vol. 2, p. 220.
98
Cairo (alab), 1309, in 4
v
. On the numerous other editions cf. Brockelmann, GAL,
vol. 1, p. 158, [and Sezgin, GAS, vol. 1, pp. 118, no. 4, etc.] The Leiden edition by L. Krehl and
Th.W. Juynboll (18621908) is not particularly good. Ever since this first attempt no Christian
scholar tried to edit a work of adth, probably for the proper realization that Orientals are
best qualified for this task. We should restrict ourselves to producing alphabetical indices to
the traditions. The Frenchtranslation, Les Traditions islamiques, by O. Houdas andW. Marais
does not conform with modern requirements.
99
EI
2
; EQ; Juynboll, Encyclopedia; Sezgin, GAS, vol. 1, pp. 136143.
100
Goldziher, Muslim studies, vol. 2, p. 226sqq. On the various editions see Sezgin, GAS,
vol. 1, pp. 136143. I quote books andchapters according toal-Nawaws usage. Only inthe case
of voluminous works I supply volume and page numbers of the edition I am usingin the
margin of the Bulaq edition of al-Qasalln on al-Bukhr, Irshd al-sr il a al-Bukhr
(Bulaq, 1303) 10
v
. [Sezgin, GAS, vol. 1, p. 123, no. 29.]
101
Cf. Brockelmann, GAL, vol. 1, p. 394; Goldziher, The hirs (1971), p. 97.
102
Juynboll, Encyclopedia, p. xxx, col. 1.
variants in the first relevant place, without later again using this material.
The work is of little merit for our purpose, since, apart from a partial sec-
tion of Kitb al-fail and isolated passages, it does not contain a historical
section. The concluding chapter, interpretation of the Koran, is a miser-
able truncation which even in the marginal edition I usedas well as the
equally long commentaryamounts to no more than ten pages. The rather
extensive introduction to the science of tradition, which precedes Muslim
b. al-ajjjs collection,
103
deserves considerable praise.
The sunan works of Ab Dd [al-Sijistn] (d. 279/892 [or 275/888]), Ibn [ii/150]
Mjah
104
(d. 273/886), and Amad b. Al b. Shuayb AL-NAS (d. 303/915)
105
display a far greater interest in canonical and ritual affairs, but they are
referred to only when they preserve a tradition that does not appear else-
where.
Abs Muammadb. s AL-TIRMIDH (d. 279/892 or 3) canpride him-
self on being the pupil of three famous traditionists: Ibn anbal, al-Bukhr,
and Ab Dd [al-Sijistn]. His work is entitled sometimes Sunan, some-
times more appropriately al-Jmi al-sunan,
106
as from its content it is more
related to Muslimb. al-ajjj. Like the latter, it also contains historical mate-
rial in the abwb al-manqib,
107
albeit with a much longer commentary on
the Koran.
108
Its special place in the critical science dealing with the isnd
is remarkable. Al-Bukhr and Muslim recognize only such authorities on
whose reliability there is unanimous agreement among scholars. Ab Dd
[al-Sijistn] andhis pupil al-Nas, however, are evencontent whenthe sole
witness is not generally rejected. Al-Tirmidh goes still one step further and
accepts any tradition which, at any time, has been the accepted canonical
practice. On the other hand, he feels obliged to grade the reliability of any
tradition he is using.
109
However, as in all Muslimcriticism, this applies only
to the form of the tradition, so that the most audacious fabrications might
receive a splendid grade.
103
In the edition which I used, vol. 1, pp. 60184.
104
EI
2
; EQ; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 1, pp. 147148.
105
Goldziher, Muslimstudies, vol. 1, p. 225sqq.; Brockelmann, loc. cit., vol. 1, p. 161 sq.
106
In the edition which I usedDelhi, 1315/1897, two folio volumesthe title page has
Jmi, whereas the body of the text has the heading, Sunan. The introduction to the science
of tradition (al-Risla f fann ul al-adth), which nowcommonly appears at the beginning
of the edition, is not from al-Tirmidh but from the well-known scholar Al b. Muammad
AL-JURJN (d. 816/1413;) EI
2
.
107
Vol. 2, pp. 201234.
108
Vol. 2, pp. 111172.
109
For example, good (asan), weak (af), sound (a), strange (gharb), sounder
(aa). Combined grades are: asan a, and asan gharb.
Al-Tirmidh also composed a short study on the person and character [ii/151]
of Muammad (al-Shamil), consisting of fifty-six chapters.
110
Monographic
editions seem to exist only in connection with commentaries.
111
The copy
I used was printed as an appendix to al-a but with separate pagina-
tion.
112
It is entitled simply al-Shamil or now al-Shamil al-muaf or al-
Shamil al-nabawiyya wa-l-khail al-muafawiyya. The isnds are numer-
ous and extensive, although, unlike al-a, without critical comments.
The content is a lively reminder of the compilation Ibn Sad placed at the
end of his Sra, entitled ifat rasl Allh on this subject,
113
and likewise con-
sists almost solely of apocryphal, useless bits of historical information. The
arrangement of the material, as well as the headings, is quite different.
The number of chaptersal-Shamil, fifty-six, and Ibn Sad, fifty-eight
is remarkably similar. The more exact relation of the two works would be
worth particular research.
Wal l-Dn Muammad b. Abd Allh [AL-KHAB] AL-TIBRZs Mict- [ii/152]
ul-mabi
114
is a 737/1336 revision of Ab Muammad al-usayn b. Masd
al-Farr AL-BAGHAWs (d. 510/1117 or 516/1122)
115
Mab al-sunna. It pur-
ports to educate non-specialists, yet it also serves specialists well. The mate-
rial is selected accordingly. What remained of a sense of historical matters
was included in the chapters Fail al-nab and Manqib.
116
In the indi-
vidual sections we find first the traditions fromal-Bukhr and Muslimb. al-
ajjj, classedas a, followedby excerpts fromthe sunanworks classedas
asan, and finally some strange (gharb) or weak (af) traditions appear on
occasion.
117
Thanks to this arrangement, the book offers convenient access
110
111
Ibid., pp. 154159.
112
30 lithographed folios of the afore-mentioned Indian lithograph.
113
Ibn Sad (al-abaqt), vol. 1, part 2: Biographies Muhammeds; Ereignisse , pp. - ;
see above, p. 325sq.
114
Mishct-ul-mabh; or, Acollectionof the most authentic traditions, regardingthe actions
and sayings of Muhammed translated by A.N. Matthews, vol. 1, pp. 510584.
115
EI
2
; EQ; Goldziher, Muslimstudies, vol. 2, p. 247sq.; Sezgin, GAS, vol. 1, p. 155.
116
In my Indian lithograph (Dehli, 1310/1892), pp. 510584.
117
According to a pencil note in Schwallys manuscript he here wanted to comment on
the sunan work to which he repeatedly referred, Kanz al-umml f sunan al-aqwl wa-l-afl
of Al al-Dn Al AL-MUTTAQ al-Hind (d. 975/1567), Hyderabad, 13121314, parts 18. (2707
pp.) This workis basedonthe following three collections of traditions: First.Jam al-jawmi
(or al-Jmi al-kabr), a work which was planned to include possibly all of the adths. Its first
part was alphabetically arranged by the first line of the sayings (aqwl) of the Prophet; the
secondpart was basically arrangedalphabetically by the names of the earliest authorities and
presents the traditions of Muammads deeds (afl); (cf. on this jj Khalfa, vol. 2, p. 614,
to extant transmissions on a given subject in the seven main collections of
adth. Given the lack of other systematic reference works, the book is quite
useful for research.
All the works treated in this chapter are preceded chronologically by
al-Muwaa of Mlik b. Anas (d. 179/795.)
118
This, however, is not a adth
book but a corpus juris which presents law and justice, rites and praxis
of the cult according to the consensus and usage at Medina. Of course
it refers to adth, although by no means in every chapter, and without
attaching decisive importance to them. For this reason, the work deserves
to be referred to only when it is necessary to prove the prevalence of a
particular transmission in such an early period.
The Biographies of the Companions of Muammad [ii/153]
The earliest sources regarding the Companions of the Prophet are his own
biographies, compendiums, and world-histories, as well as works of adth,
particularly their chapters entitled manqib or fail. Nowhere in the early
literature are the Companions treated in such detail as in Ibn Sads book
of classes (al-abaqt).
119
Although al-abar also devoted a greater work
to this subject, it seems that only an insignificant excerpt survived, which
now appears at the end of his Annales.
120
More convenient reference works
are the writings of the subsequent period, which are arranged alphabeti-
cally.
The latest of these, which is available in a printed edition, is al-Istb
f marifat al-ab, by the famous Spanish scholar, Ab Umar Ysuf IBN
ABD AL-BARR al-Qurub (d. 463/1071.)
121
More comprehensive is Usd al-
and W. Ahlwardt, Verzeichnis der arabischen Handschriften, vol. 2, p. 155). Second.al-Jmi
al-aghr, an excerpt of no. 1 (repeatedly printed in Egypt, partly with the commentary of Al
b. Amad al-AZZ or the one of Munw). Third.Zawid (or Ziydat, also Dhayl) al-Jmi
al-aghr. Al-Muttaq al-Hind in his Kanz al-umml presents the complete compilation of
the adths contained in these three collections butfor the convenience of jurisconsults
in a systematic arrangement conforming to the prevailing legal categories, however, retain-
ing the separation of aqwl and afl, and in the case of the aqwl differentiating between
those fromal-Jmi al-aghr and Zawid. [FromA. Fischers additions and corrections, Seite
ii/222223.]
118
Goldziher, Muslimstudies, vol. 2, p. 197sq.; Sezgin, GAS, vol. 1, pp. 457464.
119
See above, p. 326.
120
Cf. M.J. de Goeje in the introduction to the Leiden edition of Annales quos scripsit Abu
Djafar Mohammed Ibn Djarir At-Tabari, p. xiii sq.
121
Brockelmann, GAL, vol. 1, p. 363sq.; printed, Hyderabad, 1318/1895.
ghba f marifat al-aba (lions of the thicket)
122
of Izz al-Dn IBN AL-
ATHR (d. 630/1232) the author of a world-history
.
123
As he remarks in his
preface, he depends on the above-mentioned Ibn Abd al-Barr as well as
three Isfahani scholars, Ab Abd Allh IBN MANDAH (d. 395/1004)
124
AB
NUAYM Amad b. Abd Allh AL-IFAHN (d. 430/1038)
125
and Ab Ms
Muammad b. Ab Bakr b. Ab s [al-Madn] (d. 581/1185).
126
For this rea-
son he acknowledges at the start which articles he borrowed from one
or several of these authors, using the corresponding letters (, , , .)
It is very strange that Ibn al-Athr developed the principles of the alpha-
betical arrangement to such detail,
127
as if it were something new, when
already Ibn Abd al-Barr, who lived a hundred and fifty years before him,
was familiar with this method.
128
Of the historical works consulted by Ibn
al-Athr, I merely mention Ibn Isqs al-Maghz, which he used in the
recension of Ynus b. Bukayr [Ibn Wil al-Shaybn],
129
so that we have
here yet another aid for the reconstruction of the original text of this impor-
tant book.
130
Whereas Ibn Sads book of classes treats some 1,860 Com-
panions, this number grew to 7,554 in Ibn al-Athr.
131
This was achieved
by including a wider circle of contemporaries in addition to the Compan-
ions.
Shams al-Dn Ab Abd Allh b. Al AL-DHAHAB (d. 748/1347) produced [ii/154]
from Ibn al-Athrs work a short excerpt
132
entitled Tajrd asm al-aba,
using also other books that he meticulously enumerates in his preface,
and to whichby means of particular lettershe traces back every new
contribution to its source. In the case of women, he made particular use
of the last part of Ibn Sads famous book of classes (vol. 8: Biographien der
Frauen [biographies of the women]). In this way al-Dhahab seems to have
122
Cairo edition, 1280/1863, in 5 volumes.
123
124
EI
2
125
Brockelmann, GAL, vol. 1, p. 362; EI
2
; Juynboll, Encyclopedia, p. xxxiii; Sezgin, GAS, vol. 1,
p. 88 etc., vol. 8, pp. 83, and 273.
126
Brockelmann, GAL, vol. 1, p. 355, last line; Sezgin, GAS, vol. 1, pp. 205, 215, 504.
127
Brockelmann, GAL, vol. 1, p. 5 l 20 to p. 6 l 11.
128
Ibid., vol. 1, p. 6 l 22 and 23, would indicate that the arrangement had predecessors (and
I saw much of the latest as soon as a book was arranged alphabetically ).
129
Cf., ibid., vol. 1, p. 11 l 9sq. The same recension has partially been used also by al-Wid
in Asbb al-nuzl, cf. Cairo edition of 1315/1894, p. 165 l 12. [EI
2
; EQ; Juynboll, Encyclopedia,
s.v.; Sezgin, GAS, vol. 1, p. 289 l 3.]
130
131
According to Ibn ajar al-Asqaln, al-Iba f tamyz al-aba, vol. 1, p. 3.
132
Two volumes, Hyderabad, 1315/1897.
succeeded in increasing the number of listed entries vis--vis Ibn al-Athr
by four to five hundred.
133
Because of its extreme brevity, the book can only
serve as an index to the works he was using.
The most comprehensive of all the known collections is al-Iba f tamyz
al-aba by Ab l-Fal Muammad b. Al IBNAJAR (d. 852/1448),
134
from
al-Asqaln in Palestine. Here, the number of entries has again been in-
creased, since, as stated in the preface, vol. 1, p. 4, he also has included those
who might possibly have seen the Prophet, either shortly before his death
or during the first years of their life. The Calcutta edition,
135
despite its vol-
ume, displays not inconsiderable gaps, since no perfect manuscript could
be procured.
Since chains of authorities are rarely supplied in the later historical lit- [ii/155]
erature of the Arabs, as pointed out above, and yet Ibn al-Athr and Ibn
al-ajar al-Asqaln made aneffort to produce a coherent description;
136
the
isnds in their large biographical collections take up much space. The rea-
son for this conspicuous phenomenon is that these works are not the result
of independent historical interests; they are, rather, totally in the service of
the critical tradition.
137
The traditions on which religious judgements hinge,
the knowledge of what is permitted and what is not, as well as other mat-
ters, are considered sound only when the strands of their isnds and the
transmitters are known beforehand. First and foremost, there are the Com-
panions of the Messenger of God. When someone does not know them, his
ignorance regarding other matters will be even greater. It is therefore nec-
essary to determine their origin and their circumstances of life; otherwise
one cannot know whether the application of what the authorities transmit
is correct, and the evidence ascertained. In this respect, transmissions from
unknown persons must not be used.
138
That a book pretending to be an aid
in the criticism of isnds follows this formality also in its own presentation
hardly comes as a surprise.
133
Cf. the preface, vol. 1, p. 4 l 3, where the total is estimated to amount to be eight
thousand.
134
Cf. Brockelmann, GAL, loc. cit., vol. 2, p. 67sq. EI
2
; EQ.
135
Calcutta, 18561873, 4
v
, 4,800 p. (Bibliotheca Indica.)
136
137
Connected with this is the small biographical compendium, Mal f asm al-rijl,
of Muammad AL-KHAB AL-TIBRZ (d. 737/1336) that was destined particularly for his
adth work entitled Mishkt al-mab, and is appendixed to the Indian edition. Cf. above,
p. 330sq.
138
Ibn al-Athr, Usd al-ghba f marifat al-aba, vol. 1, p. 3 ll 1419.
The Peculiarity of the Islamic Exegesis [ii/156]
of the Koran: The Exegetic adth
The analysis of the sras in the first part of our work offers hundreds of
examples
139
that allow us to appreciate properly the Muslim interpretation
of the Koran. Here it is necessary to summarize systematically the mistakes
that, it would seem, had appeared already in the earliest period:
1. Verses are attributed to historical events on the basis of a more recent
tradition, even though they had already been revealed earlier.
140
2. Insufficient attention to the sense of a word leads to overlooking the
obvious. For example, the words of sra 7:184 qad iqtaraba ajaluhum
(that it may be their term is already nigh), which clearly indicate the
general fate awaiting men after death, is considered in al-Wqid (ed.
by A. vonKremer, p. 132) tobe a reference tothe Battle of Badr,
141
where-
as verses that undoubtedly prophesy a military successlike sra
110are interpreted as an announcement of the death of Muam-
mad.
142
3. The origin of many sayings and laws in Jewish or Christian sources is
not realized and leads to distorted explanations as well as the presup-
position of impossible situations.
143
4. The habit of attributing every single revelation to a definite event [ii/157]
of contemporary history when in fact this attribution is based on a
misunderstanding of the general sense of most of the Koranic passages.
This mistake must be considered the most serious and disastrous of all, not
only because it is the one that occurs most frequently and is the most pecu-
liar mark of the Muslim commentaries, but also because it includes what is
for us the most important part of its content, namely information on the life
of Muammad. As Aloys Sprenger once remarked, the traditions contained
in the commentaries are so numerous and detailed that apart from geneal-
ogy and military campaigns, it would almost be easier to write the life of
Prophet without the biographies than without the commentaries.
144
139
140
For example, loc. cit., p. 108, on sra 16:126sq; and pp. 147148 on sra 65:1.
141
Asimilar inaccurate interpretationby the exegetes occurs withthe imperatives adkhiln
and akhrijn in sra 17:82; cf. the Commentators.
142
The commentaries; Ibn Qutayba, p. 82; note on Ibn Hishm, p. 933; al-Suy, al-Itqn,
pp. 45 and 910.
143
For example, above, p. 161 sq. on sra 4:46.
144
Sprenger, Leben und Lehre des Mohammad, vol. 3, p. cxx.
The traditions of the Commentators that are somewhat sound can be
divided into two groups. (1) Those that relate verses to important pub-
lic events, such as, for example, parts of sra eight to the Battle of Badr;
parts of sra three to the Battle of Uud; parts of sra fifty-nine to the
expulsion of the Jewish tribe of the Nar; sra 48:1 to 7, to the Pact of
udaybiyya; sra 49:1 to 5, to the embassy of the Tamm; sra 9:1 sqq., to
the pilgrimage of 9/630; and sra 5:1 sqq., to the Farewell Pilgrimage of
10/631. (2) Those that relate to private affairs and throw an unfavourable
light on the reputation of the Prophet or members of his household and
which for this reason are considered particularly reliable. Here belong the
occasions of narration of sra 24:11 sqq.; ishas
145
escapade with afwn b.
al-Muaal of sra 33:37; Muammads marriage to Zaynab [Bint Kab b.
Ujra], the wife of his adopted son Zayd [Ibn ritha]; and sra 66:1 sqq.,
Muammads treatment of his slave girl Mary(am) inthe quarters of his wife
afa.
146
The reliability of the transmissions of these passages we may indeed
not generally doubt, even if some details might be fictitiously embellished.
On the other hand, the meaning of the first sentence of sra 3:155 is too
vague for us to give the least amount of credibility to any of the transmit-
ted traditions.
147
Muammads encounter with the blind Ibn Umm Mak-
tm
148
of a respected Meccan family, can hardly be the background to sra
eighty, since the blind man, whose despicable treatment was the reason
that the Prophet was rebuked by God, was a poor man from the lower
class. The tradition, which reproaches the Prophet for having temporarily
included the pagan goddesses in the Islamic cult, must still remain suspi-
cious although no cogent proof for its inaccuracy has been suppliedso
far. But so long as the contemporary circumstances are not better known,
and, in the final analysis, the judgement is dependent solely on the two
questions of whether or not such a crude throw-back to paganism can be
attributed to Muammad, or the fabrication of such an accusation against
Muslimsquestions whichinall appearances canequally well be answered
negatively or affirmativelyresearch will never come up with an unani-
145
According to one adth isha is affectionately reminded on her death bed of her early
accusations fromwhichshe was splendidly exculpated. But inher oldage she does not care to
hear muchof the event. Cf. IbnSad(al-abaqt vol. 8): Biographiender Frauen, p. 25; Juynboll,
Encyclopedia, p. 721, col. 1.
146
See above, p. 171, 168, and 175sq; Juynboll, Encyclopedia, p. 396, col. 1.
147
Loc. cit., pp. 139.
148
mous agreement.
149
As far as the declarations of the two respective groups
are applicable, they are based not only on accompanying traditions but also
on Koranic hints that were clear enough to protect these traditions from
falling into oblivion, and, at the same time, serve as their confirmation.
Conversely, the great majority of the original narratives is quite suspi- [ii/159]
cious, for to all appearances most of the revelations of the Koran are of a
rather general nature and give no indication that they were revealed on
a particular occasion, which, in the rarest of cases,
150
Muslim exegetes will
admit. That this cannot be called an accompanying tradition becomes evi-
dent from setting the most contradictory traditions side by side, which we
did in numerous examples previously,
151
as well as fromthe routine-like pat-
tern of fabrication. It is thus popular to present the same typical persons,
such as uma b. Ubayriq as the thief,
152
for example, or Ibn Umm Maktm
as the blind man
153
or Abd Allh b. Ubayy as the hypocrite.
Generally, the number of persons who are claimed to be the subject of
revelations is extremely large. They belong to the most diverse classes and
circles, freedmen and slaves, Meccans and Medinans, muhjirn and anr,
believers and hypocrites, Jews and pagans. Al is reported to have said that
there is no Quraysh who was not the subject of revelation.
154
Some have
attempted to explain this fact by remarking that the creators of the exegesis
of the Koran were largely slaves or freedmen who, in order to enhance the
prestige of their patrons or belittle their enemies, inserted the respective
names in the earlier adths or simply coined new ones.
The mania to discover at all costs a precise historical occasion for every
revelation was, it seems, not influenced from outside, as Jewish exegesis
of the Bible was too far away from the events of its creation to dare to
offer descriptions of situations. On the contrary, this mania is of genuine
Arab growth, with roots intimately entwined with those of adth. Since the
Koran contains a certain number of totally reliable contemporary allusions,
149
Loc. cit., pp. 7274. Apart from Leone Caetani, all scholars have opted for the tradition,
lately Th. Nldeke, Die Tradition ber das Leben Muhammeds, p. 164. I only met with
reservations from Leopold von Ranke, who says in Weltgeschichte, part 5, vol. 1, p. 64: The
narrative is in sharp contradiction with everything authentically known of Muammad so
that I dare not accept it.
150
For example, loc. cit., p. 100 on sra 17:32.
151
Cf. above, pp. 72, 7576, 101103, 104105, 108109, 119120, 121, 126, 132133, 151, 156157,
159161, 163167, 170, 173174, 176177, 178179, 182, and 184188.
152
Loc. cit. p. 145sqq., 165 on sra 4:106sqq., and 5:141.
153
Loc. cit. p. 96 on sra 80.
154
which are connected with incontestable accompanying traditions, this led
to a general exegetic principle. As a consequence, the exegetes, on the one
hand, searchfor adths that are best suitedto shedding light ona revelation
and, if need be, go even further by incorporating in the adth words from
the respective Koranic passage. On the other hand, if these attempts were
futile, people did not hesitate freely to invent suitable situations (exegetic
adth in the wider sense). So long as we lack more precise investigations,
it is difficult to determine where one or the other precondition is appli-
cable. Yet this does not mean much, since both cases are untenable and
historically worthless combinations. Our judgement attains greater reliabil-
ity only when a legend can be shown in a satisfactory way to have derived
from the text of a Koranic passage (exegetic adth in the narrow sense), as
applies to some traditions on Muammads childhood and the beginnings
of his prophetic activity. This reminds me of the legends of Muammads
cleansing of the heart and the splitting of the moon, as well as the circum-
stances of the revelation of sra 74, all of which have probably been derived
from sras 94:1, 54:1 and 74:1.
155
But these reliable cases are rare. Henri Lam-
mens
156
far overshoots the mark when he has the entire adth relating to
Muammads life and appearance originate from Koranic indications; it is,
after all, unlikely a priori that a literature so diverse in content, form, and
tendency should have sprung up from one root.
The Exegetic adth
Whereas the bulk of the exegetic adthsno matter what sort of miserable [ii/161]
invention they might bestill rests on the facts of contemporary history,
some others can occasionally be found that transcend the barriers of time
and space without hesitation, and see the occasion of revelations in the
conditions of a remote future. Strictly speaking, we are dealing here with
prophecies andnot, by any means, revelations ona particular occasion. Here
belong, for example, Ab Umma udayy al-Bbils
157
solemn declaration to
have personally heard Muammad saying that sra 54:4749 was revealed
with reference to the Qadarites.
158
Whether the claimascribed to Ibn
155
Cf. above, p. 7778, 99, and 72.
156
Qoran et tradition, comment fut compose la vie de Mahomet, according to C.H.
Becker, Islam: [bibliographischer Jahresbericht,] p. 540sq.
157
Juynboll, Encyclopedia, pp. 160, 170, 286, 627.
158
al-Wid. Asbb al-nuzl (Cairo, 1315/1897), s.v., p. 300. Cf. also Fakhr al-Dn al-Rz in
al-Kalbthat sra 6:100 goes back to the Zindq
159
reflects nothing but the
personal opinion of this scholar or is covered by a adth, I cannot say. Such
interpretations are on the same level as the numerous prophecies about the
future of the Muslim empire, which adth literature puts in the mouth of
Muammad.
160
The legends concerning the occasion of revelation are customarily intro-
duced with the formula nazalat al-ya f, the verse was revealed about.
When al-Suy believes that the narrated event was placed in merely an
approximate relation to the text of the Koran,
161
his opinion is untenable, as
it not only contradicts the wording of the phrase but would also be difficult
to reconcile with its very sense, given that this type of literature generally
does not make concessions or limitations, instead presenting even the most
audacious lies as simple and plain facts. However, we can also identify al-
Suys basic inclination, which apparently amounts to putting into a more
favourable light the responsibility of the earliest authorities, usually that of
the Companions, for what he considers to be inaccurate interpretations.
It is extremely rare that authorities present statements about the occa- [ii/162]
sion of a revelation as being nothing but their own subjective views, or limit
their comments in any way. Al-Suy points out such an instance when the
familiar al-Zubayr (Ibnal-Awwm) says withreference tosra 4:68, By God,
I reckon (asibu) that the verse was revealed only on this occasion.
162
Ibn
Abbs on sra 2:278, says: The news reached usbut God knows best
that this verse was revealed with reference to Ban Amr b. Umayr.
163
It
seems tobe equally rare that authorities confirmtheir truthfulness. The only
reference that I recall concerns the preposterous tradition that relates sra
54:4749 to the sect of the Qadarites, where each of the nine links of the
isnd substantiate their evidence with a solemn ashhadu bi-Allhi, I swear
by God.
164
his commentary, s.v. The old Persiannon-Shitecommentary of the Koranoriginating
probably fromthe end of the third century ahof Cambridge University Libraryformerly
ownedby Thomas vanErpe(nius)considers 54:53 to be fromthe Qadarites, cf. E.G. Browne,
Description of an old Persian commentary of the urn, p. 504. Juynboll, Encyclopedia, s.v.
159
Loc. cit., p. 165.
160
Goldziher, Muslimstudies, vol. 2, pp. 121125.
161
al-Suy, al-Itqn, p. 70 toward the end.
162
al-Itqn, p. 70; al-Wid, Asbb al-nuzl, s.v., p. 122; al-Suy, Lubb al-nuql f asbb
al-nuzl, in the margin of Tafsr al-Jallayn (Cairo, 1301), vol. 1, p. 3; al-Bukhr, musqt, 7;
al-Tirmidh in the tafsr, p. 65, l 2.
163
al-Wid, Asbb al-nuzl, p. 65, l 2.
164
Cf. above, p. 344 n. 158.
Whenexplanations of Koranic verses are put inthe mouthof Muammad
they deserve no more credibility than all of his apparent sayings in adth.
Some Muslimauthorities think that, apart fromthe last revealed verse (sra
2:276), he continually interpreted all of the Koran. According to others, how-
ever, this activity applies only to a certain number of verses.
165
The rather
long list that al-Suy put together in al-Itqn (pp. 918954) does not con-
tain a single remark that goes beyond the common ground of the Commen-
tators. How little comprehensive the list really is can be seen from the fact
that, of the four interpretations attributed to Muammad by al-Wid in
relation to sras 2:275, 5:71, 54:47sqq., and 73, not a single one is listed in
al-Suys al-Itqn.
The Creators of the Exegesis: Ibn Abbs and His Pupils [ii/163]
As a result of the close relationbetweenthe exegetic adth, the biographical
adth and canonical adthas well as the congruence of the content
of tradition in the widest sensethe interpreters of the Koran are, to a
large extent, identical with the leading personalities of the other two types
of adth. Consequently, Muammads earliest Companions here recede
farther off from the members of the younger generation, such as isha,
Abd Allh b. UMAR [IBN AL-KHAB], Ab Hurayra, Anas b. Mlik, and
Ibn Masd.
166
No one is mentioned more often than Abd Allh IBNABBS,
however.
A direct cousin of Muammad, he was thirteen or fifteen years old at
the time of the prophets death, or, according to others, ten years old.
167
He
never played a role in politics. The governorship of Bara, to which he was
appointedin39/659 under Al, he seems to have receivedonly onaccount of
his family relation to the ruling caliph.
168
However, he left this post one year
later, retiring to al-if inthe ijz, either intacit agreement withMuwiya,
the governor of Syria, whowas reaching for the caliphate, or toascertainhow
165
166
Ibid., al-Suy, al-Itqn, p. 908, says that ten Companions in particular made them-
selves a name in exegesis, and in this class reckon the four first caliphs, the editors of the
pre-Uthmnic recension of the Koran, Ibn Masd, Ubayy b. Kab, Ab Ms, and Zayd b.
Thbit, further Abd Allh b. al-Zubayr, and Ibn Abbs. This is a rather unfortunate selec-
tion as isha, Ab Hurayra, Ibn Umar, and Anas appear infinitely more often as sources of
exegetical opinions than the eight persons reckoned by al-Suy at the first place.
167
al-Masd, Les prairies d or, vol. 5, p. 232; Ibn Qutayba, p. 32; al-Nawaw; Ibn al-Athr,
Usd al-ghba; Ibn ajar al-Asqaln; al-Dhahab, Tadhkirat al-uff.
168
al-abar, Annales, vol. 1, p. 3449.
he might intime ingratiate himself withthe rising dynasty. There he livedfor
another twenty or thirty years, enjoying the large income he received from
the Umayyads in exchange for his betrayal of the family of the Prophet. He
devoted himself entirely to scholarship, the exegesis of the Koran and the
related historical, antique and philological studies, becoming a celebrity in
his field.
169
Despite his prominent social position, he did not disdain teach-
ing. It is said that he lectured every day on a different subject, sometimes
canonical law,
170
sometimes exegesis or Muammads campaigns or poetry
or the battle days of the pagan Arabs. Finally, mention must be made of
genealogy and pre-Islamic Arabia, which, strangely enough, he intermixed
with Koranic and biblico-apocryphal elements, tracing them back to the
time of the patriarchs.
171
Everywhere, however, his main field is always given
as the exegesis of the Koran and the related adth. Muammad himself
is said to have asked Allh to teach Ibn Abbs tawl. No wonder that he
excelled in this field and was honoured with the by-name interpreter of the
Koran.
172
Nevertheless, whenever biographers take a close look at the type
and dimension of his exegetic activity, they display no little contradiction.
It is claimed sometimes that he interpreted the second sra verse by verse
and sometimes that he did this with the entire Koran.
173
Others even have
him count the verses and letters of the Holy Book.
174
He acquired his enor-
mous proficiency in traditions by persistent interviews with the old Com-
panions who, no matter how reserved and peevish they might have been
otherwise, they readily supplied information to the cousin of the Prophet.
He received other material from Jews who frequented his home.
175
Among
his authorities we also encounter the learned Yemenite rabbi Kab [Ibn al-
Abr],
176
although we do not known whether the two men met in person.
Some writers ascribe no less than 1,660 adths to Ibn Abbs, of which
169
The particulars are based on Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod
und Bestattung Muammads [last illness and death], pp. 119124; Ibn al-Athr, Usd al-ghba;
Ibn ajar al-Asqaln; and al-Nawaw.
170
Particularly mentioned among them were law of inheritance, permissible and prohib-
ited matters as well as ceremonies of the pilgrimage.
171
As far as I canmake out, nothing onthis canbe found inthe biographies. Still, the fact is
firmly established by traditions on the legendary history of ancient Arabia in al-abar, vol. 1,
and the book of idolatry of Ibn al-Kalb in Yqt, the isnds of which generally are headed by
the name of Ibn Abbs. Cf. also J. Wellhausen, Reste arabischen Heidentums, 2nd ed., p. 15.
172
Ibn Sad (al-abaqt), loc. cit.; al-Suy, al-Itqn, p. 709sq.
173
al-Dhahab, Tadhkirat al-uff, vol. 1, p. 80, 11; Ibn Sad (al-abaqt, vol. 5): Biographien
der Nachfolger in Medina, p. 343, 22; al-Nawaw, p. 541.
174
al-Suy, al-Itqn, pp. 157 and 164.
175
al-abar, Annales, vol. 1, pp. 62sq., and 424; al-Wid, Asbb al-nuzl, p. 141 l 10.
176
al-Nawaw, p. 523 l 7.
ninety-five are transmittedfrombothal-Bukhr andMuslim(Ibnal-ajjj),
apart from one hundred and twenty and forty-nine, respectively. Some one
hundred exegetic adths are said to come from Ibn Abbs.
177
Ibn Abbs aided his memory with notes on leaves that eventually [ii/165]
amounted to an entire camels load.
178
Yet he left nothing actually written.
The results of his collecting and research are rather to be found in the books
of his pupils and their successors. Nevertheless, if we were to attempt to
reconstruct his opinion fromthe remarks of those writers who refer to him,
this attempt wouldbe condemnedtofailure, as these persons contradict one
another in nearly every verse. In order to explain this conspicuous fact, sev-
eral avenues are open to us. One might think that the teachers opinions are
either erroneously or purposely misrepresented by the pupils, or that Ibn
Abbs himself permanently changed his opinion. But one of these views is
as unlikely as the other, since nowhere are there any clues to be found for
such a senseless and arbitrary action. Thus, there remains no other answer
but to consider this reference to Ibn Abbs authority a fiction. This would
correspond to a common contemporary literary habit that exegeteseven
if not pupils of that masterbe it because of modesty or in order to honour
the oldest and most respected head of the profession, forgo their own claim
to authorship and attribute to him what they themselves devised.
There is yet another consideration that necessitates such a decisive cor- [ii/166]
rective of the Muslim tradition on Ibn Abbs. It is totally unthinkable that
he really mastered all the fields of knowledge ascribed to himcanon law,
pre-Islamic history and antiquity, philology, and poetryand cultivated
research and teaching. This would not only be beyond the ability of a sin-
gle person but would also apply to branches of knowledge that were either
not yet established or still at their inception. We must thus conclude that
for partisan motives the achievements of several younger scholars are, in
many cases, shifted back to the past and attributed to a single person. Nat-
urally, tradition could not have followed this course if Ibn Abbs had not
made himself a name in one or several of the respected fields, although on
account of his high social standing, his reputation exceeded the bounds of
his true merit. That his main field was exegesis seems to be proved by the
unanimous agreement of biographies, commentaries, and literary as well as
historical works; but the extent of this activity, on which there are variant
reports, remains to be seen.
177
al-Nawaw, p. 353.
178
Ibn Sad, (al-abaqt, vol. 8): Biographien der Nachfolger in Medina, p. 216, 16.
Whereas Ibn Abbs authority within Islam remains untouched and un- [ii/167]
shakeable tothis very day, ever since Aloys Sprenger
179
it has become custom-
ary in the West to consider hima liar. This judgement could not be accepted
even if all the lies and nonsense attributed to him were really true, since in
most cases it would be just as difficult to identify the motive of the falsifi-
cation as it would be to contest the credulity of their authors. On the other
hand, the justification of an opinion with a fictitious saying of the Prophet
or a Companion was at that time a legitimate form of voicing subjective
truths.
180
Of the immediate pupils of Ibn Abbs, the most frequently mentioned
are:
181
Sad b. Jubayr (died 95/713),
182
Mujhid b. Jabr (d. 103/721),
183
Ikrima
[al-Barbar] (d. 106/724),
184
A b. Ab Rab (d. 114/732),
185
and Ab li
Bdhm.
186
Apart from Sad b. Jubayr, they were all of slave origin. Only
Ikrima and Sad seem to have left independent works,
187
whereas the lec-
tures of the others were published as books by later editors only. There
are, for example, different commentaries that are traced back to ak
b. Muim(d. 105/723),
188
a pupil of Sad,
189
or to Ibn Jurayj (d. 150/767),
190
a
pupil of A.
191
Mujhid b. Jabrs interpretation must have been highly
179
Notes on Alfred von Kremers edition of Wakidys Campaigns, p. 72.
180
Cf. above, pp. 317 and 333.
181
A more comprehensive list can be found in Ibn al-Athr, Usd al-ghba f marifat
al-aba, vol. 3, p. 194; cf. also al-Suy, al-Itqn, p. 909sqq.
182
Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, pp. 178187; al-Dhahab, Tadhkirat
al-uff, vol. 1, pp. 6567 EI
2
; EQ; Juynboll, Encyclopedia, pp. 430431.
183
Ibn Sad (al-abaqt, vol. 5): Biographien der Nachfolger in Medina [biographies of
the Followers at Medina], p. 343sq.; al-Dhahab, Tadhkira, vol. 1, p. 80sq.; EI
2
; Juynboll,
Encyclopedia, pp. 430431; Sezgin, GAS, vol. 8, p. 22.
184
Ibn Sad (al-abaqt, vol. 5): Biographien der Nachfolger in Medina, pp. 212216, al-
abaqt, vol. 2, part 29, Letzte Krankheit, Tod und Bestattung, p. 133; al-Dhahab, Tadhkirat
al-uff, vol. 1, p. 83sq.; al-Nawaw; Juynboll, Encyclopedia, p. 241.
185
Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung, p. 133sq.;
(al-abaqt, vol. 5): Biographien der Nachfolger in Medina, pp. 344346 al-Dhahab, Tadhki-
rat al-uff, vol. 1, pp. 8587; al-Nawaw; Juynboll, Encyclopedia, pp. 139140; Sezgin, GAS,
p. 31.
186
Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, p. 207, but without date of death.
Cf. also EQ; Juynboll, Encyclopedia, p. 516, n. 4.
187
al-Fihrist, ed. Flgel, p. 34 ll 1 and 7. Regarding later recensions of Ikrima compare
A. Sprenger, Mohammad, vol. 3, p. cxiii, no. 1.
188
Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, p. 210sq.
189
al-Fihrist, p. 33, and al-Itqn, p. 914; A. Sprenger, loc. cit., p. cxvi, no. 2.
190
Ibn Sad (al-abaqt, vol. 5): Biographien der Nachfolger in Medina, p. 361; al-Dhahab,
Tadhkirat al-uff, vol. 1, p. 152; Juynboll, Encyclopedia, pp. 212225; Sezgin, GAS, vol. 1,
p. 91.
191
Sprenger, loc. cit., p. xciv. Numerous isnds in al-Wid, Asbb al-nuzl, attest to this.
appreciated, as al-Fihrist knows of three different recensions of this work
alone.
192
Let us mentiononly a fewof the Commentators of the first century whose [ii/168]
scholastic background is somewhat uncertain. asan al-Bar (d. 110/728)
193
has no documented relation to the school of Ibn Abbs, but his commen-
tary is frequently cited; in the recension of Amr b. Ubayd [d. 143/760
194
] it
is heavily used by Muammad b. Ibrhm AL-THALAB (d. 427/1035).
195
As
for Qatda b. Dima (60/679118/736),
196
born blind but endowed with a
splendid memory, the biographers are at variance as to whether or not he
heard lectures fromIbn Abbs pupils, such as Ikrima,
197
Sad b. Jubayr and
Mujhid b. Jabr.
198
His commentary circulated in several recensions.
199
Con-
versely, a man of Jewish descent, Muammad b. Kab AL-QURA (d. 118/
736),
200
whose commentary was frequently consulted by Ab Mashar [Naj
b. Abd al-Ramn al-Sind] (d. 170/786)
201
as well as by other biographers of
the Prophet,
202
is alleged to have studied under Ibn Abbs.
203
Of the representatives of the younger generation deserving of particu-
lar mention are Muammad b. Sib AL-KALB
204
(d. 146/763), Sufyn b.
Uyayna
205
(d. 198/814), Wak b. al-Jarr
206
(d. 197/812), Shuba b. al-ajjj
207
(d. 160/776), Yazd b. Hrn al-Wsi
208
(118/736206/821), Abd al-Razzq b.
192
al-Fihrist, ed. Flgel, p. 33, ll 2123; Sprenger, loc. cit., p. cxvsq.
193
Ibn Sad (al-abaqt, vol. 7, part 2): Biographien der Basrier, vol. 7, part 1, pp. 116129;
194
Sezgin, GAS, vol. 1, pp. 3132; Juynboll, Encyclopedia, p. 568, col. 2.
195
196
Juynboll, loc. cit., pp. 438449; Sezgin, GAS, vol. 1, pp. 3132.
197
al-Nawaw, p. 509, l 15.
198
al-Dhahab, Tadhkirat al-uff, vol. 1, p. 110, l 6.
199
al-Fihrist, Flgels edition, p. 34, ll 3 and 4; Sprenger, Mohammad, vol. 3, p. cxvi, no. 7.
200
Sezgin, GAS, vol. 1, p. 32; Juynboll, Encyclopedia, p. 61, col. 2, p. 505, col. 1.
201
EI
2
202
al-abar, Annales, vol. 1, p. 575, l 4, p. 1195, l 2, p. 1721, l 14; Sezgin, GAS, vol. 1, pp. 291292.
203
al-Nawaw, p. 116, l 12.
204
al-Fihrist, p. 95; Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, p. 249sq. In the
selection of the following name I am following al-Suy, al-Itqn, p. 916, centre.
205
Tadhkirat al-uff, vol. 1, p. 238sqq.; al-Nawaw, p. 289; Juynboll, Encyclopedia, 569622.
206
Ibn Sad (al-abaqt) (vol. 6): Biographien der Kufier, p. 275; al-Dhahab, Tadhkirat
al-uff, vol. 1, p. 280sqq.; al-Nawaw, p. 614sqq.; Juynboll, Encyclopedia, pp. 646649.
207
al-Dhahab, Tadhkirat al-uff, vol. 1, p. 174sqq.; al-Nawaw, p. 315sq.; EI
2
; EQ; Juynboll,
208
al-Dhahab, Tadhkirat al-uff, vol 1, p. 290sqq.; al-Nawaw, p. 636sq.; EI
2
; Juynboll,
Hammm
209
(126/749211/827), dam b. Ab Iys
210
(132/749220/835), etc.
With the exception of Sufyn b. Uyayna and dam b. Ab Iys, all these
scholars were slaves or freedmen. That they left independent works is par-
tially attested elsewhere, although none of them has come down to us.
211
Since the first somewhat reliable traces of grammatical studies among [ii/169]
the Arabs point to the middle of the second century,
212
even the latest of the
aforementioned commentaries cannot have contained much of this branch
of science. Etymology seems to have been older than actual grammar. It
is thus quite conceivable that the lexical remarks in al-Bukhrs tafsr are
correctly attributed to Mujhid (Ibn Jabr). Conversely, it is by no means
certain that Ibn Abbs was a scholar in this field.
In contrast to Ibn Abbs, whose reputation was always above suspicion,
his pupils and their successors had to suffer adverse judgement. This, how-
ever, contains little importance for us, since, as we pointed out above when
discussing adths,
213
Muslim criticism is partly personal prejudice, partly
adherence to different political or religious parties, and partly formalities of
the construction of the isnd. Thus, while Ikrima is recognized as an expert
in the Book of God,
214
his reliability is neverthless questioned,
215
probably
because his adthdoes not serve as evidence incanonlaw, since he was con-
sidered a Khrijite.
216
Ab li Bdhm was held in low esteem because he
was no reciter of the Koran.
217
Even al-Kalb, the great expert in genealogy
and pre-Islamic history, is considered weak in transmission (riwya),
218
or
even an outright liar;
219
the isnd of Muammad b. Marwn AL-SUDD
Muammad AL-KALBAb li Bdhm is called the chain of lies.
220
By
209
Tadhkirat al-uff, vol. 1, p. 333; Juynboll, Encyclopedia, pp, xxxii, 2438.
210
al-Dhahab, Tadhkirat al-uff, vol. 1, p. 375; Sezgin, GAS vol. 1, p. 102, no. 32.
211
Whether there is really a Persiantranslationof Yazdb. Hrns commentary inanIstan-
bul libraryNuruosmaniye, no. 474still needs to be investigated. al-Fihrist lists among the
aforementioned commentaries only those of al-Kalb and Wak b. al-Jarr.
212
213
Cf. above, pp. 317sq. and 333.
214
al-Dhahab, Tadhkirat al-uff, vol. 1, p. 84 l 7; Ibn Sad (al-abaqt, vol. 5): Biographien
der Nachfolger in Medina [biographies of the Followers at Medina], p. 212, l 19sqq.
215
Ibn Qutayba, p. 224 l 5, and p. 231 sq.; Ibn Sad (al-abaqt, vol. 5): Biographien der
Nachfolger in Medina, p. 213 l 9 and 12.
216
Ibn Sad (al-abaqt, vol. 5): Biographien der Nachfolger in Medina [biographies of the
Followers at Medina], p. 216, ll 6, 10, and 12.
217
O. Loth, Tabaris Korankommentar, p. 598.
218
Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, p. 250, l 9.
219
al-Qurub.
220
Silsilat al-kadhib, al-Suy, al-Itqn, p 914, beginning. Al-Suy uses equally strong
the same token and without further substantiation, the reproach of false-
hoodis brought against Muqtil b. Sulaymn
221
(d. 150/767), whotransmitted
from ak b. Muzim and, according to others, from Ab l-ajjj IBN
MUJHID as well,
222
and whose commentary is given as an independent
work.
223
The Extant Commentaries
The oldest remains of the exegetic literature can be found in the historical [ii/170]
works of IbnIsq
224
(d. 151/767) andal-Wqid (d. 207/822), who, whenrelat-
ing events, not only refer to pertinent passages of the Koranandoccasions of
revelation but also supply detailed explanations of entire sras.
225
As a rule,
these explanations consist of brief paraphrases and narratives of occasion.
Only now and thenin al-Wqid more frequently than in Ibn Isqare
rare words glossed with better known synonyms. Lexical notes of philologi-
cal relevance first appear in Ibn Hishms (d. 213/828) usually very extensive
additions and are accompanied there by examples from poetry.
The fragment of a commentary on the Koran ascribed to al-Kalb
226
offers [ii/171]
a paraphrase of the text but pays little attention to historical aspects, ignor-
ing altogether variant readings and grammar. The isnds point to an origin
in the third century ah. It still awaits more precise investigation to deter-
mine whether the work really goes back to this famous scholar. In any case,
al-Kalbs transmissions included in other exegetic works are usually longer
than this one.
language in his Lubb al-nuql f asbb al-nuzl on sra 2:13: This isnd is disabled (whin)
because al-Sudd, the Younger, is a liar, and al-Kalb likewise, and Ab li(Bdhm) is weak
(af).
221
Sezgin, GAS, vol. 1, pp. 3637; Juynboll, Encyclopedia, pp. 431432.
222
al-Nawaw, p. 574sq.; al-Khazraj, Khulat tadhhb, p. 386; EI
2
; EQ, Goldziher, Schools of
Koranic commentators, pp. 38, 39, etc.; Juynboll, Encyclopedia, pp. 431432; Sezgin, GAS, vol. 1,
pp. 3637.
223
al-Fihrist, p. 34, l 5.
224
225
For example, sra 2 (Ibn Hishm, Sra, pp. 363380); sra 3 (Ibn Hishm, pp. 403411,
592607, al-Wqid, al-Maghz, edited by A. von Kremer, pp. 310317 = Wellhausen, Muham-
med, p. 145); sra 8 (IbnHishm, pp. 476485, al-Wqid, editedby A. vonKremer, pp. 126132,
Wellhausen, p. 77sq.); sra 9 (Ibn Hishm, pp. 919929, Wqid Wellhausen, p. 415sq.); sra
18 (Ibn Hishm, pp. 193202); sra 48 (Ibn Hishm, pp. 749751, Wqid Wellhausen, p. 260).
226
Ahlwardt, Verzeichnis der arabischen Handschriften, no. 732 (= Sprenger, 404); Tafsr
al-Qurn: This riwya goes back to Hishm b. Muammad al-Kalb, d. 204/819; other Mss,
Ayasofya, 113118, and Nuruosmaniye, 167183; C. Brockelmann, Geschichte der arabischen
Literatur, vol. 1, p. 190, lists a Bombay printed edition of 1302/1884, which I have not been
able to locate; Sezgin, GAS, vol. 1, pp. 3435.
Important Writings Containing
Commentaries on History and Exegesis
The two oldest systematic collections of adth that we have, those of al-
Bukhr (d. 256/869),
227
and al-Tirmidh (d. 279/892),
228
both contain a rela-
tively extensive commentary on the Koran. They are in the same vein as the
exegeses of Ibn Isq and al-Wqid.
Al-abars (d. 310/922) great work
229
is the turning point in the history
of interpretation. Unlike its two precursors, this work does not limit itself
to advancing the comprehension of the text through convenient paraphras-
ing, lexical explanationsthe first work to do so since Ibn Hishm, and also
by means of poetic examplesand narratives of origin, but also deals with
grammatical questions and the connections betweendogma and canonlaw.
On the other hand, the book endeavours to be a synopsis of the work of ear-
lier generations and thus cites for each revelation every other interpretation
available, even listing insignificant, divergent traditions. At the same time,
when reproducing the authenticated strands of authorities he observes the
same, painful exactitude familiar to us fromhis annals. The isnds are gen-
erally the same as in his annals. In general, the traditions can be identified
with the schools of Ibn Abbswithin which Mujhid b. Jabr occupies an
independent positionof Qatda (Ibn Dima) (Muammad b. Marwn)
AL-SUDD, and Ibn Abbs (on legends). At the end of an entry al-abar
finally gives his ownopiniononthe proper or most likely interpretation.
230
In
the introduction to the commentary (vol. 1, p. 1 sqq.), he lists the vague out-
line of his plan: the language of the Koran, the seven readings, the collec-
tion, sources, and history of interpretation, the names of the Koran and its
individual parts, such as sras, verses and the logograms. He disregarded the
227
Cairo edition, 1309/1891, vol. 3, pp. 63144.
228
Delhi folio edition of 1315/1897, vol. 2, pp. 119172. Whereas al-Bukhr deals with all the
sras, al-Tirmidh lacks altogether no less than twenty-one, viz. sras 45, 65, 67, 71, 73, 76, 77,
78, 79, 82, 86, 87, 90, 100, 101, 103, 104, 105, 106, 107, and 109.
229
Commonly entitled Tafsr. In the authors world-history, vol. 1, p. 87, l 2, the title reads
Jmi al-bayn f tawl al-Qurn; cf. M.J.de Goejes edition, introduction, p. xii. This is one
of the oldest examples of a bombastic title of a book in rhymed prose I know of. The
fashionstarting with the fifth century ahwhich became firmly established in Arabic
literature, is likely to originate from Persia. The work was first published in thirty volumes
at Maymniyya Press, Cairo, 1321/1903. An improved edition appeared a fewyears later at the
bookseller Umar usayn Khashshb.
230
We have an excellent appreciation of the work from Ibn Askir by M.J. de Goeje in his
introduction to al-abars Annales, p. lxxix. Otto Loth was the first European to present an
exposition of the commentary, based on manuscript studies at the Viceroyal Library, Cairo;
cf. his Tabaris Korancommentar, p. 588sqq.
variant readings, probably because he devoted a monograph to this subject,
which unfortunately seems to have been lost. Muslims consider al-abars
commentary an unequalled achievement.
231
Given its scope, versatility, and
the reliability of its content, it is indeed the most instructive work of its
type that the Muslim world produced. Yet we must immediately voice our
reservationthat, for us, it is only useful as a collectionof data, since it is com-
pletely under the spell of dogmatic prejudice andtherefore cannot reachthe
heights of an objective historical approach. Historical criticismwas also for-
eign to Muslims in subsequent timeseven down to the present day.
The way al-abar summarized the achievements of earlier generations [ii/173]
of exegetes made his thesaurus the eternal source from which later schol-
ars derived their wisdom. Such a mammoth work, which can have circu-
lated in only a few complete copies at a time, frequently occasioned the
preparation of excerpts. One of the best known is the Tafsr (al-Qurn)
of AB LAYTH Nar b. Muammad AL-SAMARQAND, who died 373/983,
375/985, or 393/1002.
232
The work has survived in manuscripts, mostly frag-
mentary, and has not yet found an editor. Ab Isq Amad b. Muam-
mad AL-THALABs (d. 427/1035)
233
al-Kashf wa-l-bayn an tafsr al-Qurn
is based not only on al-abar but also on some one hundred other works.
Al-Thalab maintains in his preface
234
that his predecessors either followed
newer or false directions, were uncritical and restricted themselves to the
riwya, or omitted the isnd, thus raising doubts as to the reliability of their
facts. Others, like al-abar and Ab Muammad ABD ALLH IBN MID
AL-IBAHN, would present all sorts of superfluous information, intimi-
dating potential critics by sheer overload.
235
Still others offer nothing but
explanations without descending to the overwhelming factual difficulties.
Because of the lucidity of these reasons, the commentarywhich does not
spare examples frompoetry, and is of moderate volume, twice the size of al-
Baywis likely to be one of the most useful, which makes it surprising
that it has not yet been published in the Orient.
The Malimal-tanzl of usaynb. Masd al-Farr AL-BAGHAW
236
(died [ii/174]
510/1117 or 516/1122), whom we already met as the author of a clear com-
231
al-Dhahab, Tadhkirat al-uff, vol. 2, p. 278, l 10sq., alladh lam yuannaf mithluh;
al-Suy, al-Itqn, p. 916, centre, ajall al-tafsr wa-aamuh.
232
Sezgin, GAS, vol. 1, p. 446, lists a Turkish translation by Ibn Arabshh.
233
Brockelmann, GAL, vol. 1, p. 350; EQ; EI
2
.
234
Cf. Ahlwardt in his Verzeichnis der arabischen Handschriften, vol. 1, no. 739 (= cod.
Sprenger, no. 409).
235
I know no more of this author.
236
Sezgin, GAS, vol. 1, p. 155, Kommentare (2). There is a Bombay lithograph of 1269/1852.
pendium of adth,
237
is said to be an excerpt from al-Thalab. Al al-Dn
Al b. Muammad [AL-KHZIN] al-Baghdd (d. 727/1326), who in his com-
mentary made much use of the Malim al-tanzl, is full of praise for this
work.
238
It comes as no surprise that Mamd b. Umar AL-ZAMAKHSHAR (died
538/1143), the author of excellent grammatical, lexical, and stylistic works,
239
also devotes much space to these subjects in his commentary on the Koran,
al-Kashshf an aqiq al-tanzl,
240
which pays attention to the variant read-
ings. When it comes to explanations of individual words, he, like al-abar,
refers not infrequently to evidence (shawhid)
241
from poetry. Nevertheless,
he displays little interest inthe originof the legends. If he does refer to them,
it is extremely briefly, omitting the isnd by using, typically, the empty for-
mula it is transmitted (ruwiya). In rare instances, however, particularly
when he contrasts different divergent traditions, he likes to identify the
individual tradition by the name of a prominent link in the chain of author-
ities.
242
The reader looks in vain for a section that, like al-abar, discusses
introductory questions; the author does not even hint at the relation to
his predecessors. He focuses his attention on theological and philosophi-
cal matters, which he handles competently, wisely, and with spirit. These
deliberations, which take up most of the space, are unfortunately with-
out value for the exegesis, as they impose ideas from a later period onto
the Koran. It is precisely for this reason, accompanied by splendid dialec-
tics,
243
that the Kashshf attainedgreat eminence anddisplacedthe previous
commentaries, despite the fact that its author was not orthodox, being a
rather outspokenMutazilite. Consequently, the work sawfrequent editions,
excerpts, andglosses.
244
The orthodox hadno scruples whatever inmeddling
237
238
Cairo, al-Khayriyya, 1309/1891, in 4
v
. Vol. 1, p. 3, ll 912:

.
239
Sezgin, GAS, vol. 2, p. 135, etc.
240
There are a Calcutta edition of 1856, and many Egyptian editions.
241
It is for this reason that frequently books explaining these verses and supplying refer-
ences to the relevant diwans are added to the Kashshf.
242
Cf., for example, on sra 17:1.
243
Amuch liked wordingalso when discussing other kinds of questionsis the formula
fa-in qulta qultu.
244
Brockelmann, GAL, vol. 1, p. 289sqq. lists sixteen glosses alone and five excerpts. Be-
cause of the not insignificant difficulties frequently posed by the authors sophistic line of
argument, the glosses are preferably put inthe margin. For example, the editionwhichI used,
Cairo, 1308/1890, contains the glosses of Al b. Muammad AL-JURJN (d. 816/1413).
with the text in order to hide the sectarian position of the celebrated master
and to deprive the literary polemics of arguments. For example, the begin-
ning of the commentary, whichal-Zamakhshar, being a true Mutazilite had
openedwiththe words Glory toGodWhocreatedthe Koran, the wordcre-
ated was later changedtosent down.
245
As far as IbnKhaldnis concerned,
al-Zamakhshars work is a towering model far beyond the so-called tafsr
naql, the exegesis laden with traditional material.
246
Abd Allh b. Umar AL-BAYWs (died 685/1286, 692/1292, or 710/ [ii/176]
1311)
247
commentary, though largely dependent on al-Zamakhshar, evi-
dently uses so many other sources that jj Khalfa,
248
with some exagger-
ation, called him an epitomist. Unfortunately, he does not identify these
sources, neither in the brief and general preface nor anywhere else in the
book. The problem can thus be solved only by a detailed literary-historical
analysis. The bulk of the subject matter is overwhelming and covers every-
thing that needs to be considered in a commentary of the Koran. However,
both accuracy and comprehensiveness leave something to be desired. Al-
Bayw pays more attention to readings and grammar than al-Zamakh-
shar; philology is probably best represented. When it comes to informa-
tion concerning traditions, he is shorter and more casual, citing far less
frequently the authenticated strand of transmitters. The Muslim view that
considers this commentary to be the best
249
andmost sacredis unwarranted.
The merit of the first printed edition of the work is due to a Christian, [ii/177]
the great Leipzig Arabist, Heinrich L. Fleischer,
250
whereas the editions of
Blq (1282/1865) and Constantinople (1296/1878) seem to be reprints. Still,
it is unfortunate that he wasted his splendid linguistic competence on a task
that any Egyptian or Indian scholar of average intelligence could also have
achieved. As in the field of adth, Islamic science is best qualified to deal
with the traditional literature of tafsr, whereas when it comes to the edition
of secular works in history, geography, and poetry it fails completely.
Although the later commentaries are instructive for the history of exege-
sis or for theology in general, we can hardly expect them to contain new or
245
Cf. al-Kashshf, Cairo ed., 1308, vol. 1, p. 2, glosses of al-Jurjn.
246
al-Muqaddima, Beirut edition of 1886, p. 384sq.
247
Cf. Brockelmann, loc. cit., vol. 1, p. 416. The book is entitled Anwr al-tanzl wa-asrr
al-tawl, or short, Tafsr al-q.
248
G. Flgels edition, vol. 5, p. 192.
249
jj Khalfa, loc. cit., calls him the shining sun in the height of the day.
250
Two volumes, Leipzig, 18461848. Very valuable are the alphabetical indeces which
Winand Fell contributed in 1878.
unknown matters regarding tradition. Foremost among them
251
is prob-
ably the Jmi akm al-Qurn of Muammad b. Amad AL-QURUB
(d. 671/1272 or 668/1268).
252
However, none of the known collections of
manuscripts seems to have a complete copy.
We next come to two large works printed in the Orient. The first, enti-
tled The great tafsr (al-Tafsr al-kabr) or The keys to the secret (Maft
al-ghayb), is from the pen of the Persian FAKHR AL-DN Muammad b.
Umar AL-RZ
253
(d. 606/1209). According to al-Suy, al-Itqn, p. 917, it
is full of sayings from wise men and philosophers, jumping from one sub-
ject to another and leaving the reader at a loss, since he lacks a presentation
that conforms with the verses of the Koran. As AB AYYN Muammad
b. Ysuf (d. 654/1256 sic) remarks, Imm Rz combined in his commentary
several scatteredsubjects that are, unfortunately, unrelatedto exegesis. Oth-
ers even claim that this book contains everything except tafsr.
Amid such speculative excesses, the Egyptian scholar Jall al-Dn AL- [ii/178]
SUY
254
(d. 911/1505) re-established the honour of old-fashioned exegesis
based on the good old tradition. The giant work, Tarjumn al-Qurn f l-
tafsr al-musnad, seems to have been lost, although its excerpt, entitled Durr
al-manthr f l-tafsr al-mathr, has only the titles of the literary sources
instead of the isnds. The work still consists of six volumes in the only
Cairene edition (1314/1896) I am aware of.
Most widespread in the contemporary Islamic Orientparticularly
among educated laymenis a compendium entitled Tafsr of the two
Jalls.
255
In this work the exegesis of sras 18 to 114 and the Ftia is from
251
The greatest commentary of all times was probably the Tafsr of Ab Ysuf Abd al-
Ramn b. Muammad AL-QAZWN (d. 488/1095 in Egypt [EQ]) about whom nothing else
is known. The work is said to have consisted of three hundred or four hundred, according
to other sources even seven hundred volumes (mujallad) and was a waqf of the mausoleum
of Ab anfa (d. 150/767) near Cairo. Cf. Ab l-Masin IBN TAGHRBIRD, ed. by Popper,
p. 313 top. According to Ibn al-Athr, al-Kmil f l-tarkh, ed. by Tornberg, vol. 10, p. 173, the
author died in 486/1093. Typical for the way Muslims speak of the volume of a commentary
of the Koran is that Al allegedly boasted that the tafsr on the first sra alone would amount
to seventy camel loads [Goldziher, Schools of Koranic commentators, p. 143 n. 127]. Al-Suy
in al-Itqn, p. 906sq., does not even consider this an exaggeration as this sra practically pro-
vokes vast excurses.
252
253
Brockelmann, GAL, vol. 1, p. 506. The work has been repeatedly printed in the Orient,
lately at Cairo in eight volumes (13071309/18891891).
254
255
Tafsr al-Jallayn, of which there are numerous Oriental editions. The one I used was
published at Cairo in 1301/1883 in two volumes.
JALL AL-DN Muammad b. Amad AL-MAALL
256
(d. 864/1459), while
the remainder was completed by his well-known pupil, Jall al-Dn AL-
SUY. The peculiar position of the Ftia at the end of the book can
be explained by the goal of not separating the contributions of the elder
scholar. The work is far more useful thanits brevity would suggestapprox-
imately two fifths of al-Bayws commentary. Since the work supplies
not only a continuous paraphrase and grammatical, particularly syntactic,
explanations but also considers the narratives of occasion and variant read-
ings, it isespecially in comparison with the poor arrangement of the great
commentariesan excellent aid when trying to understand the Muslim
view of a particular Koranic passage.
Shite Commentaries
The earliest Shite scholar to whom a commentary of the Koran has been [ii/179]
ascribed is Muammad b. Al b. usayn b. Al b. Ab lib, called AL-
BQIR (d. 114/732, 117/735 or 118/736).
257
It is by no means certain that the
book ever existed as an independent work and not only in the recension of
his blind
258
student, Ab l-Jrd Ziyd b. al-Mundhir.
259
Somewhat younger
is Ab amza Thbit b. Dinr Ab afya,
260
who died in the reign of the
Abbsid Caliph Manr.
261
These works probably do not reflect the Shite
leanings of their authors any more than the Maghz of al-Wqid who, after
all, was also suspected of tashayyu.
262
The peculiar Shite tendency of con-
sidering the ahl al-bayt to be the only true source of all tradition and of
connecting half of the Koran with the family of Al and the creed of the sect
infiltrated into the exegesis only in later times, or at least emerged only later
in literature. For example, Al b. Ibrhm AL-QUMM,
263
the fourth-century
256
Cf. Brockelmann, loc. cit., vol. 2, p. 114; EI
2
; Sezgin, GAS, vol. 2, p. 232, no. 18.
257
al-Nawaw, p. 113; Ibn al-Nadm, Fihrist, ed. by Flgel, p. 33; Ibn Sad (abaqt, vol. 5):
Biographien der Nachfolger in Medina [biographies of the Medinan Followers], p. 235sqq.;
EI
2
258
al-s, Fihrist kutub al-Sha, p. 178.
259
Ibnal-Nadm, al-Fihrist, p. 33; al-Shahrastn, ReligionsparteienundPhilosophenschulen
[religious parties and schools of philosophy], vol. 1, p. 178; al-s, Fihrist kutub al-Sha,
no. 308; EI
2
; Sezgin, GAS, vol. 1, pp. 528, 552.
260
Ibn al-Nadm, al-Fihrist, p. 33; jj Khalfa, Lexicon, vol. 2, p. 357; al-s, Fihrist kutub
al-Sha, p. 71; Ibn Sad (al-abaqt, vol. 6): Biographien der Kufier, p. 253.
261
al-Khazraj, Khulat tadhhb al-kaml f asm al-rijl, s.v.; Tusy, no. 308.
262
Ibn al-Nadm, al-Fihrist, p. 98.
263
al-s, Fihrist kutub al-Sha, p. 209; Tafsr in Ahlwardt, Verzeichnis, no. 929 (= cod.
Sprenger, no. 406;) Sezgin, GAS, vol. 1, p. 45, no. 29.
commentator, interprets the words of sra 2:1, dhlika l-kitb, to refer to Al,
ascribes actions during the Battle of Uud to Al that in reality were per-
formed by Umar, and, in the same vein, understands the frequent Koranic
expression, munfiqn (doubters), as referring to the first caliphs, so that
Nldeke calls the book a miserable interweaving of lies and stupidity. A
Shite adth mentioned by al-Suy interprets the sun at the beginning
of sra 91 as Muammad, the moon as Al, the day as al-asan b. Al and
usayn b. Al, and, conversely, the night as the Umayyads.
264
Others apply
the kinsfolk of sras 42:22, and 59:7 to the Alid family, but the cursed
tree [zaqqm, Juynboll, Encyclopedia, p. 552] of sra 17:22 to the House of
Umayya;
265
insra 16:78, the words khayr and adl to Al, jibt and ght (sra
4:54) to Ab Bakr and Umar; and finally, the religious duties of the prayer
liturgy, alms, and pilgrimage to the accomplishments of the Imms.
266
If the benefit of Sunnite commentaries for the historical contextualiza- [ii/180]
tion of revelations turn out to be rather limited, that of the Shite counter-
parts we saw is absolutely null. In view of the eccentric allegory (tawl),
267
which totally ignores the context of the texts, one might be inclined to raise
the questionof whether audacity or stupidity predominates ineachlie. Nev-
ertheless, the extravagances of the Sha are supportedby suchsystemization
andmethodthat one is hard-pressedtochallenge their intelligence. It would
also be difficult to prove that the Shite exegetes were less honest than their
Sunnite counterparts who, when it came to fabricating traditions, were by
no means timid. If, in spite of this, the sunna does not present the facts in
quite such a distorted form, this is not due to the superior character of their
literati but rather to the sounder historical grounding of their entire school
of thought. In contrast, the starting point of the Shite interpretation was a
smack in the face to the true facts, a disadvantage which the representatives
of this sect had to compensate for with a still more fanatic presentation of
their point of view.
Muammad b. Murta AL-KSHs
268
(ca. 911/1505) book, al-f f tafsr [ii/181]
al-Qurn, deals in particular with the dogma of the ambiguity of Koranic
passages, a point on which Shites approach Sufis, whose treatment of the
264
al-Lal l-mana f l-adth al-mawa, Cairo, 1317/1899, vol. 1, p. 184, according to
I. Goldziher, Introduction to Islamic theology and law, p. 182 n. 43.
265
Goldziher, Muslimstudies, vol. 2, p. 110sq.
266
Isr. Friedlaender, Heterodoxies of the Shiites I 35 [sic]. Cf. also above, pp. 254266,
regarding Shite reproaches of the Uthmnic Koran and the apocryphal Sra of the Two
Lights.
267
al-Jurjn, Definitiones, edited by Gustav Flgel, p. 52.
268
Brockelmann, loc. cit., vol. 2, p. 200. The Berlin manuscript [Ahlwardt, Verzeichnis der
Koran al-Suy considered to be devoid of any exegetic merit.
269
The great
mystic poet, Jall al-Dn Rm (d. 672/1273), expresses this theory with the
following words:
Know that the words of the Koran are simple, yet they conceal beneath the
outward meaning an inner, secret one.
Besides this secret meaning there is yet a third one that bewilders the subtlest
intelligence.
The fourth meaning none has discerned but God, Who is beyond comparison
and is the source of all sufficiency for all.
In this way one can advance to seven meanings, one after the other.
So, my son, do not confine your view to the outward sense as the demons did
who saw only clay in Adam.
The outward sense of the Koran is like Adams body; only its exterior is visible,
but its soul is hidden.
270
One of the oldest Sufi commentaries is the aqiq al-tafsr of Ab Abd [ii/182]
al-Ramn AL-SULAM from Naysbr
271
(d. 412/1021). The earliest printed
tafsr is from Muy l-Dn IBN AL-ARAB (d. 638/1240) from Murcia.
272
In
the Middle Ages, the dogma of the ambiguity of the Scriptures was also
decisive for Christian, Biblical exegesis and dominated the field until the
Reformation.
273
It is also to be found in Jewish writings of the thirteenth
century, such as, for example, the commentary on the Pentateuch of Bay
ben Asher of Zaragoza [died 1340] and the Book Zohar.
274
It is amazing how
much spirit and sagacity mankind occasionally displayed in the name of
warding off the plain sense of religious documents.
arabischen Handschriften, no. 8607,8] (= 1 Petermann, 653) is incomplete and contains only
the interpretation of the sras 1 to 17.
269
al-Suy, al-Itqn, p. 901: amm kalmal-fiyya f l-Qurn fa-laysa bi-tafsr.
270
Mas nav (Whinfield), p. 169, after Goldziher, Introduction to Islamic theology and law,
p. 223.
271
Goldziher, Introduction to Islamic theology and law, p. 139 n. 74; Sezgin, GAS, vol. 1,
pp. 671672.
272
Goldziher, loc. cit., p. 139 n. 74.
273
The hermeneutic handbooks commonly condense this wisdom in the following verse:
littera esta docet,
quid credas, allegoria,
moralis, quid agas,
quid speres, anagogia.
Thus, for example, Jerusalem actually means the city, allegorically, the Church, morally, an
orderly community, and analogically, eternal life.
274
J. Frederic McCurdy, Mystic exegesis, in Jewish encyclopedia, vol. 3, p. 171.
Writings on the Occasion of Revelation
The writings entitled asbb al-nuzl differ from the commentaries in so far
as they contain only material relevant to the occasion of revelations. How-
ever, since this constitutes the most important religious and literary parts of
the commentaries, andis liberatedhere fromall the annoying additions that
often stand in the way of quick survey, the merit of these books for research
purposes is obvious. Muslims seem to have had less appreciation of this or
the number of known relevant works would not be so small. Ibn al-Nadms
Fihrist
275
knows of only two such works. The author of the first one, usayn
b. Ab usayn, is not otherwise known, although in the case of such names
not much reliance should be attached to a hand-written transmission. The
second work is said to have been composed by Ikrima from a sermon by
Ibn Abbs. However, given the dubious role played by the Prophets cousin
in the history of Arabic literature,
276
this statement is to be approached with
utmost caution. Al-Suy apparently does not knowof an earlier book than
that of Al b. al-Madn
277
(died 234/848), one of al-Bukhrs
278
authorities.
According to him, the most famous of its kind originates fromAl b. Amad
AL-WID of Naysbr
279
(d. 468/1075); it is also the oldest of which we
have a printed edition.
280
As the author explains in the preface (p. 3sq.), he
considers familiarity with the occasion of revelations to be the basis of exe-
gesis and, for this purpose, opposes the excessive prevarication of his time,
instead endeavouring to re-establish expertise based on the study of tradi-
tion. The sources to which he usually refers are the works of biography, exe-
gesis, and adth. Whether he concurrently resorted to actual books of the
asbb type can be brought to light only after thorough research. Wherever
he relies on literary sources, it is rarely done without supplying in each and
every case the complete strand of authorities. Al-Suy (d. 911/1505) follows
in his footsteps with his Lubb al-nuql f asbb al-nuzl.
281
As the intro-
duction boasts, the work is distinguished by excellent points. It omits all
of al-Wids material that is not strictly relevant. In exchange, it incorpo-
rates new material from other sources, adth as well as the commentaries,
275
Flgels edition, p. 38.
276
277
al-Nawaw, p. 443sq.; EI
2
; Juynboll, Encyclopedia, p. 423, col. 1, l 3.
278
al-Qasalln on al-Bukhr, vol. 1, p. 33, l 6.
279
Brockelmann, GAL, vol. 1, p. 411 sq.; EI
2
; EQ.
280
Cairo, al-Hindiyya, 1315/1897, 334 p.
281
Printed in the margin of the commentary on the Koran, Jallayn, Cairo, 1301/1883, vol. 1,
152 p., vol. 2, pp. 1144.
althoughquite remarkablynot fromthe asbb type of works. It devotes
more attention to identifying the literary sources used than the isnds, and
contains assessments of the content of the traditions listed. There cannot be
any doubt that this work represents quite a useful supplement to al-Wids
work. Conversely, we could readily do without the last of the contentions,
namely the harmonization of contradicting traditions.
The Introductions to the Koran
Whereas most commentaries proceed immediately to the actual task, fol- [ii/184]
lowing the customary praise of the Prophet and a brief exposition, there are
others that include chapters of Koran-related material. Amad b. Muam-
mad AL-THALAB,
282
for example, treats on a few pages the merits of the
Koran and its science as well as the difference between ordinary and alle-
gorical exegesis. Al al-Dn Al b. Muammad (AL-KHZIN) al-Baghdd
283
also examines the collection of the Koran, the so-called seven readings, and
the prayer of danger (alt al-khawf), used when reciting revelations. The
earliest commentary to display a comprehensive exposition of the Koranic
sciences is, as far as we know, al-abars Tafsr.
284
The work is followedby the
K. al-Mabn li-nam al-man,
285
which, according to the manuscript, was
started in 425/1033. It contains so much useful information that a printed
edition is a desideratum. Equally valuable is the introduction to K. al-Jmi
al-muarrar al-a al-wajz f tafsr al-Qurn al-azz of Abd al-aqq b.
Ab Bakr b. Abd al-Malik al-Murib al-Gharn IBNAYYA
286
(d. ca. 542/
1147), whereas al-Qurubs
287
Jmi akmal-Qurn follows it pretty closely,
even literally in some places.
Suchencyclopaedic treatises were occasionally published independently [ii/185]
as well, i.e., without connection to exegesis. If the work listed in al-Fihrist,
p. 34, l 14sq., Madkhal il l-tafsr, of Ibn [al-]Imm al-Mirwho cannot
be later than the fourth century ahis indeed such an introduction, it
apparently was of no influence upon the production of the subsequent
282
283
Vol. 1, pp. 311; cf. above, p. 355 n. 238.
284
Vol. 1, p. 1 sqq.; see above, p. 353.
285
Ahlwardt, Verzeichnis der arabischen Handschriften, no. 8790 (= cod. 1 Wetzstein, no.
94). In this manuscript the introduction consists of fol. 189
r
in ten sections (fal); this is
followed by the commentary to sras 115; the rest is wanting.
286
Brockelmann, GAL, vol. 1, p. 412; EI
2
. The introduction in the Berlin Ms., Ahlwardt,
Verzeichnis der arabischen Handschriften, no. 800 (= Sprenger, 408), consists of fol. 192.
287
Berlin, Ahlwardt, Verzeichnis, 810 (= cod. Sprenger, 436), fol. 236; cf. above, p. 356sq.
period. Al-Suy
288
is outright surprised that the early period did not pro-
duce a book on the types (anw) of Koranic science when so much atten-
tion had been paid to the science of adth. Of the works of the fifth, sixth,
and seventh centuries ah al-Suy can mention
289
only those that display
a very distant resemblance to al-Itqn. These are: Funn al-afnn f ulm
al-Qurn of Abd al-Ramn IBN Al AL-JAWZ
290
(d. 597/1200), Jaml al-
qurr of Al b. Muammad Alam al-Dn al-Sakhaw
291
(died 643/1245), al-
Murshid al-wajz li-ulm tataallaq bi-l-Qurn al-azz of Abd al-Ramn b.
Isml ABSHMAH
292
(d. 665/1266), and al-Burhn f mushkilt al-Qurn
of Ab l-Mal Azz b. Abd al-Malik SHAYDHALAH
293
(d. 494/1100). On
the other hand, al-Suy
294
considers his direct predecessors to be two later
appearances, Mawqi al-ulm min mawqi al-nujm of Jall al-Dn AL-
BULQN
295
(d. 824/1421), a brother of his teacher Alam al-Dn AL-BULQ-
N
296
(d. 868/1463), and an untitled work of Ab Abd Allh Muy l-Dn
al-Kfj (AL-KFIYAJ, d. 851/1447), of whom I cannot find a reference else-
where.
297
None of these works seems to have come down to us. It is so
much more commendable, therefore, that al-Suy supplies from each of
them the table of contents, and from the first three also part of the pref-
ace. Accordingly, the last named work consisted of only two chapters (bb),
whereas the first work comprised forty-seven sections (naw). When al-
Suys literary ambition impelled him to offer the public his own ency-
clopaedia of the Koran, he used these books as preliminary studies. Of the
work that he thus created in 872/1467,
298
Tabr f ulm al-tafsr, we know
only the headings of the one hundred and two sections (naw).
299
When al-
Suy decidedlater to compose a secondwork onthe same subject, al-Itqn
f ulm al-Qurn, he also consulted the work of a second contemporary,
288
289
al-Itqn, p. 13.
290
Brockelmann, GAL, vol. 1, p. 504, knows of an excerpt fromthis work. Sezgin, GAS, vol. 1,
p. 130, vol. 9, p. 156. [Edited by asan . Itr, Beirut, 1987/1408].
291
Brockelmann, GAL, vol. 1, p. 94, knows of a manuscript at Cairo. EI
2
.
292
Brockelmann, GAL, vol. 1, p. 317; the work seems to be lost.
293
Brockelmann, GAL, vol. 1, p. 433; the work has probably been lost. EI
2
; EQ.
294
al-Suy, al-Itqn, pp. 210.
295
Brockelmann, GAL, vol. 2, p. 112, calls him Jaml al-Dn; EI
2
.
296
Brockelmann, GAL, loc. cit., p. 96; EI
2
.
297
A. Fischer, Nachtrag: Muy l-Dn Ab Abd Allh appears eighteen times in jj
Khalfa (see index, no. 6403), furthermore Lubb al-lubb, p. 218, where al-Suy says that the
proper pronunciation of the nisba is Kfiyaj. EI
2
.
298
al-Itqn, p. 7, towards the end.
299
al-Itqn, pp. 47.
al-Burhn f ulm al-Qurn of Badr al-Dn Muammad b. Abd Allh AL-
ZARKASH
300
(d. 794/1391). The Itqn has survived in several manuscripts
and exists also in printed editions.
301
As far as one can judge fromthe differ-
ent tables of contents, he not only sticks to the framework of the subject-
based scheme, as established by his predecessors, but also follows them
partially in the arrangement as well. Naturally, nothing can be said about
the proportions, since we are only informed about the number of sections
and not about their volume.
The structure of al-Suys al-Itqn is appropriate, as the number of the [ii/187]
eighty sections (naw) can easily be divided into larger groups. The arrange-
ment is as follows: (1) the external circumstances of the revelation, nos. 117;
(2) the collection and redaction of the text, nos. 1819; (3) the reading of the
Koran, nos, 2042; (4) the style, rhetoric, and writing; nos. 4376. (5) exe-
gesis and exegetes, nos. 7780. As can be seen from the preface (pp. 1317)
and the citations, it is clear that, apart fromthose encyclopaedias, al-Suy
also utilized a great number of special works on history, tradition, exegesis,
grammar, etc. The bulk of material that displays his erudition to the reader
is immense and can, in fact, only be fully utilized once we have an alphabet-
ical list of names and subjects, including biographical and bibliographical
references. The content of such a work will leave little to be desired, leav-
ing out, for instance, only the minutiae of the older textual history, which
must have fallen into oblivion in the ninth century. The authors judgement
is generally more reasonable than can be expected from a Muslim theolo-
gian, although he, too, was unable to transcend the scholastic methodology
and the dogmatic partiality of his time. In any case, Islamdid not produce a
better handbook of Koranic sciences, making the eulogy of the vain scholar
in his postscript (p. 955) not unwarranted. Although al-Itqn was originally
issued as a monograph, the author still intended for it to be an introduc-
tion (muqaddima) to his commentary on the Koran, Majma al-barayn wa-
mala al-badrayn.
302
300
Brockelmann, GAL, vol. 2, p. 91, Suppl. 2, p. 108; recent editions in, 1957 and 1988.
301
The Indianeditionwas anidea of Aloys Sprenger. At its endwe finda remarkof al-Suy
from a different source according to which al-Itqn was completed on 13 Shawwl 878 (3
March 1474), which are my references. I also know of the 1306/1888 Cairo edition.
302
Since Brockelmann, GAL, vol. 2, p. 145sq., does not know this commentary, it must
likely be considered not to have survived. Perhaps it was never completed at all. However,
a comment on the Jallayn of al-Karkh has the identical title.
Poetry as Source of History: The Poetical Examples
in the Biographic and Exegetic Literature
A peculiarity of the Arabic books of history, which as far as we knowcannot [ii/188 ]
be found in any other literature,
303
are the numerous interspersed verses.
They are either put into the mouth of the actors or loosely inserted, serving
not only to embellish the speech but also to substantiate the subject. Of
Muammads biographers Ibn Isq is the one with the most plentiful
poetical insertions. EventhoughIbnHishmeliminateda great deal of them
in his edition, the rest still amounts to one fifth of the entire work when
the indented lines of the verses are counted as full lines.
304
In this respect,
later writers exercised more restraint. It is difficult properly to evaluate
al-Wqids method, since so far we do not have a complete edition of the
Maghz, and the elimination of many poems seems to be due to later
editors. Ibn Sad has in his Sra not even three hundred lines of poetry. Most
of themconsist of elegies on the death of Muammad and are collected in a
separate chapter at the end.
305
In al-Bukhrs chapter on the maghz these
verses take up only nineteen lines. Far more copious is al-abar, although
the number of verses (314) he includes in his section of his chronicle on
Muammads Medinan period is far less than what Ibn Hishmincludes on
the Battle of Badr alone.
As far as the conclusiveness of the poetical citations is concerned, it is
undeniable that verses that accidentally and casually refer to a certainevent
represent a very valuable testimony. But it is not unusual that verses are not
at all related to the facts that are reported. In the field of elegies in particular
one must be prepared for large-scale falsification. As far as authenticity is
concerned, particularly solid trust may be given to the rather mischievous
verses of Muammads enemies, such as Abd Allh IBNAL-ZIBAR, which
the earliest biography has preserved with praiseworthy candour.
306
Naturally, complete poems or even entire collections (dwn) of a poets [ii/189]
work are far more valuable than short fragments. By far the most important
303
Onthe other hand, it is difficult toconsider this anArabinvention. The questioncannot
be separated from the genesis of historical prose; see above, p. 318sq.
304
Nearly one third of which are elegies, namely Wstenfelds edition of Ibn Hishm,
pp. 108114, 516539, 611638, 704714, and 10221026.
305
Ibn Sad (al-abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung Muammads
[Muammads last illness and death], pp. 8998.
306
Cf. Th. Nldeke, Die Tradition ber das Leben Muhammeds, p. 160sq. Ibn al-Zibar is
frequently quoted by Ibn Isq. Other fragments can be found in al-Aghn, vol. 14, pp. 1125.
EI
2
; Sezgin, GAS, vol. 2, pp. 275276, vol. 9, p. 276.
document, the dwn of assn b. Thbit,
307
who was the poet laureate of
Muammad, is extremely rich in historical allusions. The connection with
Koranic exegesis, the particulars of the historiography, as well as the authen-
ticity of the transmitted poems, are all problems that still await a solution.
Among the men who were not from Medina but who had personal con-
tact with Muammad there are three who deserve special mention. Labd
b. Raba came with an embassy from his tribe, the Kilb, to Muammad
in Medina in 9/630 and converted to Islam.
308
Kab b. Zuhayr, a son of the
well-known poet of the muallaqa, at first made fun of Muammad in satir-
ical poems. However, when he realized that from that time forward his life
was in danger, he changed his mind and became a Muslim so that the Mes-
senger of God would forgive him. The captivating ode, Bnat Sud,
309
which
he then recited, pleased Muammad so much that he presented him with
his robe. Yet another famous ode to Muammad is from (Maymn b. Qays)
AL-ASH of al-Yamma, who had close relations to Christianity, although
we have nothing reliable about the circumstances of its origin.
310
Pre-Islamic paganpoetry, whether contemporary or prior to the advent of [ii/190]
Islam, is one of the main sources for the foundations of the cultural environ-
ment onwhich the remarkable phenomenonof Islamis based. The Jhiliyya
did not produce anything that deserves the designation of national liter-
ature. The religio-historical aspect of this literature has never been suffi-
ciently appreciated, since it presents itselfat least in the qadasclad
in images that are obscure and difficult to understand. This, for example,
applies toal-Nbigha, Antara, arafa b. al-Abd, Alqama b. Abada, andImru
al-Qays,
311
to mention merely their most prominent poets.
307
The earliest printed edition appeared in Tunis, 1281/1864; the Indian lithograph, Bom-
bay, 1281, seems to be a reprint of this edition, and with a fictitous date. I have no information
as tothe Cairoedition, 1904. [Cf. Sezgin, GAS, vol. 2, p. 292, withcommentary by Shukr Makk,
Cairo, 1321/19031904, and with commentary by al-Barqq, Cairo, 1331/19121913.] Hartwig
Hirschfelds European edition, 1910, leaves much to be desired. The text is bad. The frag-
ments and their variants in literature have not been collected. No attempt has been made
to determine the authenticity. August Fischer, Nachtrge: Apart from assn b. Thbit a
second early Medinan poet deserves mention, Qays b. al-Kham [EI
2
; Sezgin, GAS, vol. 2,
pp. 285286,] whose dwn is important for the conditions at Medina immediately before
the rise of Islam. Th. Kowalski edited the dwn together with a German translation (Leipzig,
1914).
308
J. Chalidi (1880) A. Huber and C. Brockelmann (1891) deserve credit for the publication
of his dwn.
309
Sezgin, GAS, vol. 2, 230235; R.A. Nicholson, Literary history of the Arabs, s.v.
310
Regarding the poetical heritages of Kab and al-Ash cf. Brockelmann, GAL, vol. 1,
pp. 3739; Sezgin, GAS, vol. 2, pp. 229235, and 130132 respectively. [August Fischer in
Nachtrge: cf. F. Krenkow, Tabrzs Kommentar zur Burda des Kab b. Zuhair.]
311
The Divans of the six ancient Arabic poets, edited by W. Ahlwardt (London, 1870).
Umayya b. Ab l-alt,
312
whois fromal-if, near Mecca, occupies a unique
position within this literature. He displays an unprecedented preference
for Biblical and post-Biblical subjects, not only outside the Koran but also
confessing to the Unique God and the Hereafter. Consequently, he, more
than anyone else, has attracted the attention of recent research,
313
which a
number of years ago produced an excellent edition of his extant fragments
accompanied by a translation and commentary.
314
A study on his relation-
ship to the Koran is also in preparation.
315
However, further advances cannot
be expected until other contemporary poetic works have also been investi-
gated.
Whatever ancient writers didnot gather inspecial collections of the intel- [ii/191]
lectual heritage of individual poets must be extracted fromthe entire histor-
ical, belletristic, and grammatical literature. The main sources are antholo-
gies, a favorite of the Arabs.
316
Best known among them is the amsa of
Ab Tammm abb b. Aws as well as of al-Butur,
317
al-Mufaalyt, the
Jamharat ashr al-Arab and the Dwn of the Hudhalites. Most prominent
of all is probably the Kitbal-Aghn of Abl-Faraj al-Ifahn (d. 356/966). It
must be remembered, however, that in this invaluable collection the biogra-
phies of the poets take up far more space than the poems, so that the prose
sections are generally more informative than the poetry. On the other hand,
the historical accounts of Muammad are of no interest, since the relevant
sources are now extant in the original version.
318
The historical exploitation of poems or fragments of poems runs parallel [ii/192]
to references to individual verses purporting to explain rare Koranic words,
word formations or semantics from the language of poetry. This is not done
312
EI
2
; EQ; Sezgin, GAS, vol. 2, pp. 298300, vol. 9, 277.
313
Sprenger, Leben und Lehre des Mohammad, vol. 1, pp. 7681, 110119; Cl. Huart, Une
nouvelle source du Qorn considers all Umayyahs 130 verses in al-Muahhar b. hir AL-
MAQDISs Arabic work, Le livre de lacrationet de l histoire [Sezgin, GAS, vol. 7, pp. 277288],
to be genuine, which is by no means the case. Fr. Schulthe, Umajja ibn Abi--alt must be
credited to have prepared the ground for historical criticism.
314
Umajja ibn Abi -alt; die unter seinem Namen berlieferten Gedichtsfragmente (col-
lected and translated by) Friedrich Schulthe.
315
I.G. Frank-Kamenetzki, Untersuchungen ber die dem Umajja ibn Abi l-alt zuge-
schriebenen Gedichte zum Qorn [research in Umayyads poems on the Koran].
316
Brockelmann, GAL, vol. 1, pp. 1721.
317
EI
2
; EQ; R.A. Nicholson, Literary history of the Arabs (1907); Sezgin, GAS, vol. 2, 560564.
318
Two huge historical sections regarding the battles of Badr and Uud are copied verba-
tim from al-abars chronicle, and this including the chain of transmitters: al-Aghn, vol. 4,
p. 17 l 2334, l 14 = al-abar, vol. 1, p. 1291 l 141348 l 5; al-Aghn, vol. 14, p. 12 l 125, l 6 = al-
abar, vol. 1, p. 1383 l 171430 l 12. Al-abars omitted passages and some modifications have
been disregarded.
because the interpreterscontrary to the Prophets explicit denial (sras
21:5, 36:69, 52:30, 69:41)considered the Prophet a poet, but is rather based
onwhat has beensaidrepeatedly, namely that apart frompoetry the ancient
Arabs did not possess a proper national literature.
319
We do not know who
initiated the methodology. The tradition which claims to knowfor sure that
Ibn Abbs made use of poetry to explain the Koran
320
does not deserve
unconditional confidence, given what has become known of this mans
personality.
321
Although Ibn Hishm (d. 218/833) omitted many verses when editing
Ibn Isqs Sra,
322
it is precisely he who introduced all the lexicographic
explanations of the Koran based on poetic quotations. As far as I can tell,
this methodwas not imitatedinthe exegetic comments onlater biographies
of the Prophet, although it was adopted in later commentaries, as becomes
obvious inal-abar, al-Zamakhshar, and al-Bayw. As a consequence, yet
other books appearedthat extractedthe pieces of evidence (shawhid) from
the commentaries and then explained them philologically and historically
(shar).
319
Cf. above, p. 318319sq, and 321.
320
I have it fromSadb. Jubayr andYsuf b. Mihrnthat IbnAbbs was oftenconsultedon
the Koran. He then used to reply: The matter is such and such; did you not hear the poet who
expressed himself in such and such a way? Ibn Sad (al-abaqt al-kabr, vol. 2, part 2): Letzte
Krankheit, Tod und Bestattung Muammads [Muammads last illness and death], pp. 25.
321
322
Cf. above, p. 365.
NINETEENTH-CENTURY CHRISTIAN RESEARCH
Criticismof the Systemof Tradition
Evenas Christianscholars throughout the entire Middle Ages andevenright [ii/193]
up until the modern era labelled the founder of Islam a swindler, impos-
tor and false prophet, they still never thought to challenge the credibility
of the Islamic tradition as such. The first European not only to acquire
a precise knowledge of the vast material of the Arabic tradition but also
to work successfully on the critical examination of its content was Aloys
Sprenger. His pioneering work was published in 1856 in a variety of peri-
odicals.
1
He later presented his ideas once more in the introduction to the
third volume of his great work on the life of Muammad.
2
According to
Sprenger, the systematic biography of the Prophet at its earliest stage con-
sists almost totally of legends and stories that evolved not only from nave
beliefs but also from audacious fraud. Here the reports about Muammads
Companions are more reliable than those about him. The military cam-
paigns that constitute the second part are generally not compiled, like the
others, for entertainment and edification but rather for a candid interest in
the course of events. The most important of the extant works are appropri-
ately characterized, and the most valuable materials competently singled
out. Finally, there is the recommendation to scholarship to dispose of dog-
matic biography.
3
According to Sprenger, the canonical adth or the sunna
evolved from the civil wars, but between 40/660 and 80/699 it advanced
as rapidly as the conquests had earlier. One might safely assume that at
the end of the first century ah the largest portion by far of the store of
traditions was in the hands of competent men, and had already been for-
mulated. Although the work of forgers like Ibn Abbs and Ab Hurayra
cannot be underestimated, the sunna contains more truth than falsehood
1
Journal of the Asiatic Society of Bombay, vol. 25, pp. 5374, 199220, 303329, 375381,
andZDMG, vol. 10, pp. 117. [Biographical data onall of the scholars mentioned inthis section
can be found in Concise Biographical Companion to Index Islamicus, by W.H. Behn (Leiden,
20042006).]
2
Leben und Lehre des Moammed, vol. 3 (1865), 2nd ed. (1869), pp. iclxxx.
3
Cf., loc. cit., pp. lvviii, lxi, lxiv, and lxxxvi.
370 nineteenth-century christian research
and isafter the Koran and the documentsthe most reliable of the his-
torical sources.
4
William Muir follows in the footsteps of Sprenger, both of whom were [ii/194]
members of the Anglo-Indian Civil Service. Muir, in his introduction to
the first volume of his Life of Mahomet,
5
lucidly explains the reasons that
determined or might have determined the rise and change of the traditions.
He strongly emphasizes the tendencies that grew out of party, tribal or
family policies, nationalist interests, dogmatic prejudice or Christian and
Jewish influences. Whatever he lacked in the way of geniality, imagination,
and erudition compared to his predecessor he made up for with sober
judgement, better historical methodology, andorderly presentation, looking
far more critically at the sources and abandoning many things that were
beyondthe least doubt toSprenger. Nevertheless, the confidence withwhich
Muir approached transmitted reports is still very considerable, and this
mood prevailed as a common heritage among scholars almost to the end
of the nineteenth century. As late as 1879 the important Dutch Arabist
and historian, R.P.A. Dozy, still considered nearly half of al-Bukhr to be
a historical document.
6
A decisive change, and a new course, did not occur until the publica- [ii/195]
tion of the second volume of Muhammedanische Studien
7
of the brilliant
Hungarian scholar, Ignaz Goldziher, whose erudition surpasses that of even
Sprenger, and who had mastered the Islamic history of state, culture, and
dogmas better than anyone else. Based on such a broad spectrum, he was
able not only to deepen our knowledge of the tendentious nature of adth
and illuminate with a multitude of convincing examples, but also to inves-
tigate the theoretical and practical developments of the manner of trans-
mission throughout the centuries. Every current and counter-current in the
life of Islam found expression in the form of adth. This applies to political
4
Cf., loc. cit., pp. lxxxii sq., lxxxvii, lxxxix, and civ.
5
Pp. xxviicv.
6
R.P.A. Dozy, Essai sur l histoire de l islamisme, p. 124.
7
Halle, 1890, pp. 1 to 274; English translation, Muslim studies, Chicago, 1977, pp. 1251,
under the special subtitle, On the development of the adth. The course of the study is
evident from the chapter headings: (1) adth and sunna; (2) Umayyads and Abbsids; (3)
The adth in its relation to the conflicts of the parties in Islam; (4) The reaction against
the fabrication of adths; (5) The adth as a means of edification and entertainment;
(6) alab al-adth; (7) The writing down of the adth; (8) The adth literature. A valu-
able addition is Goldzihers article Neue Materialien zur Litteratur des Ueberlieferungswe-
sens.
nineteenth-century christian research 371
parties and social schools as well as to canonical differences and dogmatic
quarrels. Up to a certain degree, Muslims of the second century admit to
the existence and justification of false adths, and that it was legitimate for
the moral benefit of the people and the advancement of piety to fabricate
and circulate sayings of the Prophet. Among the rich content of this epoch-
making work might also be noted the exposition of the basis of Islamic criti-
cismof adth, whichclings tothe formalities of the isnd, as well as the vivid
descriptionof the maincollections of adth, a task whichnoone before him
ever attempted.
8
Although Goldziher was mainly concerned with the legal
adth, his methodology could easily be applied to the historical tradition,
where it would exert an extraordinary influence and result in a complete
revolution in approach. Whereas up to his time every tradition was sound
until proven otherwise, research has gradually gotten used to the reverse
point of view.
The first scholar to apply Goldzihers methodology to prominent person- [ii/196]
alities of the early Islamic period was Theodor Nldeke.
9
In the Prolegomena
to the first volume of the Annali dellislam, Leone Caetani enlarged upon
Goldzihers principles, with special reference to historical sources. Cae-
tani paid particular attention to the isnd, investigating more deeply these
peculiar introductions to the transmissions in the most important sources
and trying to learn details about the individualities named in the trans-
mitting links in the chain of authorities, particularly the suspected forgers,
Ibn Abbs and Ab Hurayra, and their reputation as collectors and liter-
ates.
10
Based on this preliminary investigation, Caetani submitted several
incidents in the biography of the Prophet to an original and severe criticism,
even though he occasionally went a bit too far.
11
The most radical contemporary scholar inthis respect is Henri Lammens,
S.J., who continues Goldziher and Caetani, combining enormous erudition
with splendid powers of discernment. He put down his views in the follow-
ing sentences:
1. The Koran supplies the only historical basis of the sra.
2. The tradition does not offer an addition but rather an apocryphal
development.
8
Cf. pp. 125, 149, 136144, and 189251.
9
Zur tendenzisen Gestaltung der Urgeschichte des Islams.
10
Annali dellislam, vol. 1, pp. 2858.
11
Nldeke pointed out many of these exaggerations in his reviewof the first two volumes
of Caetanis work in WZKM, 21, 297312. I replied to some of Caetanis remarkable findings
above, p. 229sq. and 281. See also above, pp. 83sqq. on sras 53:19sqq., and 22:51.
3. The merit of a tradition corresponds to the extent of its independence
from the Koran.
4. As far as the Medinan period is concerned, the existence of a vague
oral tradition must be admitted.
12
These theses are partial and exaggerated because the body of traditions [ii/197]
needs to be enlarged, because there are traditions that accompany Koranic
revelations,
13
and because the fabricated traditions are so diverse that a
Koranic source alone would seem most unlikely. The investigation of Lam-
mens evidence is no more than a confirmation, for among the different
groups into which he classifies the subjects of narration, there is only one
the first of Muammads revelationswhich can be traced back to Koranic
allusions, while in the case of the otherschildhood history, age,
14
number
of sons, campaignsthere are still other non-Koranic sources to be consid-
ered, including the case of the names, wives, and personality of the Prophet
(shamil), where there is hardly any connection with the Koran at all. Lam-
mens main mistake is that, for no apparent reason, he generalizes correct,
individual observations, partly already made by others, and inexplicably
overextends them as a principle.
15
I referred previously to the narratives of
origin that are not, as they would have us believe, based on an accompa-
nying tradition; rather they are the product of the learned exegesis of the
Koran.
16
The Christian Biographers of the Prophet [ii/198]
Corresponding to the judgement of the Arab tradition, with which we have
just acquainted ourselves, the Occidental works on the life of Muammad
can be divided into three periods: (1) the unbroken rule of tradition until
the middle of the nineteenth century (Sprenger); (2) the period of incipi-
ent criticism of individual parts of the transmission; and (3) the period of
systematic criticism of the entire tradition.
12
H. Lammens, Qoran et tradition, comment fut compose la vie de Mahomet.
13
See above, p. 341 sq.
14
This topic Lammens treated a bit later in a separate article, L ge de Mohamet
15
C.H. Becker in Islam: annual bibliographic report, p. 540sq., anticipates the out-
right self-dissolution of historical criticism because of Lammens excessive scepticism. For
more detailed criticism of Lammens exaggerations see Th. Nldeke, Die Tradition ber
das Leben Muhammeds in Der Islam [the tradition about the life of Muhammad], vol. 5.
pp. 160170.
16
Cf. also above, p. 341 sqq.
I.
The first period covers the entire Middle Ages to the middle of the nine-
teenthcentury. Acomprehensive expositionof the periodwouldbe a worth-
while socio-historical research project.
17
Relevant for us are only those writ-
ers who had access to Oriental sources and who tried to reach a compe-
tent judgement free of prejudice. Among the earliest of them we count
Johann H. Hottinger of Zrich,
18
the Italian L. Marracci,
19
and the Dutchman
Reland (Reeland).
20
No one, however, regardless of personal prejudice
considering Muammad to be the most criminal of all men and the greatest
sworn enemy of Godwas more objective than the Frenchman Jean Gag-
nier.
21
He thought that he would best serve truth if he could acquaint Euro-
peans with what Muslims themselves said about the Prophet, and therefore
considered it appropriate to add a series of translations fromArabic sources.
Of course, he only hadaccess tolate writers like Abl-Fid (d. 732/1332), and
al-Jannb (d. 999/1590). He personally added nothing, neither praise nor
blame, neither doubt nor conjecture; only the connecting sentences of the
different accounts are his own. Nearly four generations of writers learned
from this wealth of information, each of them deriving what he considered
correct according to his subjective opinion and omitting what was contrary
to his views or prejudice.
The Essai sur l histoire des Arabes (18471848) of Armand P. Caussin de [ii/199]
Perceval can be considered a modern version of Gagnier. In addition to
enlarging the scope to include pre-Islamic Arabia in more detail and cov-
ering the conversion of Arabia down to the time of Ab Bakr, he also dis-
tinguishes himself from Gagnier by using more comprehensive and older
sources, which he presents with more liberty, albeit while retaining all the
peculiarities.
The first scholar to apply the historical-critical method to the history of
the Prophet is the Heidelberg Orientalist, Gustav Weil.
22
That I still consider
him part of the first period is because he knew only very little of the Arabic
17
August Fischer in Nachtrge: An exposition of the Occidental appreciation of the
Prophet as suggested by Schwally is now available from Hans Haas, Das Bild Muhammeds
im Wandel der Zeiten, [the image of Muammad in the course of time] and from Ernest
Renan, tudes d histoire religieuse, 2nd ed. (1857), p. 222sqq.
18
Historia orientalis (1651), 2nd ed. (1660); see also Concise biographical companion to
Index Islamicus, by W.H. Behn (20042006), s.v.
19
Ibid., s.v.
20
Adriaan Reland, 16761718; ibid., s.v.
21
La vie de Mahomet (1732).
22
Mohammed der Prophet, sein Leben und seine Lehre (1843).
sources, did not have a clue about the Muslim criticism of traditions, and
therefore could conceive the idea of taking a position in all of this. The
main sources he used are three late Arabic works: an excerpt from Ibn
Hishm by Imd al-Dn (Amad b. Ibrhm) AL-WSI (711/1311),
23
the
Tarkhal-khams (982/1574)
24
of al-Diyrbakr, andal-alabiyya(1043/1633)
25
of Al b. Ibrhm AL-ALAB, the last two works containing much old and
good information. Weils lasting merit is that he recognized the importance
of the Koran as a historical source for the life of the Prophet and made use
of it accordingly. Although his book represents a great advance, it became
conspicuously out-dated soon after its appearance. The reason for this is
that other biographies of the Prophet quickly appeared whose authors had
better sources at their disposal, criticizing not only the reported facts but
also making the very sources the object of their criticism, and who, last but
not least, substantially surpassed the complacent Heidelberg professor in
terms of knowledge, ability, and sense of history.
II.
These advantages become immediately apparent in Aloys Sprengers Life [ii/200]
of Mohammed, even though it covers only the period up to Muammads
appearance at Medina. Apart fromthe prime sources of Ibn Hishmand al-
abar, Sprenger knows not only all the important Arabic works enumerated
above but also many more, both from manuscripts and from lithographs
of adth collections. He displays an unrivalled knowledge of the sources,
combined with a penetrating critical approach to the literature, illuminat-
ing its genesis and character as no one had before him. Although his book in
English remained unfinished, ten years later he produced a far more com-
prehensive work in German
26
that opened a new era of biography of the
Prophet. It is of epic importance, as it is the first attempt to utilize not only
the whole spectrum of native tradition but also to determine its merit as
a source of history. Even apart from this, however, the work eclipses every-
thing hitherto written, both in form as well as content. Since he attributes
the influence of Islammainly tothe spirit of the age, he attempts tosupply as
many details as possible, and presents as many actors as possible, with their
words and actions. The only thing he seems to have learned fromWeil is the
23
Ibn Hishm, ed. F. Wstenfeld, p. xlvi; and Brockelmann, GAL, vol. 2, p. 162.
24
25
Ibid., p. 307.
26
Das Leben und die Lehre des Mohammad (18611865). See already above, p. 369.
utilizationof the Koranas the mainsource for the biography of Muammad.
Yet he goes far beyond Weil and includes nearly two thirds of the Holy Book
in translation. He was the first scholar to recognize (vol. 3, p. 20sqq., and
pp. 359475) the importance of Muammads Constitution of Medina (Ibn
Hishm, p. 341 sqq.)
27
for the development of the ecclesiastical state as well as
the usefulness of the section on Muammads diplomatic negotiations with
the Arab tribes in the Sra of Ibn Sad for the history of Islamic missionary
activity. Additionally, there is his natural perception for the driving forces of
life and history, his eminent ability to understand the soul of the actors of
the past, his startling, brilliant inspirations, and his vivid, intelligent style.
Opposing these brilliant virtues there arealbeit not as importantlystill
considerable deficiencies. Most annoying is the chequered medley of nar-
rative and critical discussion, a clear indication that the author had not yet
mastered the material. His animated spirit continuously interferes with the
course of the methodical investigation. In the case of the fundamentals of
Muslim criticism of tradition, his unbelievable familiarity with the sources
tempted him to a greater calm than permissible. His rationalism is unable
to comprehend Muammads nave, religious self-confidence. And, finally,
his philological precision when interpreting Arabic texts leaves something
to be desired.
This unbelievable erudition, which Sprenger had already displayed in
1851, when he published his Life of Mohammad, he could obtain only in a
country like India, where a goldmine of manuscripts could be had and the
most important collections of adth were readily available as lithographs.
His inquisitive character enabled him to discover in dusty libraries impor-
tant works by, among others, al-Wqid, Ibn Sad, and portions of al-abars
annals, to encourage others to publish them alone or in cooperation with
Indian scholars,
28
and then to animate the editors of these works. When
he finally returned to Europe in 1858, he brought back with him probably
the most systematic and comprehensive collection of Oriental manuscripts
and printed books that have ever come fromthe Orient. After the collection
passed to the Kniglische Bibliothek, Berlin, it was instrumental in inaugu-
rating a new era of Islamic studies in Germany.
In the same year that the first volume of Sprengers German biography of [ii/202]
Muammad was published, the lastfourthvolume of an English work
27
An English translation from the Dutch appears as an appendix to A.J. Wensincks
Muhammad and the Jews of Medina (1975), pp. 128138.
28
al-Suy, al-Itqn, 18521854; al-s, Fihrist, 1855; Ibn ajar al-Asqaln, al-Iba f
tamyz al-aba, 18561893.
onthe same subject appeared.
29
Its author, WilliamMuir, was also a member
of the Anglo-Indian Civil Service. He naturally bases his work on Sprengers
earlier work, to whomhe is also indebtedfor pointing out the classic sources
al-Wqid, Ibn Hishm, and al-abar. He otherwise pursues completely
independent research and insists on his own judgement. As far as erudition,
powers of discernment, and spirit is concerned, however, he is no rival to his
predecessor. Compensating somewhat for this inferiority, he displays more
composure in the line of argument, more methodology, and a great deal
more commonsensesupposedly so rare among scholarswhichinmany
cases gives hima better conception of events and the reliability of tradition
than Sprenger possesses.
30
For this reason Muir is the better guide for the
layman. The dogmatic prejudice that prompted him to claim in earnest
that Muammad was moved by the spirit of Satan and not of God is not
disturbing, since it has no bearing on the course of his exposition and, in
general, he interprets properly the contradictory actions of the Prophet.
Nldekes popular booklet, Das LebenMuhammeds,
31
is based onhis inde- [ii/203]
pendent research, which is explained in the first edition of his Geschichte
des Qorns. With its fortunate combination of critical approach and a plain
and charming style, it is unsurpassed to this day. As Wellhausen expressed
it at the time,
32
anyone wanting to study the historical Muammad with-
out access to the Arabic original sources would be well advised to resort
to this little book along with Wellhausens abridged German version of al-
Wqid rather than consulting Sprengers great work. It goes without saying
that after more than fifty years, and the great advances in our knowledge,
particularly in the criticism of tradition, some of the views are outdated.
Reinhart Dozy
33
has an eye for the characteristics of personality of the
historical actors, although without the ability to penetrate deeper, with the
result that too often he succumbs to tradition and frequently includes it
verbatim in his narration. He is constantly surprised that there are so many
authentic reports in the collections of adth, and he maintains that even by
29
WilliamMuir, The Life of Mahomet. See already above, p. 370. [August Fischer in Nach-
trge: William Muir, The Life of Mahomet from original sources, 2nd ed., London, 1876, 3rd
ed., London, 1894, and WilliamMuir, The Life of Mohammad fromoriginal sources, a newand
rev. ed. by T.H. Weir, Edinburgh, 1912, cxix, 556 p. constitute a one-volume excerpt from the
authors great work.]
30
Cf. Muir, Life of Mahomet, vol. 1, p. lii, and Sprenger, Leben und die Lehre, vol. 1, p. xiv.
31
Das Leben Muhammeds nach den Quellen populr dargestellt.
32
Muhammed in Medina; das ist Vaqidis Kitab al Maghazi in verkrzter deutscher Wieder-
gabe, preliminary excursus, p. 20.
33
Essai sur l histoire de l islamisme, pp. 1132.
strictest criteria half of al-Bukhr ought to be included (see already above,
p. 332). Conversely, the best of sources, like Ibn Hishm, who had long been
published, he by no means explores sufficiently.
Of his predecessors Ludolf Krehl
34
utilizes mostly Sprenger and Weil. He [ii/204]
allots a disproportionally large space to psychological reflections on the
reasons that guided the Prophet in his actions. He does not seem to have
been able to deal with the subject adequately.
Leopold von Ranke
35
makes use of not only the entire literature but also
the sources that were available in translation, and he generally presents an
accurate picture. Despite the conciseness of the presentation, however, even
our great historian was unable to advance research; due to the peculiarity
of the subject, the universal approach was of little help. Where excessive
brevity might lead can best be seen on p. 84sq., where the merciless slaugh-
ter of the Jewish Quraya
36
is considered typical of Muammads treatment
of the Jews.
Chapters two to four on Muammad in August Mllers well-known Der
Islam im Morgen- und Abendland [Islam in Orient and Occident], vol. 1,
pp. 44207, are a bright and elegant summary of earlier research rather than
a product of a thoroughstudy of the sources. Inseveral instances this offered
the clever author the opportunity to viewold facts fromnewand surprising
perspectives.
Hubert Grimme, in the first part of his Mohammed; das Leben nach den [ii/205]
Quellen,
37
is more or less limited to the Koran, taking hardly any notice of
tradition, even in the case of important and controversial problems. The
goal of reaching his own conclusion independently of predecessors is com-
mendable. Unfortunately, not all of his original interpretations are equally
well documented. This applies particularly to his mainargument that Islam
did not at all appear as a religious system but rather as a socialist attempt
to counteract a certain predominantly bad secular state of affairs.
38
He
pretty much gets the logic reversed here that fromthe beginning, and exclu-
sively, Muammads sermons at Mecca had a religious orientation, and
34
Das Leben und die Lehre des Muhammad. The projected second part of the study on
Muammads teaching has never been published.
35
Die arabische Weltherrschaft und das Reich Karls des Groen, pp. 49103.
36
Cf. Juynboll, Encyclopedia, p. 508, col. 1.
37
Darstellungen aus dem Gebiete der nichtchristlichen Religionen, no. 7: A Catholic com-
pilation.
38
Grimme, p. 14; cf. in general pp. 1421, 2931, and 39sqq. The book has been thoroughly
reviewed by Chr. Snouck Hurgronje in Revue de l histoire des religions, 30 (1894), pp. 4870,
and 149178.
that zakt (= adaqa, sra 9:60) was not introduced as communal tax until
Medina, and even in this case not mainly as a relief for the poor but in order
to cover the expenses of the military campaigns. The second part of the
authors projected work was conceived as a supplement to the biography
and would have followed up the genesis and development of Muammads
religious ideas throughout his life, and explained the inner as well as outer
causes. But instead, after a brief survey of the originof the Koran, he supplies
us with the system of Koranic theology.
39
In another monograph that was published ten years later,
40
Grimme [ii/206]
firmly maintains the erroneous opinion that Muammad was a social
reformer (pp. 48, 54, 58, 64, and 73). In the later work, however, this opinion
recedes into the background behind his newdiscovery of the South Arabian
origin of Islam: based on an image of South Arabia that is both subjective
and fantastic, and which he creates for himself, he considers Muammads
earlier Meccan ideas as a reflection of the South Arabian spirit (p. 48). His
concept of God resembles that of the South Arabian monotheists (p. 49),
whose owndesignationis preservedinthe Sabians of the Koran(p. 49). His
concept of the Other WorldParadise, Hellis neither Jewish nor Chris-
tian, but rather continues the South Arabian idea of the thither world
(p. 50). Islam and devotion are considered the renewal of the South Ara-
bian divine slavery (p. 60), zakt going back to South Arabian temple taxes
(p. 60) and the prayer ritual also going back to similar ceremonies at the
temple (p. 50). Also in other respects, Christian and Jewish influences are
considered to have been totally absent fromearly Islam(p. 53). It is peculiar,
however, that Muammad sent the hard pressed believers to Abyssinia but
not to monotheistically inclined South Arabia (p. 55).
Grimme did not supply evidence for any of these daring assertions. Of
course, centuries before Muammads appearance in the Yemen there had
been Christian and Jewish settlements, and Jewish ideas in particular left
their mark in the odd places of heathendom. It is extremely unlikely, how-
ever, that Jewish and Christian ideas did not directly influence Muam-
mad but rather reached him only in the watered down form in which they
reached Yemenite heathendom. It would be equally unbelievable that, in
this respect, Mecca was exclusively influenced from the south. After all,
there were still many other Jewish and Christian centres more readily acces-
39
Mohammed; 2nd part: Einleitung in den Koran; System der koranischen Theologie. The
pages 1 to 29 are devoted to the introduction of the Koran.
40
Mohammed; die weltgeschichtliche Bedeutung Arabiens (1904). [Muammad, the uni-
versal importance of Arabia.]
sible to the Meccans through their lively trade connections, not to mention
Syria and particularly Abyssinia, which was so easily accessible by sea. The
channels by which knowledge of the old revealed religions reached Mecca
were as numerous and manifold as the meanderings of trade to this centre
of commerce and pilgrimage. Grimmes attempt to eliminate this diversity
in favour of the one South Arabian trail cannot be justified and must be
considered a total failure.
Frants Buhl
41
very carefully utilizes European literature as well as the [ii/207]
most important Arabic sources, reaching his own conclusions based on crit-
ical verification and producing a largely penetrating, acute, and thoughtful
investigation. The first third of the work is devoted to Arab paganism; the
rest consists of an appropriate survey of the original sources. His success-
ful disentanglement of the contradictory statements regarding the Battle of
Badr, and the emigration to Abyssinia may be added in passing. Such an
objective and comprehensive book was long overdue, although its Danish
garb will always be an obstacle to its wide recognition.
David S. Margoliouth
42
makes the mistake of not appropriately consid-
ering all the important problems. The introduction of new material from
hitherto untapped sources is no substitute, since the critical edition leaves
something to be desired. On the other hand, it is an undeniable service that
hethe very first scholar as far as I can seedraws a parallel between Mor-
monism and Islam.
III.
The third period of Christian biographical writing on Muammad, which
was inaugurated by the publication of the second volume of Ignaz Goldz-
ihers Muslim Studies (1977), originally published in 1890 under the title
Muhammedanische Studien (1890),
43
can be recognized first in the section
on the Prophet in Annali dellislam of Leone Caetani, Duke of Teano. This
gigantic work
44
has a quite peculiar arrangement, combining the objectivity
of a Gagnier or a Caussin de Perceval with the critical spirit of Sprenger. The
author first of all mentions nearly all the sources in translation, printed as
well as in manuscript form, and lists parallels, including important differ-
41
Muhammeds Liv; med en Indledning om Forholdene i Arabien fr Muhammeds Optrae-
den. Kbenhavn, 1903.
42
Mohammed and the rise of Islam, 1905.
43
English translation, Muslimstudies, Chicago, 1977.
44
Annali dellislam, vol. 1 (xvi, 740 p.), vol. 2 (lxxviii, 1567 p.), vol. 3 (lxxxiii, 973 p.), vol. 4
(xxxv, 701 p.), vol. 5 (xxxvi, 532 p.), vol. 6 (viii, 218 p.), vol. 7 (lv, 600 p.).
ences. This huge collection is accompanied by pertinent explanations and,
most importantly, by thorough critical elaborations of all sorts, testifying
not only to his powers of discernment but also to his historical ingenu-
ity. It is self-evident that incontestable results cannot be attained on such
unsafe ground. Caetani supplemented and enlarged by independent obser-
vation Ignaz Goldzihers systematic theory of the criticism of tradition, as
explained above, p. 371 sq. Strictly speaking, we are not dealing with a biog-
raphy at all but rather with a preliminary study towards such an end.
45
Yet it
must not be forgotten that we are still on the threshold of the third period
of the European biographies of the Prophet.
Separate Studies in History and Interpretation
46
[ii/208]
Recent research in the relation of the Koran to Judaism begins with Abra-
ham Geigers pioneering work on Muammads borrowing from Judaism.
47
Its results were quickly accepted but, unfortunately, Jewish theologians
of the following generation did not continue the study, either because of
a lack of interest or insufficient Arabic studies or both. The first scholar
who wanted to follow in Geigers footstepsafter nearly half a century
is Hartwig Hirschfeld, whose doctoral research focused on the Jewish ele-
ments in the Koran.
48
Israel Schapiro was planning a collection of all the
Haggadic elements in the Koran but so far only the first part has been pub-
lished, i.e., his Straburg doctoral thesis onthe Sra Joseph(12).
49
Everything
that he offers does not serve to throw light on the Koran itself but only on
its interpretation. Atimely rewriting of Geigers work would be highly desir-
able. Its 1902 reprint, falsely called a revised edition, is a regrettable mistake.
The situation is far worse when it comes to Christian elements of the [ii/209]
Koran, which have not seen any comprehensive treatment at all. Karl Fried-
rich Gerocks presentation of the Christology of the Koran of 1839,
50
which
45
[August Fischer in Nachtrge: The main features of a biography of Muammad are to
be foundonpp. 1 to325 of the thirdvolume of Caetanis Studi di storiaorientale (Milano, 1914,
ix, 431 p.). The content of this volume is supplied on the title page: La biografia di Maometto
profeta ed nomo di stato.Il principio del califfato.La conquista d Arabia.]
46
In view of the large scope of the relevant writings an exact subject arrangement is
impossible. For this reason the following observations appear in chronological order.
47
Was hat Mohammed aus demJudenthume aufgenommen?
48
Jdische Elemente im Korn (1878). [August Fischer in Nachtrge: a revised and en-
larged edition entitled Beitrge zur Erklrung des orns appeared in 1886.]
49
Die haggadischen Elemente imerzhlenden Teil des Korans (1907).
50
Versuch einer Darstellung der Christologie des Koran (1839) is probably a recast of his
1833 doctoral thesis, Al-Coranus prophetici muneris Christi laudatur.
after the progress in ecclesiastic history is completely insufficient, should
have beenreplacedlong ago. It still has not dawnedontheologians, however,
that Islam is part of Church history.
An enormous advance in every respect is the work of the Dutch scholar
ChristiaanSnouckHurgronje, distinguishedby profoundacquaintance with
the sources, a sense of historicity, penetrating criticism, and persistent log-
ical line of argument. In his doctoral study on the Meccan pilgrimage
51
he
convincingly demonstrates that all Koranic passages referring to Islam as
the religion of Abraham make sense only in the context of Muammads
Medinan period. When he experienced there the disappointment that the
People of the Book, whose religion he had considered his own, did not
want to recognize him, he started to look for an alternative that, on the one
hand, did not in principle contest his earlier Meccan revelations but, on the
other hand, could not as easily be challenged by the People of the Book as
were his statements regarding Moses and Jesus. So he clung to the religionof
Abraham, who, on account of his justice and faith, was equally respected by
Jews and Christians, but who both had only vague notions of his religion. In
another important article
52
he explains the genesis of the word zakt, which
was originally used with the meaning of charity and only later applied to
the new institution of communal tax. In the detailed review of Grimmes
Mohammed,
53
Snouck Hurjronje does not limit himself to pointing out the
errors in the book but also considers also other problems of the biography
of the Prophet.
Ignaz Goldzihers pioneering Muhammedanische Studien, as well as other [ii/210]
writings that treat the bases of the contemporary criticismof tradition, have
been acknowledged above in connection with research on the reliability of
the Arabic sources.
54
To the classic researchers belongs also the late Julius Wellhausen, who
most successfully combined ingenious intuition with all conceivable mer-
its in his research and descriptive power. His very first Arabist work, the
German version of al-Wqid,
55
displays in the preface a remarkable com-
petence regarding the sources, their European editors, and the ability to
present profitable remarks even on thorny subjects like chronology. In the
51
Het Mekkaansche feest (1880), pp. 2848. Cf. also above, pp. 106, 127, 135. [An English
translation is being prepared by the translator, and expected to be published in 2012.]
52
Nieuwe bijdragen tot de kennis van den Islam, pp. 357421.
53
Une nouvelle biographie de Mohammed see above, p. 377 n. 38.
54
Muhammedanische Studien, zweiter Teil (Halle, 1890), and its English translation, Mus-
limstudies, vol. 2 (Chicago, 1977.)
55
Muhammed in Medina (1882); cf. above, p. 323.
last chapters of his study on the remnants of Arab heathendom, Reste ara-
bischen Heidentums,
56
he supplies an illustrative survey of the cultural and
religious preconditions for the development of Islam, concluding with the
proof that the most decisive influences upon Muammad at Mecca came
from Christianity and not from Judaism. The fourth part (Heft) of Skizzen
und Vorarbeiten
57
consists of three consecutively paginated fundamental
treatises. In the first of theseMedina vor dem Islam [pre-Islamic Med-
ina]he disentangles the most complicated conditions of settlement and
the political conditions of the tribes around Medina. The other two parts
Muhammads Gemeideordnung von Medina [Muammads constitution of
Medina] and Ibn Sads Seine [Muammads] Schreiben, und die Gesandt-
schaftenanihn[Muammads correspondence andofficial letters to him]
supply evidence of the authenticity of these documents and put in proper
perspective the importance of the political and diplomatic means which
Muammad used to establish the Medinan theocracy. These pioneering
works might have been still more important if his distinct artistic and aes-
thetic inclination had not led himto supply presentations of his viewrather
than investigations.
Otto Pautz displays scrupulous accuracy but little spirit in his doctoral [ii/211]
study on Muammads dogma of revelation.
58
He is able neither to master
his material nor to work out problems. The work can be recommended only
as a collection of source material.
Herman Theodorus Obbinks Utrecht doctoral study on the Holy War [ii/212]
according tothe Koran
59
relates the manifoldphases of the notionof jihdby
connecting them with Muammads changing mood in his conversion and
martial policies, and finally notices that the establishment of the obligatory
jihd for proper salvation can be observed only in the latest of the Koranic
passages.
Hartwig Hirschfeld
60
is the first scholar after Gustav Weil
61
and Theodor
Nldeke
62
to study independently and thoroughly the construction, con-
tent, and style of the sras. His sagacity, however, generally degenerates
into excessive subtlety, losing track of the simple and obvious things. Unfor-
tunately, his work also displays such a glaring lack of historicity that my
56
1st ed. (1887), pp. 171212; 2nd ed. (1897), pp. 208242.
57
Berlin, 1889.
58
Muhammeds Lehre von der Offenbarung quellenmig untersucht (Leipzig, 1898).
59
De heilige oorlog volgens den Koran.
60
Newresearches into the composition and exegesis of the Qoran.
61
Historisch-kritische Einleitung in den Koran (1902).
62
Geschichte des Qorns (1860).
checking of the first and second parts nearly always led to the rejection of
his findings.
63
However, no future researcher can ignore such a serious book
that, even despite its misleading elements, still remains instructive.
Charles Torrey
64
collects the theological expressions of the Koran derived
from the commercial language and attempts to draw religious-historical
conclusions. But since this linguistic usageas the author himself admits
is evenly distributed throughout the Koran, it was impossible to get a lead
on either the development of Koranic ideas or the chronology of the sras.
The researches in the law of family, slavery and inheritance in the Koran,
which Robert Roberts
65
pursued, would have led to the same conclusions. In
any case, he lacked knowledge as well as ability.
The first two chapters of Arent Jan Wensincks Muammad and the Jews [ii/213]
of Medina
66
(1928) produced most valuable contributions to the topography
of ancient Yathrib and Muammads constitution of Medina, the traces
of which he follows with good fortune not only in the Sra, where Leone
Caetani
67
had already achieved preliminary results, but also in the adth
collections of al-Bukhr, Muslim, al-Nas, Ab Dd [al-Sijistn], and
al-Dmir. The two subsequent chapters discuss the Jewish influence upon
the Islamic cult and subject to drastic and fruitful criticism the reports on
Muammads policy against the Jews after the Battle of Badr.
Rudolf Leszynsky, in his history of the Jews in Arabia,
68
bases his study
on original sources, to which he makes abundant references. In spite of
this, the work is a popular account. He focuses mainly on the justifica-
tion of the derogatory references to Arabian Jewry in the sources as well
as in Occidental editions. In doing so, however, he transgresses the limits
of historical objectivity and becomes such a passionate panegyrist of his co-
religionists that most of his vindications are off the mark. This applies above
all to the attemptcontrary to Sprenger, but particularly to Wellhausen
to assign once again the first place in the development of Islam to Judaism
(pp. 3646), which is made so much easier for him by the fact that he is
totally ignorant of the history of Christianity. Among the many follies to
63
Hirschfeld, loc. cit., cf. part 1, pp. 10, 31, 73sq., 81, 96, 108, 126, 155; and part 2, p. 77sq.
64
The commercial-theological terms inthe Quran (Leiden, 1892), 51 p. (Dr. phil. Universitt
Straburg).
65
Das Familien-, Sklaven- und Erbrecht im Qorn (Dr. phil., Leipzig, 1907). The English
translation, The social laws of the Qorn, considered and compared with those of the Hebrew
and other ancient codes, was published in 1971.
66
Originally published in 1908 at Leiden entitled Muhammed en de Joden te Medina.
67
Annali dellislam, vol. 1, p. 376sq.
68
Die Juden in Arabien zur Zeit Mohammeds (1910).
which his partisanship led himlet us merely add that he considered a docu-
ment foundat a synagogue inold Cairo andgranting unbelievable privileges
to the Jews of Khaybar and Maqn to be genuine even though its falseness
was materially obvious (p. 104sq).
We are obliged to Henri Lammens, S.J., professor at the Pontificio Isti- [ii/214]
tuto Biblico di Roma, for several works distinguished by splendid pow-
ers of discernment and phenomenal acquaintance with the sources, albeit
not devoid of excessive scepticism, inconsistency, and religious partiality.
Muammads sincere belief in his divine mission, which Lammens declares
to be psychologically impossible,
69
is the unconditional precondition of his
lasting success and cannot be shaken by some of his moral lapses. In the
comprehensive monograph on Fima and Muammads
70
other daugh-
ters, Lammens, although he knows perfectly well how to present the pleas-
ant and sympathetic character traits of the respective persons as tenden-
tious flattery, blindly takes over from the sources everything ghastly and
unfavourable, making Fima and Al into true caricatures. Muammad is
presented to us as a pompous oriental prince, a heavy eater, and doting on
children. This is naturally as exaggerated as the traditions that emphasize
the wretched poverty of the Prophets household. As can be seen fromthese
examples, Lammens respective works must be used with great discretion,
although even in cases where they are provocative they are yet a mine of
knowledge and stimulation. Less reproachable are the results of the first
volume of his Le Berceau de l Islam, l Arabie la veille de l hgire; le climat,
les Bdouins, which deals with the natural and cultural presuppositions of
Islam, since this work concerns not persons but things and institutions. It is
also here that the author displays his genius in brilliant colours when creat-
ing a colourful and illustrative mosaic from thousands of notes.
Eduard Meyer,
71
in his study on the origin and history of the Mormons, [ii/215]
withexcurses onthe beginnings of IslamandChristianity, draws aninforma-
tive parallel between the appearance of Muammad and Joseph Smith, the
founder of the Mormons, and with his superior historical discernment fre-
quently comprehends many elements better thanthe specialists. Yet he goes
too far when, on the basis of Smiths visions, he attempts to throwmore light
on the meaning and course of the earliest revelations of Muammad. The
statements of the Koran are not unequivocal, and Muslim tradition, which
is based solely on its interpretation, is of no independent value. When the
69
Mahomet fut-il sincre? in Recherches de science religieuses (1911).
70
Fima et les filles de Mahomet.
71
Ursprung und Geschichte der Mormonen (1912).
author, just like August Mller, considers the Garden of the Refuge (sra
53:15) to be a locality near Mecca, not only is the Arabic text of sra 79:35
contrary to this, but so also is the total absence of a Meccan local tradition.
Ininsisting onthe allegedmeaning of toread of the Arabic qaraahe went a
bit too far, since, in spite of Wellhausen, to recite is the only correct trans-
lation. The other possible translation, to read aloud, generally also goes
too far, since the Heavenly Tablets [law maf] are always in the back-
ground of Muammads revelation, although it is nowhere explicitly said
in the Koran that he himself deciphered the writing of these tablets. It is
equally unclear which original the Prophet mustin sra 96:1, the starting-
point of the entire controversyrecite or read aloud from: the Heavenly
Tablets directly or fromhis own written notes. In the latter case, which 96:4
would seem to suggest, it would be completely impossible to attribute the
sra to the early period, let alone consider it the earliest revelation of all.
The Swedish scholar Tor Andrae
72
traces the perception of Muammads [ii/216]
personalityas reflected in his book on the teaching and belief of Muam-
mads communitydown to the final stages, culminating in the cult of the
Prophet. The author deserves credit for having dealt with a subject that was
only accidentally touched upon now and then in articles and books and
that, on the other hand, presupposes a considerable acquaintance with the
Arabic literature of several centuries. Here, I must limit myself to concen-
trating on those sections that are more closely connected with the biogra-
phy of the Prophet: this is only half of the first chapter (pp. 2663), which
gathers the legendary parts of the earliest biographieschildhood, Night
Journey to Jerusalem, Ascent to Heaven [mirj], the disappearance at the
end of the earthly courseand inquires into their origins, looking at them
under the aspect of the history of religion and comparative folklore. The
thoughtful introduction (pp. 125) represents a survey of the vacillations
and contradictions of Muammads self-confidence. First of all are the diffi-
culties created by his twofold conviction that he was the godlike harbinger
of the approaching Final Judgement and the vehicle of the Heavenly Book.
Equally successful is the proof that the Medinan period does not represent a
break with Muammads religious past but rather a real religious continua-
tion of his prophetic consciousness, particularly because of his military suc-
cesses, first of all the Battle of Badr, the glorious victory of which he clearly
interpreted as Gods blessing (sra 8). Additionally, Muammads activity
as a writerwhich, by the way, had already started at Meccaand as a
72
Tor Andrae, Die Person Muhammeds in Lehre und Glauben seiner Gemeinde (1918).
law-giver, so characteristic for his Medinan period, are mentioned in con-
nection with his belief in revealed religion in a very appealing psycholog-
ical analysis. We must assume that the inspiration that initially appeared
isolated in the Prophets mental life as a super-human influence without
conscious connections in time gradually established a liaison with the nor-
mal consciousness which, to a certain measure, eventually exerted psychic
control (p. 19). This brief introduction displays such a sense of history, and
such fine religious-psychological perception, that it would be a shame if the
sketch did not lead one day to a comprehensive treatise.
The Exegesis of the Koran [ii/217]
Up until the present day, commentaries on the Koran in the proper sense
have not been written by Western scholars.
73
The exegetic results of such
scholars are embedded either in the biographies of the Prophet or in a great
variety of individual studies, as well as in the translations of the Koran,
which, as a rule, are accompanied by a larger or lesser number of foot-notes.
To the most respected works of this kind belongs the very first one, the
Refutatio Alcorani of the Italian member of a religious order, Ludovico Mar-
racci,
74
whose proficiency deserves particular praise as all of his Arabic was
self-taught. The first part of the work, entitled Prodromus, contains a biog-
raphy of the Prophet, an introduction to the Koran, and Refutationes made
up of many chapters. His deserving successor is the English translation of
the Koran by George Sale,
75
which, apart from its 143-page preliminary dis-
course in eight sections, is provided with numerous foot-notes from the
best Arabic sources available at the time. The progress of scholarship after
Marracci can best be seen in the absence of refutationes and that both Sales
preliminary discourse and the foot-notes are based on a larger historical
knowledge and more impartial judgement. Both works are the mine that
suppliedandstill supplies later writers withtheir wisdom. Until quite recent
times, Sales translation saw numerous new editions in England. Already
in 1746 it was translated into German by Theodor Arnold and stimulated
73
(August Fischer in Nachtrge: It is quite likely that Schwally here purposely omitted
E.M. Werrys mediocre Comprehensive commentary on the Qurn, comprising Sales transla-
tion and preliminary discourse, with additional notes and emendations, together with a com-
plete index. London, Trbners Oriental Series, 18821886. 4 v. viii, 391, 407, 414, 340 p. New
ed., London, 1896. 4 v.)
74
Patavii, 1698.
75
London, 1734, and often reprinted until recent times in a variety of sizes and types.
the appearance of similar works, of which the Koran of professor Samuel
F.G. Wahl
76
of Halle an der Saale is probably the most deserving achieve-
ment. The dependence of later German works on George Sale did not wane,
although the explanations deteriorated successively and the introductions
disappeared completely.
Ludwig Ullmanns literal translation was first published in 1840 and is the [ii/218]
most widespread version in Germany; it can claim no other merit than that
it lists in its notes AbrahamGeigers 1833 study on Muammads borrowings
from Judaism.
77
Some years later, the publisher commissioned the most
competent man of the day, the previously mentioned Gustav Weil,
78
to write
the missing introduction to the book, who for this purpose merely had to
re-write the final chapter of his own biography of Muammad.
79
The posthumous work of the Orientalist and poet Friedrich Rckert
80
[ii/219]
is quite an independent and elegant achievement, although his scholarly
basessince the editor, August Mller, did not bring them up-to-date
are somewhat dated, as they reflect the state of the art in the first decades
of the nineteenth century. Rckerts philological interpretation of the texts
is still under the spell of tradition, a basic mistake from which all the later
translations down to the present day suffer. On top of this flaw, furthermore,
not even the complete Koran was considered. The style leaves a mistaken
impression of the original insofar as the long-winded and heavy sentences
of the later sras are subjected to an arrangement that is foreign to the
original.
81
In spite of the great advances in Koranic research since George
76
His Der Koran oder das Gesetzt der Moslemen is a new translation based on Friedrich
E. Boysens 1773 translation, with explanations, a historical introduction, and a complete
index.
77
See above, p. 380 n. 47.
78
Historisch-kritische Einleitung in den Koran.
79
August Fischer, Nachtrge undBerichtigungen, Seite ii/224: The lost page of Schwallys
manuscript (cf. Vorwort, p. iv) is likely to have contained also the following three translations
of the Koran: Le Koran, traduction nouvelle par faite sur le texte arabe de Kazimirski (1840),
often reprinted with corrections; The Koran, translated by J.M. Rodwell (1876); The Qurn,
translated by E.H. Palmer (1880).
80
Der Koran (1888).
81
I amhere in complete disagreement to August Mller who describes Rckerts method
of translation as follows: In any case, the rhymed prose used throughout the translation is
a stroke of genius of the poet. The speech thus becomes endowed with that dignified tenor
to which the original owes its arrangement, and much better than a strictly prose translation
approaches the impression which the Koran must create when solemnly recited in religious
usage. This verdict is valid with reference to the translation of the oldest of the sras, which
already in the original consist of short verses and more poetical buoyancy. It is known that
as a translator Rckert was not always lucky. For example, he faithfully imitated metre and
Sale, there is to this day neither a translation nor interpretation in keeping
with that state of science, as the best experts have shunned this duty, either
because they were not interested in the easy parts or because the difficult
aspects seemed to be insurmountable, provided, of course, that they were
not simply discouraged a priori by the awful dreariness of large stretches of
the Holy Book.
rhyme of the poems of the amsa but at the same time he completely destroyed the power
of the originals.
THE HISTORY OF THE TEXT OF THE KORAN
1
The Consonantal Errors of the Uthmnic Text [iii/1]
From an early time Muslims themselves have realized that the text of the
Koran edited at the commission of the Caliph Uthmn was not absolutely
perfect. We possess a number of traditions that accuse it of outright errors,
2
of which the best known
3
says that Uthmn himself found incorrect expres-
sions whenlooking at a copy of the final recension(
) and
allegedly said
4

5

6

1
Under the aspect of the Koranic exegesis Goldziher studied the history of the text of the
Koran in his Schools of Koranic commentators (2006), pp. 135, and 171182.
2
Already Ab Ubayd al-Qsim b. Sallm (d. 223 or 224/838 [EI
2
; Sezgin, GAS, s.v.; Berlin
Ms., Ahlwardt, no. 451, p. 7, l 7]) in the introduction to the section dealing with such tradi-
tions of his K. al-Fail al-Qurn, deals with the collection of the Koran (fol. 35
r
, sqq.) [cf.
Das K. al-arb al-muannaf von Ab Ubaid und seine Bedeutung fr die nationalarabische
Lexikographie, by Raman Abd al-Tawwb (Mnchen, 1962)]; further al-Dn (d. 444/1052)
in Muqni (see below, p. 406) (bb 21; cf. A.I. Silvestre de Sacy, Notice du manuscrit no. 229
de la Bibliothque impriale , p. 301 sqq.); the author of K. al-Mabn li-nam al-man
(fourth paragraph of the introduction); and al-Suy in al-Iqtn (naw, 41, tanbh, 3); finally
(derived from al-Itqn), Mevzuat ul-ulm (enlarged Turkish translation of Takprlzade,
Mift al-sada (Brockelmann, GAL, vol. 2, p. 426) of his son, Kemaleddin Mehmed, Con-
stantinople (1313) vol. 2, p. 68sqq.). Among more specific sources al-Suy cites, apart from
Ab Ubayd al-Qsim b. Sallm, the K. a-Maif of AB BAKR Muammad b. Abd Allh
b. Ashtah AL-IBAHN (d. 360/970); Flgel, Die grammatischen Schulen der Araber, p. 229,
and on the nisba, al-Suy, al-Bughya, p. 59; Ibn al-Jazar, K. al-Nashr f l-qirt al-ashr, Ms.
[Ahlwardt, Verzeichnis,] no. 657, fol. 17
v
.; the author seems to be identical with [AB BAKR]
Muammad b. Abd Allh AL-IBAHN of Muqni, bb 18 (cf. also bb 2, fal 4); but first
of all the K. al-Radd al man khlafa muaf Uthmn of Ab Bakr Muammad b. al-Qsim
AL-ANBR (d. 327 or 8/938 or 9; Brockelmann, GAL, vol. 1, pp. 119; [EI
2
p. 147, no. 6]) which has also been used in al-Mabn li-nam
al-man (cf. I. Goldziher, Schools of Koranic commentators, p. 25sqq.).
3
Ab Ubayd and from there al-Itqn; Muqni; Mevzut;the first part slightly different
also al-Mabn f namal-man; Nashr (Berlin Ms., Ahlwardt, Verzeichnis, no. 657, fol. 173
v
.);
al-Muttaq al-Hind, Kanz al-umml, vol. 1, no. 4793sq.; Ibn Khallikn, no. 516; etc.the
second part, Kanz al-umml, vol. 1, no. 4795 (Kanz al-umml, both from Ibn Ab Dd
al-Sijistn, <see p. 22, l 4> and Ibn al-Anbr.)
4
Thus Ab Ubayd al-Qsim [Ibn Sallm], in other cases .
5
Variant (mentioned in the tradition itself)

, other than these two, there are differ-
ent misrepresentations.
6
al-Dn, Muqni omits the and has .
7
al-Dn, Muqni f marifat rasammaif al-amr, omits the and has .
390 the history of the text of the koran
, Do not change them because the Arabs will get
it straight with their tongues; if the writer had been from the Thaqf
8
and
the dictating person fromthe Hudhayl these forms (expressions) would not
occur there. Asecond tradition
9
has isha say the following with reference
to the three [sic] passages, sra 2:172 (for ); sra
4:160

(for

and
respectively, andalways
accusative); sra 5:73
(for ); and sra 20:66
(for ), say
10

: This is the work of

the writers; they made errors when writing.
11
Here it is linguistic consider-
ation that raised objections to the text, while in other cases it is conceptual
consideration. Thus, some considernot without likelihood
sra
24:27
12
and (= ) sra 13:30
13
as a spelling mistake for
and or
of sra 17:24 to result from running ink in .
14
It is quite bold when
, sra 24:35, because of the reservation that Allh is too great
for His light to be compared with that of a lamp, is simply termed an ortho-
graphic error for .
15
All these traditions are already based on the notion that the Uthmnic [iii/3]
recension remains an irrevocable fact, even if it might be wrong. In the
8
Alsothe Thaqfite Umayya b. Ab l-alt praises the proficiency inwriting of his clansmen
(fragment, no. 1, Schulthe.) [Sezgin, GAS, vol. 2, pp. 298300.]
9
al-abar, juz 6, p. 16, l 10; Ab Ubayd al-Qsim and from there al-Itqn, al-Muqni,
Mevzuat, and, somewhat shorter, al-Mabn li-namal-man. Another tradition, according
to which Sad b. Jubayr [EI
2
; EQ; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 1, p. 28]
considers at least

of sra 4:160 to be lan in al-Itqn and Mevzuat. On

of sra
4:160 one tradition in Ab Ubayd al-Qsim, al-abar, vol. 6, p. 16 l 7, and al-Mabn li-nam
al-man offer the explanation that the writer, after having written the preceding text, asked

, and received the reply

in the accusative as still being dependent on (in
which case the is hardly compatible) and took it down as it is.
10
Ab Ubayd al-Qsim .
11
Also, sra 23:62, isha allegedly considered orthographic errors (
)
for (al-Suy, al-Itqn).
12
Ibn Abbs (and Sad b. Jubayr) in al-abar, juz 18, p. 77, l 30; al-Zamakhshar, s.v.,
al-Suy, al-Itqn. Cf. I. Goldziher, Schools of Koranic commentators, p. 20.
13
Ibn Abbs in al-abar, juz 13, p. 91, l 7; Al, Ibn Abbs, and several other Companions
in al-Zamakhshar, s.v.; al-Suy, al-Itqn; cf. p. 49.
14
al-abar, juz 15, p. 44, l 23; al-Itqn; cf. I. Goldziher, Schools of Koranic commentators,
p. 20. Graphically this is quite possible; the differs indeed from a often only by separation
because in early Kfic writings it is written on the line.
15
al-Suy, al-Itqn. Additional passages where occasionally the possibility of a writing
error is hinted at are sras 63:10 (al-Mabn li-namal-man, and see below, p. 391 n. 22, and
352353, and sra 81:24, (for , in the tradition of imal-Jadar [d. 128/745; Sezgin,
GAS, vol. 1, p. 5, vol. 9, pp. 38 and 43] mentioned below, p. 391), furthermore sra 2:131 (
for , al-Mabn); sra 3:75 (Mujhid for , al-abar, juz 3, p. 216, l 25)
sra 21:49 (Ibn Abbs without , al-Itqn).
the history of the text of the koran 391
first of the traditions this attitude is clearly stated. At the same time, they
reflect an obvious apologetic tendency: the men responsible for the text
of the Koran, Uthmn and his Commission, but particularly the Prophet
himself, are defended against the reproach of linguistic and substantial
shortcomings by attributing these defects to the scribes. This apologetic
approach is indeed still so nave, and proceeds from such a simple, human
conception of the production of the authorized version, that we must date
the origin of this tradition to a very early period.
The more liberal, and therefore still earlier, attitude is that such errors [iii/4]
were not toleratedand were simply changed, as happenedwiththe passages
referred to as well as in hundreds of similar instances, of which at least
a few will be mentioned below. The changes that have come down to us
demonstrate that there were far more complaints, linguistically as well as
substantially, than is pointed out in the respective traditions.
If changes were out of the question, and if the objectionable text was
there to stay, there remained nothing but a compromise, namely to read
differently from what was written. This is evidently also the opinion of
the authors of the above-mentioned transmissions.
16
This opinion becomes
even clearer in a tradition from Al,
17
[when describing the joys of Paradise
in sra 56:28.
18
] im al-Jadar (d. 128/745
19
) is named the conscious repre-
sentative of this practice. It is perpetuated down into the canonical systems
of reading.
20
The practice has its counterpart on less difficult grounds in
the attitude of the readers when dealing with the peculiarities of orthogra-
phy. The well-known traditionist Ibrhm b. Yazd AL-NAKHA
21
(d. 96/715)
allegedly justified this method with specific reference to the peculiarities
of orthography: Instead of an was written in , sra 20:66, and a
in both of sra 5:73 and of sra 4:160.
22
Most of the later
16
Cf. first of all the in the tradition of Uthmn; still clearer in a considerably
qualifying variant, which expunges the expression (Itqn from [AB BAKR] Ibn Ashtah
[AL-IBAHN])

(a variant appearing between this form
and the common form, al-Muttaq, Kanz al-umml, vol. 1, no. 4792 fromIbn Ab Dwd [see
below, p. 404 n. 109], and Ibn al-Anbr).
17
al-abar, juz 27, p. 93 l 16; al-Zamakhshar on sra 56:28, according to which Al
(andaccording to al-Zamakhsharsimilarly Ibn Abbs) in this passage explicitly reject
the change from to , although he reads the latter; cf. I. Goldziher, Schools of Koranic
commentators, p. 24.
18
I. Goldziher, Schools of Koranic commentators, p. 24.
19
al-Mabn li-namal-man; Sezgin, GAS, vol. 1, p. 5, last line.
20
Above all, it is Ab Amr b. al-Al (Bara) [ca. 90/709180/796; Sezgin, GAS, vol. 1, p. 10,
no. 3; vol. 8, pp. 5051; vol. 9, pp. 4042] who reads in sra 20:66. More later on, p. 414.
21
EI
2
; EQ; Juynboll, Encyclopedia, p. 4sqq; Sezgin, GAS, vol. 1, pp. 403404.
22
al-Suy, al-Itqn.
generations found it impossible to reconcile this loophole with their vener-
ation of the Holy Book. They not only adhere in writing and pronunciation
to the Uthmnic recension,
23
but also attempt with increasing conviction to
document an increasing number of possibilities that will reconcile the text
with the requirements of language and meaning.
24
After this interpretation of the passages became generally accepted, the
old transmissions about errors in the text became most inconvenient. An
attempt had to be made to cope with such passages by means of jar (which
was difficult in the case of the tradition from isha)
25
and either give them
a new interpretation, or simply to reject them as untrustworthy. We can
give the approximate date of the beginning of these attempts: whereas Ab
Ubayd al-Qsim b. Sallm
26
(d. 223 or 4/838) simply lists the transmissions,
Ibn al-Anbr (d. 327 or 8/939) as well as al-abar (d. 310/923) attempt to
save the Uthmanic text without exception.
Variants of the Uthmnic Copies [iii/6]
As regards the fate of the four official copies of the Koran produced on
the order of the Caliph Uthmn, next to no reliable information is avail-
able,
27
and in the science of the Koran they are of almost no impor-
23
An authority was found very conveniently in the person of Zayd (Ibn Thbit) who is
made to say (al-Mabn):
(sra 20:66),

(sra 63:10).
24
Here belongs also the attempt to rescue
of sra 20:66 by the vocalization

because evidently in reality an = inna occurs only directly before and similar verbs
(also
) but not before nouns (cf. Reckendorf, Die syntaktischen Verhltnisse, p. 129;
Sibawayhs Koranic references in136citedfor the opposite must be interpretedina different
way; cf. Bergstrers particles of negation and question, Verneinungs- und Fragepartikeln,
p. 14sqq.). The old scholars of the Koran will have objected to the text not without good
reason.
25
According to al-Suy, al-Itqn (and Mevzuat) the isnd conforms with the require-
ments of al-Bukhr and Muslim.
26
EI
2
27
The earliest source for them seems to be the apology of al-Kind [GAS I, 612](used
in the London editions of 1880 and 1885: Rislat Abd Allh b. Isml al-Hshim (!) il Abd
al-Mas b. Isq (!) al-Kind yadh bi-h il l-Islm, wa-Rislat Abd al-Mas il l-Hshim
yaruddbi-halayhwa-yadhill-Narniyya) if it was indeedwrittenin204205/820821, as
P. Casanova, Mohammedet lafindumonde, 2me fasc., I (1913), p. 112, tries to prove. According
to al-Kind, p. 80, and Casanova, p. 121, respectively, the Damascus copy was at that time still
at Malaya (without saying how it got there) whereas the Meccan copy burnt during the
rebellion of Ab l-Sary, [EQ,] the Medinan copy was lost during the conquest of Medina
under YazdI in683, andthe Kfancopycontrary tothe assertionthat it was still extantin
the rebellion of Mukhtr [EI
2
]. (Thus, al-Kind, contrary to common tradition, presupposes a
different distribution of the four copies.)
tance,
28
with the sole exception of the Medinan copy, which is mostly re-
ferred to as Imm
29
muaf Uthmn. In spite of this, it is this particu-
larly copy that is shrouded in darkness. One could try somewhat to rec-
oncile the contradicting reports by assumingwith a certain degree of
probabilitythat Uthmns copy, which because of the mention in the
history of his murder attained such fame, was different from the autho-
rized Medinan copy.
30
Yet this attempt also leads nowhere. According to
Ibn Qutayba (d. 276/889),
31
it was passed on in the family, first to his son
Khlidto whom a report seems to refer
32
and later to other members.
IM b. al-Ajjj AL-JADAR (d. 128/745)
33
supplies a number of details.
34
Mlik b. Anas (d. 179/795) considered it lost (
.
35
) According to the apol-
ogy of al-Kind (see above, p. 392 n. 27) it did not disappear in the flames
until the rebellion of Ab l-Sary in 200/815. And, finally, Ab Ubayd al-
Qsim b. Sallm (d. 223 or 4/838)
36
has it fished out min ba khazin al-
umar, and finds as a sign of authenticity still traces of Uthmns blood,
37
a
28
Once (al-Dn, al-Muqni, cap. 21, beginning) the Damascene copyalso known as the
Immis mentioned on the authority of HRN b. Ms AL-AKHFASH al-Dimashq (d.
ca. 292/904, Sezgin, Geschichte, vol. 9, p. 200), as regards the text of the others we lack any
explicit and direct reference. The frequent expression, f muaf ahl al-Kfa, etc., means no
more than in the Kfan consonantal text.
29
Following sra 36:11: . (Less frequently all the Uthmnic copies
are referred to under this title.)
30
When later writers occasionally try to differentiate between Uthmns private (kh)
copy and the Medinan (mm) copy, this is based on mere misunderstanding.
31
K. al-Marif, p. 101 [EI
2
; Sezgin, GAS, vols. 7, 8, 9, s.v.,] and then Shams al-Dn AL-
SAKHW [d. 902/1497; Sezgin, GAS, vol. 8, p. 117, last line] in the commentary on al-Aqla
(see below, p. 406sq.); cf. P. Casanova, p. 130.
32
Amad b. Muammad AL DIMY AL-BANN (d. 1117/1705), Itf fual al-bashar
(see below, p. 407) on sra 63, with reference to the traces of blood still visible. EI
2
; Sezgin,
GAS, vol. 1, p. 592, no. 2.
33
Sezgin, GAS, vol. 1, p. 5, last line, etc.
34
al-Dn, al-Muqni, bb 5, fal 1, bb 19 and 21. Ab Ubayd al-Qsimb. Sallmlists them
as the authority on the Imm in the first of these passages as well as in his Fail al-Qurn,
fol. 40
r
.
35
Silvestre de Sacy Commentaire sur le pome nomm Rayya intitul Akila, p. 344.
36
Sezgin, GAS, vol. 3, p. 367; fromamong other sources in particular al-Nawaw, [Tahdhb
al-asm] The Biographical dictionary of illustrious men, edited by F. Wstenfeld, p. 744; Yqt,
Irshd al-arb, vol. 6, p. 162.
37
al-Muqni, bb 2, fal 1. AbUbaydal-Qsiminhis Fail al-Qurn, fol. 40
r
, onthe other
hand, refers only to (read:
) .
Thaghr is likely to refer to Tarsus where he spent eighteen years as a q. Furthermore, it
is noteworthy that occasionally (see below, p. 396 n. 52) the identical statement regarding
the Imm is traced back from al-Dns al-Muqni to the personal inspection of Ab Ubayd
al-Qsimb. Sallmwhohimself however has it fromimal-Jadar. All this makes any trans-
mission suspicious that states that Ab Ubayd al-Qsim himself has consulted the Imm.
recurring theme in the innumerable later reports of alleged Uthmnic
copies.
38
The tradition of the text of this copy is therefore extremely weak,
since Ab Ubayd al-Qsim is the main source in both, al-Muqni
39
and also
al-Zamakhshar.
40
This is not to say that the particulars of the tradition about the text of this [iii/8]
copy of the Koran and the three others produced on Uthmns initiative are
necessarily unreliable. The actual differences of the text of the Koran of the
four townswhence originate the variants of the Koran and to which alone
they naturally refer
41
permit safe clues regarding the original copies since
the respective local variants have been preserved with meticulous accuracy.
For everything beyond strict orthographic matters, the written transmission
of the text found solid support in the oral transmission, since the readers
of the Koran of the respective towns followed their authorized version,
and wherever this was occasionally not the case at least the awareness of
the deviation remained alive.
42
This explains why the transmission of these
38
Regarding old copies of the Koran in literature and libraries that are claimed to be the
one of the Caliph Uthmn see now P. Casanova, Mohammed et la fin du monde, pp. 129139
(sixteennumbers) as a continuationof Quatremre inJournal asiatique, Sur le got des livres
chez les Orientaux, p. 41 ff., and his Mlanges d histoire et de philologie orientale, p. 41 sqq.;
some additions in Goldziher, Schools of Koranic commentators, p. 173; A. Mez, in his Renais-
sance of Islam, p. 138, points out a reference to another such copy in 369/979 at Baghdad.
The so-called Samarqand Koran of which Serge N. Pisarev published a reproduction in 1905,
I intend to treat later at the appropriate place; for the time being cf. A.F. Shebunins article
. . . To use the near certain spuri-
ousness of all these copies as an argument against the historicity of the Uthmnic recension
as done by Casanova is of course totally out of the question.Safer still is the forgery in the
case of the alleged Korans from the hand of Al; cf. Quatremre, p. 47sqq.; N.V. Khanykov in
Mlanges de l Acadmie de St. Ptersbourg, t. 3, p. 63sqq. [not identified], a facsimile of the
sub-title of the manuscript here discussed in Berlin as Ms. 366, Ahlwardt, Verzeichnis (= cod.
Mss. Or., folio no. 532); but, most of all, I. Goldziher, Schools of Koranic commentators, p. 173sq.
39
Onceapart from the above-mentioned im al-Jadaralso AB TIM (Sahl b.
Muammad) AL-SIJISTN (Sezgin, GAS, vol. 3, pp. 367368, d. 250/864), namely at the end
of chapter 18.
40
See on sras 29:27, 33:10, 38:2. Without naming his authority, al-Zamakhshar supplies
a detail on sra 5:59 about the reading of imm, which al-Muqni, bb 21, traces back to Ab
Ubayd al-Qsim [Ibn Sallm].
41
Cf. the titles Ikhtilf maif ahl (Fihrist, p. 36, l 9, al-Kis, see below), and Ikhtilf
ahl f l-maif (ibid., l 10, Yay b. Ziyd AL-FARR d. 207/822 [Sezgin, GAS, vol. 9,
pp. 131134]). Ab Ubayd al-Qsim [Ibn Sallm] introduces his list of variants inna ahl
al-ijz wa-ahl al-Irq ikhtalafat maifuhum f dhih al-urf; only al-Mabn li-nam
al-man replace this by ikhtalafa muaf (!) ahl al-Madna wa-l-Irq , and close the
first list hkadh hijuh f l-imm muaf ahl al-Madna, whereas al-Muqni still gives the
individual variants with f maif ahl al-Madna, etc.
42
Cf. al-Dn, al-Muqni, bb 21, where Ab Amr (al-Mzin) in the case of sra 43:68
follows the Medinan variant with particular reference to the Medinan Korans.
variants
43
is subject to almost no variations; it can be considered entirely
sound. The written formwas likely established in the second century, possi-
bly first by al-Kis (d. 189/804).
44
Two ancient lists of these variants have come downto us. One is fromAb [iii/9]
Ubayd (al-Qsim b. Sallm), entitled Fail al-Qurn (Ahlwardt, Verzeich-
nis, Ms. 451, fol. 44
r
sqq.), in which he gathers the various sources. On the
authority of Isml b. Jafar AL-MADN
45
he enumerates twelve differ-
ences between Medina and Iraq, then on the authority of the Damascene
reader, Ab l-Dard (died toward the end of the caliphate of Uthmn)
as well as (Abd Allh) IBN MIR (Ibn Yazd al-Yaub, 21/641118/736)
twenty-eight peculiarities of Damascus (i.e., compared with Iraq), and final-
ly, without supplying sources, five peculiarities of Kfa vis--vis Bara. Mecca
is ignored. This list is enlarged in the first and third row by a single con-
troversial difference each. In the second row, however, the list is distorted
by the addition of several simple differences of variant readings,
46
and is
43
In what follows, variants always refers to the different orthography (of the manu-
scripts of the Koran) whereas variant readings refers to the oral recitation (qirt).
44
Fihrist, p. 36, l 9, considers only the manuscripts of Medina, Kfa, and Bara (cf. below,
p. 404 n. 107). It is unlikely that already (Abd Allh) IBN MIR (Ibn Yazd al-Yaub,
d. 118/736) [Sezgin, GAS, vol. 1, pp. 67] wrote on the differences of all the copies (ibid., l
12). As far as I know no such work is attributed to him in his biography. The reference in the
Fihrist is either a misunderstanding or it is falsely attributed to al-Kis. [EI
2
; Sezgin, GAS,
vol. 9, pp. 127131.]
45
Probably identical with al-Madin of al-Fihrist, p. 36, l 12, where a book, Ikhtilf
al-maif wa-jam (?) al-qirt, is ascribed to him.
46
Cf. the words of introduction qara Abd Allh b. mir al-Yaub This concerns the
following reports of the differences between Damascus and Iraq respectively: sras 6:52, and
18:27 against (in reality throughout , cf. al-Muqni, bb 18, which is only dif-
ferently vocalized by Abd Allh IBN MIR bi-l-ghudwa, and by the other Seven Readers,
bi-l-ghadwa); sras 24:31, 43:48, and 55:31, against (in reality throughout which Abd
Allh IBN MIR vocalizes ayyuh against ayyah of the others <see below, p. 409>); sra 27:69
(also Ab Ubayd al-Qsim b. Sallm) against this is evidently intended by the vague
references in Ab Ubayd and al-Mabn(in reality
[but without any diacritical marks

of the two middle letters], cf. al-Muqni, bb 9, which both Abd Allh IBN MIR and al-
Kis (Kfa) recite innan, the others however ainn, which might equally be expressed
;) sra 43:18 against (text [but without diatrical mark on the mid-
dle letter], which Abd Allh IBN MIR as well as NFI (b. Abd al-Ramn b. Ab Nuaym
al-Layth (Medina) [d. 169/785; EI
2
; Sezgin, GAS, vol. 1, pp. 910], and Ibn Kathr (Mecca),
45/665120/738, read against
of the others); sra 43:37 against (in reality ,

cf. Muqni, bb 2, fal 7, which might be read as dualAbd Allh IBN MIR as well as again
Nfi, and Abd Allh IBN KATHR, further Ab Bakr an im (Kfa)or as singular). Also
the differenceincluded by Nldeke in his list of variants on the authority of al-Mabn li-
nam al-man
Damascus against
the others (i.e., Iraq) sra 2:261, of which in

addition to that (except in Ibn Masd, see below) nothing at all is known, belongs here: The
reproduced in a slightly rearranged form in al-Mabn li-nam al-man
(fifth paragraph of the introduction). The other list is furnished by al-Dn,
al-Muqni (bb 21); it enumerated all the differences arranged according to
the position in the Koran, considering also Mecca.
47
That both lists are sub-
stantially in agreement increases the reliability of their content, although a
textual interdependence cannot be ascertained.
48
The following list of variants is basically that of al-Dn, al-Muqni f [iii/11]
marifat rasm maif al-amr, with recourse to those of Ab Ubayd al-
Qsim b. Sallm and al-Mabn li-nam al-man;
49
accordingly also the
Meccan variants are listed; they are likely also to go back to an old copy.
Sra 2:110, , Damascus; , the rest.
Sra 2:126, , Medina,
50
Damascus,
51
(and the Imm
52
); , the rest.
53
Sra 3:127, Medina, Damascus (and the Imm
54
), , the rest. [iii/12]
Sra 3:181, , Damascus; , the rest (controversial whether Damascus
also for ).
55
of the text both amza and al-Kis (Kfa) consider a pausal form and consequently
read in context
. Whereas the rest of them retain the h also in context. Ab Ubayd al-
Qsimb. Sallm, Fail al-Qurn, Ms. Berlin no. 451, fol. 37
r
; and after himal-abar, pp. 3, 24,
23sqq., report that the h originates from Ubayy, who was consultedthis from the identical
source also al-Itqn, naw 41, tanbh 3according to yet another version added on the order
of Uthmn. As canbe seenfromthe other passages mentioned inthe same context, this does
not at all purport to indicate a variant without h, rather it is simply an explicit observation
that the peculiar form with h is really part of the text). Further cf. below, p. 400 notes 7880,
and p. 401 sq.
47
As source might have served the above-mentioned AB TIM (Sahl b. Muammad)
AL-SIJISTNI, above, p. 394 n. 39, to which reference is made in al-Muqni, bb 13, about a
remark on Meccan manuscripts of the Koran.
48
But see below, p. 400 n. 80, on sra 43:68.
49
As a safety check the following odd variants have been taken into consideration:
al-Zamakhshar as well as the rather numerous references in Kashf an wujh al-qirt of
Makk b. Ab lib al-Qays (Ahlwardt, Verzeichnis, no. 578) further the Aqla and the Itf,
and finally the works on variant readings since generally the qira are indicative of the text
of the home region of the respective reader.
50
al-Zamakhshar vague, ijz.
51
Wanting in Ab Ubayd al-Qsim b. Sallm.
52
The fact that frequently the text of the Imm is explicitly identified parallel to that of
Medina shows again that the details regarding the variants do not refer to the original copies
but to the text of the Koran as common in the respective towns (see the previous page.) After
what has been said on p. 344sq., the references to the Imm cannot claim strictly historical
importance.
53
Both forms are common to the Koran.
54
So al-Dimy al-Bann, Itf fual al-bashar f qirt al-arbaat ashar.
55
al-Mabn, and al-Zamakhshar only the first one; Ab Ubayd al-Qsim, both; al-Dn,
Sra 5:58, , Medina, Mecca, Damascus; , the rest.
Sra 5:59, , Medina, Damascus (and the Imm); , the rest.
Sra 6:32,
, Damascus;
, the rest.
56
Sra 6:63, (i.e ),
57
Kfa; , the rest.
Sra 6:138, Damascus (i.e.
); the rest (i.e.

).
58
Sra 7:2, ,
59
Damascus; , the rest.
Sra 7:137, (i.e., )
60
Damascus; (i.e., ), the rest.
[iii/13]
Sra 9:101, , Mecca; , the rest.
61
Sra 9:108, , Medina, Damascus; , the rest.
Sra 18:34, , Medina, Mecca, Damascus; , the rest.
Sra 18:94, , Mecca; (i.e.,
), the rest.
al-Muqni, opts for bi- in both passages; Aqla, vol. 5, p. 62sq. is undecided on the question.
According to al-Dimy al-Bann, Itf, some of the Damascus manuscripts have bi- also
in the second case. As regards the variant reading of the Damascene reader, Abd Allh IBN
MIR, tradition varies.As in the word order the bi-, occurring in both of
the first parts, could also very easily enter the third partwhereas the erroneously omission
of a bi- existing inthe third place would be difficult to explainmore probable is the rejected
reference in al-Muqni which in addition is supported by HRN b. Ms AL-AKHFASHs
reference to the Damascene imm (in al-Muqni).
56
Apart from after la-, the Koran knows the attributive connection al-dr al-khira;
accordingly also , sras 12:109, and 16:32 must be viewed as a short form of (see
below, p. 425sq.). Inour passage we are thus really dealing witha purely orthographic variant.
57
When the sources occasionally speak directly of forms with this is nothing but an
inaccurate expression.
58
The Damascene variant is linguistically hardly possible.
59
So Ab Ubayd al-Qsim b. Sallm (bi-tayn); al-Muqni: bi-l-y wa-l-t, and likewise
al-Mabn.
60
When the sources occasionally speak directly of forms with this is nothing but an
inaccurate expression.
61
The Meccan variant is reminiscent of several parallel passages; the less unusual reading
without min is preferred. In other passages, too, the Meccan variant puts the more common
form over the more unusual of the other manuscripts.
Sra 23:89 and 91, , Bara; , the rest (and the Imm.
62
)
63
Sra 25:27, , Mecca; the rest.
Sra 27:21, , Mecca; (i.e., ), the rest.
64
Sra 36:35, , Kfa; , the rest.
65
Sra 39:64, , Damascus; (i.e.,
), the rest.
66
[iii/14]
Sra 40:27, , Kfa; , the rest.
67
Sra 43:71, , Medina, Damascus
68
(and the Imm); , the rest.
Sra 46:14, (i.e.,
69
); Kfa, the rest.
Sra 47:20, , Mecca (also ), ; the rest (also )
70
(disputed).
Sra 57:10, , Damascus;
, the rest.
71
62
So already im al-Jadar in Ab Ubayd al-Qsim b. Sallm, Fail al-Qurn, Ahl-
wardt, Verzeichnis 451, fol. 40
r
(cited also in al-Muqni, bb 21), and also according to Muqni,
bb 2, fal 1; bb 21 Ab Ubayd himself after personal inspection; the opposite version of Itf
fual al-bashar f qirt al-arbaat ashar is an error.
63
Verse 87 unanimously . Ab Ubayd al-Qsim, and al-Mabn mention erroneously
the passage only among the Damascene peculiarities (with three times ).The Baran text
allegedly is aninnovationof Nar b. imal-Layth [d. 89/708or 90; EI
2
; EQ; Sezgin, GAS, vol. 9,
pp. 3233] (so also Ab Ubayd, loc. cit.) or on the order of the governor of Iraq, Ubayd Allh
b. Ziyd (Muqni, bb 21, who takes exception to these traditions). Ab Ubayd, loc. cit., states
that also Ubayy allegedly had everywhere.
64
Kashf an wujh al-qirt, s.v. describes the second form as the writing of the muaf
without indicating a variant.
65
Wanting in Ab Ubayd al-Qsim b. Sallm, and in al-Mabn li-namal-man.
66
67
So also Kashf anwujhal-qirt, s.v.; according to Ab Ubayd al-Qsim, and al-Mabn
also Bara, aw; al-Zamakhshar for wa unspecific only ijz.
68
69
Mostly, precisely this is vaguely mentioned.
70
So al-Kis (inal-Muqni) and al-Zamakhshar; according to Khalaf b. Hishmal-Bazzz
[150/767229/844; EI
2
; EQ; Juynboll, Encyclopedia, p, 43, col. 1; Sezgin, GAS, vol. 1, p. 12, no. 9]
(in al-Muqni) also Kfa. Ab Ubayd al-Qsimb. Sallm, and al-Mabn li-namal-man
do not mention the difference, and thus do not knowof any Kfan variant that departs from
the other text.
71
Sra 57:24,
, Medina, Damascus;
, the rest.
72
This list of the variants which, of course, cannot claimto be comprehensive,
given that the possibility of early losses of discrepancies cannot be entirely
excluded, contains some variants that for internal reasons alone must not be
consideredtobe as reliable as those onthe other side of the spectrum.
73
Con-
versely, by looking at the reciprocal relation of the manuscripts we can
identify the original manuscript with a great degree of likelihood. When
we realize (1) that the Damascus manuscript, which represents so many
unique variant readings, always corresponds with the Medinan manuscript,
whereas the latter differs from the others, but never agrees with them
against the Medinan manuscript; (2) that furthermore the Baran never dif-
fers from all the others at the same time;
74
and (3) that finally, except for a
few very peculiar variant readings, the Kfan manuscript is identical with
the one from Bara, it would follow that under the aspect of tradition the
Medinan manuscript was most likely the original one, and that from this
copy both the Damascene and the Baran copies evolved, and in turn the
Kfan copy resulted from the latter.
75
Medina

Damascus Bara
Kfa
Most remote from the original are the Damascus variant readings, which
show quite clearly that they are not as good as the others. Only the Meccan
manuscriptwhich is less important than the others, as it does not belong
to the initial four copiesis a bit difficult, since in some places it displayes
totally indiosyncratic variant readings, but in other cases follows either the
72
The possibility of recognizing all these variants as equally divine was assured by the
dogma of the Sevenaruf (see above, p. 38sqq.; I. Goldziher, Schools of Koranic commentators,
pp. 26 and 28). Totally different is the equally apologetic explanation of al-Mabn regarding
these variants; here, only an attempt is made in each case to show the two rivalling forms to
be possible, linguistically as well as sensibly.
73
But see above, on sras 6:138, and 9:101.
74
Except sras 23:89 and 23:91, on which above.
75
This result fits perfectly the arrangement of the variants in Ab Ubayds list; cf. above,
p. 395sq.
practice of Medina and Damascus or, in still other cases, that of Bara and
Kfa. We are here looking at an eclectic text (though by no means critically
eclectic) which might have been put together quite some time after that of
the other manuscripts.
The above-mentioned list of variants, however, does not exhaust the [iii/15]
differences of the local texts of the Koran, which are not entirely negligible.
There are still two additional types of differences, standing between those
true variants enumerated with great precision and the purely orthographic
deviations.
The first group includes only a few passages that are also contained in [iii/16]
the list of variants, but not quite rightly so, particularly when considering
the point of view of at least one of the variant readings. These are the fol-
lowing sras: 10:96, Damascus, the rest;
76
17:95, Mecca and Dam-
ascus, the rest; 21:4, Kfa, the rest; 23:114, and 23:116, Kfa, the
rest.
77
Here, for example, the writing is compatible with the variant read-
ings (
) as well as
, the writing , however, only with the first of the

two.
78
Thus, the difference actually goes beyondmere orthography.
79
Tothese
passages belongs sra 43:68 Medina, Damascus, and Bara, Kfa
(Mecca uncertain):
80
it ought to have been omitted from the list because it
is controversial only from the orthographic point of view.
The second group of differences consists of passages which, when viewed [iii/17]
from the aspect of the type of variant, belong either to those of the above-
76
So Ab Ubayd al-Qsim b. Sallm; in the list of variants in al-Mabn sra 10:34 is
erroneously cited instead of 10:96. Al-Dn, al-Muqni, who does not mention the difference
in his list of variants, states in bb 17 with respect to sra 10:96 a tradition going back to Ab
l-Dard (d. 31 or 2/651) regarding the Damascene writing (plural), and his own findings in the
manuscripts of Iraq (singular with ). In both passages as well as sra 40:6 the plural is read
not only by Abd Allh IBN MIR (Damascus) but also by Nfi (Medina).
77
al-Zamakhshar mentions the variant only in the case of sra 23:114, but erroneously in
reversed order of the two orthographies. al-Mabn add sra 43:23 Damascus, the rest;
otherwise known only as a difference of variant readings. Cf. above, p. 395 n. 45.
78
But it is said that it taqdran (implicitly) fits also the second one.
79
That the numerous other passages where also or is read are missing from the list
of variants might justify the conclusion that the text originally was uniformly written and
that many of the variants outside the list, , originate from carrying the actual reading into
the text.
80
The uncertainty regarding the Meccan version might be explained in such a way that
the passage found its way by secondary means from al-Mabn li-nam al-manwhere it
is added as being controversial to the list of Ab Ubaydinto the one of al-Muqni, where
originally it had been missing. It must also be noted that al-Muqni mentions the version in
a second passage, namely bb 3, fal 1, where Ibn al-Anbr says of the controversial word:
upon which al-Dn replies:
. is offered in the Koran fragment Ms Gotha, no. 462.
mentioned list or at least to those of the first group, with less certainty, but,
more importantly, transmitted with more uncertain and more incomplete
distribution over the four (or five) towns. Here belong: from al-Muqni ,
IraqandDamascus, for rest ;
81
sra 2:92, for ;
82
sra 4:40, Kfa,
occasionally for ;
83
sra 76:4, for ;
84
sra 76:16 (not verse 15)
Bara for Medina (and ijz, respectively) and Kfa;
85
from
Itf sra 55:24, (i.e., )
86
Iraq for (i.e.,
); sra 63:10,

for
;
87
fromKashf, sra 47:16, for ;
88
fromSbawayh,
89
sra
81
AbUbaydal-Qsimb. Sallm, Fail al-Qurn(Ahlwardt, Verzeichnis, no. 451, fol. 38
r
),
claims to have found the spelling only in sra 2, which is confirmed in al-Muqni, bb
19. also (the Koran fragment) Ahlwardt, Verzeichnis, no. 307, firsthand in sra 22:43.
The shortened spelling refers to the pronunciation (so the Damascene Abd Allh IBN
MIRcertain in sra 2; other passages are controversial).
82
Muqni, bb 2, fal 1, mentions the lacking in the Imm, evidently comparing it with
(i.e., Mkilso Nfior Mkl, like the majority of the readers). (so Ab Amr)
(Ibn al-Al and af [Ibn Sulaymn, d. 180/796; EI
2
; Sezgin, GAS, vol. 1, p. 10, no. 3] an im)
is thus not attested there, but it is already presupposed as the standard spelling (assured by
the metre) in Aqla, v. 51, so that the sense of the comment is reversed. Itf fual bashar
somewhat reconstitutes the meaning by saying
. (Koran fragment),
Ahlwardt, Verzeichnis, no. 305.
83
al-Farr in the list of variants in al-Dn, al-Muqni f marifa, bb 21, but where disputed
(again against [Ab Isq Ibrhm b. Uthmn (Umar)] AL-JABAR (d. 732/1331) [Brockel-
mann, GAL, Supplement 1, p. 725, l 7, from bottom] in the Itf), and Aqla, v. 64; as a variant
reading also in al-Zamakhshar.
84
(so Flgel) exists in the written form(it is read by the majority) only on account
of Ab Ubayds, statement in al-Muqni, bb 2, fal 1, that the (final) was washed out in the
Immevidently a fabrication to justify the grammatically correct reading. , without
variant, is presupposed by al-Muqni, bb 5, fal 1, and Itf fual (printing wrong, ).
85
Bb 5, fal 1; Aqla, v. 125; cf. al-Dimy al-Bann, Itf fual bashar; also in the list of
variants of al-Mabn li-nam al-man as additional difference between Kfa and Bara as
transmitted by some scholars. According to Ab Ubayd in al-Muqni, bb, 2, fal 1, the was
erased (!) in the Imm. The distribution of the variant readings is not clear. V. 15 consistedly
(Flgel is wrong).
86
As a variant reading, amza b. abb and Sulaymn b. Mihrn AL-AMASH [Juynboll,
Encyclopedia, pp. 78126]; according to some, also Ab Bakr (Shuba) an im b. [Ab
l-Najd] Bahdala, [d. 127/745; EI
2
; EQ; Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 9, p. 42]
(all Iraq).
87
The manuscripts evidently have consistently
; this means that a difference ingeneral

is out of the question, and that most of the readers follow this orthography, though they
consider it linguistically very uncomfortable, in cases even imperfect (see above, p. 390 n. 14).
Also Ab Ubayd claims to have see it in the Imm (al-Muqni, bb 21; al-Dimy al-Bann,
Itf fual al-bashar); next to it Itf fual al-bashar, mentions about the Imm an
account of Khlid (see above, p. 393) expounding
which seems to have been fabricated

to justify the proper linguistic variant reading, wa-akna, upheldmainly by AbAmr (Bara).
88
Unknown as a variant reading.
89
EI
2
17:78, for
90
and sra 68:9, for ;
91
from al-Zamakh-
shar, Kashshf, sra 12:7, for
;
92
sra 17:40,
for ;
93
sra 20:15,
added at the end of the verse .
94
Affixed as a semi-
orthographic variant from Muqni sra 20:80, instead of .
95
The uncertainty of the transmission becomes quite evident fromthe dis-
tribution of the copies, which partly contradicts the results of their relation
as shown above. There is nearly always the chance that we are not deal-
ing with early differences of Uthmnic codices of the Koran, but with the
penetration into the written text of largely non-Uthmnic variant readings.
The number of semi-orthographic variants (predominantly with or without
) could easily be enlarged if all those would be considered about which
nothing has come down to us regarding the distribution of the different
orthographies. In the case of these there is the likelihood that the secondary
development of the unequivocal form (for example, above , ) from
the ambiguous form ( , ) is still greater when carrying the correspond-
ing variant reading into the text.
The Orthography and Its Sources [iii/19]
The problems we face in the case of the textual transmission are nothing
comparedwiththose of orthography. The number of passages where contra-
dictory spellings are transmittedis quite considerable, be this incases where
the Uthmnic copies differ, or where the early spelling was abandoned.
Yet it is not as great as might have been expected. The overall orthogra-
phy is firmly established, and in a great many cases even the distribution
of rival spellings is limited to individual Koranic passages. The variations
and uncertainties are essentially restricted to two main problems, the writ-
ing of with or without , and the separation or combination (al-maq
90
237, also as a variant reading; al-Zamakhshar, s.v., only as text of Ubayy, but Mufaal
594 as a variant reading. Otherwise not known as such.
91
241 quoted from al-Zamakhshar, s.v., and (without name) Mufaal 592; unknown
as a variant reading.
92
This variant is totally different fromall the others sothat it is likely that f baal-maif
in al-Zamakhshar is a vague expression for in the muaf of Ubayy, to whomthis variant is
usually ascribed. It is unknown as a variant reading.
93
Unknown as a variant reading.
94
The first two words also Ubayy. Shown to be a difficult passage by the use of an
inaccurate rhyme serving as explanation.
95
In a second (earlier) list of variants, bb 19, containing mainly orthographic differences.
The first form as variant reading in amza (Kfa), etc.
wa-l-mawl) of shorter words (particles, etc.). Orthography was undergo-
ing changes in both fields. In contradistinction to the underlying Aramaic
orthography it hadbecome customto treat as a vowel-letter, andto be more
observant in the word division. In both fields the number of the respec-
tive Koranic cases was so great that it was quite difficult to keep sight of
them. That also the orthographic transmission is very definite and pretty
uniform can be explained by the fact that interest in orthographic ques-
tions started so earlyin the middle of the second century ahthat there
were undoubtedly still sufficient copies fromthe time immediately after the
Uthmnic recensionavailable to settle controversial points of view. Inthese
two fields it was, of course, not possible immediately to settle all points of
contention by explicit textual evidence (na). This explains why in default
of na the ijm of the copyists of the Koran
96
was consulted, and that still
Uthmn b. Sad AL-DN (d. 444/1052)
97
was at times obliged in his studies
to refer to ancient manuscripts of the Koran (mainly from Iraq). That this
source was insufficient for the reconstruction of the earliest orthography is
evident from extant old fragments of the Koran, where none of the impor-
tant copies agree in every detail with the transmitted rules.
There were two reasons that led to the study of the orthography of the [iii/20]
Koran in the second century. There was first Mlik b. Anas (d. 179/795) deci-
sion
98

. The awareness that Arabic orthography would continue to change led

to the question whether or not the neworthography could be applied to the
Koran, and in case of its negation, could it be applied to a stricter obser-
vation of the Koranic orthography. Mliks and his supporters prohibition
was of no consequence. Whereas the traditional restraint in the develop-
ment of Koranic variant readings continued to lessenalready the Kfic
fragments of the Koran display an increase in deviations from the rules of
scholars, although again mainly in the case of the with the appearance of
modern styles in Koranic manuscripts more and more modern orthography
appeared, so that in our texts of the Koran only remnants of the old orthog-
raphy remain. This divergence of the direction of development is explicable
in so far as from the second century onward the emphasis of transmissions
of the Koran again shifted towards oral reading.
96
al-kuttb, occasionally ab al-maif; also dat al-kuttb.
97
Brockelmann, GAL, suppl. vol. 1, pp. 719720; Sezgin, GAS, vol. 1, p. 17, no. 28, etc.
98
al-Dn, al-Muqni, bb 1, andthenal-Suy, al-Itqn, naw 76, fal 2; Mevzuat, vol. 2, p. 6;
al-Damy al-Bann, Itf at the beginning of the fal, F dhikr jumla min marsm al-kha.
Similarly Amad b. anbal (Itqn, Mevzuat).
The second reason is recognizable from the transmission of the orthog-
raphy of the Koran itself; it is the question of the discrepancy between the
variant readings and the writtentext. Along list going back to the frequently
referred to Nfi b. Abd al-Ramn al-Layth (d. 169/785)
99
has been pre-
served in al-Dn, al-Muqni f al-marifa,
100
which does not contain any .
Nearly everywhere it canbe shownthat it was read as well as a.
101
The fixing
of writing without thus serves to establish the scriptual conformity of the
reading with . It is rare to find the reverse relation, namely that the estab-
lishment of a spelling serves to reject a reading which cannot be reconciled
with a variant reading.
102
Already Nfi b. Abd al-Ramns contemporary, Yay b. al-rith al- [iii/21]
Dhimr (d. 145/762)
103
wrote on spelling (hij) in the Koran.
104
Al-Dn
105
frequently refers to a related book dealing particularly with Medinan manu-
scripts of the Koran by (Ab Muammad) AL-GHZ IBNQAYS al-Andalus
(d. 199/815) entitled K. Hij al-sunna.
106
Books onikhtilf al-maif,
107
which
were mentioned above on p. 345sq., on ittifq al-maif,
108
on gharb al-
maif
109
or simply al-maif
110
also deal with orthography apart from
99
One of the seven canonical readers, [EI
2
; Sezgin, GAS, vol. 1, pp. 910].
100
Bb 2, fal 1.
101
In the very isolated instances when our research aids fail to provide proof we might be
more successful if we were better acquainted with the great old works of the variant readings.
102
Examples mainly in bb 5: bb dhikr m rusima bi-ithbt al-alif li-manan, and similar
bb 7 with reference to y. Cf. also al-Suy, al-Itqn, naw 76, fal 2, qida 6: f-m fh
qiratn fa-kutibat al idhum.
103
EI
2
104
Fihrist, pp. 36 l 24. He is casually referred to in al-Dn, Muqni, bb 18, toward the end.
105
Bb 2, fal 4, bb 5, fal 4, fal, bb 7, 8, 9, 12 and 16, and thereafter Aqla, v. 187, and
repeatedly. Itf (of al-Dimy al-Bann) seems to have used it directly, in any case not by
way of al-Dn, al-Muqni; cf., for example, the remark on sra 30, which apparently cannot
be located there. On the author cf. al-Suy, Bughya, p. 371.
106
107
Fromamong the authors of such books listed in the Fihrist, p. 36, l 8sqq., al-Dn refers
to some of them in his al-Muqni, although he most likely did not use them directly: al-Kis
(particularly bb 13, 16, 19), al-Farr (mainly bb 4 and 21), and Khalaf b. Hishm al-Bazzr
[Sezgin, GAS, vol. 1, p. 12, no. 9] (bb 5, fal 1, bb 21).
108
From [Ab l-Mundhir] NUAYR b. Ysuf [AL-RZ, d. ca. 240/854]; (who according to
Fihrist, p. 30, l 8, transmittedfromal-Kis andthus is likely tobe identical withhis pupil, Nar
b. Ysuf [Sezgin, GAS, vol. 8, p. 119], <see, for example, Flgel, Die grammatischen Schulen der
Araber, p. 128>); al-Muqni, bb 9, 12, 16 (here the title) 18 (f m ijtamaa alayh al-maif),
19 and 20; he takes into consideration also Baghdad (but not Mecca), and includes the
differences.
109
Fihrist, p. 35, l 5 and 8, probably from the author of the one hij book, ibid., p. 36, l 24.
110
E.g., Fihrist, p. 33, l 6. Regarding the book with this title by [AB BAKR] b. Ashtah [AL-
IBAHN (d. 360)], which was used by al-Suy in al-Itqn, naw 76, fal 1, and elsewhere,
other subjects. Special treatment is accorded to the subject of separation
and combination of words (al-maq wa-l-mawl), which has already been
described above on p. 352 as a particularly difficult field. The first writer to
deal with the subject was probably amza b. abb (80/699156/773),
111
to
whom also al-Dn refers
112
in the relevant section.
113
Another question
treated in a monograph is the writing of the feminine as ; the paragraph
in al-Muqni regarding this orthography
114
is likely based on the K. al-Ht f
kitb Allh
115
of Ab Bakr Muammad b. al-Qsim(IBN) AL-ANBR (d. 327
or 8/939 or 940).
116
The earliest extant survey of the entire field of rasm
(or marsm al-kha) is the K. al-Muqni f marifat kha maif al-amr
allat jumiat f zamn Uthmn b. Affn
117
of Ab Amr Uthmn b. Sad
AL-DN (d. 444/1052),
118
which Silvestre de Sacy
119
studied in detail. The
Muqni is unlike al-Dns works on the variant readingsand particularly
the Taysr, a precisely arranged and minutely constructed compendium.
see above, p. 389 n. 2. Of a different character seems to be the book with identical title of
Ab Bakr Abd Allh IBN AB DWD Sulaymn b. al-Ashath AL-SIJISTN (d. 316) [EI
2
; EQ;
Sezgin, GAS, vol. 1, pp. 174175] (d. 316/929), which is mentioned in Fihrist, p. 233, l 1 (proba-
bly identical with the K. Ikhtilf al-maif li-Ab Dwd al-Sijistn, and which is the main
source of Itqn, naw 76, fal 3: F adab kitbat al-muaf).
111
[amza b. abbb. Umr al-Kf; EI
2
; Sezgin, GAS, vol. 1, p, 9.] Fihrist, p. 36, l 26, for what
is here said regarding the book of (Abd Allh) IBNMIR (AL-YAUB) applies to what has
been said above, p. 395 n. 44.
112
The work with the identical title by al-Kisif the title variant, al-Mawl lafan
wa-l-mafl manan (cf. Flgel, Die grammatischen Schulen der Araber, p. 125) proves to be
correctdealt with the questions discussed in al-Suy, al-Itqn, naw 29.
113
al-Muqni, bb 16.
114
Bb 17.
115
Which in any case was a small work.
116
Sezgin, GAS, vol. 8, pp. 151154, and vol. 9, pp. 144147. al-Muqni in the respective
paragraph mentions his name twice, but not the title of the work. Also the other writings
on ht and yt of the Koran (Fihrist, pp. 31, 18, 32, and 21) are likely to have been on
the subject of orthography. Al-Kiss work on the ht (Flgel, Die grammatischen Schulen,
p. 126), however, dealt with grammar, provided the addition, al-mukann bi-h, is correct,
and likewise all the works on lmt (Fihrist, pp. 35, and 27sqq.; cf. also G. Bergstrer, Das
Kitb al-Lmt, p. 77sqq.). Other important authorities of the Muqni that have not yet been
mentioned are MUAMMAD IBN S b. Ibrhm b. Razn AL-IBAHN (d. 253 or 243/867
or 857); cf. for example, al-Suy, Bughya, p. 88) (bb 5, fal 1 and 2, bb 9, 12, 16 and 18), by
whom, among others, excerpts from Nuayr b. Ysuf have been transmitted, and Ab af
[Sezgin, GAS, vol. 8, p. 103 (f.), no. 4] (?) (bb 12, 13, 16 and 18).
117
So the title in A.I. Silvestre de Sacy (see below); otherwise shortened and rasm or
marsminsteadof kha. Witha slightly different sub-title inthe Constantinople manuscript,
Waqf Ibrahim, 31, fol. 81
v
: Muqni f hij al-maif (cf. above, pp. 404405).Utilized in
al-Dns Berlin manuscript, Ahlwardt, Verzeichnis, no. 419.
118
Sezgin, GAS, vol. 9, p. 42, no. 2, and foot-note 1.
119
Commentaire sur le pome nomm Rayya, pp. 290306, 330332.
Rather, it is more like a collection of sources for such a book, a differ-
ence which perfectly displays the insufficient development of the science
of Koranic orthographywhich for practical purpose had nearly fallen into
oblivionwhen compared with the science of the variant readings. Even
when by and large there is a subject classification of the material, the sub-
jects nevertheless frequently overlap, and large parts are taken over in toto
from the original, interspersed only by isolated critical remarks. Several of
the sources have already been mentioned. The most important authority
seems to benot only for the alleged copy of the Caliph Uthmnthe
above-mentioned Ab Ubayd.
120
Unfortunately, from this title none of his
works we know of can be attributed to the subject under discussion. The
same applies to Muammad IBN al-Qsim AL-ANBR, unless his writings
mentioned on p. 340 n. 2, might be considered. Entirely different from al-
Muqni was the related book of Ab l-Abbs al-Marrkush entitled Unwn
al-dall f marsm kha al-tanzl. According to the excerpts supplied by al-
Suy
121
it dealt with the proof of an alleged meaning of the orthographic
differences.
The Muqni is the main basis of a work with an unknown title, but of [iii/24]
similar arrangement, by Abd al-Aad b. Muammad al-anbal al-Harrn
(8th cent.),
122
and most of all the versified treatment of the subject by Ab
l-Qsim al-Qsim b. Firruh AL-SHIB (died in 590/1193)
123
Aqlat atrb al-
qaid f asn l-maqid (because of the rhyme also called al-Riyya),
124
but
other sources have been consulted as well (Aqla, verse 46). From its com-
mentary by ALAM AL-DN Ab l-asan Al b. Muammad AL-SAKHW
(d. 643/1245)
125
Silvestre de Sacy, too, published excerpts with translation in
his Mmoire sur l origine et les anciens monuments de la littrature parmi
les Arabes.
126
The Aqlawiththe same commentary andanadditional one
120
al-Zamakhshar is likely to have made direct use of him in the Kashshf. It is nearly
certain that al-Zamakhshar did not make use of al-Muqni; cf. below, p. 409 n. 143, and also
below, p. 423 n. 225; and also frequently above in the second section.
121
al-Suy, al-Itqn, naw 76, fal 2. AlsomentionedinMevzuat, vol. 2, p. 7, andHa Qalfa
[sic,] both without further details.
122
Nuzhat al-lim f qirat im, Brockelmann, GAL, vol. 2, p. 165. Used by Nldeke in
his first edition of the present work from the only known Ms., Leiden, no. 1640, [al-arrn,
Nuzhat al-limf qirt im.]
123
Sezgin, GAS, vol. 9, p. 41, no. II.
124
Used in the printed edition in Majma f l-qirt, Cairo, 1329/1911.
125
126
pp. 419434, 327342; cf. further his Recueil de diffrens traits relatifs l orthographie
et la lecture de l Alcoran, pp. 333354.
by AbIsq Ibrhmb. Uthmn[or Umar] AL-JABAR (d. 732/1331)
127
(and
also probably al-Muqni itself)constitutes the basis of the brief but very
useful studyin the work on variant readings, Itf fual al-bashar f
qirt al-arbaat ashar
128
by Amad b. Muammad AL-DIMY, called
al-Bann (d. 1117/1705)
129
in a chapter of the introduction, and contin-
ued with one section each at the end of the discussion of the individual
sras.
130
An even more drastic rearrangement of the subject into a contin-
uous orthographic commentary on the Koran was produced by Mumin b.
Al b. Muammad al-Rm al-Qalkbdh (d. 799/1396) in his Jmi al-kalm
f rasm muaf al-imm,
131
which also considers the alleged reason for the
orthographic differences. Al-Suy in his Itqn (naw 76, fal 2) supplies a
brief survey, mainly based on al-Muqni, though arranged in a better order.
According to Ibn Khaldn,
132
al-Muqni and Aqla were displaced in the
Maghreb by the rajaz-poem of Ab Abd Allh Muammad b. Muammad
AL-KHARRZ() [d. 711/1311], on which several commentaries have come
down to us.
133
127
Brockelmann, GAL, vol. 2, p. 164 (with a different kunya [Burhn al-Dn Ab l-Abbs]).
128
Printed in Cairo, 1317/1899. The title of the print is erroneously f l-qirt. [Edited by
Shabn Muammad Isml, Beirut and Cairo, 1407/1987].
129
130
Other works on the variant readings as well as the commentaries on the Koran, as far
as I know, consider orthography only occasionally, and supply only useless references to
the sources. The short presentation of the rasmin vol. 1, pp. 2834, in the margin of al-
abars commentary on Gharib al-Qurn wa-raghib al-furqn [ed. by I. A. Awa, Cairo,
19621971] of Nim al-Dn Ab l-Qsim al-asan b. Muammad al-Qumm AL-NSBR
(d. probably 706/1306, cf. P. Schwarz, Wann lebte der Verfasser, p. 300sqq.), which Schwarz
used as the basis for his article, Der sprachgeschichtliche Wert einiger lterer Wortschrei-
bungen im Korn, p. 46sq.does not, as Schwarz thinks (p. 47), reflect the seventh century
condition of the manuscripts of the Koran in Khursn, rather it is a peculiarly arranged,
and partly imperfect, collection (see below, p. 416 n. 179, p. 419 n. 203) from the same mate-
rial which is available in a more complete and correct form in Muqni, etc. The quite iso-
lated instances of details (see below, p. 423 n. 234) missing from Muqni and not obviously
incorrect shows that al-Muqni cannot be the sole source. It is more likely that a Dama-
scene secondary transmission (see below, p. 423 n. 234) has been used, which we also find
in al-Sakhw (see below, p. 424 n. 245) and in Nashr of Ibn al-Jazar (see below, p. 426
n. 254).
131
Constantinople, Wakf Ibrahim, Ms. 31, fol. 139 (which contains from fol. 81 onward
some related smaller writings of more recent times). Related is the somewhat more recent
work [on the orthography of the Koran], Berlin Ms., Ahlwardt, Verzeichnis, no. 429. Yet
another brief anonymous work on the orthography of the Koran without title is the Con-
stantinople Ms., Fatih Camii Ktphanesi, no. 73, fol. 21
v
sqq. Cf. also Silvestre de Sacy,
Recueil de diffrens traits relatifs l orthographie et la lecture de l Alcoran, pp. 355359;
and Ibn Alawn al-Muqri in: Wilh. Ahlwardt, Verzeichnis, no. 420.
132
al-Muqaddima, fal 6, section 5.
133
The Cardinal Orthographic Peculiarities of the Uthmnic Text [iii/26]
As the early orthography of the Koran is of prime importance for the his-
tory of the textwhich to a large extent is made up of such triflesand
allows not insignificant deductions about the language of the Koran and
the origin of the new orthography, and, furthermore, as it is interesting
to observe how people handled an imperfect script, which was adapted
from a foreign peoplealthough from the outset some improvements had
been introducedone still notices howclumsily this was occasionally han-
dled.
134
Here, the main cases of this orthography have been listed essentially
according to the details in al-Muqni in the light of the spelling in the extant
Kfic fragments of the Koran, where they either confirmpeculiar references
in literature or else depart from the following six rules of scholars.
135
First.In Arabic orthography words generally
136
do not appear in the [iii/27]
formthey take in context but in the pausal form.
137
But Koranic orthography
occasionally retains the form in context.
138
It is thus written:
(a) instead of in:
in eleven cases; in seven cases; in seven cases;
in five
cases; sras 3:54 and 24:7; sra 58:9 and 58:10; sra 7:133;

sra 11:87;
sra 28:8;
sra 30:29; sra 44:43;
sra
56:88; sra 66:12.
In other cases it is disputed whether the singular or the plural is meant
(which in most cases is indicated merely by instead of ). For example,
134
Later Islamic interpretation recognizes in every clumsiness a profound wisdom (cf.
above, pp. 406 and 407sq.). This opinion is strictly opposed by Ibn Khaldn in his Muqad-
dima, fal 5, section 30. The attempt of P. Schwarz in his article Der sprachgeschichtliche
Wert (above, p. 407 n. 129) to recognize in most of the written peculiarities an expression of
oral phenomena presupposes among the earliest scribes of the Koran a sensitivity for sounds
and a precision and consistency in the rendering of sounds which was hardly attained by
Arab grammarians centuries later.
135
In citations from Kfic manuscripts diacritical signs have been added, except in cases
when more accuracy is required. [Here this is limited to the possibilities of the computer.]
136
An exception are words like
which in analogy to the genitive nunation are written,

although the pausal form is hdbut see below, this page and pp. 362363.
137
Therefore for
*
,
for
*
, for
*
, etc. For this reason the Sandhi
phenomena are not expressed in writing, for example,
not
*
, and
*

, not
*
(only when written in one word ).
138
In such instances Koranic scholars disagree as to what happens to a word not written
in pausal form when it occurs in pause (what even happens when spoken separately); does
it receive the ordinary pausal form or not? For example, in pause, is it to be pronounced
ramah or ramat? ( .) The respective rules constitute the main part of
the chapter F l-waqf al marsmal-kha of the books on variant readings.
, sras 6:115 and 10:96 (see above, p. 400); and frequently,
sra 29:49;
sra 41:47, etc.
After : sra 2:203, and quite frequently; sra 53:19 (instead [iii/28]
< < ), , .
139
(b) The consonantal letters and are frequently omitted if the vowel is
shortenedbecause of a following wal (inwhichcase it must be remembered
that the final frequently also in other cases remains unwritten, see below,
pp. 413414):
in fifteen cases:
140

, sra 4:145;
, sra 6:57;
141

, sra 10:103;
, sras 20:12, 28:30, 79:16,
, sra 27:18;
, sra 22:53; and
, sra 30:52;
142
, sra 37:163;
,
sra 81:16;
, sra 39:19;
, sra 27:36;
, sra 36:22.
: , sra 17:12; , sra 54:6; , sra 96:18;
, sra 42:23.
143
According to al-Zamakhshar it is permitted to put here also
, sra 66:4 (for ), and indeed this is certainly correct.
as a vowel letter in the identical case is omitted only occasionally in ,
namely before , sras 24:31, 43:48, and 55:31 (cf. also above, p. 395 note 46).
(c) The nunation is written as in (i.e.,
or ),
144
its origin from
+
[iii/29]
has been lost.
(d) Regarding the way to express hamza at the end of a word cf. below,
p. 421 sqq.
Second.Small particles are assimilated in speech as well as in writing.
In the Koran this is more common than in later orthography. There are,
however, no set rules, and many occurrences are uncertain. The Koran
writes:
139
This word ought not have been included as by nature it cannot occur in pause and thus
cannot be written in pausal form (irrespective of the late artificial form , the being).
140
al-Muqni, bb 6, fal 2; still, in the various sections of the Muqni I have been able to
find only the above-mentioned fourteen passages.
141
Others read in this case
.
142
Others read
[the first and the last words have the vowel marks in].
143
al-Zamakhshar onsra 42:23 omits the passage insra 54:6 and states that also inother
cases spellings with occur.According to al-Farr also sra 59:19 is to be written
without , which al-Dn contests. According to the reading of asanal-Bar lu (for l),
sra 37:163 ought to be here.
144
In poetry kin is far more frequently encountered than kaayyin. The word appears
also in other forms ( , Ibn Muayin, [d. 123/740, EI
2
; Sezgin, GAS, vol. 9, pp. 202203] and
occasionally al-asan al-Bar; cf. al-Dimy al-Bann, Itf on sra 3:140).
, but three times
.
145
, but twice .
), but sra 7:166, .

, but mainly .
146

, but ten times .

147

for
, but
, sra 11:17.
148

, but four times .

(also sra 6:144 = or what).

, but sra 13:40,
. [iii/30]
(as well = only, as = verily, was,) but sra 6:134,
.
149
, but in some passages
.
150
.

, somewhat infrequently .
151
.
152
, but several times .
153

.
Twice , four times ( , sra 59:7).
145
Sra 63:10 is controversial (Ms no. 1 in: J.G. Ch. Adler, Descriptio cod. quorundamcufico-
rum (1780), ).
146
The distribution is very uncertain since according to some is found in eleven
passages, whereas according to others, only once.
147
Sra 21:87 is controversial.
148
In the Kfic Koran fragment, Ahlwardt, Verzeichnis, no. 328, it is written in one word.
149
Sra 16:97, controversial (al-Dimy al-Bann, Itf).
150
The references to some of the passages vary.
151
Ibid.
152
Only sra 14:37, and according to some, sras 4:93, 23:46, and 67:8, .
153
The references to some of the passages vary.
is always affixed (as simple ) to the following vocative.
154
The assim-
ilation proceeds even one step further in
155
for
, sra 20:95, so
that here not only is connected with the following word but also
full
brother is written like a single word.
A peculiar writing is
, sra 4:80; , sra 18:47;

, sra 25:8; , sra 70:36; here, the preposition
, which can-
not stand alone, is inappropriately attached to the preceding word (assimi-
lation with the noun ) instead of the following word.
Stranger still is the writing for , sra 38:2, which Ab [iii/31]
Ubayd [al-Qsim] claims to have seen in his copy of the Uthmnic recen-
sion. Al-Zamakhshar does not doubt its existence, but al-Dn says it is not
found in any manuscript, and it also does not seem to occur in extant Kfic
fragments.
Third.The systematic use of vowel letters to indicate the long vowels
gives Arabic orthography a great advantage over the other Western Semitic
orthographies, with the exception of Ethiopic, but they are still now and
then omitted.
(a) The , which the old orthographies never use medially as a simple vowel
signor only incertaincasesis frequently omitted inthe Koranas well as
in other ancient Arabic antiquities. This omission is the first orthographic
peculiarity which the reader of Kfic texts encounters. Although Koranic
scholars attempted to introduce certain rules according to which was
used now and then, their recommendations are precarious and at times
contradictory, and in addition, Kfic Korans frequently depart from the
rule.
156
Of course, those manuscripts which, with the exception of the words
that also later omit the (like , )
157
and always use the ,
158
originate
from a somewhat later time, and are rewritten in accordance with more
154
Likewise , not only in , etc. but also in = sra 3:59, and frequently.
155
The Gotha Koran fragment, W. Pertsch, Die arabischen Handschriften der Herzoglichen
Bibliothek zu Gotha, no. 451, , and so also occasionally in literature (unspecific).
156
Karabaceks statement in his Ein Koranfragment des neunten Jahrhunderts, p. 36sq.,
that with respect to the use of in the Viennese Ms., Ser. nova, no. 4742, concurs with the rules
of al-Muqni, is open to discussion.
157
But even is to be found in Kfic Korans, for example, the Gotha fragment of the
Koran, W. Pertschs Verzeichnis, no. 427; J. Ch. Lindberg, Lettre sur quelques mdailles cufiques,
table 12; D.S. Margoliouth, Mohammed and the rise of Islam, plate to p. 218.
158
Nearly so the magnificent Berlin fragment of the Koran, Ahlwardt, Verzeichnis, no. 301;
B. Moritz, Arabic palaeography, plates 3136; and in addition the Ahlwardt manuscripts,
numbers 303, 306, 309, and 315, from the transition stage to naskh.
recent rules. Still, also many others that do not display intentional deviation
from ancient usage frequently put an where according to those recom-
mendations it ought to be omitted, for example in that according to
scholars ought to be written with in four instances only, but everywhere
else always . Conversely, the case that a required is omitted is rare
159
as
most of the manuscripts demonstrate the attempt of its increasing usage.
Moreover, these rules that go back to goodoldcopies are generally more reli-
able than our extant manuscripts, which are far from being carefully edited
model codices, but rather are mainly calligraphic artifacts.
This is not the place to discuss all the cases where is either written [iii/32]
or omitted. Let us only present a few principal examples. Naturally, is
omitted where its omission is customary, or at least tolerated, also in later
orthography; thus, for example, also for
. Following that tradition, it is

mostly omitted also in the plural ending t, in the dual ending ni, in the
afformative -n, when it is followed by a suffix (e.g. = ), in the forms
(thus )but not in and in most of the words with
between the second and third radicalafter an l (in order to avoid the
conglomeration of similar long letters), in the vocative particle ( ,
,
etc.), in the particle of invocation (
= , sra 3:59 and frequently),

etc. An interesting isolated case is ,
160
the Medinan and Iraqi spelling of
the relative pronoun
(on see below, p. 421 sq.).

Still, this learned tradition overlooked several words where has been
omitted, although it is the bearer of a resting hamza. The omission of the
also in this case must be explained by the fact that the Meccans, and many
other Arabs, abandoned the laryngeal in such cases and pronounced a as
(see below, p. 420sq.). This orthography is to be found in the following cases:
instead
, sra 2:286, in B. Moritz, Arabic palaeography, plate 19;

instead
, sra 4:104, in W. Ahlwardt, Verzeichnis der arabischen
Handschriften, no. 313, and Wilh. Pertsch, Die arabische Handschriften zu
Gotha, no. 433; instead
, sra 6:42, inAhlwardt, no. 314; instead

, sra 12:102, Ahlwardt, no. 331; instead
, sra 28:26, in
B. Moritz, Arabic palaeography, plate 39; instead
, sra 33:5, in
Ahlwardt, no. 349; instead
, sra 33:13, inthe same manuscript;

159
For example, Wm. Wright, Facsimiles of manuscripts and inscriptions, plate 59, where
sra 27:1 reads although the passage is one of the four where according to al-Muqni
ought to be written.
160
So the Berlin Koran fragment, Ahlwardt, Verzeichnis, no. 350, sra 33:4; conversely, the
Berlin Mss., nos. 349, and 351, at the same place .
instead
, sra 34:29, in W. Pertsch, Gotha, no. 457.

161
This
peculiar orthography of partially very old manuscripts might be ascribed
without hesitation to the earliest time. Literature mentions only two cases
of this kind: sra 2:67, instead of , and sra 50:29, instead of

(in most of the manuscripts).

162
(b) The designating the vowel is omitted
163
when two come together,
[iii/33]
for example in instead of . Exceptions are sra 83:18,
, and the
forms of the causative of with afformative or suffix,
164
for example,
,
as well as the corresponding , sra 50:14.
This purely graphic loss of a medial is juxtaposed with the dropping
of the final conditioned by the pronunciation. One not only suppresses
many a final in writing and speech in pausefrequently for the sake of the
rhyme (for example,
, sra 13:10;
, six times in sra 54;
, sra
43:26, etc.)and shortens the possessive suffix of the first person singular
inthe vocativea phenomenonthat according to the other rules governing
the Arabic vocative can easily be explained(e.g.,
, sra 2:51)
165
as well
as final generally before wal (also in writing, see above, p. 409). Rather,
the shortening goes far beyond these limits. From the well-known Quraysh
name instead of , and possibly also from the existence of similar
forms in adth, it follows that this pronunciation was common to the
Quraysh dialect. We thus find in the Koran
in sras 2:182, 54:6, 54:8;

in sra 11:107;
in sras 17:99, and 18:16;
in sra 18:63; in
sra 34:12;
in sra 50:40, etc.; with the nominal suffix of the first person
singular, in sra 14:42; and, above all, with the verbal suffix of the first
person singular, for example,
in sra 2:182, in sra 2:193,
in
sra 11:48, etc. Particularly frequent is this omission (for similar reasons as
with the vocative) with the imperative and jussive.
161
Further
instead , sra 24:27, in A.S. Lewis Sinai Palimpsest, A(see below,
p. 426sq.).
162
Further , sra 106:1, in the variant reading (Ikrima).
163
The spelling instead , W. Pertsch, Die arabischen Handschriften, Gotha Ms,
no. 460, sra 37:111, is a spelling mistake; but cf. Lewis Sinai Palimpsest, below, p. 426sqq.,
section The orthography of Agnes S. Lewis Sinai Palimpsest.
164
The sources are vague and contradictory about the orthography of the forms without
ending.
165
The only exception is (which apparently means the pronunciation ibdiya)
sras 29:56, and 39:54 (here thus also in the Berlin Ms., Ahlwardt, no. 352); as far as the
contentious passage sra 43:68 is concerned, see already above, p. 400.In poetry the
shortened form is not unusual, for example in
.
In all these cases which, taken as a whole, elide or shorten
166
a great many [iii/34]
of the final ,
167
we must recognize the orthography as the expression of the
pronunciation. Also the inconsistency of orthography as far as writing with
or without in completely parallel cases is nothing but a reflection of the
actual variation in the pronunciation.
Most of the readers of the Koran recognize the disappearance of the [iii/35]
only in rhyme and pause, and restore it in context in a larger or smaller
number of cases; others even do this in pause.
168
(c) Except before wal, as a sign of is elided only in the case when two
meet, for example, in for
, and in the word for

<
.
169
(d) The original long final vowels of the pronominal suffixes
and
170
fol-
lowing pausal pronunciation
are written defectively.

171
166
In some of the cases the statements are contradictory.
167
Itf offers at the beginning of the chapter F l-yt al-zawid (p. 71, bottom) the
following statistics: Apart from rhyme, is missing in thirty-five cases, twenty-two of which
with the suffix of the first pers. sing.; in rhyme, in eighty-six cases, eighty-one of which with
the suffix of the first pers. sing.; thus, altogether in 121 cases. This statistic, however, considers
only those cases with contentious pronunciation; their total number is considerably greater
(Itf, p. 7, bottom: excluding the vocatives with suffix of the first pers. sing., 133). I am not
aware of a statistic of the cases where is not dropped; Itf enumerates at the beginning of
the chapter F l-yt al-zawid (p. 68sq.) 796 cases with written of the suffix, although it
must be observed that in 230 cases also the pronunciation -iya is maintained.
168
Most daring in the restitution is Yaqb al-aram (Bara), [EQ; Sezgin, GAS vol. 1,
pp. 1112] who even in rhyme mostly again inserts (with -iyah as context pronunciation
of the suffix); most closely related to him are Ibn Kathr (Mecca)who treats similarly a
considerable part of these placesand Ab Amr b. al-Al (Bara) who, still somewhat more
frequently than Ibn Kathr, restitutes at least in context. The Kfans (im al-Jadar,
amza b. abb, al-Kis, Khalaf b. Hishm al-Bazzr) are the most rigid followers of the
consonantal text, who, apart from a few places, shorten the vowel in context and pause
whenever it is written defectively.
169
The only explanationfor this peculiar writing seems tobe that inthe dialect of the ijz
hamza had so early and completely disappeared that the word could be effected by the rules
of sound i i . Less likely is the assumption that for the word by nature in the dialect of the
ijz the nominal form ful instead the nominal form fil was the custom. In any case, the
writing must be considered ry; this is indeed transmitted (Itf on sra 12:5) as a variant
reading of Ab Jafar Yazd IBN AL-QAQ AL-MAKHZM (Medina, d. 130/747; Sezgin, GAS,
vol. 1, p. 9, l 5). The writing can be found in the Berlin Ms., no. 363, on sra 48:27, and in the
Gotha Ms., no. 460, on sra 37:105.
170
Cf. now the detailed study of A. Fischer, Die Quantitt des Vokals des arabischen
Pronominalsuffixes hu.
171
The readers of the Koranunanimously retainthe lengthexcept after resting consonants
(i.e., mainly after a long vowel anddiphthong) where only IbnKathr andIbnMuayin(both
Also the old forms, hum -hum (-him -him) -tum -kum, were [iii/36]
changedat least in pausein ordinary language to the shorter hum-hum
(-him) -tum -kum which are likewise found in all poets, although not as fre-
quently as the long ones; and for such reason the Koran writes .
172
Whether the in indicates only the pausal formor whether in the ijz
the old length, which is only occasionaly found in the poets,
173
was still alive
cannot be determined with certainty.
174
Fourth.The long vowel which we read is frequently expressed by or
in the Koran.
Wherever is the third radical (as in ) or appears in inflexion (as in
, 2ndpers., ; , dual, ), the final stemwhennot followed
by
175
is expressedby ; andthis not only at the endof a wordbut also before
suffixes. Further written with is the ending - in the feminine ending and
in
, etc., and in a number of particles ( etc.,
[]).
This practice cannot be explained from any consideration for etymol- [iii/37]
ogy because, if for some strange reason this would have been effective,
also
*
for ought to have been written. This practice must rather be
explained by a particular pronunciation of the vowel, namely words like
were not enounced with a pure but approaching (imla nawal-y),
thus an with or without a stroke above. This explanation is supported not
only by orthography but also by rhyme. If we consider the large number of
verses terminating in , , and it cannot be considered an accident that
only very few rhyme with .
176
These few cases, however, do not matter
very much given the well-known inaccuracy of the Koranic rhyme, which
takes various liberties.
177
Mecca) pronounce -h and -h respectively (chapter h al-kinya of the works on variant
readings).
172
But Nfi (according to parts of the transmission) and Ab Jafar Yazd IBN AL-QAQ
AL-MAKHZM (Medina) as well as Ibn Kathr and Ibn Muayin (Mecca)thus, precisely
the ijzi readersread the forms with final - (Itf, etc., on sra 1:6).
173
Nldeke, Zur Grammatikdes classischenArabisch, p. 14, 13. Cf. alsoJ. Barth, Die Pronom-
inalbildung in den semitischen Sprachen [the pronominal formation in the Semitic lan-
guages], p. 3.
174
An is considered a pausal formin Koranic reading as well as in Nfi b. Abd al-Ramn
al-Layth and Ab Jafar Yazd (Medina) before hamza (al-Dimy al-Bann, Itf, etc. on
sra 2:260). Therefore also the writing
, Berlin Ms., 338, sra 18:37.

175
in contrast to ( ).
176
Sra 18:12 (where the Damascenes do not assume the end of a verse); 65:6; 99:5 (where
the vowel occurs in the antepenultimate, where the rules of rhyme are less strict). Sra 20:125
would not even rhyme perfectly if it were pronounced . (Cf. below, p. 417sq).
177
In any case, the rhyme from to with a stroke above is easier than the ordinary rhyme
from to .
The imla is of great importance also in the recitation of the Koran but
there it does not merely follow the apportionment of the writings of and
.
178
This way of writing is changed only in the following instances: [iii /38]
(a) Occasionally, when a wal follows and the vowel is shortened so that
the difference disappears:
179
, sra 12:25 (disputed sra
40:18);
, sra 17:1,
180
, sra 28:19, and 36:19; , sra
69:11; further disputed , sra 55:54;
181
or before nunation
where the same happens as in
, sra 20:12 (although controversial).

But here as everywhere else orthography expresses the pausal form.
(b) When another precedes or follows (in order to avoid the repetition
of the same vowel letter), for example, , = ; never-
theless (as a name as well as a verbal form, in contrast to other
178
Cf. M.Th. Grnert, Die Imla, der Umlaut im Arabischen, particularly pp. 36sqq., and
84sqq. (Excerpt fromthe relevant naw 30 of al-Itqn, which, however, is by no means exhaus-
tive).Orthography is most strictly observed by the Kfans (amza b. abb, al-Kis,
al-Amash, andKhalaf b. Hishm); there are, however, considerable differences betweenthem
so that they go far beyond the limits of orthography in many ways, occasionally also failing to
attain them(cf. below, p. 418 n. 193, p. 417 n. 186 and 187). amza and al-Amash read ten per-
fect tenses of verbs II Inf. with imla. On the other hand, the imla of the Barans, Ab Amr
and [Ab Muammad] al-Yazd, [d. 202/817; Sezgin, GAS, vol. 9, pp. 6364] is based strictly
on (combinative) phonetics, conditioned mainly by an r neighbouring (similarly, going
beyond the orthographic imla, Ab Umar AL-DR [d. 240/854; EQ; Sezgin, GAS, vol. 1, p. 13
no. 11] an al-Kis and, to a lesser extent, according to some transmitters [Uthmn b. Sad]
WARSH, d. 197/812 [EQ; Sezgin, GAS, vol. 1, p. 11] an Nfi). The remaining readers have the
imla only to a lesser extent. The details of grammarians regarding the imla, whichvary con-
siderably, deviate very considerably from the practice of all the readers of the Koran. In this
field a particularly great diversity of dialects attests to this. (Cf. also K. Vollers, Volkssprache
undSchriftsprache imaltenArabien, p. 101 sq.; Chr. Sarauw, Die altarabische Dialektspaltung,
p. 31 sqq., particulary, pp. 3342, attempts to interpret the Koranic orthography with as
rendering ay and respectively, but it is the result of erroneous assumptions regarding their
distribution).
179
Recognized also in the recitation of the Koran.The different explanation of P.
Schwarz, Der sprachgeschichtliche Wert einiger lterer Wortschreibungen imKoran [philo-
logical value of select early orthography in the Koran], p. 52 (cf. above, p. 417 n. 134)that it
produces the effect of impeding imla in the preceding consonantsis untenable given the
great number of s after the identical consonant in the Koran. The explanation of the of
fromthe particle-nature of the word is in contradiction to the writing with in all the
other places. As far as I know, there is no word which is referred to in order to explain .
The spelling cited with reference to al-Naysbr, 1, 32, and 40, sra 43:7, in my opinion,
must be considered a mistake, copied by Schwarz from a manuscript like the Berlin Ms. 354
(see below, last line) since it appears nowhere else in literature.
180
Unlike Schwarz, loc. cit., only in al-Naysbr but also in al-Muqni, bb, 19.
181
So the Berlin Mss. 305, and 337.
forms of the verb like ) and
, sra 91:13 (because of the unin-

terrupted rhyme with ).
(c) The verb is written ; only sra 53:11 and 18 write . Some cases [iii/39]
can be explained with rule a (and thus it reads in sra
26:61 for ) but this does not apply to all, and for such reason one
must probably imply an assimilation of this verb to r,
182
which thus
frequently looses its .In the same way the verb
is changed to
in sras 17:85 and 41:51.
183
(d) In some isolated words: , sra 22:4, , sra 48:29, as well as
controversial , sra 5:57.
184
The Kfic manuscripts go one step further when replacing with ; so
Berlin Ms. 333 (sra 15:99); Gotha Ms. 460 (38:31); Berlin Mss. 323
(sra 8:51), and 362 (often), Gotha Ms. 458 (often);
185
Berlin Ms. 346 (sra
26:207); BerlinMs. 354 (sra 43:7); BerlinMs. 301 (sra 16:9), Ms. 305
(sra 2:181); Berlin Ms. 301 (sra 16:39); Berlin Ms. 346 (sra 27:36);
Gotha Ms. 447 (sra 14:6); BerlinMs. 333 (sra 17:14); Wright, Fac-
similies of manuscripts, plate 59 (sra 26:218); B. Moritz, Arabic palaeog-
raphy, plate 31 (sra 2:286); etc. This usage is to be explained by the prepon-
derance of another pronunciation over the transmitted orthography;
186
this
is later found particularly in manuscripts from the Maghreb.
On the other hand, only in isolated instances do we find where the [iii/40]
inflexion has , because from among the relevant cases cited in support of
this, the roots of
, of , sra 79:30, and of
, sra 91:6, also have

as a third radical, not only ; and in the case of
, sra 24:21, perhaps a

change effectedby
must be presupposed
187
so that only two cases remain,
, sra 91:2, and
, sra 93:2, both occasioned by the rule of rhyme.

In the middle of a word we find for regularly only in the foreign word

, where the explanation by means of a different vocalization cannot be

182
This is actually how al-asan al-Bar reads and, according to others, also (Uthmn b.
Sad) WARSH [Sezgin, GAS, vol. 1, p. 11] an Nfi (Medina). Cf. Itf on sra 12:28, and 27:10.
183
There are, though, the secondary forms ra and na (this also as variant reading).
184
Further mentioned sra 23:46, controversial as to variant reading and meaning,
andto rule a , sra 18:31, fromthe erroneous assumption that this is a feminine
form, ending in (in respect of imla, though, it is treated as such).
185
Several times in this fragment in Joh. H. Mller, Palaeographische Beitrge aus den
herzoglichen Sammlungen in Gotha, plate 12, no. 2 (1844).
186
As far as the particles are concerned even amza b. abb follows this pronunciation;
cf. M.Th. Grnert, Die Imla, p. 78sqq.
187
Although also the representatives of the most far-reaching imla exclude precisely this
word (a single opposite transmission regarding amza and al-Kis is rejected in Itf, s.v.).
excluded;
188
further, in

189
sra 3:27 (against sra 3:97, or );
190

,
sra 12:88 (Iraq).
191
The feminine ending is in these forms treated in anal-
ogy with the suffixes (see above, p. 415sq.). Added to this are two transmis-
sions not recognized by al-Dn: imal-Jadar reports fromthe Immthe
spelling for sra 4:3, and Ab tim al-Sijistn (d. 250/864) from
Meccan manuscripts the spelling for , etc.
192
Both orthographies agree
with the extended imlalikely unjustified in the Koranas advocated
among the readers of the Koran by amza.
193
The representation of by has as an alternative , but this is limited [iii/41]
to a few particular words. Since the grammarians explicitly state that the
pronunciation of these words is broader ( , ) and leaning to (imla
naw al-ww) in the ijz, we must assume that the vowel was here pro-
nounced like (a long) or . The words are the following: , ,
194
,
, sra 24:35,
195
, sra 40:44, and , sra 53:20,
196
as well as .
197
But
writing with applies only when the word has no suffix, whereas when it
does have one, the vowel is indicated by an or written defectively.
198
The Kfic manuscripts of the Koran occasionally restitute the , for exam-
ple, Berlin Ms. no. 352 (sra 39:27); Gotha Ms. no. 442 (sra 9:38), and Ms.
no. 446 (sra 14:3); , B. Moritz, Arabic palaeography, plate 33 (sra 8:3).
Fifth.Every final is followed by an .
199
The phonetic explanation for
this spellingthe expresses the broader soundof the final andawnot
only fails owing to the improbability of the identical phonetic appearance
in the case of the long vowel , and the diphthong aw, but more so because
188
Cf. the Aramaic form [vowels omitted by the editor], and similar forms (Schwal-
ly).
189
Which might also indicate the pronunciation taqiyyatan supported by YAQB b. Isq
AL-ARAM [EQ; Sezgin, GAS, vol. 1, p. 11, no. 6] (Bara) and al-asan al-Bar.
190
Although in the Berlin Ms. no. 308 ; but in Berlin Ms. no. 305, made to read .
191
So also Gotha Ms. no. 445.
192
al-Kis transmits this orthography also from the Koran of Ubayy [Ibn Kab] (likewise
for ).
193
Cf. above, p. 415sq., and Chr. Sarauw, Die altarabische Dialektspaltung, p. 35sq.
194
In these two words the vowel is likely to be influenced by the Aramaic basic forms
(Fr. Schwally); cf. Nldeke, Neue Beitrge zur semitischen Sprachwissenschaft
[new contributions to Semitic philology], pp. 25, and 29.
195
Ethiopic maskt (orignally probably mashkt), Nldeke, Neue Beitrge, p. 51.
196
Also Nabatan (Schwally).
197
Some transmit sra 30:38
(the only place with nunation; cf. above, p. 415). Regarding

see above, p. 395 n. 46.
198
Sras 9:104, and11:89, where is written, andlikewise sra 23:9( ) the common
reading is the plural.
199
The old Koranic orthography thus does not differentiate between , Sing. Ind. or
Subj., and Plur. Subj., and apocope.
this is also found in which ends in (long) - (-), in the subjunctives in
200
as well as in many forms with final hamza (see below, p. 422sq.).
201
Thus,
we are here dealing with a purely graphic phenomenonprobably a rem-
nant of a word divisionwhich survived particularly after and might be
explained from the fact that the importance had already occurred to Arab
grammarians that usually is to be attached to the following word.
Most of the exceptions to the rule can easily be explained: In , , , [iii/42]
sra 2:226,
202
the final is omitted because of the preceding one, in ,
sra 59:9, because of the immediately following ; in
203
because of the
elimination of the word division on account of its proclitic circumstances
in
, sra 25:23,
204
possibly because of the inconsistency with the
otherwise identically looking following word.
205
There remains only
,
sra 34:5, , sra 4:100,
206
and the disputed passage
, sra 33:69.
Sixth.The words with hamza pose the greatest difficulties. In this case
there was a time when the pronunciation frequently varied greatly, and
there were intermediate stages, from full glottal stop to its total loss, which
are difficult to comprehend. Further, there was no unambiguous sign avail-
able for cases when hamza was pronounced as a full consonant since the
actually designated increasingly lost its purpose because of its use as a
vowel letter. Thus, the clumsiness of the ancient Koranic scribes becomes
particularly apparent in the graphic representation of hamza.
When looking at the old orthography one must start from the principle [iii/43]
that only where is written a hamza may be added but that the existence of
an is no guaranty that hamza is pronounced. Looking at the old orthog-
raphy of the Koran from this point of view it follows that their authors
pronounced hamza only in very few instances, and that in many cases it
had completely disappeared, and in others, it had been replaced either by
or , or by respective intermediate sounds between these consonants and
hamza. This conclusionis incomplete agreement withthe statements of the
200
Where, though, the spelling is likely to constitute the pausal form in -.
201
The attempt of P. Schwarz, loc. cit., p. 54sqq., phonetically to explain this, sets out from
classical orthography and the insufficient details of al-Naysbr instead the old Koranic way
of writing.
202
And likewise
sra 40:84, Berlin Ms. no. 354, sra 62:11, and Ms. no. 1 in J.G. Adler,
Descriptio codicvm;
sra 63:5; ibid.

203
Conversely the plural with .al-Naysbrs statement, pp. 1, 31, and 40 (Paul
Schwarz, loc. cit., p. 56) that is written everywhere with except in six places stands
completely alone.
204
Berlin Ms., 345 .
205
P. Schwarz, loc. cit., p. 57.
206
But sra 2:238
, which Berlin Ms. no. 313 offers also in the other passage.
grammarians about the ijz dialect in general and the Quraysh dialect in
particular.
After vowelless consonants is omitted; it is thus written for
,
etc.,
207
for milun, milin, , sra 48:29, for
; and also after a diph-

thong, , sra 81:8,
208
for , and this again for al-mawdatu. That also
in the disappeared, and that the is thus nothing but a vowel letter,
is evident from the spelling in sras 12:2, and 43:2
209
(by omission of the
in the middle of a word because of the beginning of the nunation), to
which al-Dnprobably unjustlyobjects because of the strength of Iraqi
manuscripts.
210
As exceptionis listed =
sra 29:19, andfrequently, which, however, [iii/44]

as already al-Dn noted, represents rather the other variant readings ,
and
, sra 18:57, where between the diphthong and the heterogeneous

vowel possibly also another phonetic treatment occurred. Further, initial
is retained after the article for consideration for the form without article,
except in the enigmatic words , sra 26:176, and 38:12 (next to
,
sras 15:78, and 50:13), and =
.
211
Well-known is the disappearance of the vowelless hamza with auxiliary
prolongation of the preceding vowel; it is recognizable after a by the occa-
sional omission of the (see above, p. 413), after i and u by the spelling with
and (with a : , sra 19:75, for

thus pronounced ryan; with a : :

sra 33:51 [similarly 70:13] for , thus pronounced tw).
Hamza can occasionally disappear between vowels when they both have
an a sound. Thus are explainedalthough by no means unanimouslythe
transmitted spellings for
, sra 7:17 and often, , sra 10:7,

for
,
212
, sra 39:46, for
213
as well as partially for
,
214
and before for
, sra 55:24 (see above, p. 400sq.),

207
Sra 33:20 transmits also the spelling . Gotha Ms. no. 443, writes sra 10:94,
.
208
Thus here B. Moritz, Arabic palaeography, plate 30.
209
In the Berlin Ms. no. 305, we also find , to which only a later hand added an ; further
, Wm. Wright, Facsimilies and manuscripts and inscriptions, plate 59 (sra 27:1). also
in A.S. Lewis Sinai Palimpsest A (see below, Abschnitt c, p. 375) sra 41:2, Sinai Palimpsest B
sra 13:30. Ibn Kathr (Mecca) reads everywhere qurn (al-Dn, Taysr, etc. on sra 2:181).
210
has even disappeared in the space after aw in for , sra 7:94, Berlin Ms. 305,
although a later hand resurrected the common form.
211
Sra 72:9 has (according to Itf in some manuscripts only). Cf. in this respect also
or , which is cited by al-Zamakhshar on sra 53:51 as a variant to
.
212
So here also Ahlwardt, Verzeichnis, Ms., no. 325.
213
In addition for
, sra 4:118, Ahlwardt, Verzeichnis, Ms. 313.

214
So in Ahlwardt, Verzeichnis, Mss. no. 307, 1st hand (sra 10:60), no. 327 (same place),
no. 359 (sra 46:3), and no. 364 (sra 67:28).
further, the forms and treated above on p. 365. The strict spellings,
without exception, for (with the interrogative a-) and even
for

, sra 7:120 (and often), for

, sra 43:58, etc., permit also the
interpretationless likelyof the mere graphic omission of one .
The same uncertain explanation applies to the rendering of the combi- [iii/45]
nations i in for , and for , etc.
Generally, the orthography among different vowels corresponds to what
later became customary; the following exceptions must be noted:
After prefixes the initial remains unchanged; only ai- becomes ayi- (
, sra 37:84), as well as occasionally au to awu (
, sra 3:13). But

since in a number of passages, where also the interrogative particle would
be expected before and ,
215
only is written,
216
also the retention of
hamza ( for )
217
must likely have been possible. In other cases, is
found in the sound sequence ai only in ; -ia- becomes -iya- only
in the combination =
, which is considered one word. In sra 106:1 sq.

the consonantal outline was ; it is completely uncertain how this
is to be vocalized. Cf. below, p. 423sq.
In the combination i of Form III verbs, the hamza is omitted, for
example,
for
, sra 10:54. Here we are dealing with a far
advanced transition of Form III verbs into the inflexion of Form III verbs
Infinitive. This also determines the interpretation of forms like
(pro-
nounced probably , not iyu).
Also after or hamza remains unspoken, and has thus very likely dis- [iii/46]
appeared. An exception is , sra 30:9,
218
but not for , sra 5:32,
and for
, sra 28:76,
219
where it concerns only the word division after
a final .
220
The words ending in-u, -i, and -a (including nunation) ought to have
only an at the end; but they are frequently written in the nominative with
, and in the genitive with . For example, , sras 5:32 and 37, 42:38,
59:17;
221
, sra 10:16, , sra 16:92; , sra 20:130,
222
, sra 42:50
215
In several instances it is debatable whether this constitutes a question or a predicate.
216
Accordingly , sra 27:61 = , etc.
217
This actually occurs inAhlwardt, Verzeichnis, Ms. 349(sra 32:9), andinA.S. Lewis Sinai
Palimpsest, B (see below, p. 426sqq.) sra 17:52.
218
in Flgel is wrong.
219
So here also Ahlwardt, Verzeichnis, Ms. 348.
220
Already al-Suyu recognized this, al-Itqn, naw 76, fal 2, qida 3, at the end.
221
Tradition is ambiguous about some other cases. to be found in Ahlwardt, Verzeich-
nis, Ms. no. 313 (sra 5:32), Ms. no. 316 (same sra) Ms. no. 317 (sra 5:88), Ms. no. 355 (sra
41:28), Ms. no. 356 (same sra, and often), first Ms. in J. Adler, Descripio codicvm (sra 59:17).
222
Cf. sra 3:109, Ahlwardt, Verzeichnis, Ms. no. 308(for
); against sra 20:130the Berlin

Ms. 341 reads .
as well as the above-mentioned (p. 361 n. 160) . Here the proper orthogra-
phy of the medial letter (e.g., before a suffix: , sra 8:34, awliywuhu <
awliyuhu) is transferred to the final letter. In contrast, in the medial posi-
tion both, and , are occasionally missing; thus with suffix ( etc.)
inthe nominative as well as inthe genitive
223
everywhere, except inthe place
referred to, certainly in Iraqi manuscripts.
224
This probably purports to indi-
cate a contractible pronunciation awliyhum, etc., which is perhaps derived
from the accusative awliyahum < awliyhum (see above).
In order to render a final -u use is made of but also of its reverse, [iii/47]
:
225
, sras 2:268, 14:24, and 40:50, , sras 6:5 and 26:5, , sras
6:94
226
and 42:20, , sra 11:89, , sra 26:197,
227
, sra 30:12, ,
sra 37:106, and , sra 44:32, , sra 40:53, (for
) sra 60:4.
228
Here
the is to be considered a word divider (see above, p. 418sq.) that displaced
the vowel letter which is presupposed before the , whereas in the spelling
the opposite happened.
229
Thus, easily intelligible is the identical spelling (insteadof ) for the final
-au with short vowels in the verbal forms for ; , sra 12:85; ,
sra 16:50; , sra 20:19; , sra 20:117; , sra 24:8; , sra 25:77;
, sra 43:17; , sra 75:13; and in the nouns , sras 23:24, 27:29, 37:32
and 37:38;
230
and for , sras 14:9,
231
38:20,
232
and 64:5 as well as for , sra
38:67.
223
Thus , Ahlwardt, Verzeichnis, Ms. no. 349, sra 33:6.
224
al-Dimy al-Bann, Itf on sras 2 and 6.
225
This spelling also caught the attention of al-Zamakhshar; he discusses this with ref-
erence to sras 14:24, 26:197, and 30:12.The reciprocal delimitation of the spellings and
is not entirely certain since al-Dn in al-Muqni leaves it at the fact whether or not with
, without explicitly stating the order but, on the other hand, the Berlin manuscript of al-
Muqni itself is not reliable in its orthography. Al-Dimy al-Bann, Itf has through-
out, which according to the Kfic fragments is certainly wrong, at least in the case of
jazu.
226
So here also Ahlwardt, Verzeichnis, Ms. no. 314.
227
Regarding the orthography of sra 35:25 the details (al-Dimy al-Bann, Itf) vary.
228
In addition controversial sra 5:21, and so here Ahlwardt, Verzeichnis, Ms. 316 and
W. Pertsch, Die arabischen Handschriften, Gotha Ms. no. 437.
229
There is much to be said for this opinion, particularly since the writing for -i lacks
a respective conterpart .
230
Against Ahlwardt, Verzeichnis, Ms. 346, sra 27:29 , sra 27: 32 and 38 but .
231
So here Gotha Ms., W. Pertsch, Arabische Handschriften, no. 447 (also in J. Mller,
Palaeographische Beitrge, plate 7, no. 2). This is also the passage referred to by al-Naysbr,
pp. 1, 32, 35, not sra 9:71 (so P. Schwarz, loc. cit., p. 57).
232
According to al-Dimy al-Bann, Itf only in some manuscripts.
Completely according to rule is the identical group of signs in for [iii/48]
sra 22:23,
233
and in for
sra 4:175; the spelling here represents

simply the pausal formwith - and the usual alif of separationafter this final
vowel.
It is not quite as simple to explain some additional spellings of the inter-
vocal hamza that appear next to one anotherand in partially unexpected
sequence, and , respectively (or eventwo s). They canbe dividedinto
two groups.
(1) Prefixes ending in -a before initial written with , and this in both
cases, before a-:
, sra 27:21, for
, sra 9:47,
234
as well as
before i:
235
(for
), sras 3:138
236
and 21:35;
(for
) sras 3:152 and

37:66; and before u:
237
, sras 7:142 and 21:38;
238
, sras 20:74
and 26:49.
239
In all these cases we may assume an actual vowel lengthening
under the influence of the following hamza
240
although passing over to
some kind of i and , respectivelyas it seems to be metrically documented
in similar instances in poetry, but particularly in the Meccan Umar b. Ab
Raba
241
after prefixes or in the word interval before hamza.
(2) In the second group, on the other hand, it indeed seems to be a purely [iii/49]
graphic particularity when here the sound sequence aiin for
, sra
6:34; for
or reversed ia , also in the later orthogra-

phy; for
sra 51:47, for
, sra 68:6; and in many manuscripts

233
So also in the Imm according to im al-Jadar. Likewise in the manuscripts of Kfa
and Medina in the parallel passage, sra 35:30. In the two passages referred to, orthography
promptedsome of the Koranic readers tothe affectedpronunciationas accusative. According
to Ab Ubayd, already the grammarianand reader Ab Amr b. al-Al (d. ca. 154/770 [Sezgin,
GAS, vol. 8, pp. 5051]) supplied the correct explanation for the ; also al-Kis considered the
to be purely graphic. (al-Dn, al-Muqni, bb 5, fal 1.)
234
Controversial.According to al-Naysbr, 1, 32, 2, the Damascene manuscripts add
sra 2:220 (Schwarz, loc. cit., p. 50). Also (sra 10:17) Ibn Kathr considers to be
.
235
Not all the passages entirely in agreement.
236
So here Gotha Ms. no. 433; plate in J. von Karabacek, Ein Koran Fragment. In addition
(for ) sra 19:36, Berlin Ms., Ahlwardt, Verzeichnis, no. 339, left hand.
237
Not all passages entirely in agreement.
238
Others, though, read sarkum (from Form IV) or even .
239
In the Berlin Ms. no. 305, here without (in agreement with sra 7:121). According to
al-Zamakhshar, inversely the Imm, sra 75:1, has
for .
240
Paul Schwarz, loc. cit., p. 49sq.
241
Paul Schwarz, Der Diwan des Umar Ibn Abi Rebia, part 4, p. 174.

242
for
and for
sra 14:5or evena for
, sras 39:69
and 89:24 in Andalusian manuscripts allegedly modelled after the Medinan
muafis rendered or the order of the letters and is changed =
ai- instead aiin for
, sra 18:23
243
and the forms of the Imperfect
of :
244
, sra 12:87, etc.
245
Whether according to these spellings in , , , , , it
[iii/50]
is intended to express a simple - by means of or whether indeed - was
pronounced, can hardly be determined; in poetry the long first syllable does
not seem to occur.
The initial sound appears generally in the form it would have in the
isolated word; deviations from the principle are found only in isolated in-
stances.
246
It is the rule only after and , e. g.,
, sra 2:185.
In Koranic reading the treatment of hamza is by far the most difficult
chapter, complex as such, and heightened by the reciprocal relation to
the orthography of the Koran, going hand in hand with the phonetical
and purely graphic appearance. But the assumption that the presumed
242
al-Dn considers this spelling to be the less common, but inour extant Kfic fragments
it is the one that abounds, even in the singular: , Berlin Ms. no. 310 (sra 3:44) and Gotha
Ms. no. 446 (sra 13:38). On the other hand in the headings of the sras (i.e. without
preceding bi-) in the Berlin Ms. no. 305, as well as in others, is an erroneous generalization
of the old orthography (possible only after bi-). The orthography of the plural without is
extremely rare in true Kfic manuscripts; I find this only in the Berlin Ms. no. 305 ( , sra
2:37), and Gotha Ms. no. 446 ( , sra 14:5). In the somewhat later Berlin Ms. no. 303, sra
23:47 is corrected .
243
To this sra 16:37, Ahlwardt, Verzeichnis, Berlin Ms. no. 335; sra 4:62, B. Moritz,
Arabic palaeography, plate 44. According to Muammad b. s in al-Muqni, bb 5, fal 2, Ibn
Masd always wrote . I cannot find any reference in the literature on the variants for the
pronunciation of ; however, the Berlin Ms. no. 338, has [with a over each ]. Cf.
further below, p. 428sq.
244
To which there is the secondary form .
245
According to some, equally in Form X , sra 12:80, , v. 110. It is read
yyas, etc. by (Ab l-asan Amad b. Muammad b. Abd Allh b. Ab Bazza) AL-BAZZ
(d. 250/864) an Ibn Kathr (Mecca)disputed by some; equally according to some in the
waqf amza(see below, p. 425 n. 247).The alleged omission of and as the bearer
of hamza next to a second and respectively (in cases when not this other or can
be the one that was omitted because it is a full consonant) is based solely on the examples
mentionedabove onpp. 368 and369, andonthe spelling
, but where hamzais expressed

by the . In the singular
, and the same way in
sra 18:9,
sra 18:15 the

tradition differs only with these spellings and such with (
, etc.) the latter of which is

quoted by al-Ghz [Ibn Qays, Sezgin, GAS, vol. 2, p. 661] (p. 21) from Medinan manuscripts,
and ascertained by al-Sakhw (in-Itf, chapter marsmal-kha, section adhf al-y) from
the Damascene muaf.
246
As example serves , sra 18:95 insteadof
(sosra 12:59), and ibid.

vv. 9495. Althoughhere too it is read
, and it is probably this whichthe consonantal form

wants to express.
pronunciation based on the Koranic ways of writing is most likely to be
discoveredinthe ijz local qirt is less disappointing inthis chapter than
in its treatment of the final and , and the imla.
247
Seventh.Apeculiar formof writing is which ought to be read
[iii/51]
in sra 12:110 likely so, and in sra 21:88, safely so; added to this, according
to some statements, are the spellings for , sra 10:15, and for
, sra 40:54 (Medina), which, however, al-Dn did not find in any
manuscript;
248
cf. also the variant to sra 25:27, above, p. 398. Inthese cases it
is unlikely tobe simply a carelessness inwriting but a dissimilative reduction
of the second n preceding a consonant.
249
Eighth.The of the article is surpressed not only in all the relative [iii/52]
pronouns (and this in all forms)thus , etc., , = but also
in for ;
250
in addition probably for
(see above, p. 397).

The connecting alif is dropped: in the frequent formula (for );
after
before an article (e.g., for )similarly after
in
,
sra 18:76
251
after the interrogative particle
(e.g.,
sra 2:74); in

etc. For
,
252
and in the above-mentioned spellings (see above,
247
Of course, most perfect to the orthography of the Koran fits the one of the two sys-
tems of the pausal pronunciation going back to amza b. abb (Kfa) which received
its designation rasm precisely for this reason. How unnatural this system is can best be
seen from the endeavour to represent the entire consonantal outline, including the purely
graphic peculiarities by the pausal pronunciation. Also the second system of pausal pro-
nunciation (tarf)which alone Ab Amr AL-DN considers in his Taysr (f l-qirt al-
sab)is still close to the Koranic orthographyless so the systems of Hishm b. Ammr
al-Sulam (153/770245/859) [Sezgin, GAS, vol. 1, pp. 111112] an Ibn mir (Damascus), and
al-Amash (Kfa.) All these systemstreated in the works on variant readings in the chapter
entitled madhhab amza wa-Hishm f l-waqf al l-hamza or similarlyrepresent a com-
promise between the consonantal outline and the actual pronunciation of the respective
readers, whereby with the help of the greater licence which pause accords, the former is
helped to its right. Warsh an Nfi (Medina) is the one who most strictly observes the actual
(darj)-pronunciation, and follows the true phonetic peculiarities of the text of the Koran; he
is the only one who recognizes the disappearance of hamza after a consonant; in some other
details he is still surpassedby another Medinanscholar, AbJafar IBNAL-QAQ. Apart from
this, a mitigation of the hamza occurs in [Ab Ms s b. Min b. Wirdn] QLN[EQ; Sez-
gin, GAS, vol. 1, p. 12, no. 8] an Nfi (Medina), Ibn Kathr (Mecca), and, most of all, Ab Amr
b. al-Al, and Ab Muammad al-Yazd (Bara), who distinguish themselves by the greatest
abandonment of vowelless hamza.
248
So far they have not been identified in the extant Kfic manuscripts either.
249
Similarly P. Schwarz, loc. cit., p. 48.
250
The Berlin Ms. no. 301 [Ahlwardt, Verzeichnis] has exceptionally (sra 16:12).
251
In which case, though, some read
(from a secondary verb ).In Ahlwardt,

Verzeichnis, Berlin Ms. no. 331 also for , sra 13:17.
252
The transmitted spelling naturally indicates the pronunciation wa-sal.
p. 360), and (see above, p. 420), (cf. see above, p. 420sq.
253
),
(see
above, p. 424);
254
but not dropped in ( ).
Ninth.There are some irregularities to be found in the orthography of
the sibilants where on account of assimilation is replaced by ( ;
, sra 2:246, and , sra 7:67;
255
, sras 52:37 and 88:22), or that
and are mixed up ( , sra 81:24, for ).
256
Tenth.Finally, it must be noted that although the somewhat changed
pausal pronunciationof the rhyme is mostly also expressedinthe script, like
in , sra 33:10, v. 66, v. 67; , sra 69:29, etc., but not
always (e.g., , sra 74:15, azd).
If we add to all these spelling possibilities the fact that the script did not [iii/53]
generally
257
differentiate between in the initial and medial
positions, andbetween at the endof a word, betweenandinthe
initial and the medial positions, , and , and , and , and
, and , and , and that, moreover, next to no signs assisted in the
pronunciation, and that the words were insufficiently separated from one
another,
258
we come to the conclusion that this graphic representation of
the Holy Text was still extremely imperfect.
The Orthography of Agnes S. Lewis Sinai Palimpsests
The orthography of Lewis Koran palimpsests
259
requires detailed treatment
since, by and large, it corresponds to the transmitted orthography of
253
Ibid.
254
Further, , sra 27:25, if you read previously
(instead
) and then consider

to be . According to a note in Nashr, the form with connecting alif, , is
said to have been in the Imm, and in Damascene manuscripts, and was fabricated solely in
order to support this variant (cf. al-Dimy al-Bann, Itf, s.v.).
255
So here also the Gotha Ms. no. 441.
256
Ab Ubayd al-Qsim b. Sallm in al-Dimy al-Bann, Itf, s.v., points out the insig-
nificance of the difference in the Kfic script.
257
But as final letter not only and but also and have their individual shapes that
cannot be confused with one another.
258
Kfic writing does not leave a larger space between words as is customary between the
average ligatures of the same word so that only the sense can tell whether one ought to read
or or .
259
Leaves fromthree ancient Qorns, edited by A.S. Lewis and A. Mingana. The leaves con-
sist of three groups, A, B, and C, which according to their style are uniform(A and B possibly
not entirely). According to the likely accurate opinion of the editors, A and B are approxi-
mately of the same period, B somewhat later. However, the editors misjudged the charac-
ter of the orthography of their textswhich could have easily been accurately determined
had they consulted Nldekes Geschichte des Qornsas well as many details. In addition,
the Koran
260
(particularly its sparse use of the prolongation of the alif which
likely goes beyond the literary transmission and the Kfic Korans)
261
but
still shows a number of deviations which are listed here in the order of the
section on the orthography of the Uthmnic Korans.
II. The writing in one word goes considerably further, namely words with [iii/54]
final -m are preferably connected to the following word (A: sras 41:40 and
41:41, C: sra 7:146 and 7:155, and often), once even and still attached to the
preceding word (A: sra 41:38).
262
On the other hand, at the end of a line, even after letters that potention-
ally can connect two letters, the word can be separated; thus in A after ,
sra 45:12; in B after , sra 11:33 (?); after , sra 13:36; after , sra 16:17;
after , sra 16:26; after , sra 17:16; in C after , sras 7:153, 154, 156, and
9:37; after , sra 9:72. That the phenomenon is relatively rare in A is occa-
sioned partly by the fact that here many ends of line and beginnings of line
are illegible. But alsoinother respects, the three groups display considerable
differences.
III. As is the case with , for no apparent reason also and are occasion-
ally missing as vowel letter: Thus, A: , sra 44:44, for ; B: , sra
15:91, for ; , sra 16:92, for ; C: , sra 7:154, for ; , sra 7:154,
for ( );B: , sra 16:12, for ; C: , sra 9:26, for .
263
IV. Instead of before a suffix we frequently encounter , which later
became accepted practice (or, what amounts to the same, entirely without
a vowel letter); thus, A with , sra 44:56, without vowel letter, sra 29:23; B:
with alif, sra 11:29 and 30, and without vowel letter, sras 13:18, 14:6 and 17:14
(but with sra 11:31, 11:37 and often); C: with alif, sra 7:139, and without
vowel letter, sras 7:156 and 9:51 and often (so also , sra 7:156); occa-
sionally also at the end instead of ; so A:
, sra 41:5; B:
, sra 16:94.
264
some of their readings are extremely unlikely. The edition needs expert verification. The
use of brackets, square brackets or none at all when referring to the three groups has been
retained from the German edition.
260
Also the writing of sra 16:14 (B) for
and sra 9:24 and 9:37 (C) for , both

with connecting alif, fit here; see above, p. 409.
261
E.g., in group B, where still dominates, sra has 13:38 , where the transmission
demands ; although conversely sra 15:87
and sra 17:14 . C always writes

contrary to the rule , e.g., sra 7:157 and 9:23.
262
Conversely B , sra 13:40, for
.
263
The alleged spellings , sra 13:43 (B), and , sra 9:72 (C) whichincontrast ought
to have a superfluous still need to be confirmed.
264
Conversely B: , sra 17:1, where it ought to be .The apparent imla
in the middle of the word in
, sra 16:82 (B) for
( ) and , sra 9:64 (C) for

( ) where, as far as I know, no reader reads with imla, seems as equally incredible as
the spellings mentioned above in note 263.
V. The alif of separation is more frequently missing before a word that [iii/55]
begins with alif; thus A: sra 24:22; B: sra 13:19; C: sras 9:34 and 9:68.
265
VI. The hamza remains still more frequently unmarked, namely in the
middle of the word also in the connection ai ( for sra 24:25 [A];
16:89[B], for sra 29:22 [A], , sra 13:28[B], and , sra 16:108
[B] for
and
), i ( , sra 41:13 [A] for , , sra 16:9 [B] for

); ia ( sra 13:22 for , , sra 17:40 [B] for ), i ( , sra
45:20 [A], 16:36 [B], 7:152 [C] for
); inadditionalso u in , sra 16:88

[B] for , cf. above, p. 421 sq.
266
Not all of the peculiarities of the hamza orthography are repeated; it is
written , sra 16:92 [B] for , not ;
, sra 7:142 (C), not ;

, sra 9:47 (C), not ; , etc. A sras 29:22, 41:14, and often, B
sras 14:5, 16:106, and often, C sras 7:144 and 7:155 for
, not .
Conversely, the spelling for
, which occurs only once in the trans-

[iii/56]
mitted orthography of the Koran, is here the rule(A: sra 29:19; B: sras
16:91, 17:13, andoften; C: sras 7:142 and7:155; only Bsra 16:37)whereas
for , sra 13:30 (B)
267
(e.g.
) is written.
The context form at the beginning of the word also
, sra 41:10
(A: for
).
268
VII. A simplified writing of two ns is , sra 41:4 (A) for
, reproduc-
ing the pronunciation dhnn, which is represented by al-Muawwi an
al-Amash (Kfa). Corresponding abbreviations can be found in the Uth-
mnic consonantal text as variants to sras 18:94, 27:21, and 39:64; see above,
p. 397sq.
VIII. B writes in sras 16:12 and 17:13 .The connecting alif is several
times ostentatiously omitted: A,
, sra 44:40; B, , sra 11:24

(for
).
269
A quick glance will indicate that here we are apparently dealing with a
mixture of discrepancies in the Palimpsests from the orthography of other
oldKorans, displaying traits of great antiquity interspersedwithquite recent
265
In addition irregularly once B , sra 13:36.
266
In addition perhaps after the article in , sra 11:24 (B) for
(); see below,
p. 428 n. 268.
267
That this spelling is very oldis evident fromthe transmissionreferredto above onp. 341,
according to whichyayas is a mistake for yatabayyan, whichpresupposes the graphic outline
for the former.
268
The identical graphic outline is presupposed by al-Zamakhshars reading
; per-
haps this applies also in this case.
269
If it is not rather a case of a variant:
; in support of this is perhaps the fact that,

unlike the previous example, the m is written together with the second word.
ones, the most ostentatious being the preference for
instead in a medial
position, corresponding entirely to later orthographic usage. In this case we
would be dealing with a text that is not really old but has adapted some
antique habits, probably fromnon-Uthmnic secondary sources. Fromthis
point of view, however, it remains inexplicable that particularly the more
recent group B displays a decisive retreat of
in favour of . This makes

sense only if AC represents an old parallel text to the Uthmnic text, which
is already incorporated in B. Thus, we here find the converse of the devel-
opment also in the field of orthography, which we shall encounter later
repeatedly not only in the field of the variant readings but also in the field
of orthography.
There is one peculiarity that points in the same direction as these con- [iii/57]
siderations, namely the frequency of the almost exclusive writing of in-
stead of ,
270
which has been documented precisely in the Koran of Ibn
Masd (see above, p. 424 n. 243). But it is hardly conceivable that the
Palimpsests belong directly to this recension. This, at least, is the result of
the examination of the variants. We must conclude that Ibn Masd is not
the only one withthis orthography so that inthe Uthmnic Koranis not
an accidental peculiarity, rather it is the casual adaptation of a rarely used
but otherwise quite common orthography.
When we thus arrive at an early origin for the Palimpsests we must con-
sider variants like
for to be the influence of a somewhat different dialec-

tical basis or, at least, a different interpretation of the identical dialectical
basis. The same conclusionwill apply tothe disappearance of hamzainwrit-
ing; age alone is no explanation.
The Non-Uthmnic Orthographic
Variants and Readings: The Sources
We have seen that the Uthmnic text itself was not completely uniform
because of the multitude of equally recognized copies; it contained already
a number of variants, even a still greater variety of orthography, and at least
the latter could easily multiply by copying. In spite of this, the Uthmnic
text represents a relatively compact unity in view of the considerably large
number of consonantal variants or variant readings which presupposed the
existence of a divergent consonantal text.
270
It is typical that again occurs only in B.That the absence of the Uthmnic writing
is no indication of the age has already been determined above, p. 428 n. 267.
The earliest extant collection of such variants is contained in the sec- [iii/58]
tion, Bb al-zawid min al-urf
271
allat khlifa bi-h l-kha of Ab Ubayd
al-Qsim b. Sallm (d. 223 or 4/838) in his above-mentioned K. al-Fail
al-Qurn (Berlin Ms. no. 451, fol. 37
r
sqq.). It contains slightly more than
one hundred variants and variant readings that depart from the Uthmnic
consonantal text. In strictly traditional habit, nearly all of them are traced
back to Companions, and thus partly to the Prophet. The copies of the
Koran of Ibn Masd and Ubayy b. Kab are frequently mentioned, with-
out being favoured above the other authorities. The collections of tradi-
tions,
272
although they include special sections on variant readingsor at
least commentaries of the Korancontain relatively little pertinent mate-
rial andapart from their more or less rigid critical standardsrecognize
only what has been ascribed to the Prophet and, less so, to his most promi-
nent Companions, particularly Umar. Al-abar, on the other hand, in-
cluded a great deal of such material in his commentary on the Koran.
From among the later commentators it is al-Zamakhshar
273
who stands
out with his extraordinary mine of relevant details, which offers far more
than he could have adapted from al-abar, whom he undoubtedly utilized.
It is likely that Ab Ubayd al-Qsim b. Sallm also resorted to an older
shawdhdhwork; (see more details, below).
274
Inso far as the non-Uthmnic
readings belong to closed systems of Koranic recitation of which we have
descriptions, they represent a valuable addition to the data recorded by al-
Zamakhshar. Isolated references to non-Uthmnic variant readings are to
be found not only in many passages in the literature of traditions, commen-
taries, grammar, and lexicography,
275
but also in relatively extensive refer-
ences in works on variant readings that are not limited to listing them, but
271
The list is not at all limited to additions to the Uthmnic text.
272
Cf. the summary in the chapter on the variant readings in al-Muttaq, Kanz al-umml,
vol. 1, p. 284sqq., where are consideredapart from the literature of tradition, and also
al-abars commentary on the Koran as well as several special relevant writingsfirst of
all al-Fail of Ab Ubayd al-Qsim b. Sallm, and the maif-books of Ibn Ab Dwd
al-Sijistn, and Ibn al-Anbr (see above, p. 405 n. 110, and p. 389 n. 2).
273
Being a Mutazilite, he seems to have listed with a certain purpose extra-canonical
variant readings. Unfortunately, his statements are by no means complete, neither as far as
the variants are concerned, nor the names of the readers. The lacunae increase in the second
part of his work.
274
Most likely from the one of Amad b. Ms IBN MUJHID [Sezgin, GAS, vol. 1, p. 14]
because onsra 19:85 he refers to Mutasab (thus, not Mutasib, the only Ms. inConstantino-
ple, Ragb Paa Ktphanesi, no. 13) of Uthmn IBN JINN [EI
2
; EQ; Sezgin, GAS, vol. 9,
pp. 173182], to whom he frequently refers. Whether this might be the sole source can be
determined only after scrutiny of the manuscript.
275
I did not list the material scattered throughout this literature, except the Mufaal
are searching for their underlying causes, and for this purpose utilize also
non-Uthmnic variant readings that concur with them.
276
For the first sra
some passages are added, that serve as a model for a rich apparatus of vari-
ants.
277
The Text of Ibn Masd
278
[iii/60]
Among the mass of non-Uthmnic variants and variant readings there
are above all two groups that deserve a separate treatment, namely those
ascribed to Abd Allh IBNMASDand Ubayy b. Kab, because we have the
reliable transmission that both men had their own collection of the Koran
(muaf; cf. above, p. 234sq.). The question therefore poses itself whether
those variants and variant readings
279
ascribed to them might derive from
their very own recensions of the Koran.
In order to get at the core of this question it is necessary first of all
to list
280
the transmitted variants and variant readings in so far as they
differ
281
from the Uthmnic consonantal text, and briefly explain them. We
begin with those of Ibn Masd.
of both Sibawayh and al-Zamakhshar. Whatever I collected offers nothing new from the
commentaries or works on tradition, and variant readings.
276
From among them I utilized mainly the Kitb al-Kashf an wujh al-qirt wa-ilalih
wa-ujajih of Makk b. Ab lib al-Qays (d. 437/1045; Brockelmann, GAL, vol. 1, p. 406) in
Ahlwardt, Verzeichnis, Berlin Ms., 578 (Petermann, no. 17). It represents a commentary on the
same authors Tabira; [edited at Damascus, 1974, by Muy l-Dn Raman].
277
Makk b. Ab lib al-Qays, Man l-qirt (appendixed to his Kashf an wujh al-
Qurn), Berlin Ms., Ahlwardt, Verzeichnis, no. 578, p. 517sq.; al-Jazar, al-Nashr f l-qirt
al-ashr; Berlin Ms., Ahlwardt, Verzeichnis, no. 657, fols. 18
r
19
r
.
278
About him, cf. above, p. 235sq.; and also A.J. Wensinck in E.I., s.v.; Goldziher, Schools of
Koranic commentators, p. 5; Jeffery, Materials for the history of the text, pp. 114116; Juynboll,
Encyclopedia, pp. 78; Caetani, Annali, vol. 7, year 32, nos. 126144.
279
No differentiation is here made between these two types since also the variant readings
if, indeed, they can be traced back to Ibn Masd and Ubayy b. Kab, respectivelymust
have been in their respective copies (the later separation of writing and reading cannot be
presupposed in this case) thus being true variants at the same time. Tradition usually does
not differentiate, it rather uses completely synonymous expressions like f qirat Ibn Masd,
f arf Ibn Masd, f muaf Ibn Masd, and similar ones, referring to himand Ubayy b. Kab.
280
This list and the following list are based on al-Zamakhshars Kashshf; other lists
are explicitly referred to as the case may be, but, conversely, the absence of a variant in
al-Zamakhshar is particularly mentioned.I think I can offer the material of the Kashshf
pretty comprehensively. I cannot dare claim to have exhausted al-abar with any degree of
completeness. Other commentators have not been systematically consulted.
281
Whereby also those deviations have been omitted that agree with the Uthmnic text
not lafan, but as the scholars of the Koran say, taqdran, and this in the way that thanks to
the liberties of the old Koranic orthography an acceptable spelling for both variant readings
could become possible.
Sra 1:5 : ;
282
| 1:6 :
.
Sra 2:19 : , cf. 2:153
283
(and similarly, 180):
(besides vari-
[iii/61]
ant reading
):
; 2:222, : ;
284
6:125, : ; 4:142,
: ; then also 7:170, :
(Imper.), but cf. 10:25, :

. In this case the Uthmnic text permits everywhere the variant read-
ing as reflexive (

, etc.), which Ibn Masd demands; thus, his

writing is clearer. This explanation, however, is not entirely satisfactory;
it is impossible in the case of the parallel variant 2:22 : ; and
even more so in the reverse relation, 23:103, :

285
| 2:29 ,
namely the objects authenticated by name ( ): ,
286
referring to
; smoother, but wrong, thus unwarranted correction. | 2:34,
or
similar: ; explaining. | 2:46, : . | 2:58, : ,
287
thus,
the literary form of the word for the dialectical or individual form. | 2:58,
,
288
the correct form. | 2:63, : . | 2:77, : , clear
prohibition. | 2:94, : ,
289
better known synonym. | 2:98
: .
290
|
2:100,
:
291

plain text and easy to understand of the very controversial passage. |
2:113, : , to prevent the vocalization
.
292
| 2:122, :

,
293
(namely the descendants, in the
mouth of Abraham and Ishmael). | 2:126, : then added
as intro-
duction of the direct speech, likewise 79:17; conversely
of the Uthmnic
text omitted, 41:30,
294
55:7, 68:24, 71:1.
295
| 2:131, : ,
296
dogmatic correc-
tion (Goldziher, Schools of Koranic Commentators, p. 16). | 2:153,
;
297
282
Also appendix to Makk b. Ab lib, Kashf an wujh al-qirt; Ibn al-Jazar, Nashr f
l-qirt al-ashr; etc.
283
Also al-abar, juz/vol. 2, p. 30, l 19; al-Dimy al-Bann, Itf.
284
Also Makk b. Ab lib, Kashf, al-Dimy al-Bann, Itf.
285
In addition also sra 16:39
: = , one of the attempts to re-interpret this

passage (cf. below, Ubayy b. Kab, s.v., whose correction produces the same artificial sentence
construction as well as the variant reading
).
286
Also al-abar, juz/vol. 1, p. 167, l 5.
287
288
Also al-Dimy al-Bann, Itf.
289
Also al-abar, juz/vol. 1, p. 333, l 22, from Ab Ubayd al-Qsim b. Sallm.
290
291
So al-abar, juz/vol. 1, p. 359, l 16; al-Zamakhshar has only the first part of the sentence.
292
293
294
Also al-abar, vol. 24, 67, 13 [sic].
295
Also al-abar, vol. 29, 49, 35 [sic].
296
al-abar, vol. 1, p. 422, l 16 only in Ibn Abbs, whom also al-Zamakhshar mentions.
297
Also al-abar, vol. 2, 29, 8.
correction to safeguard the former licence against the slowly penetrating
opinion that the ceremony of say between al-af and al-Marwa is indis-
pensable. | 2:172,
,
298
elimination of the odd construction. | 2:183,
:
.
299
| 2:194, behind added ;
300
illustrative
addition. | 2:209, after added ,
301
explanatory addition going back
to the parallel passage from sra 10:20 (Goldziher, Schools, pp. 78). | 2:214,
before added ;
302
repetitionof the prepositionbefore the appositionto
determine the grammatical relation, cf. the quite frequent repetitions after
: 2:239, 4:1 and 5:62 (see below), and cf. 51:46, : , after v. 43. |
2:226, :
; equally perfect instead of imperfect after 3:20

303
( for
the second , where the past tense is substantially more appropriate).
7:169 (see below), and after
79:36. | 2:226 after added ,

304
explana-
tory addition. | 2:229 : ,
305
parallel to the following ; smoothing
correction. | 2:241 :
,
306
elimination of
a hard ellipsis by shortening rearrangement | 2:261,
: .
2:261,
, the standard form (cf. above, p. 395 n. 46). | 2:269,

:

,
307
the original and literary form. | 2:276 after
complementary addi-
tion, .
308
| 2:281,
, necessarily to be read as passive. | 2:282,

:
309
, clearly apocope. | 2:285, : , following preceding 3. pers.
plural.
Sra 3:1 :
310
(i.e. .) | 3:5, :

311
in
[iii/63]
order to prevent that the following is contracted as a second subject
to . | 3:16,
312
(before direct speech.) | 3:16, :
313
a l very difficult
to accommodate. | 3:43 :
314
with better attachment to the
preceding and the clearer ruling (inflate). | 3:44 after added
298
Also Makk b. Ab lib, Kashf.
299
300
Ab Ubayd; not al-Zamakhshar.
301
Probably also means abar, vol. 2, 188, 30.
302
303
Also al-abar, juz/vol. 3, 132, 28; Makk b. Ab lib, Kashf.
304
By Ab Ubayd apparently rather ascribed to Ubayy b. Kab.
305
Also al-abar, 2, 261, 22.
306
Cf. also al-abar, 2, 338, 16.
307
308
Ab Ubayd al-Qsim b. Sallm; not al-Zamakhshar.
309
al-abar, 3, 83, 15; al-Zamakhshar without names.
310
So, explicitly as writing, Ab Ubayd; al-abar, 3, 101, 8 . Not in al-Zamakhshar.
311
312
313
314
al-abar, 3, 173, 2, leaves it open whether Ibn Masd or Ubayy (f id al-qiratayn).
as explanation , and after added (cf.
Goldziher, Schools, pp. 78). | 3:75 : ,
315
whichmuchbetter
fits the context, which for the prophets would result in an unfounded
criticism. | 3:86 : , clearer. | 3:114 : , immediately preceding
feminine; conversely, in the identical case feminine instead of masculine of
the Uthmnic text, 68:49. | 3:127 : . | 3:165
: ,
316
clearer;
the same 8:19,
317
and cf. 6:154 (see below). | 3:169
, whereby the
misunderstanding shall be avoided that
shall be frightened (instead

of becoming the object of frightening). | 3:177 [the letter after has two
dots underneath and one above] : (and evidently before,
) so that
Allh is not directly the subject; similarly 50:29.
Sra 4:19 : , after the more ordinary construction; cf. 6:57 [iii/64]
:
318
after
= (whereby at the same time excluding the reading

), 27:84
[anna or inna]:
after (whereby at the same time the

conjunction is identified as , and the verb as II. Form = to speak), and
49:2 : after . | 4:38 : sound plural. |
4:38 after the second added , complement to the ellipsis
parallel to the Imperfect of the second half of the verse. | 4:44
: , better
known synonym. | 4:56 : ,
319
after
. | 4:81 after added

,
320
to mitigate the rather strong indeterminate expression
from yourself. | 4:102 missing,
321
the following
than = so that
not. The when you fear of the Uthmnic text might be from the parallel
passage 2:240. | 4:109 : : unwarranted change in favour of reference to
a particular person. | 4:127 : ,
322
whereby the governing becomes
definite by
. | 4:134 :
, after which would otherwise

be without subject; similarly 10:2 : . | 4:160

:

,
323
correction
of a famous mistake (see above, p. 390sq.)
Sra 5:2
, before object and genitive, respectively; conversely,

[iii/65]
19:94
, 22:36

:

. | 5:3 : ,
324
after
[alif with both,

fata and kasra,] this as
identifying (
= therefore that would require

315
See above, p. 390, n. 14; Mujhid b. Jabr refers to Ibn Masds reading.
316
317
al-abar, 9, 131, 12, rather the equivalent
.
318
Also Ab Ubayd.
319
Cf. Ubayy on 17:78.
320
Ab Ubayd (the manuscript has thus at best also might be read); not
al-Zamakhshar.
321
Probably by mistake ascribed to Ibn Masd; al-abar, 5, 144, 7, rather Ubayy.
322
Makk b. Ab lib al-Qays, Kashf; not al-Zamakhshar.
323
324
perf.) | 5:4 : , eliminating the grammatical objection, which is
occasioned by the passive fal with feminine termination; conversely 5:69
: ,
325
the more idiomatic and possibly more original expres-
sion. | 5:42 collective singular : ,
326
cf. 6:155
: ,
327
here at the same time excluding the true sin-
gular and elative. | 5:42 :
,
328
more closely identified (Goldziher,
Schools, p. 10). 5:60
: ; conversely, 47:12 : .
329
| 5:62 :
,
330
correction because unbelievers would include once more
also the aforementioned People of the Book. | 5:65 :
,
331
pick-
ing up the relative and, as in the preceding words, constructed according to
sense. | 5:73
: , whereby eliminating the serious grammatical objec-

tion of the verse (see above, p. 390sq.). | 5:91 after added ,
332
better
defining and settling a controversy (cf. Goldziher, Schools of Koranic Com-
mentators, pp. 910). | 5:96 : ,
333
clearer and linguistically smoother. |
5:114 (as ifa) : (as concluding sentence to imperative).
Sra 6:16, after added ,
334
clearer than
without subject
[iii/66]
or passive
. | 6:70

, only by artificial explanation:

.
335
| 6:94
after added , whereby the hard substantival usage of the following
is elided. | 6:105 : ,
336
which excludes the interpretation of
the much disputed word as 2. person. | 6:140 , vaguely congruant with
instead of with : .
337
| 6:154 [ or] :
,
338
rejecting the reading , andcorresponding tocontinuing in3. person.
325
Also Ab Ubayd al-Qsim b. Sallm.
326
Also al-abar, 6, 133, 1 [sic].
327
Also Ab Ubayd al-Qsim b. Sallm; al-abar, 8, 61, 20.
328
Also al-abar, 6, 133, 2sqq.; al-Zamakhshar, Mufaal, 233. The Cairo Kashshf has
, probably an unwarranted correction.
329
Both also al-abar, 26, 27, 31 sq.
330
Also al-abar, 6, 166, 27, from Ab Ubayd; not in his Fail.
331
al-abar, 6, 169, 3, but without , identical with the text of Ubayy b. Kab.
332
Also al-abar, juz/vol. 7, p. 19, ll 11 sqq.; cf. Ab Ubayd: Ibrhm (al-Nakha), see at the
end of the section.
333
334
Makk b. Ab lib al-Qays, Kashf; not al-Zamakhshar, where only Ubayy b. Kab
maintains the variant, who is mentioned in the Kashf.
335
Ab Ubayd; al-abar, 7, 142, 18sqq., among others also Ab Ubayd; not al-Zamakh-
shar.
336
al-abar, 7, 189, ll 13, 21, here from Ab Ubayd (thus to be read); but next to it l 16,
where the same opinion is maintained, but which is in agreement with the reading
of
the Uthmnic consonantal text. Not al-Zamakhshar.
337
338
In al-abar, 8, 60, 16sqq.; Ibn Masd, however, cites the verse in the usual form.
Sra 7:19,
: , dialectically different. | 7:25, missing before ;

it is redundant and annoying since it is immediately repeated; its dele-
tion was quite obvious. Cf. the deletion of the first 23:37,
339
of the first
33:49,
340
and the restituting 62:8. | 7:38 after added .
341
| 7:38
: . | 7:103 missing before
; after it does not fit the con-

struction, a difficulty which it was tried to avoid by the vocalization
. |
7:169 :

.
Sra 8:1 missing;
342
it is then intended that
is the sub- [iii/67]

ject of
but the word apparently does not have this meaning. In

reality
is likely to be meant; in this case the Uthmnic text quali-

fied this rather hard construction by inserting , without changing the
meaning. | 8:2 : ; common synonym. | 8:39 : the punctuation
of the preformative must be changed accordingly.) | 8:61 before add
,
343
which after superordinated could easily be inserted before per-
fect.
Sra 9:51 : . | 9:107 : , whereby the preceding
prevails over . | 9:111 : ,
344
stronger. | 9:118 /
: ; the Uthmnic text is a moderation!
Sra 10:12 (active or passive): (clearly active, although hardly
fitting the preceding subject ). | 10:81 : ,
345
as predicate more
common. 10:98 :
,
346
whereby the former is used as explanation in the
commentaries.
Sra 11:30 see in Ubayy b. Kab. | 11:60 : (continuation of a [iii/68]
conditional subordinate clause beginning with ). | 11:74 after addition

347
(cf. Goldziher, Schools, pp. 89). | 11:75 : ,
348
this is what
makes more sense. | 11:83 missing is ;
349
the sentence might have
come here fromthe parallel passage 15:65. | 11:113
, very difficult to explain:

. | 11:113 : more unequivocal and common, .
339
340
341
Ab Ubayd, and al-abar, 8, 121, 4; not al-Zamakhshar.
342
Also al-abar 9, 110, 19sqq.
343
344
345
346
Also al-abar, 11, 111, 11; the sentence that follows there, and despite the similar intro-
duction, cannot be intended as a variant but only as an explanation.
347
al-abar, 12, 41, 21 .
348
Sbawayh 119; al-Zamakhshar without name.
349
Also al-abar, 12, 53, 24sqq., from Ab Ubayd (not in his Fail).
Sra 12:31 , which as predicate of is suspect because of its accusative
which is seldom found in the Koran: ;
350
cf. the other passage with accu-
sative (others, nominative) after 58:2 : . | 12:35 :
351
con-
sidered a dialectical form of the Hudhayl (Ibn Masd was a Hudhal) but
appears at this place evidently under the dissimilative influence of the fol-
lowing . | 12:36 : ,
352
after more obvious. | 12:64 (and

respectively) : ,
353
likely an unwarranted assimilation to the
following . | 12:70 : . | 12:105
: but then preced-

ing required; the inclusion of the other verb safeguards this reading
which was purported also in other ways.
Sra 14:47 : ,
354
a moderation of the sense which results from the
interpretation of
as conditional.
Sra 15:66 before
added
355
in order to qualify the strong direct [iii/69]
connection of with dependent
.
Sra 16:9 : ,
356
more likely. | 16:78
.
Sra 17:1 next to tautologic:
357
(partitive). | 17:24 .
considered a spelling mistake by some (see above, p. 390sq.): .
358
| 17:95
: ;
359
cf. Goldziher, Schools, p. 10sq.
Sra 18:14 before added ,
360
dogmatic correction because of the
in the continuation. | 18:31 : , and then
,
eliminating the disagreement between and the singular predicate. | 18:36
, peculiar form :
. | 18:36 : , an insertion of the

shahda formula which is difficult to understand. | 18:62 before
,
361
the more natural position. | 18:76 : ,
362
clearer spelling
350
Also al-Mabn li-namal-man, section 4 of the introduction.
351
Also al-Muttaq, Kanz al-umml, vol. 1, no. 4821. Umar allegedly protests against the
pronunciation in a letter to Ibn Masd!
352
Also al-abar, juz/vol. 12, p. 119, l 7sqq.
353
Makk b. Ab lib, Kashf an wujh al-qirt; al-Zamakhshar, rather Ab Hurayra.
354
So al-abar, vol. 13, p. 147, l 2 and 8 (also intended, 145, 28 and 146, 2, where, as
frequently in this section, by mistake instead has been printed), Ab Ubayd al-Qsim
b. Sallm, and Makk b. Ab lib Kashf; not al-Zamakhshar, after Ibn Masd with instead
excludes the certainly correct conditional interpretation. This variantand the negative
interpretation in generalowe the origin to the same attempt to moderate, like the variant
supported by many old authorities.
355
356
Also al-abar, vol. 14, p. 54, l 4.
357
Also al-abar, juz/vol. 15, 3, 7.
358
al-abar, vol. 15, p. 44, l 18; not al-Zamakhshar.
359
Ab Ubayd al-Qsim b. Sallm; al-abar, vol. 15, p. 102, l 5sqq.; not al-Zamakhshar.
360
al-abar, vol. 15, p. 142, l 3; not al-Zamakhshar.
361
So al-abar, vol. 15, p. 164 l 27; al-Zamakhshar not this but only instead of ,
hardly correct.
362
(cf. above, p. 425 n. 251). | 18:78 after add ,
363
motivating addition. |
18:102 :
, rejecting the reading
.
Sra 19:35 : ,
364
allegedly synonym. | 19:65 : , likely intention- [iii/70]
ally more general.
Sra 20:32 : , suffixed to the preceding Imperatives and thus
itself identified as Imperative (not also read as such, 1. pers. singl. apocopa-
tive.) | 20:66 :
365
(and preceding
as introduction of the direct

speech); eliminationof a serious grammatical offence (see above, p. 390sq). |
20:83 :
. | 20:96 after added , corresponding to the more com-

mon variant of the narrative. | 20:97 : .
366
Sra 22:28 : . | 22:37 :
,
367
probably correct, and only
incorrectly written in the Uthmnic text (
instead of the pausal form

). | 22:45 (conditional pronoun): ,
368
obvious correction of the
feminine explainable by attraction.
Sra 23:20 : ,
369
verb better fits the object. 23:20
: , determination because of preceding .
Sra 24:14 : , unusual synonym! | 24:37 (rather recognized
mistake, see above, p. 390sq.): and this positioned after .
370
| 24:35
behind added ;
371
cf. above, p. 390sq.
Sra 26:19, : ,
372
weakening. | 26:166, poorly fits as
[iii/71]
object : .
373
Sra 27:25 : ; cf. p. 373 n. 247. | 27:25
: (i.e.,
likely
by retaining hamza).
374
| 27:32 : , a synonymwhich is a bit
easier tounderstand. | 27:36 : , corresponding tothe preceding plural. |
363
Also al-abar, juz/vol. 16, p. 3 l 6.
364
So al-abar, vol. 16, p. 55 l 20 (cf. also line 10); in al-Zamakhshar this is followed by
.
365
366
al-abar, juz/vol. 16, p. 138 l 12, for .
367
Ab Ubayd al-Qsim b. Sallm; al-abar, 17, 107, 28; al-Zamakhshar without name.
, where in Ab Ubayd begins the quotation fromthe Koran, is likely to be a simple error
for .
368
Also al-abar, 17, 117, 10.
369
370
Also al-abar, 18, 78, 10, from Ab Ubayd; not in his Fail.
371
Nldeke in the first edition of this work on p. 273, in a source unknown to me.
372
Also Ab Ubayd al-Qsim b. Sallm; al-abar, juz/vol. 19, 38, 33.
373
374
al-Zamakhshar considers the spelling to be rather the rendering of a pausal form
with . This can only mean that the transmission is based on a written copy of the Koran,
with an orthography that already in its beginning was no longer comprehensible.
27:37 relating to : . | 27:40 after added .
375
| 27:83
: ?
376
Sra 28:8 before whereby this receives a bearer. | 28:14 :
. | 28:28 : ; cf. 71:25 : .
Sra 29:24 before added is , whereby the at the beginning be-
comes inna + m what.
Sra 31:26 : ,
377
peculiar and therefore perhaps original.
Sra 33:6 behind added is ;
378
cf. Goldziher, Schools, pp. 78.
| 33:40 [second last vowel a or i] : ,
379
simplification in the
sense of the reading
. | 33:49 before added is ,

380
difficult with
regard to content, which can hardly be attributed to an intentional change. |
33:49
missing before (which is followed immediately by a second

parallel
).
381
| 33:51 before , the usual position.
Sra 34:13 : ,
382
conformity with the common form of [iii/72]
narrative.
Sra 35:41 : ,
383
which is more likely, particularly next
to indeterminate .
Sra 36:7 : ,
384
supplies a plastic illustration. | 36:28 :
,
385
unusual synonym! | 36:30 :
.
386
| 36:38
.
387
| 36:52
:
,
388
unusual synonym! | 36:56 : , both equally possible;
375
376
For
, 27:94, al-Zamakhshar has , Ab Ubayd, (which is no variant, cf. above,

p. 418sq.); common base possibly the Imperative
.
377
Makk b. Ab lib, al-Kashfprobably wrongrather ascribed to Ubayy b. Kab.
378
Cf. al-abar, 21, 70, 8sqq., where Ibn Masd is not mentioned, and where the variant
of asan al-Bar is considered the original text ( ).
379
380
al-abar, juz/vol. 22, p. 14, l 7sqq.; not al-Zamakhshar.
381
Also al-abar, vol. 22, p. 14, l 19.
382
According to al-abar, 22, 45, 8sq., it is rather the text of Ibn Abbs. For Ibn Masd
al-abar, line 27, supplies the qira
, without indicating the

passage in the regular text. What follows is a paraphrase of the
text of Ibn Abbs, which thus also presupposes this tradition in Ibn Masd.
383
384
385
Also Ab Ubayd al-Qsim b. Sallm; al-Mabn li-nam al-man, section 4 of the
introduction.
386
387
A tradition in al-Bukhr, Kitb al-tawd, bb 22 (Goldziher, Schools, p. 7, note 33),
rejects this statement; Muslim, Kitb al-mn, bb 71, which ends: thumma qaraa (the Proph-
et) dhlika mustaqarrun lah f qirat Abd Allh; but al-abar, juz/vol. 23, 4, 35, it says in
the same tradition, merely as a continuation of the words of the Prophet, wa-dhlika.
388
cf. 59:17, conversely : .
389
| 36:58 : ,
390
grammatically
equally difficult, but next to the accusative is still more likely.
Sra 37:45 : ,
391
cf. Goldziher, Schools, pp. 1112. | 37:54 : [iii/73]
| 37:66 : .
392
| 37:123 : (andaccordingly v. 130, :
),
393
| 37:171 : ,
394
clearer. 37:177 : .
Sra 38:5
.
395
| 38:22 after
396
added is
, peculiar. | 38:38 after ;
397
extremely unusual word order.
Sra 39:4 after added is ,
398
as introduction to the following direct
speech.
Sra 40:16 : , without clear connection. | 40:37 :
,
399
excluding the attributive connection of with .
Sra 42:1 : .
400
Sra 43:58, : ,
401
ascertaining the connection with Muammad. |
43:77, : .
Sra 44:54,
: ,
402
thus, hardly a more common synonym, which,
particularly in this frequently cited expression, must be very old. | 44:57
before added is , that better fits as object of the verb .
Sra 45:23, : ,
403
the more natural order. [iii/74]
Sra 47:22, for as explanation .
404
Sra 48:9, : ,
405
to exclude connection with . | 48:26
: ,
406
perhaps inversion to express a comparison.
389
al-abar, juz/vol. 28, 33, 4, drawing on the Kfan grammatical tradition, rather gives
, and then for .
390
391
al-abar, 23, 31, 18sqq.; not in al-Zamakhshar.
392
Ab Ubayd al-Qsim b. Sallm; al-abar, 23, 38, 6; not in al-Zamakhshar.
393
al-Zamakhshar mentions the name of Ibn Masd only in the first instance, and al-
abar only inthe second one (23,55, 35.) According to al-abar, 7, 158, 14, belongs IbnMasd
to those who consider and to be identical. Cf. Goldziher, Schools, p. 11.
394
395
396
According to al-abar, juz/vol. 23, p. 81, l 24 rather after the first .
397
398
Also Sbawayh 269; al-abar, 23, 110, 26 and 28.
399
Ab Ubayd and from him al-abar, 24, 38, 17sqq.; not al-Zamakhshar.
400
Also Fihrist, p. 26, l 29; al-abar, 25, 5, 11.
401
According to al-abar 25, 47, 27sqq., rather the text of Ubayy b. Kab.
402
Also al-abar, 25, 75, 12sqq.
403
404
405
Ab Ubayd al-Qsim b. Sallm, not al-Zamakhshar; al-abar, 26, 43, 19sqq. without
mentioning names. at the beginning of the quotation from the Koran in Ab Ubayd is
likely a mistake.
406
According to al-Zamakhshar this is how it was written in the manuscript of al-rith
Sra 49:2, (after expressionof prohibition) : ,
407
clearer. | 49:11,
: and , respectively (both types occur in the Koran). | 49:17,
before : .
Sra 50:18, : ,
408
more natural.
Sra 51:58 : ,
409
continuing the 1. Pers. of the preceding verses.
Sra 55:6, : .
410
| 55:8 : ;
411
cf. Goldziher, Schools,
pp. 1011. | 55:27, : ;
412
both possible, cf. verse 78 where both as variant
in the Uthmnic text (see above, p. 398sq.). | 55:43, which
violates the rhyme: *
413
with a dual that
is difficult to understand (where the continuation fits only when in verse 44
it is read instead ).
Sra 57:23, , I. or IV. Form with vague subject:
, clearly IV. Form.

[iii/75]
Sra 58:8, (three times): , clearer. | 58:8 after added
, pedantic filling of an apparent gap. | 58:8 : . | 58:8 after
added is
414
or ,
415
cf. Goldziher, Schools, p. 7sq.
Sra 60:11, : , more suitable to the continuation ; similarly
92:3, : .
416
Sra 61:11, :
,
417
clearer. | 61:14, after added
to intensify ; cf. 87:16, before added is ,
418
here, at the same time,
safeguarding the 2. Pers.
b. Suwayd [Juynboll, Encyclopedia, p. 8, col. 1, p. 95, col. 2, p. 96, col. 2] (see section c at the
end, note); al-abar, juz/vol. 26, 61, 26, says only f qirat Abd Allh.
407
408
Also Makk b. Ab Tlib, Kashf an wujh al-qirt, appendix; apparently also al-abar,
26, 91, 14.
409
al-Dhahab, Tadhkirat al-uff, vol. 1, abaqa, 7, no. 76 (Goldziher, Schools of Koranic
commentators, p. 29, note 47); al-Zamakhshar only as qirat al-nab and as such, however,
transmitted by Ibn Masd, also Ab Dwd al-Sijistn, Sunan, kitb al-urf, no. 25, and
al-Tirmidh, Abwb al-qirt an rasl Allh, no. 14.
410
411
Goldziher from al-Ghazzl, Iy ulm al-dn, K. al-kasb wa-l-mash, bb 3, qism 2,
no. 3; not in al-Zamakhshar.
412
Also al-abar, vol. 27, p. 70, l 25; Makk b. Ab lib, Kashf an wujh on 55:78.
413
Also al-abar, 27, 75, 19. Al-Zamakhshar has, probably by mistake, instead .
414
al-Zamakhshar only this.
415
Goldziher from Fakhr al-Dn al-Rz, s.v.
416
So al-Zamakhshar. Although according to the prevailing traditionAb Ubayd al-
Qsim b. Sallm; al-Bukhr, Kitb fail al-ab, bb 27, and Kitb al-tafsr, s.v. (Goldziher,
Schools, p. 7 n. 38); al-Tirmidh, loc. cit., no. 15; al-abar, 30, 119, 33sqq.; Ibn al-Jazar, Nashr,
Berlin Ms. no. 657, fol. 6
r
the reading of Ibn Masd agrees rather with the one of Ab l-
Dard, which he traces back to the Prophet, namely (without ). al-abar,
30, 120, 20, offers for as the interpretation of asan al-Bar.
417
The first part also al-abar, 28, 54, 27.
418
According to al-abar, 30, 86, 30, rather the text of Ubayy b. Kab.
Sra 62:9, : ,
419
correction because hastening ( ) to prayer
was considered objectionable.
Sra 66:4, : .
420
| 66:12, : , as it says more appropriately in the
parallel passage 21:91.
Sra 68:24,
(continuation of the direct speech already introduced by [iii/76]
verse 22, but interrupted by verse 23) missing | 68:51, : .

421
Sra 69:9, : , better suits the context.
Sra 74:6, before added is
;
422
an elucidation interfering with the
poetic expression. | 74:43, at the beginning of the verse added ,
423
as
an explanation. | 74:43, : .
424
Sra 76:30 : .
425
| 76:31, : ,
426
picked up again with
what in the case of is grammatically hardly possible.
Sra 77:17, [amma and sukn onthe ] : , to reject the reading
with .
Sra 81:11, : ,
427
dialectical variant; cf. conversely, 93:9, :
.
428
| 81:24, : , the correct form (see above, p. 425sq.).
429
Sra 89:29, before (so) added , elucidation which, at the same
time, safeguards the reading of the following word as a singular, but reverses
the meaning.
Sra 93:5, :
.
430
Sra 94:6, missing, as repetition of verse 5 it could easily be omitted; this
way, however, verse 5 would lose its corresponding rhyme.
Sra 95:2, :
;
431
against the rhyme, like 23:20. | 95:5, : ,
conforming to regular Koranic usage.
Sra 96:15,
, less well fitting to verse 18.

419
Also Ab Ubayd al-Qsim b. Sallm; al-abar, juz/vol. 28, p. 60, l 32; and 28, 61, 9sqq;
al-Zurqn on al-Muwaa (which even Ibn Masd himself does not mention), vol. 1, p. 197;
and others.
420
421
AbUbaydal-Qsimb. Sallm; al-abar, juz/vol. 29, p. 26 l 20; al-Zamakhshar without
names.
422
423
Ab Ubayd al-Qsim b. Sallm; missing in al-Zamakhshar.
424
Ibid.
425
So al-abar, vol. 29, p. 122, l 24; al-Zamakhshar probably by mistake .
426
427
428
Also al-abar, 30, 128, 31.
429
Also al-Dimy al-Bann, Itf, s.v.
430
al-Zamakhshar on sra 19:67; not s.v.
431
al-Muttaq, Kanz al-umml, vol. 1, no. 4813; not al-Zamakhshar.
Sra 98:2, : , grammatically better following the preceding text. | [iii/77]
98:4 : ,
432
perhaps the original (because of the rhyme
instead
of Masc.).
Sra 99:4, : .
433
Sra 101:4, : ,
434
better known synonym.
Sra 103, totally different text:
* *

*
435
Sra 107:1 : .
436
| 107:7, : .
Sra 109:1 : ,
437
which does not fit the rhyme, but
is unlikely to be intended to be an independent verse.
Sra 111:1
,
438
thus a clear statement in contrast to the wish
.
Sra 112:12 : .
439
As an example of orthographic peculiarity cf. the spelling instead of
, above, p. 424 n. 243.
Goldziher considered all of Ibn Masds different variantslike Koranic
readings in generalunder the aspect of the deviations from the genuine
text of the Koran. Indeed, evenamong boththe variants andvariant readings
that can be traced back to Ibn Masd, there are plenty of instances in which
the Uthmnic text has erroneously been changed,
440
or where at least a
motive for the deviationfromthe Uthmnic text canbe recognized, the text
of Ibn Masd thus being of secondary importance. For Goldziher the most
important of these motivesalthough not the most frequent oneis prob-
ably the elimination of substantial offences,
441
or factual comments,
442
or
432
al-abar, 30, 145, 31 doubtful ( ); al-Zamakhshar without names.
433
Also al-abar, 30, 147, 18.
434
435
Fihrist, p. 26, l 23; not al-Zamakhshar. Verses 1 to3 alsoKanz al-umml f sunanal-aqwl
wa-l-afl, vol. 1, no. 4770, but there verse 2, . is listed in al-abar,
30, 160, 19sqq., as the text of Al as well as l 23 as anonymous.
436
Apparently meant also al-abar, 30, 173, 9.
437
al-Fihrist, p. 26, l 25; not al-Zamakhshar.
438
Also al-Fihrist, p. 26, l 25; note on Ibn Hishm, Sra, p. 231; al-abar, 30, 190, 26.
439
Soal-Fihrist, p. 26, l 26; according toal-Zamakhshar only is missing, whichis perhaps
a confusion with the text of Ubayy b. Kab, which he supplies at the same time.
440
See on sras 2:29, 4:109, 14:47, 74:6, 89:27, 95:2; cf. 94:6, where there is hardly an inten-
tional change, but still an obviously corrupted text.
441
See on sras 2:131 and 153, 3:75 and 177, 4:109, 5:62, 9:107, 14:47, 17:1, 18:14, 24:27, 34:13, 62:9;
by additions to the text, see on sras 4:81, 24:5, 58:8. Included are also passages that merely
purport to smoothen contextual roughness. [Cf. Goldziher, Schools, pp. 2224.]
442
For explanatory, also descriptive touches, supplemental, smoothing additions see on
sras 2:194, 209, 226 and 276, 3:44, 4:38, 11:74, 18:78, 20:97, 27:40, 33:6, 44:57, 58:8, 74:43, 89:29;
linguistic elucidation
443
of the text; added to this are the removal of peculiar-
ities or mistakes,
444
stylistic heaviness, and a general polishing and improve-
ment.
445
But Ibn Masds text must not be considered a correction even
where it reads smoother than the Uthmnic text. A study of the synonyms
used for the individual words of the Uthmnic text alonealso Goldzi-
her points out the great number of synonyms in the non-Uthmnic variant
readingsshows that frequently the word used by Ibn Masd was the bet-
ter known and more convenient expression,
446
but not always.
447
There is
no other explanation than that in the latter cases the Uthmnic text, con-
versely, is secondary with regard to that of Ibn Masd, or more properly
speaking, that many Koranic passages were circulating in different versions
differingamong other thingsby their respective use of synonyms, and
that Ibn Masds textor both texts directly and independentlydrawon
this transmission. Still more probable is the direct take-over from the oral
transmission in most of the cases, in which Ibn Masd offers
448
an unam-
biguous
449
alternative vis--vis the ambiguous form or orthography of the
Uthmnic text. In this case the relation between the text and the variants is
further, a decisive addition settling a controversy, sra 5:91; another one effecting a variant
to a narrative, sra 20:96; averting an ambiguity by textual changes, see on sras 3:5 and 169,
43:58, 48:9.
443
See on sras 2:214, 3:86, 5:42 and 65, 6:16, 11:113, 20:83, 27:25, 49:2, 58:8, 61:11, 74:6.
444
See on sras 2:58, 261 and 269, 4:134 and 160, 5:4 and 73, 6:140, 11:113, 17:24, 18:36, 20:66,
22:45, 76:31, 81:24.
445
See on sras 2:29, 100, 172, 229, 241 and 285, 3:16 and 20 (in 2:226) as well as 43, 4:19, 5:65
and 96, 6:70, 94 and 154, 7:25 and 103, 8:61, 10:81, 11:75, 12:64, 15:66, 16:9, 18:31 and 62, 23:20, 27:36,
28:8, 33:40, 49 and 51, 35:41, 36:58 (?), 39:4, 45:23, 48:26 (?), 50:18, 51:58, 57:23, 60:11, 61:14, 66:12,
68:24, 69:9, 95:5, 98:2.
446
See onsras 2:34 and 94, 4:44, 8:2, 10:98, 12:36, 17:95, 23:20, 26:166, 27:32, 28:14, 36:7, 47:22,
101:4; twisting see on sras 19:65, 26:19; qualification with regard to content, see on sra 5:42.
Further, elimination of ambiguities, see on sras 2:281, 12:105, 18:102.
447
No noticable improvement results from the use of synonyms in sras 1:5, 2:46, 63 and
100, 3:127, 33:40, 37:54 and 177, 55:6, 58:8, 66:4, 68:51, 99:4, 107:7; more difficult to explain are
Ibn Masds synonyms in sras 24:14, 36:28 and 52, 44:54, deviation from subject, sra 37:45,
and 55:8.
448
Even in cases in which Ibn Masd certainly made changes, this need not necessarily
be based on an already fixed written Uthmnic text; also in such instances it might have
been derived directly fromthe oral transmission. Graphicalbut perhaps accidentallyare
possibly the variants to sras 28:14, 68:51, and 107:7.
449
Form, see on sras 2:77 and 282, 6:155 (in 5:42), 11:113, 29:24, 37:171, 57:23, 111:1 (cf. also
2:214); orthography, see on sras 2:19, and 18:76, in particular remoulding in a way that
the existing possibility of double vocalization (or punctuation) in the Uthmnic text is
eliminated, see on sras 2:113 and 281, 3:165, 4:127, 5:3, 6:57 (in 4:19), 105 and 154, 10:12, 12:105,
18:102, 20:32, 27:84 (in 4:19), 77:17, 87:16 (in 61:14), 89:29. Establishing unambiguous syntactic
connections by change, see on sra 40:37.
not that the originator of IbnMasds variants hadthe actual Uthmnic text
at his disposal, realized its ambiguities, and settled for one particular inter-
pretation which he then put down in script. Rather, it is that he was vividly
aware of the meaning of the passage and tried to reproduce it as clearly as
possible in writing.
450
He distinguished himself fromthe authors of the Uth-
mnic text by a stronger determinationto a clear writtenexpression, at least
as far as this was possible in view of the imperfection of the script. In the
case of the numerous variants, which are too insignificant and unimportant
to be interpreted as deliberate changes
451
(including the purely orthographic
cases),
452
one must ultimately assume that one is dealing with an indepen-
dent concurrent transmission of the Uthmnic text, displaying the obvious
mark of either originality
453
orparticularly with some greater differences
in the last srasof total independence from the Uthmnic text.
454
If thus
the origin of many a variant from the oral transmission is to be admitted,
the possibility must be recognized that also more elegant and easier, nay,
even linguistically more correct variant readings might have been derived
from it. Of course, a safe verdict regarding the individual passages cannot
be reached.
It is probable that the greater part of the variants traced back to Ibn [iii/81]
Masd is of a relatively uniform origin. This forms a unit by the frequent
repetition of similar traits in different passages.
455
In this process, however,
no principle is artificially established that would have been followed con-
sistently. However, we are unable to determine if this form of the text goes
back to Ibn Masd, and to what extent he himself really took down gen-
uine oral transmissions in the form that has come down to us, and, at the
450
This circumstance gives the text of Ibn Masdapart from the deliberate addtions
of explanation, etc., pointed out by Goldziherthe character of a commentary on the
Uthmnic text. Cf. the alleged saying of Mujhid
(note to Ibn Hishm Sra, p. 231.)
451
See on sra 1:6; 2:126, 183 and 226; 3:16 and 114; 4:38 and 56; 5:2, 60 and 114; 7:38 and 169;
9:51; 12:70; 20:83; 22:28; 23:103 (in 2:19); 27:37; 36:30; 38:56; 43:77; 49:11; 55:27; 74:43; 76:30; 93:5;
107:1; in addition dialectical deviations in the phonetic field, see on sras 7:19; 12:35; 27:25;
66:4 (?); 81:11; cf. also on sras 2:58 and 269; and 81:24.
452
There is the theoretical possibility that some copies of the Koran with considerable
differences to the Uthmnic text might have erroneously been considered to be the famous
recension of Ibn Masd; in this case, the alleged variants of Ibn Masd could partially
constitute ordinary mistakes of the copyist. But chances that this is the case are negligible.
453
See on sras (4:102), 5:69 (in 5:4); 8:1; 9:118; 11:83; 12:31; 22:37; 31:26; 98:4; cf. further the
peculiareven if not originalvariants of sras 11:60; 33:49; 38:22 and 38; 40:16; 109:1.
454
See on sras 2:22 (in 2:19), 98, 122 and 261, 3:1, 8:39, 9:111, 16:78, 18:36, 19:35, 27:83, 28:28,
37:123, 38:5, 42:1, 49:17, 96:15; 103; 112:12; and the additions 7:38, and 55:43.
455
See sras 2:19, 126, 214 and 226; 3:165 and 177; 4:19 and 134; 5:2 and 42; 7:25; 28:28; 60:11.
same time, also recording changes in the true text, indeed, perhaps even
occasionally making changes himself. At any rate, his authorship cannot be
excluded. First of all, there are no chronological objections because the ter-
minus ante quem applies to the variants produced by the deliberate change
of the text, because this emerged for us from all the non-Uthmnic vari-
ants (see above, p. 391). The only aid available to ascertain the authenticity,
the isndswhich Ab Ubayd al-Qsim b. Sallm supplies for all variants,
and al-abar for roughly half of themand other information regarding
their origin, naturally cannot produce reliable results, particularly as the
variants, certainly in instances containing difficulties or giving rise to dif-
ferences of opinion, are closely linked with the entire exegetic material of
tradition, and thus subject to the identical criteria. Still, the isnds leave
a favourable impression since they are not traced back too far, only to the
second century ah; and their critical investigation
456
confirms this impres-
sion. The isnds of both Ab Ubayd, in his Fail al-Qurn and in al-abar
generally, lead through ajjj b. Muammad al-Awar (d. 206/821)
457
to Ab
Abd Allh HRNIBNMSal-Awar
458
al-Azd in Bara (grammarian and
expert in variant readings, particularly the uncanonical ones, d. ca. 170/786)
or, less frequently, to the expert in the Koran (Abd al-Malik b. Abd al-Azz)
IBNJURAYJ, lately of Bara (d. 149 or 151/ca. 767);
459
another Baran authority
is Qatda b. Dima (60/679118/736),
460
who does not appear in Ab Ubayd
al-Qsimb. Sallm. Next tothis Barangroupthere is yet another groupfrom
Kfa, neglected by Ab Ubayd al-Qsim, but prevailing in al-abar, which
includes Ibrhmb. YazdAL-NAKHA,
461
the famous traditionist (d. 96/715),
Ab Isq Amr b. Abd Allh al-Hamdn AL-SAB (d. ca. 127/745),
462
and
(Isml b. Abd al-Ramn) AL-SUDD (d. 127/745);
463
they are all equally
renowned exegetes.
464
If in the case of Ab Abd Allh HRN (IBN MS)
456
It cannot be their task to reach results regarding the individual passages by checking
the isnds, rather this must be limited to the general character of the transmission.
457
Juynboll, Encyclopedia, p. 153sq.; Sezgin, GAS, vols. 1 and 9.
458
EQ; Sezgin, GAS, vol. 9, pp. 4344; Goldziher, Schools of Koranic commentaries, p. 26.
459
Who frequently refers to older authorities, particularly Mujhid, who also inother ways
appears as a follower of Ibn Masd. Juynboll, Encyclopedia, pp. 438449, Sezgin, GAS, vol. 1,
p. 31.
460
In al-abar there are several additional Baran readers who are not mentioned by
name.
461
Who occasionally refers to his maternal uncle Alqama b. Qays al-Nakha (d. 62/681);
(Sezgin, GAS, vol. 1, p. 398).
462
EI
2
; Juynboll, Encyclopedia, pp. 4750; Sezgin, GAS, vol. 1, p. 283, no. 12.
463
EI
2
; Juynboll, Encyclopedia, pp. 1, 470, 630, 640; Sezgin, GAS, vol. 1, pp. 3233.
464
In addition again occasionally undisclosed readers and grammarians.
we find ourselves on more solid interest in the text of the Koran and out-
side the dubious subject of exegetic adth; the Kfan authorities, on the
other hand, lead us to a place and time when the qira that followed in Ibn
Masds footsteps was still alive.
465
In the case of the great mass of traditions
without isnd in al-abar, but most of all in al-Zamakhshar as well as Sb-
awayh
466
(d. before 180/796), we are probably dealing withanoriginfromthe
living practice or fromKoranic-grammatical transmission, bothof whichare
likely to deserve more confidence than the transmission fromthe science of
tradition.
467
There is certainly the possibility, if not probability, that a not inconsid- [iii/83]
erable core of variants traced back to Ibn Masd do indeed go back to his
copy of the Koran. It needs no proof that at best we know only an insignifi-
cant portion of the deviations of his text from that of Uthmn.
The Text of Ubayy b. Kab
468
We again start with a summary and short explanation of the transmitted [iii/83]
material.
Sra 1:4, at the beginning of the verse is added as the beginning of a
new sra.
469
Sra 2:19, (I. or VIII. Form, cf. IbnMasd): ; cf. belowon2:261. |
2:19, :
470
(?), accumulation of synonyms. | 2:29,
: ;
471
cf. Ibn Masd. | 2:105, : singular and
.
472
| 2:113, : ;
473
cf. IbnMasd. | 2:120,

and
(of God) : 1. pers.

plur., whereby excluding reading as imperative.
474
| 2:131, : ; cf. Ibn
465
Cf. al-abar, juz/vol. 6, p. 133, l 2, and juz/vol. 7, p. 19, l 13 f l-qiratin, l 14 f qirat
ab Abd Allh (i.e. Ibn Masd), l 18, and juz/vol. 9, p. 110, l 21, kna abu Abd Allhi
yaqrana(h); juz/vol. 16, p. 55, l 10, Kn yaqlna f qirati Abd Allhi.
466
467
Some substantiations canbe derivedfromthe totally isolatedtransmissioninthe Fihrist
which makes a reliable impression.
468
For details about himsee above, p. 235sq.; Goldziher, Schools of Koranic commentators,
p. 5; L. Caetani, Annali, year 19, no. 91101; A. Jeffery, Materials for the history, pp. 114116;
Juynboll, Encyclopedia, p. 93, col. 2, p. 265, col. 2, p. 479, col. 2.
469
Fihrist, p. 27, l 16; not al-Zamakhshar.
470
al-Suy, al-Itqn, naw, 16, masala 3, qawl 9; not al-Zamakhshar.
471
472
473
al-abar, 1, 389, 12 only .
474
al-abar, juz/vol. 1, p. 407, l 14 lists Ubayy as authority only when considered 1. Pers.,
but not for a variant.
Masd. | 2:139, : , explanation. | 2:143, : , the same. |
475
2:180
after added , likewise 2:192 after ; cf. Ibn Masd (and Ubayy
himself) sra 5:91. | 2:181, : , the reading
is excluded. |
2:200, :
, whereby the reading
(instead the usual )

is precluded. | 2:201, : ,
476
with resumption of the conjunction
after ; cf. above, Ibn Masd on 2:214. | 2:216, : ; the Uthmnic text
is easier but suspected to be influenced by the preceding . | 2:228,
: . | 2:229, : ,
477
more suitable in the context. | 2: 239, after
added is ;
478
addition which purports to settle a related con-
troversy (Goldziher, Schools, p. 9sq.). | 2:241, : (according to others,
), probably outright to exclude the idea of an extended application. |
2:249, : (cf. above, p. 259). | 2:250, : , constructed according
to sense, contrary to the general grammatical rule. | 2:261,
,
479
but
in reverse 9:57, : , 19:68, 25:63, : , 57:17, :
,
480
63:10, : , all safeguarding the derivation
from the V. Form; cf. above on sra 2:19, and in addition Ibn Masd. | 2:276,
immediately before
: . | 2:280, : ;
481
linguistically more nat-
ural. | 2:281 : ; cf. Ibn Masd. | 2:283,
which produces a
very heavy construction.
Sra 3:5, : ;
482
cf. Ibn Masd. | 3:17, :
[iii/84]
, fixing the vocalization
(not
, dependent on ). | 3:20,
: ,
483
thus elimination of
the obvious and establishing the meaning to kill (not to fight,
cf. Ibn Masd on 2:226) also in the case of the second object. | 3:32, :
in order to eliminate the reading as I. Form. | 3:147 after added
is ;
484
explanatory addition which identifies the verb as IV. Form(not
I. Form). | 3:185, : , a clear synonym which at the same time excludes
the variant (Passive).
475
2:172, Makk b. Ab lib, Kashf an wujh has the variant , which otherwise derived
only from Ibn Masd.
476
al-abar, 2, 179, 8; not al-Zamakhshar.
477
478
Ab Ubayd al-Qsim b. Sallm; not al-Zamakhshar. Cf., however, al-abar, 2, 327, 15
and 328, 23, where Ubayy rather approves of the variant .
479
Cf. in this connection the account above, p. 395 n. 46.
480
Makk b. Ab lib, Kashf an wujh al-qirt; al-Zamakhshar without name.
481
al-abar, juz/vol. 3, p. 67, l 29; not al-Zamakhshar.
482
Also al-abar, 3, 113, 27; Itqn, naw 43, fal 2 (according to al-Farr).
483
In al-Zamakhshar the first is missing, probably unintentionally.
484
Sra 4:15, after added is , a necessary clarification of the
[iii/85]
meaning and therefore easily added. | 4:92, missing, which makes the
heavy construction of the Uthmnic text even more difficult. | 4:102 see Ibn
Masd. | 4:128, : , one of the interpretations of the disputed
word.
485
4:134 ,
486
referring to two duals: . | 4:157, : ,
487
to be
vocalized according to the following (and then previously
); this
form of the text excludes the reference of the suffix to s, as it is often
claimed. | 4:162, an accusative (twice): ,
488
which is out of place in
this construction, andcanbe explainedonly as a preceding object pickedup
againby a suffix, thus either a combinednominal sentence or a continuation
of the preceding , etc.
Sra 5:49, : , a better continuation of the
preceding , and avoiding an incorrect reference of the suffix. | 5:49,
:
, with clear expression of the dependency. | 5:51, :
,
489
whereby both the character of summons and the dependency on the
preceding sentence is to be made explicit. | 5:62, : ,
490
with the
resumption of the preposition after ; cf. above, Ibn Masd on 2:214. | 5:73,
: ;
491
cf. Ibn Masd, and above, p. 397sq.
Sra 6:16, see in Ibn Masd. | 6:74,
,
492
clearly vocative. | 6:109, [iii/86]
: ,
493
better fits the context. | 6:111, : ,
494
establishing the interpre-
tation of the word (like 17:94) and better fits the preceding singular than a
plural
.
495
| 6:144, : . | 6:154, [with fata or kasra on
the first alif] : ; cf. Ibn Masd.
Sra 7:103, : ; cf. Ibn Masd. | 7:124, : ,
496
daring correction of a rather obscure text. | 7:169, : , after (cf.
Ibn Masd on 2:226), clearly II. Form, not IV. Form.
485
As such, e. g., al-abar, juz/vol. 5, p. 187, l 16sq.
486
487
488
Also al-abar, 6, 18, 16; al-Zamakhshar says nothing about the vocalization, al-abar
presupposes a nominative.
489
Also al-abar, 6, 153, 24, who, however, explicitly doubts the reliability of the transmis-
sion.
490
491
According to Ab Ubayd al-Qsim b. Sallm, fol. 37
r
, rather , and for ;
this, however, is the variant that is normally ascribed to Ibn Masd!
492
Itf, s.v.; not al-Zamakhshar.
493
Also al-Dimy al-Bann, Itf, s.v.
494
495
Ab Ubayds Fail in the Berlin Ms., no. 451, has as a variant , sra 6:139,
but which is identical with the Uthmnic text, probably a mistake in the manuscript.
496
Also Ab Ubayd, and from him al-abar, juz^/vol. 9, p. 16, l 4; not al-Zamakhshar.
Sra 9:49, : , corresponding to the preceding singular .
Sra 10:2 ( )
; cf. below on 20:66. | 10:25,

added .
497
| 10:28,
,
498
active insteadof the requiredpassive construction. | 10:59,
; ,
499
ascertaining the interpretation as 2. pers. (= .) | 10:72,

: , since poorly suited as object of the preceding

verb. | 10:81, : ,
500
and reverse, 21:48, : which, of course, at
the same time rejects the punctuation , as well as the interpretation as
IVth form.
Sra 11:30, :
,
501
establishing the reading as II. Form (and this [iii/87]
way Allh as the originator).
Sra 12:7,
:
502
(cf. above, p. 402 n. 92). | 12:18, (nominative
without coherence of the sentence, difficult to understand) : . |
12:90,
503
(predicate) :
504
(the surprised question that better
fits the context).
Sra 13:2, : (referring to which in this case is the only possi-
bility, and this considered singular).
Sra 14:43,
, excluding the readings
and . | 14:47,
: , and : ;
505
very daring correction of the
passage which in Ibn Masd is only lightly softened.
Sra 16:39 :
(for Allhhas no leader, whom He leads

astray, i.e., whomAllhleads astray, he has no leader,) whereas inthe Uth-
mnic text Allh is concurrently the subject of the non-rightful-guidance
and the leading astray. (Cf. Ibn Masd, p. 379, note 285). | 16:39, :
,
506
rejecting the reading as Ist form which serves the same purpose as
just mentioned.
497
So Ab Ubayd and al-abar, 11, 65, 22, here, after of the text; al-abar, juz/vol. 11,
p. 65, l 15sqq., but rather , etc., after of the text; not al-Zamakhshar.
498
499
According to Ab Dwd al-Sijistn, Sunan, kitb al-urf, no. 12; al-abar, juz/vol. 11,
p. 80, l 6sqq., and al-Dimy al-Bann, Itf, s.v., rather , in agreement with the
Uthmnic consonants.
500
501
Also al-Farr; Man l-Qurn (Ms., Constantinople, Vehbi, 60), s.v.; according to al-
abar, 12, 17, 21, rather the text of Ibn Masd.
502
503
So the Uthmnic consonantal text even though most people read , in which case,
however, is hardly suitable.
504
505
Makk b. Ab lib, Kashf anwujh; not al-Zamakhshar. At the beginning of the Koranic
passage the Ms. has instead ; probably by mistake.
506
The wording in al-Zamakhshar is not quite clear.
Sra 17:14, : (?)
507
| 17:17, :
[iii/88]
;
508
likely a moderation of the hazardous idea that Allh induces sin in
order then to be able to punish it. | 17:35, [first letter t or y] : ,
plural 2. pers. as before and after, whereas the Uthmnic text suggests the
wording as 3. pers. sing., referring to the preceding
. |
509
17:78, : ,
after
, cf. above, p. 402 n. 90, and Ibn Masd on sra 4:56. | 17:104,

, with the good old construction of
(see
above, p. 392 n. 24).
Sra 18:24, , after contradicting the general rule: . | 18:76,
:
,
510
introduction of a synonym, probably because of the pecu-
liarity of the spelling (cf. Ibn Masd).
Sra 19:35, after added in order to establish the as 3.
pers., and past tense. | 19:37, [with hamza above and kasra belowthe alif]
:
,
511
excluding the reading .
Sra 20:15, at the end of the verse added ,
512
cf. above, p. 402, n. 94. |
20:32 and 20:33 reversed, and the before verse 32 (cf. in this connection Ibn
Masd). | 20:66,
,
513
see above on 10:2; cf. Ibn
Masd.
Sra 22:77, : , preventing the reference to Abraham.
Sra 23:20, :

; cf. Ibn Masd. [iii/89]
Sra 24:14, : ,
514
the more explicit form. | 24:35, :

(or instead of only );
515
cf. Ibn Masd.
Sra 26:129, : ,
516
explanation. | 26:202,
: .
Sra 27:8, : , or ,
517
both serving to
eliminate
which could only refer to Allh, and then affect a new inter-
pretation. | 27:25, : ,
518
better suitable to the IVth form of
507
508
509
To , 17:40, which many people in order to attain consistency with read (cf.
also the variant above, p. 401 sq.), Ab Ubayd refers to a variant which the Ms. suggests as
; meant is perhaps al-Zamakhshars anonymous reference to the reading
.
510
511
512
al-abar, 16, 99, 17, anonymous (f ba al-urf).
513
514
al-abar, 18, 69, 30; al-Zamakhshar, anonymous.
515
al-Zamakhshar, and al-abar, juz/vol. 18, p. 94, l 33, only , Ab Ubayd al-Qsim b.
Sallm, has both. Al-abar, however, juz/vol. 18, p. 94, l 28, seems to offer .
516
al-abar, 19, 54, 26, anonymous (f ba al-urf).
517
Thus two forms of the text excluding one another.
518
al-abar, juz/vol. 19, p. 85, l 15, cites as the text from Ubayy , so that thus
would be missing; likely a mistake.
the verb . | 27:25, : ,
519
a more explicit synonym which, how-
ever, destroys the parallelism. | 27:30, [with hamza above and below ]
:
, better continuing the preceding letter. | 27:84, :

,
520
cf. Ibn Masd (on 4:19). | 27:94,
: ;
521
cf. also above,
p. 439 n. 376.
Sra 34:13, : , whereby the allusion is eliminated, which impedes
the appreciation of a legend that is no closer identified, but destroys the
sense of the narration; cf. Ibn Masd. | 34:23, : , more forcefully
emphasizing the disjunction (followed by ).
Sra 36:29, is missing,
522
whereby also the explanation, which also
appears
523
as a variant, , is alluded to. | 36:72, :
;
524
more explicit ( = , not ).
Sra 39:4, :
525
(and further preformative instead ), consis-
[iii/90]
tently keeping the direct speech.
526
Sra 43:58, see inIbnMasd. | 43:61,
,
527
a difficult wordreplaced
by another one that is hardly better.
Sra 48:26, after added is ,
528
an
additional explanation of the situation.
Sra 53:29, (vague, referring to ) : .
Sra 55:31, : , in order to provide an acceptable meaning to the
governing verb, .
Sra 56:22, :
,
529
very peculiar.
Sra 57:29, : ,
530
to avoid the construction of with indic. imperf.
Sra 65:1, : , more definite (against you, not against Allh).
Sra 72:21, :
, better fitted for the contradistinction .

Sra 74:39, : , grammatically more transparent.
Sra 87:16, see in Ibn Masd on 61:14.
519
Identical reference to the previous note.
520
Also Ab Ubayd al-Qsim b. Sallm.
521
Also Ab Ubayd, but without . It is not quite clear howthe variant is supposed to fit
the context.
522
Ab Ubayd al-Qsim b. Sallm; al-Zamakhshar without name.
523
al-abar, juz/vol. 23, p. 3, l 8 and 11.
524
Ab Ubayd; al-Zamakhshar, anonymous.
525
Also al-abar, 23, 110, 26.
526
Cf. Ibn Masd who has precede.
527
528
al-Mabn li-nam al-man, section 4 of the introdution; al-Muttaq, Kanz al-umml
f sunan, vol. 1, no. 4823sq.; not al-Zamakhshar.
529
Sbawayh, 27; not al-Zamakhshar.
530
Sbawayh, 276; not al-Zamakhshar.
Sra 89:27, before ,
is added as an explanation.
531
| 89:28, :

, intentionally more general.

Sra 112:1, missing; cf. Ibn Masd.
In addition there is a number of passages, where Ubayys text is identical [iii/91]
withthat of IbnMasd:
532
Sra 2:58 ( ),
533
2:77,
534
2:126 and2:172,
535
2:209,
536
2:222 (see above, p. 432, l 12);
537
3:75,
538
3:127; 4:81,
539
5:65 (only without
),
540
5:91,
541
6:105,
542
6:154 (but instead ), 7:25, 10:81,
543
10:98,
544
11:113
(only ), 17:24,
545
18:36 (only ), 18:78,
546
24:27 (without change),
547
27:25
(only ),
548
69:9.
It must be observed that Ubayy frequently appears as the defender of the
Uthmnic text against deviations; so in the case of sras 4:160,
549
9:101 ( ,
not ),
550
62:9,
551
and 81:24. Here also belongs the report
552
that during the
collation ( ) of the copies of the Koran Uthmn sent his client, Hni
al-Yazd, to Ubayy with a shoulder-blade of a sheep, containing the three
passages, sras 2:261 (see above, p. 395 n. 46, at the end) 30:29, and 86:17,
and he changed to , to , and to , i.e., restoring
the Uthmnic form of the text.
553
531
Also al-abar, vol. 30, p. 105, l 16sqq.
532
Only suchpassages have beenlisted where at least one source documents the same text
for both so that in case there should have occurred a confusion this must be older than the
source.
533
al-Dimy al-Bann, Itf; not al-Zamakhshar.
534
Also al-abar,1, 295, 13.
535
Makk b. Ab lib, Kashf an wujh; not al-Zamakhshar.
536
al-abar, 2, 188, 16; cf. 29; not al-Zamakhshar.
537
Makk b. Ab lib, Kashf an wujh, and Itf; not al-Zamakhshar.
538
539
Ab Ubayd al-Qsim b. Sallm (see above, Seite 64, note 310); not al-Zamakhshar.
540
541
Also al-abar, 7, 19, 7sqq.
542
543
Also Ab Ubayd; al-abar, 11, 94, 31.
544
Also al-abar, 11, 109, 10.
545
546
547
Alsoal-abar, 18, 78, 3. Fakhr al-Dnal-Rz, s.v., has a different text for the entire second
part of the verse
. This is an explanatory
version of the Uthmnic text that violates the rhyme.
548
549
550
Also Ab Ubayd; al-abar, 11, 6, 14 and 29 (here from Ab Ubayd); al-Muttaq, Kanz
al-umml, vol. 1, nos. 4831, and 4866sq.
551
Also Ab Ubayd; al-abar, 28, 60, 22; Fakhr al-Dn al-Rz, s.v.; al-Muttaq, Kanz al-um-
ml, vol. 1, nos. 4816, and 4830.
552
553
On the other hand, Umar opposes Ubayy on several occasions, both when deviating
It is reported that Ubayys orthography went further than the Uthmnic [iii/92]
imla, recognizing in the middle of a word the spelling ; e.g. (instead
of ), (instead of ), and (instead of );
554
(cf. above, p.
417sq.).
The impressionwe get fromthe text going backtoUbayy is less favourable
than that of Ibn Masds text. Even a quick look reveals the close relation
of the two texts,
555
particularly in cases where this is secondary. That both
recensions should have so frequently arrived at the same changes indepen-
dently fromone another is hardly less unlikely thanthat they independently
drew on the same differences from the Uthmnic text from oral transmis-
sions. But if one of the recensions is dependent on the other, it is likely to be
Ubayys from the one of Ibn Masd, which is the richer and safer transmis-
sion.
When attempting to characterize the variants going back to Ubayy, and
disregarding all those that touch upon Ibn Masd, we first find as the
most prominent trait the endeavour to attain a smoother and more cor-
rect linguistic expression;
556
secondly, a more precise and distinct formu-
lation or the resolution of possible doubts
557
(and most of all a decision in
the case of rivalling vocalizations, etc.)
558
and accordingly the elimination
of substantial offences and objections
559
or, less frequently, more precise
definitions by means of additions,
560
which otherwise also serve as mere
explanation.
561
The frequent use of synonyms also makes for an easier and
simpler
562
or a clearer, more explicit and appropriate
563
text. In all these
instances the incentive for change is evident, and therefore the change,
the non-originality of the variant, is likely; occasionally the text obtained
from the Uthmnic text (see on sra 48:26), and when approving it (see on sras 9:101 and
62:9); cf. further, Kanz, vol. 1, no. 4827. Umar says of him (regarding sra 62:9)

.
554
al-Dn, al-Muqni, bb 13.
555
Apart fromthe above list of literal agreements there are contacts in sra 2:19, 29, 113, 131,
201 and 281; 3:5; 5:62, 73 and 154; 7:103; 9:57, etc; (in sra 2:261); 14:47 and 66; 23:20; 24:35; 27:84;
34:13; 112:1.
556
Sras 2:105, 229 and 276, 3:20; 4:134, 157 and 162; 5:49; 6:109 and 111; 9:49; 10:68; 12:18 and
90; 13:2; 17:35; 18:24 and 27; 25:30; 39:4; 53:29; 55:31; 57:29; 74:39.
557
Sras 5:49, 51 and 157; 6:16 and 111; 17:35; 19:35; 22:77; 34:23; 36:29; (43:58.)
558
Sras 2:120, 181 and 200; 3:17; 20 and 32; 6:74; 7:169; 10:59; 11:30; 16:39; 19:35 and 37; 24:14;
36:72; (87:16).
559
Sras 16:39, 17:17, 27:8.
560
Sras 2:180 and 239, 4:15, 19:35.
561
Sras 7:124, 10:25 and 72, 20:15, 48:26; 89:27.
562
Sras 2:139 and 143; 4:128; 18:76.
563
Sra 2:241; 3:185; 26:129; 27:25; 65:1; 72:21; 89:28; exclusion of possibilities in vocalization,
sras 3:147; 14:43; 21:48 (in 10:81).
can be demonstrated to be outright wrong.
564
It is peculiar that compared
with Ibn Masd the motives with regard to content recede considerably,
whereas the linguistic-stylistic aspects very much come to the fore. Signs of
truly oral transmission are much less numerous and less clearly to be found
than in Ibn Masd. There remains, though, a remainder of grammatically
difficult
565
and unintelligible
566
variants as well as difficult synonyms,
567
and
a still somewhat larger residue of variations
568
and replacements of syn-
onyms,
569
for which at least no plausible motivation can be found. But this
rest is not considerable, and, most of all, hardly a single instance can be
found where the text going back to Ubayy would have in its favour the like-
lihood of greater originality vis--vis the Uthmnic text.
570
One must add
that this recension is void of internal cohesion, which we have been able to
identify in Ibn Masd: there is next to no trace of homogeneous treatment
of different passages,
571
and one of the very fewis even congruent with a case
that is typical for Ibn Masd.
572
The conclusion that the form of the text ascribed to Ubayy falls far short [iii/94]
of the originality and coherence of Ibn Masds text has in itself no bearing
onthe questionof whether it really originates fromUbayy.
573
But the external
evidence is also weaker. Ab Ubayd al-Qsim b. Sallm relies almost exclu-
sively on HRN (b. Ms) AL-AKHFASH who was also the main author-
ity for Ibn Masd; in al-abar, who in approximately half of his examples
refers to the latters authorities, this isnd is dominated by another per-
son, leading through the intermediary of AB JAFAR s b. Abd Allh
AL-RZ, from Merv, living at al-Rayy (d. ca. 160/776) through al-Rab b.
Anas [al-Bakr] from Bara, living in Khursn (d. 140/757 or earlier),
574
and
occasionally through the famous scholar of the Koran, AB AL-LIYAH
Rufay b. Mihrn al-Riy at Bara (d. 90/708 or later).
575
Compared with the
rich transmission of Ibn Masdone portion of which was derived from
Kfan readers, grammarians, and traditionists of the circle of the immediate
564
By mistake in the rhyme, sra 20:15; by interference with the parallelism, sra 27:25.
565
Sra 2:250 and 283, 4:92, and 56:22.
566
Sra 1:4, 2:249, 10:2, 17:14, 26:202, and 27:94.
567
Sra 2:216, and 17:104.
568
Sra 2:228 and 261, 6:144, and 20:32sq.
569
Sra 2:19, 10:81, 12:7, and 43:61.
570
Possibly sra (4:102), and 17:104.
571
Still, see on sra 2:180 and 261, and 10:2.
572
See on sra 2:261.
573
Cf. the opposite view, A. Jeffery, Materials for the history of the text of the Qurn, p. 116.
574
575
EI
2
followers of Ibn Masds textwe here have an insufficient, purely Baran
transmission, collected far from its origin, and which does not display any
repercussions of a living practice.
576
Thus, the probability that our variants derived from Ubayy represent [iii/95]
genuine remnants of his codex are rather negligible. It is of course possible
that a part of the variantsand also those which appear in Ibn Masd in
the same or similar formreally derive from them. But we have no way of
ascertaining this.
577
This difference in the character of transmission from Ibn Masd and
Ubayy is well foundedinthe difference of the external conditions for the sur-
vival of the two recensions. As can be seen from the uncertainty of Ubayys
date of death, he played no important role after the death of Muammad,
and soon disappeared from the political scene, either because of an early
death or because of other circumstances. The distribution of his codex must
have remained a strictly private matter. Ibn Masd, on the other hand, had
the opportunity, as the governor of Kfa, to enjoy the official sanction of
his codex of the Koran; and he made good use of it.
578
Accordingly, also the
fate of these two copies of the Koran seems to have been dissimilar. Ubayys
copy vanished early; copies were hardly ever made.
579
On the other hand,
576
Also Muslim criticism values the transmission of the text of Ibn Masd higher than
Ubayy b. Kabs text; according to Sad al-Dn Masd b. Umar AL-TAFTAZN (d. 792/1390;
Sezgin, GAS, vol. 1, p. 427: e) the muaf Ubayy is transmitted bi-arq al-d, the muaf b.
Masd however (bi-arq) al-shuhra (al-Talw, qism1, rukn 1, beginning (Cairo ed., 1327/1909,
pp. 1, 27, note 3)).
577
Most authentic are perhaps those that are derived fromstatements of a special arrange-
ment of the Fihrist (see above, p. 237sq.; fromthis originates also the reference to the variant
on sra 1:4).
578
That the qira of Ibn Masd was legitimized at Kfa (or in Iraq in general) is pre-
supposed by the attempts to derive the claim of im b. al-Jadar from it (see below), and
documented by accounts regarding the motivation for Uthmns text (see above, p. 251 sq.),
and regarding Ab l-Dards reading of sra 92:3 (see above, p. 441 n. 416). This account in its
earliest version (Ab Ubayd, Fail, Berlin Ms., Ahlwardt, no. 451, fol. 42
r
) reads as follows:
an Alqamata qla: laqtu Ab l-Dardi fa-qla l: mimman anta? qultu: min Ahl al-Irq.
qla: a-taqrana al qirat Abd Allh? qultu: naam. qla: fa-iqra wa-l-layli idh yaghsh!
fa-qaratu: wa-l-layli idh yaghsh, wa-l-nahr idh tajall, wa-l-dhakari wa-l-unth. qla: fa-
aika wa-qla: hkadh samitu rasl Allh yaqrauh (on this several parallel versions in
Ab Ubayd); further, by the statement that the mother and the maternal grandfather respec-
tively of Sufyn b. Unayna (d. 198/813814) followed the text of Ibn Masd (Ibn Mujhid
in Ibn Jinn, Mutasab, on sra 24:14, andprobably from thereal-Zamakhshar, s.v.). Cf.
finally, above, p. 447 n. 465.
579
Reliable relatedreports have come downonly throughal-Kis (see above, p. 418 n. 192),
and an otherwise unknown Muammad b. Abd al-Malik al-Anr who, because of the
remark rawaynh (i.e., the muaf Ubayy) an bin is to be looked for among Ubayys
descendants (Fihrist, p. 27, l 3). These facts are in contrast to the information derived from
copies of Ibn Masds text are definitely known to have existed for quite
some time.
580
[After I completed the preceding sections, new material regarding the [iii/97]
text of the Koran became available to me, which I published in 1933 under
the title Nichtkanonische KoranleseartenimMutasab des Ibninn [unca-
nonical variant readings in Ibn Jinns al-Mutasab]. As far as the text of Ibn
Masd and Ubayy is concerned, I refer to the index of that publication.]
The Text of the Lewis Palimpsests
From among other pre-Uthmnic recensions of the Koran mentioned by
tradition
581
none, or hardly any, have left traces in literature. The manu-
scripts of the Koran, as far as they have been investigated, all display the
Uthmnic consonantal text,
582
the only exception being the afore-men-
tioned Lewis palimpsests (see above, p. 53). In the same way that they dis-
Ubayys son, Muammad, that Uthmn confiscated (qabaah) Ubayys copy (Ab Ubayd,
Fail [Berlin Ms. no. 451, fol. 36
v
; Kanz, vol. 1, no. 4789].Additional claims of personal
inspection of Ubayys codex in section 3 of the introduction to al-Mabn li-nam al-man
are questioned by the author with reference to a possible falsification (
). al-abar, 15, 44, 20, mentions an alleged copy (muaf al
qirat Ubayy) fromthe property of Ibn Abbs, which still Yay b. s (d. 201/816817) claims
to have seen.
580
That Ibn Masd successfully refused to hand over his copy to Uthmn is attested by
Ab Ubayd (Fail al-Qurn, Berlin Ms. no. 451, fol. 36
v
) and the apology of al-Kind (see
above, p. 392 n. 27, p. 444sq., and below, p. 486, respectively), according to which, still at
the time of composition (204205?/819821), it was passed down in the family (pp. 446sq.,
and 488 respectively). A copy of it is likely to be the manuscript of the ib of Ibn Masd,
al-rith b. Suwayd (d. after 70/689), which, according to al-Zamakhshar on sra 48:26, was
hidden in the ground during the time of ajjj b. Ysuf [EQ; EI
2
, d. 95/714]. A manuscript
allegedly containing a copy of Ibn Masds codex, to which the Baghdd Shites refer, was
burnt in 398/10071008; (cf. Goldziher, Schools, p. 172; A. Mez, The Renaissance of Islam,
pp. 195196). Ibn al-Nadm has seen a number of manuscripts (Fihrist, p. 26, l 29); others are
mentioned in Muqni (see above, p. 424 n. 242), and al-Mabn li-nam al-man (section 3
of the introduction). The manuscripts that Ibn al-Nadm has seen differ considerably from
one another, but this does not come as a surprise, because the codex of Ibn Masd was not
copied for scholarly interest but for practical purposes, and could hardly escape mixture with
other forms of the text, primarily with the Uthmnic codex.
581
See above, 192sqq., and 206sqq.; in addition the version of Uqba b. mir al-Juhan [EI
2
;
Juynboll, Encyclopedia, p. 15sqq.; Sezgin, GAS, vol. 1, p. 341, d. 58/678]; cf. al-Suy, al-Itqn,
naw 20 [Sprenger ed., p. 169, l 22] which allegedly had been in the possession of Ab l-Qsim
Al b. al-asan IBN QUDAYD (d. 312/925) [Sezgin, GAS, vol. 1, p. 356]. Cf. R. Guest in the
introduction to al-Kinds The Governors and judges, p. 18; quoted by A. Mingana in his An
ancient Syriac translation of the Kurn, p. 9.
582
The occasional differences are throughout erroneous misspellings of the Uthmnic
text. For details see below, the chapter on manuscripts of the Koran.
play an older orthography compared with that of the Uthmnic text as well
as a different dialectical basis, they also differ in content. If we ignore the
mere orthographic differences of Minganas list (p. xxxvii sqq.),
583
the follow-
ing variants remain:
Fragment A: Sra 29:24, our text : palimpsest ,
584
insignificant. |
41:4, : <>; senseless, possibly misreading. | 41:10, : ; fits poorly,
perhaps the of the text has been interpreted incorrectly.
585
| 45:18,

: (or ) ; completely senseless and impossible,

586
probably an
erroneous reading.
Fragment B: sra 11:25, : , vary between Ist (IInd) and IVth form [iii/98]
as so frequently in the literature of the variant readings; however, the Ist
(IInd) form is here very unlikely. | 11:34, : ; somewhat hard, but
not impossible, and therefore noteworthy. | 13:26, : ; hardly fitting,
probably a mistake incopying. | 13:33, : , hardly possible after ; expla-
nation? | 14:3, : , misreading for = ? | 15:94, : ;
insignificant, perhaps wrong reading (copying). | 16:17, : , insignifi-
cant. | 16:30, : , very inappropriate and is likely a mistake in copying. |
16:36, : , insignificant. | 16:38, : , insignificant. 16:87
and 16:88, : ; the two sentences are mistakenly assimilated to the narra-
tives of the Prophet introducedby . | 16:95, : , insignificant. | 16:95,
after added , results from the irregular positionit ought to stand
before recognizable as an addition. | 16:112, : ; the more com-
mon syntax, but differing from the Koranic linguistic usage. | 16:124, :
;
587
might be an orthographic difference, but also constitute the form
which occurs in the Uthmnic text (see above, p. 401 n. 82). | 17:24,

: , whereby the preceding loses its indispensable object; likely a

mistake in copying.
Fragment C: Sra 7:153, : , one of the typical confusionof synonyms;
objections to the Uthmnic text might be that its is a common
formula, and thus might possibly be an adaptation from another passage;
but ( ) is too much out of place in this context. | 9:23 : ,
583
To these also belongs sra 40:85 (A) against of the Uthmnic text: The
scriptio continua is incorrectly divided; further, likely also the difference not included
in Minganas list , sra 16:32 (B) instead (cf. above, p. 427, section III.) sra 9:75
(C) instead Mingana himself seems to have considered an error.
584
The diacritical marks are by the editor [of the text.]
585
Cf. above, n. 583.
586
The attempts of interpretation by Mingana, p. xxxvii, are as useless as those of W.S.C.
Tisdall in New light on the text of the Qurn in Moslemworld, 5 (1915), p. 149.
587
Missing from Minganas list.
insignificant. | 9:33, added only by another hand, but it is indispensable. |
9:36, : , the more common syntax. | 9:36, the first is missing; next
to the second one it could have been omitted by accident. | 9:37, :
,
588
wrong interpretation of the orthography of , which in this case
really expresses al-nas. | 9:38, which is indispensable for the idea is
missing. |9:43, : Mingana correctly interprets as
;
589
remarkable
variant. | 9:54, : , insignificant.
The gain in important variants is thus limited. Most of the deviations [iii/99]
are of the kind that can be sufficiently explained as accidental variations
of the oral or written transmission; some of them clearly indicate a writ-
ten original. This agrees with the fact that according to the character of the
writing the Palimpsest itself does not seem to be as old as the character of
the orthography of the text
590
on which it is based. The few more serious
variants make it at least likely that we are not merely dealing with defor-
mations of the Uthmnic text but with repercussions of a non-Uthmnic
transmission.
591
Ahistorical connection with it cannot be established on the
basis of the literature available to us. In any case, there is no trace of the
text of either Ibn Masd or Ubayy b. Kab in the Palimpsests. In group A,
sras 24:27, and 44:54 against Ibn Masd; group B, sras 16:9, 17:1, and 17:24
against Ibn Masd, and 11:30, 16:39 (twice), 17:17 and 17:35 against Ubayy;
group C, sra 9:51 against Ibn Masd, and 9:49 and 9:57 against Ubayy but
agreeing with the Uthmnic text. The fact that the Koranic excerpts con-
tained in the Palimpsestseven in our fragmentary transmissionconsist
of such a large number of important variants of Ibn Masd and Ubayy also
shows how relatively close Lewis Palimpsests are to the authorized text of
the Koran.
The Alleged Syriac Translation of a Non-Uthmnic Text of the Koran [iii/100]
Mingana
592
speculates that he has unearthedlarge pieces of a pre-Uthmnic
text of the Koran at an unexpected place, going beyond the literary trans-
588
Missing from Minganas list.
589
This requires as continuation instead of the Uthmnic ; the passage in the
Palimpsest does not seem to be legible.
590
Cf. Mingana, p. xxxii sq. about the palaeological character.
591
Differences of the character of the text of the three groups, A, B, and C cannot be
identified with certainty; striking is the large number of omissions in the relatively short
group C.
592
An ancient Syriac translation of the urn , with facsimile of the relevant sections
(fol. 73a to 84b) of the Ms. Mingana, no. 89.
mission. He claims that the anti-Islamic polemics of Dionysios Bar alb
(d. 565/1171),
593
the Jacobin polyhistorian of the late period of Syriac litera-
ture, contain remnants of a Syriac translation of the Koran from about the
time of the Umayyad Abd al-Malik b. Marwn (seventh century ad) and at
any rate represent a quite different text than ours. Bar alb arranges chap-
ters 25 to 30 of his work, which constitute the last of the three mmr, in two
columns: The first of which contains almost exclusively quotations fromthe
Koran, and the second one now and then mostly brief refutations. Rather
than subjecting Minganas interpretation of these quotations to negative
criticism, pointing out the individual errors and the historical impossibility
of his assumptions, we shall immediately proceed to the positive presenta-
tion of the facts.
The quotations have not been taken froma Syriac Koran at some indeter-
minable time. It is rather that an Arabic-writing Christian apologist incor-
porated them in the original version in polemics against Christianity as
documentation. They were only later taken froma Syriac translation or edi-
tion of the polemics and put together separately as a kind of Koranic flo-
rilegium.
594
This becomes sufficiently clear from the arrangement and the
type of passages, but particularly fromthe abruptness of many of the adduc-
tions, which show that they are ruthlessly taken out of context for which
they were tailored from the undivided whole of the text of the Koran. How-
ever, evidence beyond doubt is the fact that on several occasions the iden-
tical verse of the Koran is referred to in different passages and in different
translations.
595
Indeed, even in one instance a verse is followed immediately
by a translation variant
596
with the introduction in a different place it is
written (!). All this makes sense only if the translator encountered the same
verse several times in his original text, and on the second occasion trans-
latedanewwithout regardfor the first one. The originof the translationfrom
polemics against Islam, whichas is common with such writingsare not
particularly conscientiously documented, explains not only the deviations
593
A. Baumstark, Geschichte der syrischen Literatur, p. 297, note 1.
594
Not to be excluded, but it is far more unlikely that the quotations were first collected
and later translated into Syriac.
595
Sra 2:130 infol. 77a, l 10sqq., andfol. 81a, l 13sqq.; sra 5:77 infol. 82b, l 13sq., andfol. 84b,
l 5sq.; sra 10:94 in fol. 79a, l 24sqq., fol. 83a, l 11 sqq., and fol. 84b, l 16sq.; sra 17:96 in fol. 83b,
l 7sqq., and fol. 84a, l 20sq.; sra 41:10 in fol. 77b, l 16, and fol. 84b, l 2; sra 66:12 in fol. 82b, l
1 sqq., and fol. 84a, l 27sqq.
596
Sra 11:9 in fol. 77b, l 5sqq.; the first of the two variants is an inaccurate quotation
influencedby sra 41:8. WithMingana, p. 20, tofindanotherwise unknownverse of the Koran
here is superfluous.
from the Arabic text but also the fact that among the genuine quotations
from the Koran all sorts of things from tradition have been intermingled.
597
Thus, these quotations say nothing about the earliest history of the text
of the Koran. We can leave them aside, although they deserve an unbiased
and competent investigation.
598
The Victory of the Uthmnic Text
The variants mentioned in the preceding sections by no means exhaust the [iii/102]
store of transmission; the rest of them, however, cannot be grouped coher-
ently. They are traced back to a large number of the Companions of the
Prophet as well as to members of the following generation.
599
The names
of the authorities vary frequently, sometimes in a way that one source lists
a younger authority whereas another one lists an older authority to whose
pupils and transmitters the former belongs, so that both references must
originate from the identical isnd, differing only in their place in the strand
of transmission. According to the more traditionally minded interpretation
of later times, essentially all variants originate from the Prophet,
600
or else
597
Mingana, pp. 2122, who, following his general attitude, assumes that also these sen-
tences hadbeenpart of the underlying basis of the text of the Koran, althoughinone instance
(fol. 82b, l 14) the introductory formula, Muammad has said, is used. Of the four sentences
referred to by Mingana (pp. 1920) that cannot be ascertained in either Koran or tradition,
the first one (Mingana, no. 4) has been discussed in the preceding note no. 594; the second
one (no. 3), fol. 84a, l 1 sq., is sra 3:5, which Mingana did not recognize because, against the
punctuation, he applied the words of the Syrian to what was said before; the third one (no. 2),
fol. 84b, l 3sq., seems to be an inaccurate quotation from sra 16:104, mixed up with 5:50,
which explains the connection with passages like sras 2:81, 2:254, and 5:109; and the last one
(no. 1), fol. 84a, l 7sqq., carriesas Mingana himself admitsthe distinct mark of tradition,
even if until then it had not been documented in adth (the Koranic model is sra 17:90).
Moreover, it would in no way weaken the above-mentioned opinion if also some material
would be found that could not be documented elsewhere.
598
Nldeke in a letter of 13 April, 1925, gives the following opinion. I have now looked
carefully at the deviations of the Syriac text that Baralb presents to us, but have not found
anything that supports the existence of a variant in the Arabic original of the Syrian author.
This Syrian rather added some things for the better explanation of the meaning, even if in
most cases unnecessarily so; and some things he simply misunderstood. But that Baralb
inserts many brief, but also some longer, additions not derived from the Koraneither
adths or entirely different materialadds to the confusion of everything. In any case, the
Syriac text that the person used is not from the early period of Islam The language of the
Syriac Koran does not leave me with the impression of great antiquity.
599
Cf. the list inthe first editionof this book Seite 267sq., andthe index to my (Bergstrer)
Nichtkoranische Koranlesearten imMutasab des Ibn inn.
600
Cf., for example, Ibn Jinn in his al-Mutasab on sra 9:57: hir hdh anna l-salafa
they would not have been included in the Koran. It thus does not come
as a surprise, and did not create astonishment in Islam, that some vari-
ants that deviate markedly fromthe Uthmnic text were explicitly ascribed
to the Prophet himself.
601
It is strange that Uthmn himself is mentioned
as authority for forms of the text that are in contradiction to his recen-
sion;
602
but no one considered it necessary to reconcile the contradiction.
Even if Uthmn might have made an attempt to destroy the copies [iii/103]
of deviating recensions,
603
and even if al-ajjj b. Ysuf might have been
more successful in this respect,
604
the total destruction of all the copies
kn yaqrana l-arf makn narih min ghayr an tataqaddamal-qiratu bi-dhlik, lkinnah
li-muwfaqat ibih f l-man; wa-hdh mawi yajid al-inu bih idh kn hkadh al
l-qirati maanan wa-yaql: laysat hdhih al-urf kulluh an al-nab , wa law knat anh
la-m sgha ibdl lafin makna lafin ; ill usna l-anni bi-Anas yad il itiqadih taqad-
dum al-qirati bi-hdhih al-aruf al-thaltha fa-naql iqra bi-ayyih shita, fa-jamuh
qira masma an l-nab; Anas would not have read this way if he did not think that also
this formwas derived fromthe Prophet (although he does not refer to this, only to the equal-
ity of the importance, see below, p. 463!)
601
Cf, for example, Ab Ubayd al-Qsim b. Sallm, Fail (Berlin Ms., Ahlwardt, no. 451)
38
v
to 42
r
(on sra 5:85, qisssna : iddqna; 24:64, almun : barun; 39:54; 65:1; 92:3, see to
this above, p. 441 n. 416, and Seite 95
3
[but there is no foot-note no. 3 in the German original]).
Ab Dwd al-Sijistn, Sunan, Kitb al-urf wa-l-qirt, trad. 24 (on sra 43:77, trad. 25
(on sra 51:58; both like Ibn Masd), trad. 27 on sra 104:3, yasibu: asibu;) al-Muttaq,
Kanz al-umml, vol. 1, no. 4823sq. (on sra 48:26), no. 4880 (on sra 18:76), and no. 4883 (on
sra 18:78); al-Zamakhshar on sra 39:54, 51:58, 56:81, 65:1, 91:15, 92:3, and 112:1; index to my
(Bergstrer) Nichtkanonische Koranlesearten.
602
Cf., for example, Ab Ubayd, Fail (Berlin Ms., Ahlwardt, no. 451) 38
v
39
v
(on sra
5:120 as conclusionwal-ari wa-Allhsam bar; 18:78 after safnatinaddedliatinintro-
duced by kataba Uthmn, or something similar); al-Muttaq, Kanz al-umml, vol. 1, no. 4836
(on sra 5:120); al-Zamakhshar on sra 2:280; index to my (Bergstrer) Nichtkanonische
Lesearten; and see p. 412 n. 528 (on sra 3:100).
603
604
It is the undeniable merit of P. Casanova (Mohammed et la fin du monde, 2me fasc.,
vol. 1 (1913), pp. 110 and 121129) vigorously to have pointed out that similar to Uthmns mea-
sures aiming at uniformity of the Koransuchendeavours are reportedtohave beenmade also
by al-ajjj b. Ysuf [EI
2
]. But the conclusion that la recension d al-adjdjdj a exist; celle
de Outhmn est une fable (p. 127) far overshoots the mark. The endeavours at vocaliza-
tion going back to al-ajjjs suggestionto which we shall return laterpresuppose the
existence of a written consonantal text; this is the result of the state of the transmission
of the Koran as we have been able to observe in the qira of al-ajjjs contemporary, al-
asan al-Bar (see G. Bergstrer, Die Koranlesung des Hasan, pp. 51 and 54). Apart from
these endeavours, it is a fact that al-ajjj tried to suppress the text and the qira of Ibn
Masd (Casanova, p. 128; and cf. above, p. 457 n. 581); a report from a local Egyptian tradi-
tion with unsuspicious connection makes it rather certain that he dispatched a model copy
of the Koran to Egypt (Casanova, p. 124sqq.; Mingana, An ancient Syriac translation of the
urn, p. 16). All this is a plausible continuation of Uthmns efforts which the Umayyad
administration followed throughout. The dispatch of a model copy makes sense since al-
ajjj was governor of Kfa, the centre of the qurr and the home of the Kfan Korans,
of non-Uthmnic variantswhich had certainly never been achieved
would not have totally eliminated all the copies of non-Uthmnic variants
since they themselves were far fromhaving disappeared. This became possi-
ble only whenalsothe recitationof the Uthmnic recensionwas recognized
as binding; but such a theoretical recognition of the Uthmnic text could
have practical repercussions only when the careless licence of the treat-
ment of the textcharacteristic of the earliest periodwas abandoned,
and respect for the minute rendering of the word of God was awakened.
The Companion of the Prophet, Anas b. Mlik, is reported as saying [iii/105]
that in sra 73:6 he recited awabu instead of aqwamu and when this was
criticized he replied: aqwamu wa-awabu wa-ahyau widun.
605
The same
leniency is expressed in the tradition inna l-Qurn kullah awb m lam
yajal ramatan adhban aw adhba ramatan,
606
or in the account that
the Prophet left it to the copyist to choose from among the formula type of
verse endings like azz akm or sam alm or azz alm,
607
culminating in
the classic example in the anecdote of Ibn Masd and his pupil who cannot
pronounce th.
608
That this was indeed howpeople felt is quite evident from
and, furthermore, that Egypt had not received a copy of the Uthmnic recensions (see above,
p. 305sqq.), and generallyalso laterwas behind the other amr as far as Koranic studies
was concerned. For al-ajjjs subsequent and more basic activity there is only one rather
dismal source in the formof al-Kinds apology (see above, p. 392 n. 27) fromwhich Casanova
derives too much. Al-Kinds account that al-ajjj jamaa l-maif (1885 edition, p. 131, l 12,
similarly p. 137, l 6)not the Koranis explained by the continuation, wa-asqaa min-h
ashya kathratan (he collected the copies of the Koran in order to delete passages) and
at the same time mark them as tendentious lies. The report that he dispatched six copies to
different cities (p. 137, l 10sqq.), is a duplication of Uthmns account.Alphonse Mingana,
The transmission of the urn, follows Casanova.
605
al-abar, juz/vol. 1, p. 17, l 8 (here instead ) and s.v. 29, 71, 24sqq.; Ibn Mujhid in
the Mutasab of Ibn Jinn, s.v., to which Ibn Jinn procures a parallel tradition on Ab Sarrr
al-Ghanaw on sra 17:5 (Mutasab, s.v., rather Ab l-Samml [of Ban Asad; Sezgin, GAS,
vol. 2, pp. 228229; EI
2
]) then both also al-Zamakhshar on sra 73:6; similarly on Anas b.
Mlik on sra 9:57: Ibn Mujhid in the Mutasab, s.v., and afterwards al-Zamakhshar, s.v.
606
al-abar, vol. 1, p. 10, 19, following the seven aruf tradition; etc.
607
al-abar, vol. 1, p. 17, l 25sqq.
608
See above, p. 41 n. 131, and apart fromthe passage cited there, for example, Ab Ubayd,
Fail (cites al-Suy, al-Itqn, naw 16, qawl 9 [Sprenger ed., p. 109, l 11;] thanks to the
kindness of Herrn Studienrat A. Schachner, Schwetzingen, who is preparing an edition of
the Fail, the passage is missing in the Berlin Ms., Ahlwardt, no. 451); (Amad b. Al)
AL-JA (d. 370/981, Sezgin, GAS, vol. 1, pp. 444445), Akm al-Qurn (Constantino-
ple, 1335), vol. 3, p. 445, l 3. Elsewhere the same is reported from Ab l-Dard; al-abar,
juz/vol. 25, p. 71, vol. 29, p. 33, and subsequently al-Zamakhshar on sra 44:44. In the
Mabn li-nam al-man passage cited by Schwally (section 9 of the introduction), apart
from these two, also Ubayy is mentioned; but he alone, al-Muttaq, Kanz al-umml, vol. 1,
no. 4884.Schwally did not properly understand the passage; the pupil, a Persian, lightens
not only the hamzathis also many Arabs did, and this was not particuarly noticedrather,
the large number of old variants,
609
despite Ab l-Khayr IBN AL-JAZARs
(d. 833/1429
610
) strong opposition: wa-amm man yaql: inna baa l-aaba
ka-Ibn Masd kna yujz al-qira bi-l-man fa-qad kadhaba alayhi.
611
It
was possible to reflect upon these matters only when they were no longer
obvious. Thus, the reports cited are likely to originate froma time when the
controversy on the genuine Koran broke out. This was precisely the reason
that the authorizedversionwas established. Whereas Uthmnandal-ajjj
radically deprived the controversy of its basisthe differences of the texts
of the Koranthere were other men who tried to achieve this by cultivating
tolerance.
612
This was the aim of tradition and related reports, namely that
the Koran had been revealed in seven aruf and that the Prophet allegedly
declined to mediate between contradictory versions of the text.
613
Bothof these contradictory tendencies achievedtheir goal; inpractice the [iii/107]
Uthmnic text prevails, andintheory alsothe ancient non-Uthmnic forms
most important, he could not pronounce th so that al-athm in his pronunciation sounded
like al-yatm; to avoid the annoying wrong meaning, a word with the proper meaning had
therefore to be inserted so that he, too, could pronounce (like fjir, lim).
609
Here belongs also the report that Ab anfa gave permission to recite the Koran in
Persian (al-Zamakhshar on sra 44:44; al-Suy, al-Itqn, naw 35, masala 20 [Sprenger
ed., p. 256, l 14]; in the anafite fiqh literature, for example, al-Ksn [sic, d. 587/11911192],
Badi ani [Cairo, 1327/1909], vol. 1, p. 112, l 24sqq.), one of the archaic characteristics of
his teaching. Howit actually happened that verses of the Koran were recited in Persian is not
clear from the juridical discussion.
610
EI
2
.
611
Ibn al-Jazar, al-Nashr (Damascus, 1345/1926from here onwards all citations refer
to this edition) vol. 1, p. 31, l 21, cites al-Suy, Itqn, naw 27 (Sprenger ed., p. 182, l 16.)
Ibn Jinn, on the other hand, states completely correctly in Mutasab on sra 73:6: hdh
yuannis bi-anna l-qawmkn yatabirina l-maniya wa-yukhlidna ilayh, fa-idh aalha
wa-aanh sma f l-ibrt anh.
612
Ibn Masd might have been one of them; in his farewell message to the Kfans, which
at the core might be genuine, he displays a lenient disposition (al-abar, vol. 1, p. 10, l 25sqq.;
cf. Makk b. Ab lib, Ibna, [Berlin Ms., Ahlwardt, no. 578], p. 515).
613
See above, p. 38sqq.; the material could easily be enlarged. The interest of these tradi-
tions, particularly the latter group, is abstract and political, by no means specifically directed
to the Koran; proof is that they never cite the passage of the Koran which is the source of the
controversy, but mentiononly the name of the sra. They clearly distinguishthemselves from
those transmissions which, although involved themselves in the controversy on the text of
the Koran, report differences of opinion among old authorities regarding a certain passage.
Best known are those on sra 62:9 (cf. above, p. 441 n. 409), and the accounts of the dispute
betweenUmar andUbayy (see above, p. 453 n. 553); further, for example, according to Makk,
Kashf, s.v. (cf. al-abar, juz/vol. 6, p. 72, l 20), Al in sra 5:8 reprimands his sons for reading
arjulikum and demands the accusative. It is rare that the Prophet himself makes a decision;
according to Makk, Kashf; al-Dn, Taysr f l-Qurn; Ibn al-Jazar, Nashr on sra 30:53 (also
Amad IBN ANBAL, Musnad, vol. 2, p. 58, l 26sqq.), here, the Prophet rejects Abd Allh b.
Umars vocalization, af and insists on uf.
of the text have been recognized as divine, i.e., as Koran. [Ab Amr] IRR
IBN Umar [i.e., AMR] is accused of heresy for dogmatically rejecting the
texts of Ibn Masd and Ubayy b. Kab.
614
The doctrine of naskh, namely that
the non-Uthmnic forms of the Koranare manskh, abrogated, offeredthe
possibility of mediating between the two schools.
615
According to the most
radical interpretation, this had already been the case after the ar
616
akhr:
During every Raman the Prophet recited to Gabriel what had been
revealed of the Koran up to that time; in the year before his death he did
this twice. According to a more tolerant interpretation it was only the free
choice (ikhtiyr) of the surviving Companions of the Prophet, whoagreedon
the arf laid down in Uthmns Koranthus the ijmthat the remaining
six aruf were abrogated.
617
Their modest fragments have been transmitted,
and even retained a certain importance, if not for the qira but at least for
the tafsr: After all, the dogma of the Seven aruf presupposes that in prin-
ciple their sense is identical. Thus the task emerged to prove this in the
non-Uthmnic forms of the text in each and every case and, at the same
time, there arose the possibility of using them as a means to interpret the
canonical text, which indeed facilitated their disposition.
618
This is essen-
tially the spirit of al-abars treatment of the variants of the Uthmnic text.
Occasionally people went further and claimed that fromthe outset the vari-
ants were intended to be nothing but explanations,
619
but the science of the
Koran abandoned this attitude.
614
al-Shahrastn, translated by Haarbrcker, vol. 1, p. 95 = Cairo printed edition, 13171321
(in the margin of Ibn azm), vol. 1, p. 115, l 15; [EI
2
; EQ; Sezgin, GAS, vols. 1 & 5 (brief)].
615
Already in the saying of Umar, above, p. 453 n. 553.
616
Teaching term, see below, p. 494 n. 157.
617
Both are possible interpretations, Ibn al-Jazar (d. 833/14291430), Nashr f l-qirt,
vol. 1, p. 14, l 16 (both connected, vol. 1, p. 31, l 19sq.); only the second one in the reverse
(maar) of Ab l-asan Muammad b. Amad IBN SHANNABDH of 323/935 ( muafi
Uthmna al-mujmai alayhi; Yqt, Irshd, vol. 6, p. 302, l 20) and the q Isml b. Isq
al-Mlik, d. 282/895896), in Makk, Ibna (Ms Berlin, Ahlwardt, no. 578), p. 501 (without
using the expression manskh) as well as Makk himself, ibid., p. 497 (he says cautiously
ka-annah manskhatun and points out the difficulty that al-naskh f l-Qurn bi-l-ijm fhi
ikhtilf); cf. Ab Jafar Amad b. Muammad AL-AW (d. 321/933, Sezgin, GAS, vol. 1,
pp. 439442), Ab Umar Ysuf IBN ABD AL-BARR (d. 463/10701071, Brockelmann, GAL,
vol. 1, p. 367), and the q Ab Bakr Muammad b. al-ayyib AL-BQILLN [EI
2
; EQ; Sezgin,
GAS, vol. 1, p. 608] (d. 403/10121013) in al-Suy, al-Itqn, naw 16, masala 3, qawl 9 (Sprenger
ed., p. 109, l 11).
618
See above, pp. 444sq., 445 n. 450, 454sq.
619
Amad b. Al AL-JA (305/917370/981), Akm al-Qurn, vol. 3, p. 445, l 3; Fakhr
al-Dn al-Rz, Tafsr on sra 73:6; cf. also below, p. 488sq. (The citation from Ibn Jinn in the
Tafsr of Fakhr al-Dn al-Rz, loc. cit., to which Goldziher refers in his Muslim studies, vol. 2,
p. 224, note 2, does not belong here; the passage has been supplied above, p. 464 n. 611.)
Mlik b. Anas (d. 179/795
620
) is the first person strictly to reject the use of [iii/108]
Ibn Masds recension in prayer. Later fuqah are influenced by this point
of view since the question changed: It was no longer a question whether
or not traditional forms of the text are admissable, but rather whether
intentional or unintentional deviation from the Uthmnic textus recep-
tuspresumed to be binding in principleinvalidates the prayer. For a
long time, at least the anafites were surprisingly tolerant
621
on this ques-
tion, despite the strict conditions imposed on the recitation of the Koran in
ritual prayer as well as in private. From among the earliest Koranic scholars
whom we can identify is al-q (Ab Isq) ISML (IBN ISQal-Mlik
al-Azd al-Baghdd, d. 282/895896) who for his time rejects the non-Uth-
mnic forms of the text;
622
al-abar (d. 310/923) shares this opinion.
623
By
that time, and in face of the realities of cult and teaching, they are likely
to have fallen into oblivion. One stage on this road becomes apparent from
asan al-Bars (d. 110/728) reading of the Koran: Here, a considerable num-
ber of variants is still to be found, but the typical interchange of synonyms
is almost totally absent. Most of the variants are limited to one letter of the
word, and very many amount to no more than simple orthographic differ-
620
al-Mudawwana, section al-alt khalf ahl al-al (Cairo, 1324/1906, vol. 1, p. 84, l
14sqq.): suila Mlik amman all khalfa rajulin yaqrau bi-qirat Ibn Masd; qla yakhruju
wa-yadauhu wa-l yatammu bi-hi (cf. also below, p. 499). According to Ibn al-Jazar,
al-Nashr, vol. 1, p. 14, l 12sqq., also the reverse is transmitted from him, likewise also both
views are represented in the other madhhib. The use of non-Uthmnic variants in and out
of prayer is prohibited by the Shfiite Taq l-Dn AL-SUBK (d. 756/1355; Brockelmann, GAL,
vol. 2, p. 86; EI
2
) by Ibn al-Jazar, Nashr, vol. 1, p. 44, l 5 (cited by al-Suy, al-Itqn, naw 2227,
tanbh 3 [Sprenger ed., p. 191, l 20]).
621
The decisive criteria for the judgement of khaa in prayer is that only such a change
in the meaning of the recited passage of the Koranwhich in itself represents neither
du nor thanannuls the prayer. (The most detailed casuistry I know of is al-Fatw
l-khniyya [Blq, 1310/1892 in the margin of al-Fatw al-lamgriyya, vol. 1, pp. 139161].)
The recensions of Ibn Masd and Ubayy are still referred to (Khniyya, vol. 1, p. 156, l 1 sqq.;
lamgriyya, vol. 1, p. 82, l 1 sqq.) but only by way of example; there is no uniform opinion,
rather the rules governing deviations are applied to all deviations from the Uthmnic text
that might be contained in them. The axiom cited nearly amounts to the toleration of the
qira bi-l-man.; this becomes quite obvious from photocopies in my possession of the end
(Leiden Ms. 1861, fol. 131
v
sqq.) of the furq book of al-Naysbr al-Qumm, which Schacht
analysed in Aus zwei arabischen Furq-Bchern, p. 505sqq.
622
Makk, Ibna (Berlin Ms., Ahlwardt, no. 578), p. 501, from his Kitb al-Qirt.
623
In the introduction to his commentary on the Koran, vol. 1, p. 21, l 20sqq. (l 26: fa-l
qirat al-yawm lil-Muslimn ill bi-l-arf al-wid alladh ikhtrah la-hum immuhum,
namely Uthmn) and in the commentary itself (vol. 1, p. 112, l 28, p. 239, l 31; vol. 2, p. 30, l
16; etc.), as well as in his work on variant readings, al-Bayn; (cf. Makk, Ibna [Berlin Ms.,
Ahlwardt, no. 578,] p. 500).
ences.
624
Within the transmission of Hasan al-Bars reading the tendency
further to push back the variants asserts itself.
625
Quite some time after the controversy had been decided against the [iii/110]
non-Uthmnic variants, a respected
626
teacher (muqri) of the Koran at
Baghdad (Ab l-asan Muammad b. Amad b. Ayyb b. al-alt) IBN
SHANNABDH(d. 328/939) attempted
627
once more to defend their correct
status by using them in the ritual prayer (f l-mirbi) while imm,
628
but
624
Bergstrer, Koranlesung des asan von Bara, p. 50.
625
Ibid., p. 48.
626
Ibn al-Jazar, abaqt: shaykh al-iqri bi-l-Irqi ustdhan kabran, and in the brief
biography, al-Nashr, vol. 1, p. 122: kna imman shahran wa-ustdhan kabran; he is one of
the two transmitters of [AbAmr Muammadb. Abdal-Ramnb. Muammad, d. 280/893]
QUNBUL an Ibn Kathr (Ibn al-Jazar, al-Nashr, vol. 1, p. 118sqq.). The unfavourable opinons
of himfi-hi umq, kathr al-lanqall al-ilm(in the abaqt Ibn al-Jazar says explicitly
maa al-ilm)are likely derived from his enemies.
627
Sources for what follows: Fihrist, p. 31 sq.; al-Samn, al-Ansb, fol. 339
r
; Yqt, Irshd
al-arb, vol. 6, p. 300sqq.; Ibn Khallikn, no. 639; al-Dhahab, abaqt (Berlin Ms. Or. Fol. 3140
(not yet catalogued in 2008) fol. 42
v
sq.); Ab l-Masin IBN TAGHRBIRD, ed. T.G.J. Juyn-
boll, vol. 2, p. 266sq.; Ibn al-Jazar, abaqt (Berlin, Staatsbibliothek, Ms., cod. simul. 55),
176
v
sq. (repeatedly mentioned in the Nashr, particularly vol. 1, p. 39, l 17, where al-q Iy
b. Ms al-Yaub [d. 544/1149, al-Shif, vol. 2, p. 290, EI
2
; Sezgin, GAS, vol. 1, pp. 136, etc.]
is cited); Tj al-ars, s.v.; I. Goldziher, Muslim studies, vol. 2, p. 222, and Schools of Koranic
commentators, p. 30sq.; A. Wiener, Die Fara bad a-idda-Literatur, p. 292; A. Mez, Renais-
sance of Islam, p. 194sq. In Tj al-ars also the spelling Shannabdh, and commonly Shan-
bdh.
628
He seems to have collected them personally; thus, in his Kitb Infirdtihi (Irshd,
vol. 6, p. 302, 3). The list of the reading violations he was charged with can still be found
in its original form in the minutes of the proceedings in Ibn Khallikn and al-Dhahab,
introduced by suila Ibn Shannabdh amm ukiya an-hu anna-hu yaqrauhu, and with
fa-tarafa bi-hi after each passage; further, without this framework in the Fihrist, in the Irshd,
and in the abaqt of Ibn al-Jazar; some portions also in Ab l-Masin IBN TAGHRBIRD.
Verification shows that Ibn Shannabdh followed the variants of Ibn Masd in sras 18:78
(not Ibn Khallikn), 34:13 (not Irshd; al-Fihrist erroneously instead [thus var.],
Abl-MasinIBNTAGHRBIRD, likely alsoerroneously, tayaqqanat insteadof tabayyanat);
sras 62:9, 92:3 (see above, p. 441 n. 416), 101:4 (not al-Fihrist), 111:1 (neither al-Fihrist nor
al-Dhahab); further, those of Ibn Abbs in sra 18:78 (ammahum instead of warahum, cf.
al-abar, vol. 16, p. 2, l 11; neither al-Dhahab nor Ibn al-Jazar); 25:77 (kadhdhaba l-kfirna
instead of kadhdhabtum, cf. Ibn Mujhid in Ibn Jinn, Mutasab s.v.; Ibn Abbs and Ibn
al-Zubayr); 34:13 (after labith add awlan, cf. al-abar, vol. 22, p. 45, l 9; not Irshd); 56:81
(shukrakum instead of rizqakum, cf. Ibn Mujhid in Ibn Jinn, Mutasab s.v.: Al, Ibn Abbs
and the Prophet, not al-Fihrist); and, finally, an addition of Uthmn and Ibn al-Zubayr in
sra 3:100 (after al-munkari added wa-yastanna Allh al m abahum, see al-abar,
vol. 4, p. 24, l 19sqq.; al-Muttaq, Kanz, vol. 1, no. 4833;in the Fihrist, nhna is an error
in the preceding part of the verse, and the continuation is a misplaced portion of sra 16:80;
not Irshd). Only two deviations cannot be documented, but are undoubtedly based on old
authorities: sra 8:74, arun instead of kabrun, and sra 5:118only Irshd, vol. 6, p. 304, l
1, and al-Dhahab outside the listal-ghafr al-ramu instead of al-azzu l-akmu (cf. to it,
without avail. In 323/934 he was summoned before a special court of justice,
consisting of qs, fuqah, andqurr presidedover by the wazr al-asanb.
Al IBNMUQLAH,
629
who demanded repentance (tawba). When he refused,
he was subjected to corporal punishment until he finally signed the reverse
(maar),
630
promising to respect the Uthmnic text. The populace, un-
doubtedly incited, bore so much resentment against him that under the
cover of darkness he secretly had to leave
631
the house of the wazr, where he
had been held, and took temporary refuge at Madin (or Bara). The person
who brought him to court and pursued his punishment was his old oppo-
nent,
632
Ibn Mujhid (d. 324/935),
633
the most successful of all the teachers of
the reading of the Koran, and the founder of strict orthodoxy in the field of
the science of variant readings.
Only very few and rather unimportant deviations from the Uthmnic [iii/112]
text in their canonical form
634
found their way into the systems of the Seven
Readersapart from the extensive use made of variants as well as ortho-
graphic licence. Some readers
635
did not recognize in sra 8:44 that
636
is
constructed according to FormII and replaced it with ayiya. This is compa-
rable with the case when some scribes
637
read tniyah Allhu in sra 27:36,
although the text has = tni <tn (see above, p. 413sq.).
638
In the case
of sra 18:95 it was obvious to replace the second
with Form I; this was

above, p. 463, andIbnMasdonsra 9:107). One deviationremains not clear insra 10:92; the
lists are different, and the literature of the variant readings does not permit a sound verdict.
629
EI
2
; EQ; Sezgin, GAS, vol. 1, p. 386, l 3 from bottom.
630
The text has survived, the most detailed version of which with signatures, is in the
Irshd.
631
This is in contradiction to the report in the Fihrist, which is also mentioned by Ibn
Khallikn, with wa-qla, that he died in prison; this report might explain that the year of his
death is also given as 323/934 instead of 328/939.
632
Details are reported by Ibn al-Jazar, but first of all by Yqt, Irshd, vol. 6, p. 207, l
8sqq., as well as al-Dhahab, according to whom Ibn Shannabdh did not accept students
who had taken classes with Ibn Mujhid. Another opponent was Ab Bakr IBN AL-ANBR
(d. 327/938); he had polemized against him (al-Samn; Ansb; Sezgin, GAS, vol. 1, pp. 347
and 353.; Irshd, vol. 6, p. 300, l 5, and vol. 7, p. 77, l 2;the name Ibn Shannabdh is mis-
represented in the title of the work in al-Fihrist, p. 75, l 23, and G. Flgel, Die grammatischen
Schulen, p. 170, no. 10).
633
Most clearly in Irshd, vol. 6, p. 302, l 7.
634
Additional deviations fromthe consonantal text beyond these canonical forms of their
qirt are ascribed to some of the Seven; cf., for example, the index to my (Bergstrers)
Nichtkanonische Koranlesearten.
635
Nfi b. Abd al-Ramn, al-Bazz an Ibn Kathr, Ab Bakr SHUBAH (EI
2
; Sezgin, GAS,
vol. 1, pp. 1011) an im (Ibn Ab l-Najjd).
636
Thus al-Dn, al-Muqni, bb 8.
637
Nfi al-Layth, Ab Amr b. al-Al, af b. Sulaymn an im (Ibn Ab l-Najjd).
638
Thus al-Muqni, bb 3, fal 1.
partly done also in the case of the first

639
where Form I is ill-suited. In
sras 11:71, 25:40, 29:37 and 53:52 the form
640
(triptote) violated the com-
mon rule so that it was changed to Thamda.
641
All this is within the bounds
of orthographic licence. Stronger interferences are found most frequently
in Ab Amr b. al-Al, whose text, despite all purges, retains remnants of
its sometime originality. All this remains on the border of what was ortho-
graphically possible. Operating on the very fringe of what is perhaps ortho-
graphically still tolerable, he interprets
642
the of the text
643
(sra 2:100)
as nansh = nansah we postpone itthe verse of the Korani.e., we
do not yet make it binding, in order to avoid the offensive common reading
nunsih we consign the verse of the Koran to oblivion. Also transgressing
these last orthographic limits, he corrects the grammatical error , sra
20:66 (see above, pp. 390 and 391 n. 20), deletes in sra 77:11 the dialectal
form
644
for wuqqitat, and honours Allhs privileges in sra 19:19 by li-
yahaba may He grant you a pure boy
645
instead the first Person of the
text.
646
In accordance with the majority of the readers
647
he interprets the
most obscure , sra 106:12 (see above, p. 421) by sheer force as
li-lfi lfihim.
In addition to these few passages the canonical interpretation of the [iii/113]
Koran insists on the licence of the old compromise to read differently from
what is written (see above, p. 418sq.). Everywhere else the text of the Koran
came out as the winner. Ifas is nearly always the caseone does not
adapt the reading to the text, at least one invents an alleged consonantal
variant in support of the variant reading.
648
The subjugation to the letter of
639
In the second case amza, and according to most scholars Ab Bakr an im, in the
first case only this one. Cf. above, p. 424 n. 244.
640
Thus al-Dn, al-Muqni, bb 5, fal 1.
641
imal-Jadar (only in the first three cases, according to the transmission of af [Ibn
Sulaymn] and amza b. abb).
642
The same way here Ibn Kathr.
643
Thus Muqni, bb 21.
644
Thus Muqni, bb 21.
645
Here likewise Nfi b. Abdal-Ramn; (Qlns transmissionfromhimis controversial).
646
Thus Muqni, bb 5, fal 2.
647
Only Abd Allh IBN MIR al-Yaub [Sezgin, GAS, vol. 1, pp. 67] reads li-lfi
lfihim (sic!)
648
Thus sra 63:10 (see above, p. 401 n. 87), where again Ab Amr b. al-Al corrected the
text; sra 76:4 (see above, p. 401 n. 84), where Ibn Kathr, Ab Amr b. al-Al, Ibn Dhakwn
an Ibn mir, af b. Sulaymn an im b. al-Ajjj and amza b. abb read salsila, but
Ab Amr b. al-Al and, according to a widely differing transmission, also part of the others
recognize the of the true text, at least in pausal pronunciation salsil; probably also
sra 76:16 (see above, p. 399 n. 79), where Ibn Kathr, Ab Amr, Ibn mir, af an im, and
the word reaches its apex when the ambiguity of the script is disregarded
and, by slavishly following the closest interpretation, senseless forms of the
text are construed that have never existed in genuine oral transmission.
649
Generally speaking, Ibn mirs variant reading is the most dependent on [iii/115]
the consonantal text. In comparison, the most independent one is that of
Ab Amr b. al-Al, followed by that of amza b. abb. Some remnants of
liberty can still be found in the only two remaining systems today, af b.
Sulaymn an IM b. al-Ajjj AL-JADAR, and Uthmn b. Sad WARSH
an Nfi b. Abd al-Ramn al-Layth: Even in our days, the Uthmnic
consonantal text has not yet been totally established.
amza assume , although Hishmb. Ammr an Ibn mir only for the pronunciation in
context (verse 15 has at the end of the verse, might thus be considered a rhyme; in spite
of this, amza ignores the even in pause). As to the possiblity of using the changes in pausal
pronunciation for a closer connection to the consonantal text, see above, p. 425 n. 247. There
are numerous similar cases, even if less spectacular.Also in sra 23:89 and 91, the Baran
insteadof everywhere else seems tosupport AbAmrsjustifiedcorrection, Allhu;
cf. above, p. 398 n. 63.
649
Sra 6:32, see above, p. 397 n. 56 (Ibn mir reads la-dru l-khirati); Sra 6:57, see
above, p. 409 n. 141 (meant was yaq, but Nfi Ibn Kathr and im read yaquu); sras
12:110, and 21:88 (see above, p. 425) Ibn mir and im read nujjiya (in the second passage
only according to the transmission of Ab Bakr); sra 18:36, Ibn mir reads also in context
lkinn although the final of is likely to correspond to the one of (lkinna <lkin ana,
cf. above, p. 437sq.) the others, - only in pause; sras 22:23, and 35:30, see above, p. 423 n. 233
(accusative read Nfi and im); sra 27:25, the , which must somehow include an
affirmative invitation in a peculiar form of speaking and writing (cf. above, p. 438sq.), which
the majority interpretedas allyasjudinorder not toprostrate, andfromamong the Seven
Readers only by al-Kis interpreted as a-l y isjud, prostrate! (see above, p. 426 n. 254);
sra 49:14, six of the Seveners are misledby the spelling , andonly AbAmr b. al-Al has
the correct form, ylitkum= yalitkum(see Nldeke inC.C. Torrey, The commercial-theological
terms, p. 33 note 1); sra 75:1, Ibn Kathr reads la-uqsimu because of the spelling (see
above, p. 424 n. 239). Here, too, belong the forms yyas, etc., sra 12:80, etc. in al-Bazz an
Ibn Kathr (see above, p. 424 n. 245), even if indeed there had been beside yaisa a secondary
form ayisa, its Imperfect ought to read in Ibn Kathr
*
yayas, since the deletion of hamza at
the end of the syllable is unknownto him. Most likely also the reading la-takhidhta, sra 18:76
(Ibn Kathr, Ab Amr) ought to be here (cf. above, p. 425 n. 251).
THE VARIANT READINGS
1
Basic Problems: The Sources
The question whether deviations from the Uthmnic consonantal text are [iii/116]
permitted is only one of the many basic problems which developed early on
from the practice of reciting of the Koranbeginning approximately with
the tradition of the Seven aruf
2
and its family. The changing response to
this question became in turn a permanent influence upon its praxis. It is
such problems and their history that determine the place that the history of
the Koran takes within Islamic history of ideas, and which, beyond the value
of preliminary work for the establishment of the text of the Koran, creates a
scientific merit of its own.
Discussions of principles found their literary expression in introductions
to commentaries of the Korancertainly since al-abar, but not yet in the
time of Yay b. Ziyd AL-FARR (d. 207/822)
3
and in the works on vari-
ant readings. Our most precious source is the introduction to one of them,
the great al-Nashr f l-qirt al-ashr, by Ibn al-Jazar (d. 833/1429),
4
whose
wealth of citations from relevant older literature make it particularly valu-
able for historical research.
Ibn al-Jazar previously
5
treated the subject in a monograph entitled
Munjid al-muqrin
6
wa-murshid al-libn;
7
the following chapter headings
8
might serve to illuminate the range of the relevant questions: (1) qirt,
muqri and qri as well as their functions; (2) al-qira al-mutawtira, al-
aa, and al-shdhdha, scholars differences of opinion, and the truth
1
Although I am quite aware of the imperfection of this translation I continue to use it
for lack of something better, equally convenient. It must be observed that qira is used in the
double sense, for the individual type of variant reading (reading of one particular passage) as
well as the totality of variant readings and pronunciations of one reader (qri); in the latter
meaning the translation (Koranic) reading is been used. Muqri is rendered also teacher of
the Koran.
2
3
Sezgin, GAS, vol. 7, pp. 343344, vol. 8, pp. 123125, vol. 9, pp. 131134.
4
Printed, Damascus, 1345/1926; there, vol. 1, pp. 153; Brockelmann, GAL, vol. 2, p. 201.
5
See below, p. 504 n. 219.
6
Muqarrabn is taf.
7
Manuscripts also in Istanbul, Ragp Paa Ktphanesi, 14, 2
v
27
v
, and 15, 219
v
sqq.
8
With the exception of the first one, also in W. Ahlwardt, Verzeichnis der arabischen
Handschriften der Kniglichen Bibliothek zu Berlin, no. 656.
472 the variant readings
about the matter; (3) the Ten Readings have been recognized (mashhra)
from their origin down to the present; (4) a list of the famous readers who
recited and taught according to the Ten Readings; (5) citations of scholars
comments on those; (6) the Ten Readings constitute only one part of the
Seven aruf; (7) scholars who rejected the limitation to the Seven, and who
therefore reproved Ibn Mujhid.
The survey shows that Ibn al-Jazars purpose is not purely academic but
rather thatas in al-Nashr f l-qirt al-ashrhe struggles for the equal
treatment of the three readings beyond the seven readings. This saves his
discussion from scholastic dryness.
The second part of the title seems to allude to a work of Ab Shma [iii/117]
(d. 665/1266),
9
which, despite its more comprehensive title, one tends to as-
sociate with the Munjid al-muqrin wa-murshid al-libn precisely because
Ibn al-Jazar repeatedly cites
10
al-Murshid al-wajz f ulm tataallaq bi-l-
Qurn al-azz. As for the range and tenor of the subject, a precursor
probably the earliestis Makk b. Ab libs
11
Kitb al-Ibna,
12
which was
heavily used by Ibn al-Jazar. As for substance, it remains far behind the
introduction to al-Nashr f l-qirt al-ashr, but is the next most important
source on account of its greater age.
The Nashr itself is the main source of the relevant chapters
13
of al-Suys [iii/118]
(d. 911/1505) al-Itqn f ulmal-Qurn, although it is not the sole source.
The Connection with the Consonantal Text
Even if most of the Koran were written down during the lifetime of the
Prophet,
14
indeed even on his orders, nevertheless the Koran still had to
remain first of all the orally revealed Word of God, knowledge of which
9
Actually Shihb al-Dn Ab l-Qsim Abd al-Ramn b. Isml al-Maqdis al-Dimashq
(d. 665/1266), Brockelmann, GAL, vol 1, p. 317.
10
al-Nashr f l-qirt al-ashr, vol. 1, p. 9, l 22; vol. 13, p. 9; and vol. 38, p. 9.
11
Full name: Ab Muammad MAKK IBN AB LIB ammsh al-Qays, d. 437/1045,
Sezgin, GAS, vol. 8, p. 251; and vol. 9, pp. 214215.
12
The title is not quite certain. On the one hand, Ibn al-Jazar cites in his al-Nashr, vol. 1,
p. 46, l 17, a passage entitled al-Ibna which is contained in W. Ahlwardt, Verzeichnis, Berlin
Ms. no. 578 on page 517sq.; on the other hand, he everywhere else cites the introductory work
of Makk b. Ab lib, without specific title, as supplement to al-Kashf an wujh al-qirt
which actually is in the Berlin Ms. (pp. 494519). On page 491, Makk explicitly states that
it was published separately, although without supplying its title. The content of al-Ibna is
givenby Haj Qalfa [sic], s.v.: F man l-Qurn; Yqt, The Irshad, ed. D.S. Margoliouth, vol. 7,
p. 174, l 19, however, has Man l-qira, andprecisely man l-qirt is according toKashf, p. 2,
the content of the supplement.
13
Naw 2227, and parts before naw 16, 18, 20, and 21.
14
See vol. 2, p. 1 sq.
the variant readings 473
was passed on by word of mouth. That this was actually the case becomes
evident fromthe many deviations fromthe Uthmnic text originating from
the oral traditionand even in written forms fixed at an early date.
15
In order to collect the revelations, one could have memorized them as a
rw does poetry. Yet this task would have been far more difficult; it thus
makes sense that, according to tradition, Zayd b. Thbit preferred to use
writtensources
16
whenproducing his first collectionof the Koranunder Ab
Bakr or Umar. The other collectors are likely to have followed suit.
The emphasis of the transmission of the Koran thus shifted toward writ-
ten books. The Uthmnic recension must have followed the same method,
particularly as it claimed to be the officially sanctioned Koran. The oral
transmission alone would have hardly been sufficient to spread knowledge
of the Koran. Among the old followers of the Prophet there can hardly have
been more men who had memorized considerable portions of the Koran
than there had been among the Arabs who transmitted the ancient poems
and narratives of the ayym al-Arab. The number of men who knew indi-
vidual parts of the Koran must have varied greatly, although on average it
was probably not considerable, and in the case of less important parts it
could well have been extremely small. During the turbulent years of Ab
Bakr and Umar, the activities of the Prophets Companions did not allow
them to pass on their Koranic knowledge to many other men. In Medina
the source of oral transmissions might have been more abundant, but in the
new centres of Islam in the conquered territories it would have been diffi-
cult for men desiring to know all of the Koran to find an expert in all of its
parts. At the same time, it was precisely in those centres of Kfa, Bara, and
Damascus that the urgent needfor competence inthe Koranmust have been
felt, given the aimof establishing an Islamic communal life in close contact,
and in rivalry with, the People of the Book. This was facilitated by the writ-
ten collectionsin Kfa the one of Ibn Masdbut, most of all, Uthmns
recension.
17
The Battle of iffn (37/657) illuminates the situation: some five
years after Uthmns recension was issued, the Syrians are in possession of
maif which they fixed on their lances;
18
and among the Irqs the qurr
constitute an influential party.
15
See above, p. 444sqq., and 454sq.
16
See vol. 2, p. 13sq.; A. Jeffery, Materials for the history of the text of the Qurn, p. 223;
17
One may ask whether this was not the purpose or at least one of their purposes next to
the uniformity of the text.
18
This canhardly have beenmeant literally; a Kfanmuaf is usually rather voluminous.
A number of facts suggest that the oral distribution of the Koran during [iii/120]
the early period was followed by a period when the study of the written
text prevailed. The variants of the Uthmnic copies survive in the qirt of
those cities to which these copies had been dispatched.
19
Some mistakes of
the copyist, for example, in the qira of asan al-Bar (d. 110/728),
20
appear
in the text, but, most importantly, a vast number of variant readings devel-
oped that interpret identical consonantal forms in different ways. Naturally,
it is conceivable that multiple versions of the text appeared when orally
transmitted but their differences remained unexpressed in the consonantal
script, with the result that, for example, someone who had heard sra 11:48,
inna-hu amalun ghayru liin, would say instead inna-hu amila ghayra
liin or vice versa. Yet it is infinitely more probable here that the same con-
sonantal form is read differently, and that conversely, in the
case of oral variation, duplicates appear that also vary in consonantal form,
as is the case, for example, in the interchange of synonyms. There is a whole
group of variant readings in which the origin from the consonantal text is
made still more likely. These are those that obviously result fromthe search
for, and joy in, the unexpected aspects of the consonantal text. These types
of variant readings are clearly visible in asan al-Bar (d. 110/728).
21
The
half century that separates asans prime of life from Uthmns recension
(approximately 32/652) is likely to have been the period when the bulk of
variant readings were created on the basis of the written text.
In the first half of the second century ah, the Baran grammarian (Ab
Umar) S IBN UMAR AL-THAQAF
22
(d. 149/766) could still attempt to
establish
23
a strictly purist reading of the Koran (al qiys [or madhhib]
al-Arabiyya),
24
which undoubtedly means that he wanted to introduce new
19
20
G. Bergstrer, Die Koranlesung des asan von Bara, p. 51.
21
Ibid., p. 54.
22
EI
2
; Sezgin, GAS, vol. 9, pp. 3739, etc.
23
Ibn al-Jazar, abaqt, p. 161
r
(Berlin, Staatsbibliothek, Ms. simulata orientalia, no. 55).
Ab Ubayd (al-Qsim b. Sallm) finds it typical for this reading of the Koran that Ibn al-
Jazar prefers the accusative, for example, sra 111:4, ammlata (among the Seven only
im al-Jadar, and therefore in the common text); sra 24:2, al-zniyata wa-l-zniya; sra
5:42, wa-l-sriqa wa-l-sriqata; sra 11:80, ahara. The passages have been the subject of
further discussion at Bara; still Sbawayh treats them (33, 116, and regarding the lat-
ter passage cf. al-Zamakhshar) and says (33): wa-lkin abati l-mmatu ill l-qirata bi-l-
rafi.
24
Similarly characterized is the much older reading of Ibn Muayin (d. 122/739 or
123/740) inIbnal-Jazar, al-abaqt, p. 207
r
(Berlin, Staatsbibliothek, Mss. simulata orientalia,
no. 55).
types of variant readings that better suited his linguistic inclination, and
which were a novelty to tradition. In the second century the reverse ten-
dency, the restrictionof licence, alsosets ininthis city. Inthe thirdcentury, it
is aggravated by the renewed battle against ray in fiqh and Koranic exegesis,
initiated by Dwd (Ibn Khalaf) al-hir, d. 270/884.
25
The introduction of
newtypes of variant readings, too, was nothing but an exercise in free judge-
ment.
26
Al-Mubarrad (d. 285/898), as an outsider, was still assuming that the
mature reader was permitted to use discretion;
27
al-abar (d. 310/923), on
the other hand, also shares the opinion of all orthodoxy in this case.
28
There
are two main currents in this direction that run parallel: concurrently with
the disappearance of the non-Uthmnic variants, independent interpreta-
tion of the consonantal text begins to wane. The latter current comes to an
end shortly before the first one in a surprisingly similar fashion, namely the
proceedings against Ibn Miqsam,
29
a pupil
30
of Ibn Shannabdh, and the last
representative of non-Uthmnic forms of the Koran.
Ab Bakr (Muammad b. al-asan b. Yaqb b. al-asan) IBN MIQSAM [iii/122]
al-Ar (d. 354/965 in Baghdad), a respected teacher of the Koran and
25
EI
2
; I. Goldziher, The hirs, p. 27sqq.; Sezgin, GAS, vol. 1, p. 521, no. 10.
26
al-abar, juz/vol. 1, p. 59, l 20, states: m shadhdha min al-qirti amm jat bi-hi
l-ummatu naqlan hiran mustafan, fa-rayun li-l-aqqi mukhlifun ; and regarding (Mu-
ammad b. al-asan) IBN MIQSAM (265/878354/965, Sezgin, GAS, vol. 9, pp. 149150)
his contemporary AB HIR (Abd al-Wid b. Umar b. Muammad) b. Ab Hshim
(d. 349/960) AL-BAZZR [Sezgin, GAS, vol. 9, pp. 167168] and a pupil of Ibn Mujhid says:
jaala li-ahli l-ildi f dni Allhi bi-sayyii rayihi arqan yatakhayyaru l-qirti min jihati
l-bathi wa-l-istikhrji bi-l-ri dna l-itimi wa-l-tamassuki bi-l-athari l-muftarai (Yqt,
al-Irshd al-arib il marifat al-adib, vol. 6, p. 499, l 18sqq.).
27
His statement about sra 2:172 (in al-Zamakhshar, s.v.), law kuntu mimman yaqrau
l-Qurna la-qaratu wa-lkinna l-barra (instead of birra), Ibn al-Munayyir (d. 683/1284, EI
2
)
properly interpreted in his K. al-Intif (in the margin of the Cairene Kashshf) anna ikhtilfa
wujhi l-qirati mawklun il l-ijtihdi wa-annahu mahma iqtahu qiysu l-lughati jzati
l-qiratu bi-hi li-man yuaddu ahlan lil-ijtihdi f l-Arabyati wa-l-lughati and strictly rejected
it. Cf. Goldziher, Schools of Koranic commentators, p. 32sq.
28
The recognition of independent types of variant readings would be as little in agree-
ment with the ijm that he always emphasizes as is the case with the naql; cf. above, p. 475
n. 26, and p. 484sq., and also p. 477 n. 42.
29
Sources for what follows are: Ibn al-Anbr, Nuzhat al-alibb f abaqt al-udab,
p. 360sqq. (with wrong name); Yqt, Irshd, vol. 6, p. 498sqq. (excerpts: al-Suy, Bughyat
al-wut f abaqt al-lughawiyyn wa-l-nut; s.n. [not identified]); Ibn al-Jazar, al-Nashr,
vol. 1, p. 17, l 4sqq., p. 167, and abaqt (Berlin Ms. cod. Or. simul. 55), fol. 195
v
sq.; al-Fihrist,
p. 33, has only a brief note (withthe differing year of death, 362). Cf. Goldziher, Muslimstudies,
vol. 2, pp. 221222, his Schools of Koranic commentators, pp. 3132; A. Mez, The Renaissance
of Islam, pp. 195196; G. Flgel, Die grammatischen Schulen der Araber, p. 179sq., does not
mention the proceedings against him; Sezgin, GAS, vol. 9, p. 20.
30
Ibn al-Jazar, abaqt, Berlin Ms. cod. simul. 55, fol. 177
r
.
grammarian of the Kfan School,
31
was of the opinion that any faithful and
grammatically correct type of variant reading in agreement with the con-
sonantal text is permissible, even if none of the ancestors ever read it this
way. The sultan summoned him in 322/933 to appear before an assem-
bly of fuqah and qurr who unanimously disapproved of his teaching,
and threatened to punish him; Ibn Miqsam then recanted and signed a
maar, promising to abandonhis particular types of variant readings.
32
But
he later resumedhis way of reading. The only example of this variant reading
together with the subsequent discussion that has come down to us
33
shows
that it was not only the theory that caused the ill-feeling against him, but
also the way it was applied: Ibn Miqsam read sra 12:80, nujaba instead
najyan, which his opponents declared senseless, and disapproved of partic-
ularly because not only didthe vocalizationdiffer fromthe commonone but
also the diacritical marks were displaced (taf).
34
Strictly speaking the des-
ignation bida did not apply to Ibn Miqsams teaching because his applica-
tionof irtijl
35
the liberal inventionof possible readings of the consonantal
scripthad been common practice long before his time;
36
but it seems to
have been considerably different
37
from the practice of younger predeces-
sors to whom he refers as authorities
38
Ab Ubayd al-Qsim b. Sallm
(d. 223 or 24/838), Khalaf b. Hishmal-Bazzr (d. 229/844), and Ab Jafar al-
arr MUAMMADIBNSADN
39
(d. 231/846). The man who opposed him,
and who started the proceedings against him, was again Ibn Mujhid.
40
It
was probably after the latters death in 324/936 that Ibn Miqsam reportedly
returned to his early variant readings.
31
Different fromIbn Shannabdhs case, the verdict regarding Ibn Miqsamwas generally
appreciative. Apart from his writings on the science of the types of variant readings, he is
important for his role as one of the four transmitters of Idrs al-addd an Khalaf b. Hishm
an amza (Ibn al-Jazar, al-Nashr, vol. 1, p. 55, l 13, and p. 159sq.).
32
According to Miskawayh, Tajrib, vol. 1, p. 285, l 13; and Izz al-Dn IBN AL-ATHR,
al-Kmil f l-tarkh, year 322ah, also his books were burned.
33
Yqt, al-Irshd, vol. 6, p. 499, l 1 sqq.; the variant reading is derived from Ibn Miqsams
Kitb al-Itijj lil-qurr.
34
Cf. below, p. 490.
35
Ibn Jinn, Mutasab, on sra 9:42.
36
37
38
Yqt, Irshd, vol. 6, p. 500, l 13.
39
40
One tends to consider Ibn Mujhids interference, and possible the entire proceedings
against Ibn Miqsam, a duplicate to the reports about Ibn Shannabdh if it had not already
been witnessed by one of Ibn Mujhids younger contemporaries and pupil, AB HIR b.
Ab HshimAL-BAZZR (d. 349/960); in Yqt, Irshd, vol. 6, p. 499, l 12sqq.; cf. above, p. 475
n. 26.
The independent interpretation of the graphic outline of the script re- [iii/123]
gardless of the existing oral transmission is recommended by statements
mostly ascribed to Ibn Masdlike dhakkir al-Qurn, i.e., in case of doubt
particularly when it concerns the imperfect preformativesone ought
to prefer the masculine to the feminine.
41
Frequently, hypothetical forms for
ones own readings, not taken from tradition, are supplied, e.g., lawquria
(la-) kna jayyidan (awban, jizan, awjaha) and similar ones.
42
Linguistic Accuracy
Of the two conditions for the admission of a Koranic reading established [iii/124]
by Ibn Miqsam,
43
linguistic accuracy and the agreement with the Uth-
mnic muaf,
44
the second one has already been discussed;
45
and the first
of them is of no importance. In the face of the huge influx of mawl into
Islam mistakes in the recital of the Koran must have occurred frequently
but could never claim recognition, and could hardly inadvertently become
established, given the number of genuine Arabs,
46
and the fact that their lin-
guistic self-confidence was too strong. Arabs, too, were prone to mistakes,
41
Makk b. Ab lib, al-Kashf, on sra 2:45 (Ab Ubayd from Ibn Masd; in a slightly
different form from Ibn Masd and Ibn Abbs) and more often; al-Suy, al-Itqn, naw 35,
masala 18 (Sprenger ed., p. 255, l. 4sqq.). al-Suy (Sprenger ed., p. 256, l 7) points out that
the qirt of Ibn Masds followers conforms with this rule. According to another version
(al-Itqn, loc. cit., Sprenger ed. p. 255, l 9sqq.) Ibn Mujhid (!) suggests that in case of doubt
to prefer the formwith preformative y-to that with preformative t-, the one without hamz to
the one with hamz, the one with wal to the one with qal, the one with madd to the one with
qar, and the one with fat to the one with imla.
42
Sbawayh 264 (twice); Yay b. Ziyd AL-FARR (d. 207/822 [sic]), Man l-Qurn
on sra 11:29 and 11:37, and often; [N. Kinberg, A Lexicon of al-Farrs terminology]; abar,
juz/vol. 1, p. 295, l 14 (more explicit: law kna maqran ka-dhlika); Ibn Jinn, Mutasab
often (cf. the introduction to my [Bergstrers] Nichtkoranische Lesearten); al-Zamakhshar
on sra 25:22 and 34:19, and often. Also la-kna jizan of Ibn Miqsam(Yqt, al-Irshd, vol. 6,
p. 499, l 2) is likely to presuppose a law.Already Ab Amr b. al-Al is ascribed to have said:
lawlanlaysal anaqraaillbi-mquria, la-qaratuka-dhwa-ka-dh (al-Dhahab, abaqt,
Berlin Ms. Or. fol. 3140, 14
r
, twice; Ibn al-Jazar, abaqt, Berlin Ms., cod. simul. 55, 76
v
;Ibn
al-Jazar reports something similar also about Nfi al-Layth in al-Nashr, vol. 1, p. 17, l 16). The
absence of a reference (cf. above, p. 464 n. 613) clearly indicates the tendency which is trying
to present Ab Amr b. al-Al (Nfi) as the champion of the principle of tradition.
43
See above, p. 476.
44
The third type, faithfulness, can be ignored as being obvious.
45
Page 342sqq.; and p. 407sqq.
46
From among the Seven Readers Ab Amr b. al-Al and Ibn mir al-Yaub are still
true Arabs (al-Dn, al-Taysr, introduction).
either through inadvertence in difficult passages
47
or through pressure of
particular readings or interpretations;
48
more frequently stilland particu-
larly in face of the great variety of the dialectswhat appeared to be correct
for one person someone else might have considered wrong.
49
There were
thus plenty of variant reading possibilities open to linguistic discussion. But
this controversy never attained fundamental importance.
Readers and grammarians are sometimes at variance in their approach. [iii/125]
In the early period, at Kfa as well as at Bara, several readers were gram-
marians at the same time;
50
whoever wanted to go beyond the mere rou-
tine of Koranic teaching turned to grammar or exegesis.
51
When people
turned to collecting the variant readings as a purpose of itself, and trans-
mitted themthus establishing a proper science of variant readings
the confidence of the qurr was strengthened, and the preponderance of
grammar declined. The outstanding exponent of this stage is Ibn Mujhid
(d. 324/936).
52
The same way that he is aware of his superior familiarity inthe
field of transmission compared with that of the grammarian and exegete
al-Farr (d. 209 [sic]);
53
conversely Ibn Jinn (d. 392/1002) looks down on
Ibn Mujhids linguistic judgements
54
with the proud conviction of newly
established virtuosity of linguistic explanation which makes it possible to
recognize much that was previously difficult to accommodate in grammar
but, at the same time, obliged one to reject other things that one previously
47
al-ajjj b. Ysuf [Sezgin, GAS, vol. 5, pp. 225226], who was proud of his faa [elo-
quence,] once read in sra 9:24, aabbu instead of aabbaseparated from the governing
kna by twelve words(Yqt, Irshd, vol. 7, p. 296, l 17sqq.).
48
For example, li-ittib al-muaf; the most glaring case being al-takhff al-rasm of
amza b. abb (see above, p. 425 n. 247) to which Ibn al-Jazar replies: l yajz f wajh min
wujh al-Arabiyya (al-Nashr f l-qirt, vol. 1, p. 16, l 23).
49
In the passage of the first edition of this book (Seite 285, note 1) al-Suy, al-Itqn, naw
2227, tanbh 6 (Sprenger ed., p. 195, l 1 sqq.), it reads irbni two vowel possibilities for case
and modus.
50
See below, p. 510sqq. Later, for example, Muammadb. Sadn(d. 231/846, [Sezgin, GAS,
vol. 9, pp. 135136]; below, p. 484 n. 96), Ab l-Abbs AL-FAL IBN IBRHM (p. 430), Ibn
Miqsam (p. 420sq.)
51
Differences of opinion regarding linguistic accuracy among grammarians and readers
of the Koran existed already at that time; cf. above, p. 474sq.
52
His elder contemporary, al-abar (d. 310/922), also knows his grammariansit comes
as no surprise in view of his comprehensive scholarshipand frequently compares Kfan
and Baran grammarians (for example, juz/vol. 1, p. 59, l 28sqq.).
53
He says, for example (in Ibn Jinn, Mutasab, on sra 2:19) about a variant reading
which al-Farr had referred to as Medinan: wa-l nalamu anna hdhihi l-qira ruwiyat an
ahl al-Madna.
54
See the introduction to my (Bergstrer) Nichtkanonische Lesearten im Mutasab des
Ibn inn.
had blindly accepted; still, he considered himself phonetically superior to
the readers.
55
The same feeling of superiority over the simple qurr is also
evident in al-Dn (d. 444/1052),
56
although he himself is a representative
of the science of variant readings. In the intervening period this science
went beyond the mere familiarity with the text and the variant readings,
and began to incorporate grammar as a propaedeutic science.
57
In the same way that some deviations from the Uthmnic muaf were [iii/126]
perpetuated in the Seven Readings, this also happened to some linguistc
mistakes,
58
with the exception that they are far more difficult to identify. It
was obvious when something deviated from the muaf, but that a variant
reading was linguistically wrong could always be challenged. It was hardly
possible to agree beyond the fact that something was correct to a greater or
lesser degree.
Perhaps the following variant readings might be cited as being pretty
much unanimously rejected by the grammarians: Sra 2:58, etc., nabna
instead of nabiyyna, etc., and to this, sra 98:5, al-baratu instead of al-
bariyyatu (Nfi [b. Abd al-Ramn al-Layth], the second one also Ibn
Dhakwn an Ibn mir al-Yaub),
59
sra 2:247, and frequently, astum
instead of asaytum (Nfi);
60
sra 10:5, and often, iun instead iyun
(Muammad b. Abd al-Ramn QUNBUL [d. 291/904, Sezgin, Geschichte
des arabischen Schrifttums, vol. 9, p. 200, l 16] an Ibn Kathr al-Kinn);
61
55
With this point of view Ibn Jinn is not the only one among the grammarians; already
Sbawayh reproached the rw of Ab Amr b. al-Al to have considered ikhtils (reduction)
of a vowel to be taskn (elision; Ab l-Baq AL-UKBAR on sra 2:51); similar renarks can be
found in al-Zamakhshar (for example sras 49:9, and 94:1, both not taken from Ibn Jinn).
56
He only compares the nawiyyn with the udhdhq min al-muqrin (al-Taysr, chapter
al-idghm al-kabr) and criticizes the phonetics of the readers (ibid., further on sras 41:15,
69:9, and more often).
57
Thus Makk b. Ab lib (d. 437/1045) in his al-Kashf an wujh al-qirt quotes gram-
marians on a large scale, particularly Sbawayh and al-Akhfash, further, from the Barans,
Ab Amr b. al-Al, al-Khall b. Amad [al-Farhid, died ca. 715/791; EI
2
p. 613, vol. 8, pp. 5156], Ynus b. abb, died 182/798 [EI
2
; EQ; Sezgin, GAS, vol. 8, pp. 5758],
Muammad b. al-Mustanr QURUB [died 206/821; Sezgin, GAS, vol. 9, pp. 6167], Ab Zayd
(al-Anr Sad b. Aws b. Thbit [Sezgin, GAS, vol. 9, pp. 6768]), Ab Ubayda (Mamar b. al-
Muthann), Ab Uthmn Bakr b. Muammad AL-MZIN [Sezgin, GAS, vol. 9, pp. 7576],
and Muammad b. Amad IBNKAYSN, d. 299/911 [EI
2
; Sezgin, GAS, vol. 9, pp. 158160], and
from the Kfans al-Kis, and Yay b. Ziyd AL-FARR.
58
59
Sbawayh 411: qallumradun.
60
al-Zamakhshar on sra 2:247: afatum, on sra 47:24: gharbun; al-Baghaw on sra
2:247: wa-hiya (asaytum) al-lughatu l-faatu.
61
Ibn Jinn, Mutasab (Ragb Paa Ktphanesi, Istanbul, Ms 13, p. 2) as an example for
afu irbin.
sra 24:35, durrun instead of durriyyun or dirrun (Ab Bakr an IM
(b. al-Ajjj AL-JADAR) and amza (b. abb)),
62
sra 34:9, yakhsibbi-
him instead of y/nakhsif bi-him (al-Kis),
63
in addition two variant read-
ings, which at least the Baran grammarians unanimously reject: sra 4:1,
wa-l-armi instead of wa-l-arma (amza),
64
and sra 6:138, zuyyina
qatlu awldahum shurakihim instead of zayyana qatla awldihim shu-
rakuhum (Ibn mir al-Yaub).
65
The Principle of Tradition
Ibn Miqsams principles, which we discussed in the preceding pages, are [iii/127]
almost more important from their negative point of view than their posi-
tive aspects. By their claim to uniqueness they exclude other claims regard-
ing permissible types of variant readings, particularly the one represented
by Ibn Mujhidand finally establishedthat all must be traced back
through an isnd to old authorities, and finally to the Prophet. Through
the introduction of the principle of tradition evolves the classic dogma of
the three criteria to which the variant readings must conform. Makk b. Ab
lib (d. 437/1045) formulates this as follows:
66
an yunqala ani l-thiqt ani
l-nabiyyi , wa-yakna wajhuh f l-Arabiyya sighan, wa-yakna muw-
fiqan li-kha al-muaf. Here, the dogma of tradition has first priority, right-
fully
67
so as far as importance is concerned, but not age. The dogma of al
62
al-Baghaw: qla akthar al-nut: huwa lanun li-annahu laysa f kalm al-Arab
fulun ; ql Ab Ubayda: wa-ana ar la-h wajhan
63
al-Zamakhshar: laysat bi-qawiyyatin.
64
Regarding the accusative, al-abar says in juz/vol. 1, p. 141, l 31: al-qira allat l nastajz
al-qria an yaqraa ghayra-h, and al-Baghaw: afau; with reference to al-Mubarrad and
others; al-arr dismisses the genitive (al-arr, Durrat al-ghaww, ed. Thorbecke, p. 62, l
5sqq.), al-Zamakhshar calls it laysa bi-shadd, and even al-Bayw, daf. Cf. Ibn al-Anbr,
K. al-Inf, ed. G. Weil, p. 192, l 7sqq.
65
Cf. above, p. 397 n. 56 (where the non-Damascene variant is unfortunately miss-
ing); al-abar, juz/vol. 8, p. 31, l 5, of the common reading: al-qirat allat l astajz ghayra-
h;the variant reading of Ibn mir (al-Yaub) in Ibn Jinn like above, foot-note 60, in
al-Bayw, f f l-Arabiyya; al-Zamakhshar says about it fa-shay law kna f makn al-
arrt wa huwa l-shir la-kna samijan mardudan , fa-kayf bi-hi f l-kalmi l-manthri fa-
kayfa bi-hi f l-Qurni ? Cf. Ibn al-Anbr, Inf, p. 179, l 17sqq.; cf also above, p. 389sqq.;
p. 468sqq.
66
al-Ibna (Berlin Ms., no. 578), p. 500, cites Ibn al-Jazar, al-Nashr, vol. 1, p. 13, l 21 sq.; in
detail about the whole subject, al-Nashr, vol. 1, p. 9sqq. (excerpt in al-Suy, al-Itqn, naw
2227).
67
The two other criteria must accept limitations; Ibn al-Jazar defines the three criteria:
kull qira wfaqati l-Arabiyya wa-law bi-wajh, wa-wfaqat aad al-maif al-Uthmniyya
wa-lawitimlan (= taqdran, see above, p. 400 n. 78, and p. 431 n. 281) wa-aasanadu-h, fa
qira sunna muttabaa
68
in this sense is to be found already in Sbawayh
69
(d.
ca. 177/793). EvenAbAmr b. al-Al (d. ca. 154/771) allegedly made a similar
comment;
70
but it became the rule only gradually, as is evident from Ibn
Miqsams opposition still in the fourth century ah. For Ibn Jinn, who died
in 392/1001, it already seems a matter of course that he dismisses strong
linguistic reservations in its favour.
71
The more scholars felt inclined towards tradition, the more they had to
forgotheir personal criticismthat the other twocriteria offered. Ibnal-Jazar
(d. 833/1429), who displays everywhere a true scholarly attitude, in theory
upholds the three criteria.
72
Al-Jabar (d. 732/1331) had already stated that
the two other factors are included in the principle of conformity to tradi-
tion,
73
and Al b. Muammad AL-NR AL-AFQUS (1053/16431118/1706)
limited himself explicitly to the latter.
74
Makk b. Ab lib
75
contrasts the three variant readings that are in accor- [iii/129]
dance with the three criteria with two other less reliable groups, i.e., variant
readings that are transmitted an al-d and, although grammatically cor-
rect, do not conform to the muaf, and those that are either not transmit-
ted by reliable traditionists (thiqa) or grammatically wrong. These variant
readings are to be rejected (l yuqbal) even if they are identical with the
muaf. Those of the preceding group must be recognized (yuqbal), but are
practically not used (l yuqrau bi-hi); only those that meet all three require-
ments are perfect.
76
In this division another fact clashes with the principle
hiyal-qiraal-aa(al-Nashr, vol. 1, p. 9, l 12, citedby al-Suy, al-Itqn, naw 22 [Sprenger
ed., p. 176, l 12sqq]).
68
An alleged comment by Zayd b. Thbit, see al-Suy, al-Itqn, naw 22 (Sprenger ed.,
p. 176, l 16sq.).
69
34: ill anna l-qira l tukhlaf li-anna-h l-sunna. [Sezgin, GAS, vol. 9, pp. 5163].
70
71
Mutasab on sra 55:76, about the reading abqiriyya, transmitted from the Prophet
and other men: amm tarku arfi abqiriyya, fa-shdhdhun f l-qiys, wa-laysa lan an
natalaqq qirata rasl Allh ill bi-qablih; similarly on sra 36:52 concerning the
reading of Ubayy b. Kab, habban; and similarly often.
72
See above, notes 67 and 68.
73
In the introduction to his commentary on the Shaibiyya; cited by Ibn al-Jazar, al-
Nashr, vol. 1, p. 13, l 14sqq., and then al-Suy, al-Itqn, naw 22 (Sprenger ed., p. 179, l 19sqq.).
74
Ghayth al-naf f l-qirt al-sab (printed, Cairo, 1341/1922, in the margin of Ab l-Qsim
Al b. Uthmn IBN AL-QSIs commentary on the Shibiyya), p. 6, l 1 sqq.; (Brockelmann,
GAL, vol. 2, p. 461; Sezgin, GAS, vol. 2, 74).
75
Ibna (Berlin Ms no. 578), p. 500, cited in Ibn al-Jazar, al-Nashr, vol. 1, p. 13, l 17sqq.,
abbreviated in al-Suy, al-Itqn, naw 22 (Sprenger ed., p. 179, l 21 sqq.).
76
According to this classification grammatical accuracy would be the decisive crite-
rion (variant readings which are grammatically wrong belong to the lowest group without
of tradition, which is alluded to in the mention of d, but is expressed in
an aside, ukhidha an ijm; this is the ijm.
The Principle of Majority
The canon of the three criteria for the admission of a reading variant is at
variance with another canon of three criteria for preferenceikhtiyr
which Makk b. Ab lib
77
ascribes to the two main representatives, more
properly speaking, the founders of the science of variant readings in the
early third century, Ab Ubayd al-Qsim b. Sallm (d. 223 or 4/838)
78
and
ABTIM(Sahl b. Muammad) AL-SIJISTN.
79
The criteria of grammat-
ical accuracy and congruence with the muaf are common to both; the
role of conformity with tradition is subsumed in the second criterion, ijtim
al-mma alayhi, namely that the majority opted for the particular variant
reading.
Certainly down to the fourth century al-mma
80
also named al-ja- [iii/130]
ma,
81
al-kffa,
82
al-jumhr,
83
al-ns
84
means inthe science of variant read-
consideration for their other qualities). Agreement with the muaf is the least important
element (its deficiency cannot even be compensated by the most reliable chain of transmis-
sion). It is obvious that the following three conclusions: (1) yuqrau bi-hi (2) yuqbalu wa-l
yuqrau (3) l yuqbalu, result in a clearer picture than the preconditions, namely that the
purpose is not to investigate how variant readings of diverse quality are to be judged, rather
on the basis of the relevant criteria they must be assigned somehowto one of the three estab-
lished grades of recognition.Cf. also al-Suy, al-Itqn, naw 2225.
77
Ibna (Berlin Ms no. 578), p. 500.
78
79
Sezgin, GAS, vol. 3, pp. 367368, and others, particularly Yqt, al-Irshd, vol. 4, p. 258.
80
Already Sbawayh, see above, p. 474 n. 23; Yay b. ZiydAL-FARR (d. 207/822), Man
l-Qurn on sra 11:30, and often (mmat al-qurr on sra 11:48); Ab Ubayd al-Qsim b.
Sallm in Ibn al-Jazar, abaqt (Berlin Ms, cod. simul. 55), fol. 161
r
; Ab tim al-Sijistn
in Makk b. Ab lib, al-Kashf on sra 2:12, and often; al-abar, vol. 1, p. 238, l 20, and often:
mmat al-qurr (mmat al-amr f jam al-aqr, vol. 1, p. 306, l 18); Ibn Jinn, Mutasab
(cf. the introduction to my [Bergstrer] Nichtkanonische Koranlesearten). The word is used
with a different meaning in Yqt, al-Irshd, vol. 2, p. 118, l 8; here iftata al-qira al rasm
al-mma means inthe commonway not according to IbnMujhid before whomthe recital
was performed.
81
Ibn Qutayba (d. 276/889) in Makk b. Ab lib, al-Kashf on sra 16:30; al-q Isml b.
Isq al-Mlik (d. 282/895) in Makk b. Ab lib, Ibna (Berlin Ms. no. 578), p. 501; al-abar,
vol. 1, p. 307, l 26; Ibn Mujhid in Ibn Jinn, Mutasab, and this itself, ibid.
82
Ibn Jinn, Mutasab.
83
Ibn al-Srf (d. 368/979) in Yqt, Irshd, vol. 6, p. 301, l 7; al-Dimy al-Bann, Itf
fual al-bashar on sra 2:6, and often. Cf. EI
2
84
IbnMujhid inIbnJinn, Mutasab, and this itself, ibid.; cf. Ahl al-Islm, al-abar, vol. 1,
p. 296, l 8.
ings not the totality but the majority, and accordingly al-ijm,
85
al-ijtim,
86
al-ittifq,
87
not unanimity but majority vote.
88
Unanimity
89
is a borderline
case and not even particularly important. As majority vote ijm was able
to attain far-reaching significance for standardizing the text of the Koran
for it could be used to displace small minority readings completely. But as
unanimity it can have served only to state that such variant readings had
already completely disappeared, and to prevent their reappearance, which
in view of the convergent tendency of the Islamic development was in any
case unlikely to happen. In order to arbitrate in cases when an overwhelm-
ing majority was in opposition to an imperceptible minority, Ab Ubayd
al-Qsim b. Sallm, Ab tim al-Sijistn and others set up special guide-
lines as to what constituteda majority.
90
These guidelines recognize not only
the diverse local variant readings and readers but also their importance.
Thus, who should be considered as mma: both the Medinans and Kfans
together or the Medinans and the Meccans (ahl al-aramayn) or Nfi al-
Layth as well as IM b. al-Ajjj AL-JADAR?
91
85
al-abar, juz/vol. 1, p. 230, l 15 and p. 285, l 17: ijm al-qurr; vol. 1, p. 87, l 15: ijm
al-ujja, vol. 1, p. 266, l 15: inadditionminal-qurr (thus tobe read), vol. 1, p. 187, l 5: inaddition
still wa-ahl al-tawl min ulam al-salaf wa-l-khalaf (al-ujja, the proper authorities, is a pet
phrase of al-abar, which in connection with ijm emerges more strongly also in his Ikhtilf
al-fuqah, for example, [ed. by Friedrich Kern], vol. 1, p. 1, l 5; p. 11, l 10; p. 24, l 18; p. 44, l 1,
etc.)Yqt, al-Irshd, vol. 6, p. 498, l 20; and p. 499, l 3 and 7, in the version of Ibn Miqsam
(in a somewhat more general sense);cf. below, p. 496.
86
Ab Ubayd al-Qsim b. Sallm and Ab tim al-Sijistn see previous page; abar,
vol. 1, p. 307, l 24: ijtim al-ujja min al-qurr wa-ahl al-tawl.
87
al-abar, juz/vol. 1, p. 240, l 2: ittifq qirat al-qurr, vol. 1, p. 87, l 14: ittifq al-ujja min
al-qurr wa-l-ulam al-umma.
88
One is tempted to interpret the frequent comparison of individual readers (identified
by name, baal-qurr, or similarly) onthe one side, andal-mma, al-ijma, etc., onthe other,
to mean that an earlier minority variant reading might have been repealed by a later ijm =
agreement of all the authoritative readers. This fails in cases like abar, vol. 1, p. 295, l 23,
where the mmat qurr al-Kfa is confronted precisely by IM b. al-Ajjj AL-JADAR,
the future authoritative Kfan reader; or vol. 1, p. 187, l 5, where Ibn Kathr al-Kinns variant
reading is rejected by ijm (sra 2:35, dama kalimtun instead damu kalimtin), or
vol. 1, p. 285, l. 17, where it is the reading of Ab Jafar Yazd IBNAL-QAQ (sra 2:73, amniya
instead amniyya).
89
For example, al-abar, vol. 1, p. 51, l 16: ijm jam al-ujja min al-qurr wa-ulam, and
similarly, vol. 1, p. 169, l 22.
90
Makk b. Ab lib, Ibna (Berlin Ms. no. 578), p. 509.
91
Similar rules are applied also elsewhere, for example, by Ibn Qutayba (d. 276/889;
Brockelmann, GAL, vol. 1, p. 120), in Makk b. Ab lib, al-Kashf on sra 16:30; by Makk
himself, al-Kashf (Berlin Ms. no. 578) p. 28; also by al-Zamakhshar (likely froman old source)
on sra 1:3.
The principle of the majority attained its greatest significance precisely [iii/132]
since Ab Ubayd al-Qsimb. Sallmand Ab timal-Sijistn made it the
basis of their ikhtiyr, i.e., their selectionfromamong different possible vari-
ant readings. FromMakk b. Ab libs al-Kashf an wujh al-qirt
92
we are
familiar with this ikhtiyr, which in its totality constitutes a complete read-
ing of the Koran. That it was applied in practice is evident fromal-Maqdiss
statement written in the 370s ah
93
that the province of Jibl read accord-
ing to the ikhtiyr of Ab Ubayd (al-Qsim b. Sallm) and Ab tim (al-
Sijistn). Conversely, it is purely hypothetical when in a quite similar sense
still AB HIR b. Ab Hshim [al-Baghdd AL-BAZZR]
94
(d. 349/960)
has an ikhtiyr, and even Makk b. Ab lib (d. 437/1045) in al-Kashf rec-
ognizes the variant reading of the mma and his own ikhtiyr and defends
this in cases of doubt. Certainly since Ibn Mujhid (d. 324/936) in theory
and soon thereafter in practice the principle of tradition in a new, more
unrestricted form carried the day over the principle of majority. Even later
on there is occasionally mention of the majority.
95
It remains moderately
important in the layout of many works on variant readings, for example, al-
Dns (d. 444/1052) al-Taysr f l-qirt al-sab: Only the champions of the
minority variant readings are mentioned, but the majority variant readings
are treated as miscellaneous, the bqn-system.
Like Ab Ubayd (al-Qsim b. Sallm) and Ab tim (al-Sijistn) also
Ibn Qutayba practised ikhtiyr mainly according to the majority of the
readers, and concurrently according to practical-linguistic considerations.
Makk b. Ab lib
96
lists Ibn Qutaybas occasional deviations from the for-
92
Makk b. Ab lib states it frequently in cases of doubt. Both of themare nearly always
in agreement; the only exception being sra 2:9 where Ab Ubayd al-Qsim b. Sallm
as well as al-abar, vol. 1, p. 95, l 30opt for yakdhibna whereas Ab tim al-Sijistn,
for yukadhdhibna. Al-abars detailed and objective arguments make it probable that in
this case also Ab Ubayd followed objective considerations. The majority is undoubtedly
against yakdhibna, which from among the Seven is held only by the Kfans. The principle
of majority is thus for Ab Ubaydand equally for Ab timal-Sijistnthe most impor-
tant aspect for the formation of ikhtiyr, although not the only one.
93
al-Muqaddas/al-Maqdis, ed. de Goeje (1906). Brockelmann, GAL, vol. 1, p. 230.
94
See above, p. 474 n. 23. In the Kashf of Makk b. Ab lib he mostly goes together with
Ab Ubayd al-Qsim b. Sallm.
95
al-mma also in Tafsr al-Jallayn, for example, on sra 31:2, further, see above, p. 482
n. 83; ijm frequently also in Ibn al-Jazar, for example, al-Nashr, vol. 2, p. 212, l 15.
96
For example, sras 14:2, 20:12; 26:176, and 38:12 (in sra 15:78). It is precisely because
such occasional deviations are emphasized that it is probable that the basic reason remains
the same.From the period between Ab Ubayd and al-abar we know still some other
mukhtrn: Muammad b. Sadn (d. 231/846) [Sezgin, GAS, vol. 9, pp. 135136]; Yqt, al-
Irshd, vol. 1, p. 7, l 12; Ibn al-Jazar, abaqt (Berlin Ms. cod. simul. 55), fol. 200
v
; Ab Abd
mer two, partly supplying the reason. Al-abar
97
(d. 310/923) takes a simi-
lar position as well,
98
but, following his excessive inclination towards har-
monizing, he amalgamates ijm with naql to create a uniform, sublime
dogma.
99
He thus paves the way for the tawtur teaching of later ortho-
doxy.
100
However in other cases, when there is nearly an equilibriumof opin-
ions, he dispenses with decisiveness and pronounces both variant readings
to be equal.
101
Yay b. Ziyd AL-FARR
102
(died 207/822) seems to have
been the same kind of precursor as Ab Ubayd was.
Allh MUAMMAD IBN S b. Ibrhm b. Razn AL-IBAHN (d. ca. 242253/856867);
Ibn al-Jazar, al-Nashr, vol. 1, p. 177, l 17, and p. 180, l 15; abaqt, fol. 222
v
. Nothing is known
about their ikhtiyr; an investigation of (Ibn) al-afrws [d. 636/1238] Taqrb wa-l-bayn f
marifat shawdhdh al-Qurn (a fragment, Berlin Ms. no. 613), where both are considered,
might be helpful.
97
In his commentary on the Koran al-abar ends every information about variant read-
ings by stating which type of reading is to be preferred, thus his ikhtiyr. He occasionally
says outright ikhtartu (for example, vol. 1, p. 296, l 6). He states it quite clearly in the work
on the types of variant readings that he himself established his ikhtiyr (vol. 1, p. 49, l 5);
we may assume that the same way his Ikhtilf al-fuqah is no mere learned and compar-
ative work but serves the establishment of his own madhhab so that the establishment
of his Koranic reading ikhtiyr was the actual purpose of his work on the variant read-
ings.
98
He says (vol. 1, p. 307, l 25): wa-qad dalaln al anna m jat bi-hi l-ujjat muttafiqatan
alayhi ujja al man balaghahu wa-m ja bi-hi l-munfarid fa-ghayr jizi l-itir bi-hi al
m jat bi-hi l-jama allat taqmbi-h l-ujja naqlan wa qawlan wa-amalan f ghayri hdh
l-mawii. The reference seems toapply toa passage where al-abar elaboratedonthe dogma
of ijm in general, not limited to the variant readings, namely very likely a passage in the
Ikhtilf al-fuqah. According to this basic point of view, al-abar is quite outspoken in his
rejection of minority variant readings; for example, vol. 1, p. 240, l 1, al-qira allat l
yajzuind ghayruh (similarly vol. 1, p. 285, l 16, and p. 307, l 23, etc.; cf. above, p. 480 n. 64
and 65).
99
For example, vol. 1, p. 266, l 2: al-qirati l- jiyat maj al-ujja bi-naql man l yajz
alayhi f-m naqalhu mujmina alayhi l-khaau wa-l-sahw wa-l-kadhib; further, cf. the fre-
quently repeated requirement of transmission that it must be mustaf, widely-held. (For
example, above, p. 475 n. 26, and below, p. 485 n. 101).
100
See below, p. 502sqq.
101
For example, vol. 1, p. 300, l 30, lughatn marfatn wa-qiratn mustafatn (cf.
p. 428 n. 96) f amr al Islm; vol. 28, p. 30, l 17, qiratn marfatn aat l-man;
vol. 13, p. 105, l 28, qiratn mashhratn qad qaraa bi-kull wida min-hum aimma min
al- qurr man-hum wid (on sra 14:2; it is typical that this is one of the passages where
Ibn Qutayba differs from Ab Ubayd al-Qsim b. Sallm, see above, p. 484 n. 96); cf. further,
vol. 1, p. 214, l 6, and p. 293, l 23, etc.That the decision is optional (takhyr) happens
also in older authorities, for example, Ab Amr in AL-DN, Taysr f l-qirt al-sab on
sra 89 at the end; al-Kis, according to some of the transmitters, in Makk b. Ab lib
al-Qays, al-Kashf an wujh al-qirt wa-ilalih wa-ujajih, on sras 1:3, 55:56, 67:11, and
79:11.
102
For example, he states (Man l-Qurn on sra 2:1) that sra 3:1, al-qira bi-ar
al-hamz.
Ikhtiyr
103
is not necessarily tied to the dogma of the majority, and does [iii/134]
not even mean selection from a more narrow circle of variant readings.
104
The older technical meaning of the word refers to a reader who is mainly fol-
lowing anolder authority, but departs fromit insome isolated instances and
follows his own way. We thus hear of the ikhtiyr of Ab Jafar Muammad
b. Ab Sra AL-RUS
105
and (Ab Muammad) al-Yazd (d. 202/817
106
) who
generally followAb Amr (b. al-Al), or of the ikhtiyr of Khalaf (b. Hishm
al-Bazzr, d. 229/843)
107
(based on amza b. abb),
108
and of (Ab l-Abbs)
AL-FAL IBN IBRHM al-Naw al-Kf (based on al-Kis).
109
The verb
ikhtra also applies to transmitters like Warsh an Nfi
110
or af (b. Sulay-
mn) an IM (b. al-Ajjj AL-JADAR)
111
who more seriously modify the
qira of their teacher. In a wider sense ikhtiyr is used nearly synonymously
with qira
112
= the closed and independent Koranic reading of a particular
authority. The only difference seems to be the low opinion associated with
the expression ikhtiyr.
113
Ijm and ikhtiyr are common terms of the science of variant read- [iii/135]
ings as well as of fiqh, and originate most likely from the latter field. The
technical terms of this provenance include the afore-mentioned ray
114
(its
103
Or its equivalent, takhayyur, e.g., al-Dhahab, abaqt (Berlin Ms. Or. folio, 1340), 18
r
of
al-Kis; Yqt, al-Irshd, vol. 6, p. 500, l 1, from Ibn Miqsam.
104
E.g., also Ibn Miqsam (see above, p. 475sq.) considers his procedure to be ikhtiyr as is
evident from Yqt, al-Irshd, vol. 6, p. 500, l 13.
105
Fl. 170/786193/809; a well-known Kfan grammarian; cf. Ibn al-Jazar, abaqt (Berlin
Ms., cod. simul., 55) 193
v
. [EI
2
; Sezgin, GAS, vol. 9, pp. 125126].
106
E.g., Ibn al-Jazar, abaqt (Berlin Ms., cod. simul. 55), fol. 266
r
.
107
E.g., his expression is used by Ibn Miqdam in Yqt, al-Irshd, vol. 6, p. 500, l 13.
108
Because of the dependency on amza b. abb it is also said of al-Kis ikhtra
and takhayyara, respectively (al-Dhahab, abaqt [Berlin Ms. or., folio 3140] 17
v
, and see
below, p. 486 n. 103; and below, p. 490). The same way the reading of amza represents the
background for the ikhtiyr of Muammad b. Sadn (see above, p. 484 n. 96sq., and Sezgin,
GAS, vol. 9, pp. 135136), and still the one of al-abar.
109
Yqt, al-Irshd, vol. 6, p. 134; cf. Ibn al-Jazar, abaqt (Berlin Ms, cod. simul. 55), fol.
165 r.
110
E.g., al-Dn, al-Taysr f l-qirt al-sab on sra 6, end.
111
E.g., al-Dn, Taysr on sra 30:53.Other mukhtrn see below, p. 510sqq.
112
E.g., Ibn al-Jazar (abaqt [Berlin Ms., cod. simul. 55] fol. 85
r
) considers (Ab aywa)
Shuray b. Yazd al-aram (d. 203/818) the ib al-qira al-shdhdha, and then speaks
about his ikhtiyr. Cf. also below, p. 494sq.
113
Later it was differentiated between the recognized readings, qira or riwya, and
the non-recognized readings called ikhtiyr; thus Abd al-Ramn (b.) AL-AFRW (died
636/1238 [Brockelmann, GAL, suppl. 2, p. 727]), Taqrb (Berlin Ms no. 613), fol. 4 vsqq.; Nim
al-Dn al-asan AL-NSBR (d. ca. 706/1306; see above, p. 406 n. 129) in the section dhikr
al-aimma al-mukhtrn in the introduction to his commentaries on the Koran.
114
verbis alsotobe found
115
), further, istibb,
116
synonymous withikhtiyr, then
qiys,
117
and the verb akhadha (with bi-
118
) with the meaning to accept a
variant reading, etc., to decide onit, thus, to exercise ikhtiyr to ones advan-
tage. The term ijtihd,
119
as might have been expected, is missing. This is
based on the fundamental difference between fiqh and the Koranic read-
ings. In fiqh there are given facts, and on the basis of ul the mujtahid
passes the appropriate ukm, but still on the basis of his individual judge-
ment. If the decision consists of choosing between different possibilities,
these themselves are thus construed by himand his equals at the time when
the problemis considered. In the science of Koranic variant readings differ-
ent possibilities exist a priori, which, according to the prevalent dogma, are
all equally divine
120
so that the reader need do no more than choose from
among them.
121
Part of the dogma of majority is the divisionof variant readings intomash- [iii/136]
hr
122
and shdhdh,
123
canonical and uncanonical. Shdhdh might be con-
sidered a borrowing from grammar, where the expression plays an impor-
tant role. However, there it is a relative concept that becomes meaningful
only by its complement: In grammar shdhdh an al-qiys contrary to anal-
ogy, in the science of variant readingsas al-abar puts it
124
(shdhdh)
an qirat al-amr remaining outside the variant readings recognized by
the leading Islamic centres. Possibly even older is the expression, al-qirat
115
E.g., al-Dn (d. 444/1052), Taysr, chapter al-idghm al-kabr, section al-arfayn al-
mutaqribayn says: kna Ibn Mujhidn jar l-idghm f , he considered it to be right. I
have not come across older references.
116
al-Dn, Taysr, chapter al-waqf al awkhir al-kalim.
117
See below, p. 487.
118
E.g., al-Dn, al-Taysr f l-qirt al-sab on sras 41, 44, and 89 end.
119
Isolated instance: Anyone could say, al m huwa l-asan indahu ijtihdan (Nldeke
in the first edition of this book, Seite 279, from a source unknown to me [Bergstrer]). As a
rule, the application of ijtihd to the variant readings of the Koran is explicitly rejected (see
above, p. 475 n. 27). In the above-mentioned passage (on sra 6:138), p. 424, Ibn al-Munayyir
blames al-Zamakhshar for the conception, anna l-qurr aimmat al-wujh al-saba ikhtra
kull minhumarfan qaraa bi-hi ijtihdan l naqlan wa-saman.
120
In so far as they are not forged (maw); cf. al-Suy, al-Itqn, naw 26.
121
Cf. Ibn al-Jazar, al-Nashr, vol. 1, p. 51, l 17sqq.
122
An approximate synonym for marf, see, e.g., above, p. 485 n. 101.
123
Already AbUbaydal-Qsimb. Sallm(d. 223/838); further, e.g., Yqt, al-Irshd, vol. 6,
p. 302, l 11 (from older source).
124
E.g., abar, juz/vol. 8, p. 61, l 4; cf. vol. 13, p. 147, l 25, and see above, p. 475 n. 26. Al-abar
also applies the word to the reader, not the variant reading shudhdhal-qri ammalayhi
l-ujja mujmiatunf dhlika, vol. 1, p. 285, l 18; cf. vol. 1, p. 87, l 15; and p. 307, l 24, etc. (similarly
Nfi, below, p. 489).
al-shdhdha al-khrija an al-muaf:
125
deviations from the Uthmnic text
that were first excluded by ijm.
126
When shdhdh is substantially related to the notion of deviation from [iii/137]
the Uthmnic muafbut more than thisone must expect that the
approach (more precisely the ukm) of the shawdhdh, in so far as they
do not deviate from the muaf, will be related to the non-Uthmnic vari-
ants
127
but more lenient. Indeed, even later generations prohibit al-qira
bi-l-shawdhdh only at the ritual prayer, but otherwise tolerate it;
128
like the
non-Uthmnic versions of the text they might serve as explanations of the
recognized readings.
Standardization of Variant Readings
The elimination of the variants of the Uthmnic muaf, as well as the
reading variants freely construedwithout considerationfor tradition, is only
part of the great process of the standardization of the text of the Koran, its
reading, and the establishment of a textus receptus. The driving force in this
endeavour is the doctrine of the majority, or more generally, the catholic
tendency, the convergence within the development of Islam. The process of
standardization prevails in the history of the text of the Koran throughout
the first centuries, and is nearly completed at the time of Ibn Mujhid
(died 324/936), when the rigid form of traditionalism was triumphant.
Even apart from the non-Uthmnic variants, the most superficial com- [iii/138]
parison of the superabundance of uncanonical variant readings in a list,
129
and the inadequacy of the variants in one of the well-known works on
125
Ibn al-Jazar, al-Nashr, vol. 1, p. 39, l 19, referring to Ibn Shannabdh (d. 328/939; see
above, p. 467sqq.); in Yqt, al-Irshd, vol. 6, p. 302, l 10, Ibn Shannabdh himself promises:
l ukhlif muaf Uthmn wa-l aqrau ill bi-m fhi min al-qirat al-mashhra.
126
127
128
Cf. al-Suy, al-Itqn, naw, 35, masala 21 (ed. Sprenger, p. 256, l 22), and further Taq
l-Dn AL-SUBK (d. 756; Brockelmann, GAL, vol. 2, p. 86); in Ibn al-Jazar, al-Nashr, vol. 1, p. 44,
l 6, cites al-Suy, al-Itqn, naw 2227, tanbh 3 (ed. Sprenger, p. 191, l 22). The question
whether or not it is permissible to read what deviates from the muaf is frequently not
explicitly answered. (e.g., al-Nashr, vol. 1, p. 1, l 14sq.)
129
The material contained in the Mutasab of Ibn Jinn (mainly from Ibn Mujhid) and
in the Shawdhdh of al-usayn b. Amad IBN KHLAWAYH (d. 370/980) [EI
2
; EQ; Sezgin,
GAS, vol. 9, pp. 169171], I [Bergstrer] intend to present shortly in the Nichtkanonischen
Koranlesearten. As a complement must be considered first of all the second Seven since they
are available in the Itf of al-Dimy al-Bann on the Fourteen (the Three after the Seven
also in Ibn al-Jazars al-Nashr, etc.).
the Seven Readings shows how enormously the reading of the text was
standardized. It is particularly significant that not only were variant read-
ings suppressed that touched upon the linguistic or essential interpreta-
tion of the textone might be tempted to assume that they belong to the
freely forged (maw) variant readings that were abandoned as a mat-
ter of principlebut also purely dialectical differences were reduced to
a minimum. The literature of variant readings alone makes this obvious,
but the standardization is in reality still stronger than one would expect,
because the lists of uncanonical variant readings undoubtedly also repre-
sents nothing but a poor selection from among the former multiplicity of
readings. A striking example follows: The pronunciation bi-h, f-hu (f-h),
etc. instead bi-h, f-hi (f-h) etc., as far as I can make out, has completely dis-
appeared
130
fromthe literature of variant readings.
131
Sbawayh, however, says
(503): wa-ahl al-ijz yaqlnamarartubi-huqabl wa-laday-huml wa-
yaqrana fa-khasafn bi-hu wa-bi-drihi l-ar (sra 28:81). This is in accor-
dance with the extraordinarily frequent vocalization in the Kf fragments
of the Koran, a fact that we shall deal with later.
The standardizationcanbe observedalsowithinthe better knownKoran- [iii/139]
ic readings, namely in the reciprocal relation of older and newer readings of
identical provenance. Among those cases, e.g., in which the Medinan Nfi
b. Abd al-Ramn (d. 169/785) deviates from his predecessor, Ab Jafar
Yazd IBNAL-QAQ al-Makhzm fromMedina (died 130/747 or 138/755), a
great portion is made up of those in which Nfi abandons isolated readings.
Extremely rare are those in which he conversely changes from a common
reading to a less commonone. The rest consists of cases inwhichNfi mixes
knownvariant readings.
132
This cannot be anaccident. All sorts of statements
130
Anexceptionare only cases inwhichthe following wordbegins withwal; here (accord-
ing to al-Dimy al-Bann, Itf, chapter h al-kinya) Ibn Muayin pronounces bihunur
bihullhu etc. (the first one, sra 6:46), also Warsh an Nfi according to one transmission.
131
Also K. Vollers, Volkssprache und Schriftsprache im alten Arabien, does not mention
them in the section on suffixes (p. 144sqq.)
132
E.g., inthe second sra Nfi abandons the following variant readings of Ab Jafar Yazd
IBN AL-QQ which none of the Seven advocates: 2:32 (and similar ones frequently) lil-
malikatu isjud (instead of malikati isjud); 2:73 (and similar ones frequently) amniya
(insteadof amniyya); 2:160 twice innainsteadof anna; 2:168 (andsimilar ones frequently) al-
mayyitata(insteadof al-maytata); ibid., manuirra(insteadof mani/uurra); 2:181 (andsim-
ilar ones frequently) al-yusura, al-usura (instead of al-yusra, al-usra); 2:206 wa-l-malikati
(instead of wa-l-malikatu); 2:209 (and similar ones frequently) li-yukama (instead of li-
yakuma); 2:233 (and similar ones frequently) tur (not quite certain, instead of turru
and turra respectively). Only in three passages, and only Nfi, conversely changes from
Ab Jafar IBN AL-QQs common variant reading to a variant reading that makes him the
transmitted fromhimwould indicate that Nfi indeed assimilated his read-
ing to that of the majority. He is said to have stated: qaratu al sabna min
al-tbina fa-m ijtamaa alayh ithnni akhadhtuhu wa-m shadhdha f-hi
widuntaraktuhuatt allaftuhdhihi l-qira;
133
and questioned about the
pronunciation of dhib bir he is said to have renounced his own dialec-
tal pronunciation dhb br in favour of the more common pronunciation
with hamzah by replying knat al-Arab tahmizu-h fa-ihmiz-h.
134
There is
a similar report by Ibn Mujhid through al-Kis (d. 189/804): Ikhtra min
qirati amzatawa-qirati ghayrihi qiratanmutawassiatanghayrakhri-
jatin min thri man taqaddama min al-aimmati.
135
Apart fromthe script itself, the first thing agreeduponwere the diacritical [iii/140]
marks of the consonantal letters.
136
The Koranic reading of asan al-Bar
(died 110/728) contained quite a number of peculiar pointings,
137
although
eventhere they were seemingly onthe way out. Differences inpointing were
quite rare in the Seven Readings.
138
In passages in which the different vocalizations reflect the difference of
both linguistic and substantial interpretations, the standardization of the
reading did not in every case lead to a uniformity of opinion. A progressive
technique of interpretation made it now possible to interpret the standard
sole person among the Seven: In 2:113 Nfi reads tasal (instead of tusalu); 2:210, yaqla
(instead of yaqlu); and 2:247 (and similar ones frequently) astum (instead of asaytum).
133
Makk b. Ab lib, Ibna (Berlin Ms. 578), p. 499sq. (cf. Nldeke in the first edition
of this book, Seite 285sq); similarly al-Dhahab, abaqt (Berlin Ms. cod. Or. folio, no. 3140)
fol. 15
v
.
134
al-Dhahab, loc. cit.
135
Ibn al-Jazar abaqt (Berlin Ms. cod. simul., 55) 141
r
= Takprlzade, Mift al-sada
(Hyderabad, 1328/1910), vol. 1, p. 380, l 9sqq.; similarly already Ab Ubayd al-Qsimb. Sallm,
ibid.
136
Particularly the punctuation of the preformatives of the imperfect was for a long
time the source of disagreement (cf. above, p. 476); but also between -tu- and -n- before
a suffix (the was not written, see above, p. 412), between -- and -at- before the suf-
fix of III. Form verbs y (the was written , see above, p. 415sq.; examples, sras 3:33,
ndahu: ndathu; 6:61, tawaffhu: tawaffathu; 6:70, istahwhu: istahwathu), the Seven still
vary between feminine ending and suffix (examples, sras 17:40, sayyiatan: sayyiuhu; 31:19,
nimatan: niamahu); between bi-hd and tahd, sra 27:83; or yatanjawna and yantajna,
sra 58:9.
137
Bergstrer, Die Koranlesung des asan von Bara, p. 51.
138
Sra 2:216, amza b. abb and al-Kis read kathrun, the others kabrun; sra 2:261,
the Kfans, nunshizuh, the others, nunshiruh; sra 4:96 (twice). Sra 49:6, amza and
al-Kis, tathabbat the others tabayyan, the others, tabayyan; sras 7:55, 25:50, 27:64,
imal-Yadar, bushrn, the others forms of NSHR; sra 10:31, amza andal-Kis tatl and
the others tabl; sra 29:58, the same two, nuthwiyannahum, the others nubawwiannahum;
sra 43:18, Nfi, Ibn Kathr al-Kinn, and Ibn mir, inda, the others, ibdu. On sra 6:57 see
above, p. 494.
text differently or, by way of acrobatic interpretations, achieve the same end
for which one had abandoned the most obvious vocalization or, in early
times, even changed consonants. One way or the other, shara, the Islamic
canon law, as well as faith, often established their own self-interests against
contradictory verses of the Koran. For both cases we will give one example
each.
First: In sra 5:8,
139
it reads: fa-ghsil wujhakum wa-aydiyakum il l- [iii/141]
marfiqi wa-msa bi-rusikum wa-arjulikum il l-kabayni; though it is
required, it suffices when the feet are wiped off.
140
From an early time the
stricter practice was favoured, namely that the feet be washed. This could
be read into the verse by adopting the strained reading, wa-arjulakum and
by considering wa-imsa bi-rusikum a parenthesis. This controversy can
be dated by the fact that already asan al-Bar (d. 110/728) offered a com-
promise. He read wa-arjulukum so that this would begin a new sentence
according to taste, with the predicate of either ghsilh or imsabi-h. The
canonical schools of fiqh unanimously adopted the dogma that washing is
far. Still, half of the Seven retain the variant reading wa-arjulikum.
141
The
apparent contradiction is harmonized by explanation; al-Bayw says: wa-
jarrahul-bqnaall-jiwri. The genitive is tobe explainedby the attraction
of the case to the neighbouring rusikum.
Second: In sra 11:44, the sinner whom tradition names Kann becomes
the son of Noah by the words wa-nd Nun ibnahu; this was considered
incompatible with his dignity of a prophet. It was first corrected
[ibna-
ha]: he was merely his foster son. Later, without change of the consonantal
text, it was merely vocalized differently, ibnaha al adhf al-alif. The Seven
unanimously read ibnahu; one was content to interpret ibn as rabb.
142
The crystallization of the textus receptus probably developed in two [iii/142]
stages which, however, overlapped in time: At first, coming to an agreement
within each mir,
143
and then among the amr. The first step had already
139
Cf. to the following, Bergstrer, Die Koranlesung des asan von Bara, p. 24sq.
in addition also al-Jas [EI
2
; Sezgin, GAS, vol. 1, pp. 444445], Akm al-Qurn, vol. 2,
p. 345sqq.
140
The passage has actually nothing to do with the mas al l-khuffayn, the wiping over
the feet. I. Goldziher, Introduction to Islamic theology, p. 207, and his Schools of Koranic
commentators, pp. 45.
141
Nfi, Ibnmir, andafs b. Sulaymnanimal-Jadar andal-Kis reada; whereas
Ibn Kathr al-Kinn, Ab Amr b. al-Al, and Ab Bakr Amad b. al-usayn IBN MIHRN
an im and amza b. abb read i.
142
E.g., al-Bayw, s.v.
143
A contrasting example is Ibn Muayin (d. 122/739 or 123/740) of whom Ibn Mujhid
says (in Ibn al-Jazar, abaqt [Berlin Ms., cod. simul. 55], 207
r
): kna li-Ibn Muayin ikhtiyr
been completed at the time of Sbawayh (d. 177/793 or 180/796),
144
as he
seems to presume a Baran standard reading;
145
he also knows of local read-
ings in other centres,
146
with the exception of Damascus, which does not
seemto have been considered.
147
The ikhtiyr of Ab Ubayd al-Qsimb. Sal-
lmandAbtimal-Sijistn (see above, p. 482sqq.) completes the second
step. If this development had continued unabated, it would have to lead to
the acceptance of a text in the entire Orient and Europe, which would have
been far less accidental than the eventually canonized text of Ab Umar
AF IBN SULAYMN an IM b. al-Ajjj al-JADAR. The development
came to a halt because with Ibn Mujhid a narrow traditionalism arose
which had been bred within the school of Koran and which did not permit
combining variant readings of different provenance;
148
rather, it demanded
that every Koranic reading be passed on unchanged in its entirety. Although
f l-qirat fa-kharaja bihi an ijm ahl baladihi fa-raghiba l-ns an qiratihi wa-ajma al
qirat Ibn Kathr. This establishes the development of a local textus receptus of Mecca at a
very early time.
144
Similarly also al-Farr (d. 207/822) in Man l-Qurn [Sezgin, GAS, vol. 8, p. 124], e.g.,
ahl al-Madna on sra 11:43, and often; but also al-abar (d. 310/923) who, however, with his
pedantic conscientiousness mostly emphasizes that it is only a majority of the readers of
the respective city that reads this way: muaam ahl al-Kfa ([abar,] juz/vol. 1, p. 95, l 2)
and al-Madna wa-l-ijz wa-l-Bara, l 3, mmat qurr al-Kfa (vol. 1, p. 295, l 23) and
al-Madna, l 24, mmat qurr al-Madna wa-l-Sham (vol. 13, p. 105, l 18) and ahl al-Irq
wa-l-Kfa wa-l-Bara, l 19, etc.; but also simply ahl al-Sham(vol. 8, p. 31, l 3) and al-qurr min
al-ijz wa-l-Irq wa-l-Sham, l 26, and similarly often.
145
This can be seen from the fact that he does not know the Barans as a unit, rather he
ocassionally mentions individual Baran readers, most frequently Ab Amr (151, 304, and
often), less frequently Abd Allh IBN AB ISQ AL-ARAM, d. 117/735 or 127/745 [EQ;
Sezgin, GAS, vol. 8, pp. 3637]; (see below, p. 513) and s b. Umar al-Thaqaf (see above,
p. 474sq., [269]). The conjecture ought to be verified by an investigation of the numerous
Koranic citations and discussions of anonymous variant readings in Sbawayh.
146
Ahl al-Madna 136, 243, and often; ahl al-Makka 503, 506 (twice) and ahl al-ijz
240, 503; ahl al-Kfa 244, 568; and al-Kfiyyn 222. In the last passage Sbawayh cites
as authority for the Kfans Ab Abd Allh HRN IBN MS al-Awar (see above, p. 446
[and Sezgin, GAS, vol. 9, pp. 4344]) who provides himwith information on variant readings
elsewhere (241, 503, 568.)
147
It ought to be investigated whether Sbawayh without providing names or locations
considered Damascene variant readings.When later in the works on variant readings
Kfans are mentioned, this has no longer anything to do with the directly identified local
variant readings, rather it is an abridgement for the congruity of im al-Jadar, amza b.
abb and al-Kis, respectively, as well as other nisbas.
148
The question of tarkb continues to be debated in the later science of variant readings
(cf. e.g., Ibn al-Jazar, al-Nashr, vol. 1, p. 18, l 12sqq., and from this, al-Suy, al-Itqn, naw 35,
masala 23 [Sprenger ed., p. 258, l 10sqq.]). But here it is no longer the question whether it is
permisible to choose ones own ikhtiyr from among the variant readings of different prove-
nance, but only in which case it is permitted or prohibited to change from the recognized
readings to yet another one.
efforts of the science of variant readings to evaluate of the individual read-
ings criticallyeven if not critical in our sense of the wordcontinued for
some time, they remained without any outside effect.
Koranic Teaching and the Variant Readings
The teaching of the variant readings probably saw its most schematized [iii/143]
formation in the fifth and sixth centuries. Al-Qsim b. Firruh AL-SHIB
(d. 590/1193), for example, demanded from a pupil who likewise wanted to
become a teacher of the Koran, that for the study of each individual qira
of the Seven he had to recite the entire Koran (khatma) three times, once
each according to one of the two canonical transmissions (riwya), and
then once again according to both together (jam).
149
But with this accom-
plished, the pupil had acquired the knowledge of the readings of only the
Shibiyya; in the study of each additional school manual the Koran had
again to be recited according to this qira (bi-muammanihi) more or less
often. Later the requirements somewhat lessened. At any rate, a man like
Ibn al-Jazar, who dedicated his life to the science of variant readings and
studied all the available relevant worksaccording to the list of isnds
150
in
his main workmust have studied the Koran extraordinarily often accord-
ing to all possible variants. Before the time of al-Shib the requirements
were, if anything, higher: Al b. Amad AL-WID
151
(d. 468/1075) after
innumerable khatmas still did not master the whole transmission (arqa)
of Ab Bakr Amad IBNMIHRN(295/908381/991);
152
Ab l-asan Al AL-
UR (d. 488/1095),
153
needed seventy khatmas to study the Seven with a
teacher.
154
Already Ibn Mujhid (d. 324/936) did not permit a pupil to get
beyond the reading of IM b. al-Ajjj AL-JADAR for years in numer-
ous khatmas.
155
But particularly in earlier days there was no lack of devia-
tions. AL-DN (d. 444/1052) did not consider it superfluous to boast
156
that
149
Ibn al-Jazar, al-Nashr, vol. 2, p. 188, 7sqq., in a very interesting chapter devoted entirely
to teaching, entiled Bayn ifrd al-qirt wa-jamih (study of every individual reading for
itself or several together), vol. 2, pp. 187198 (very brief also in al-Suy, al-Itqn, naw 34, fal
4 [Sprenger ed., p. 239sqq.]).
150
al-Nashr, vol. 1, pp. 5798.
151
EQ; EI
2
; Sezgin, GAS, vol. 2, pp. 112 and 113.
152
Yqt, al-Irshd, vol. 5, p. 101, 3; Sezgin, GAS, vol. 9 (1984), pp. 191192.
153
EI
2
.
154
Ibn al-Jazar, al-Nashr, vol. 2, p. 187, l 11.
155
Ibid., l 16.
156
Jmi al-bayn (Istanbul Ms, Nuruosmaniye Ktphanesi, 62), 2
r
: wa-afradtu qirat
he accepted the transmission of only such men who had acquired the
respective recital through practical experience, but not through theoretical
instruction or the study of manuals.
The teaching methods as describedparticularly if the recital of the [iii/144]
Koran as demanded by Ibn al-Jazar
157
was prepared by learning by heart a
particular work on variant readingscould indeed guarantee the unbroken
transmission of every individual variant reading and peculiarity of pronun-
ciation; and this independent of written fixation, but how much more so
with such aid. But these methods did not exist from the very beginning.
They are unlikely to go back to before the time of Ibn Mujhid, and are
probably created only by him, who is generally remembered for his rigorous
Koranic instruction.
158
That a similarly rigid tradition does not even go back
to the authors of the Koranic readings that carry their name, emerges from
their radical differencesnot only in the subtleties of pronunciation
between both Ibn mirs canonical riwya of the reading and particularly
that of im al-Jadar.
159
Whatever we known of the teaching methods of
the second and third centuries ahconfirms and explains these phenomena.
Muammad b. Abd al-Ramn IBNAB LAYLal-Anr
160
does not dare to
correct the mistake of a pupil, fearing that he might be right.
161
Ab Amr b.
al-Al
162
(d. 145/762), Nfi b. Abd al-Ramn
163
(d. 169/785), and more so Al
b. amza AL-KIS
164
(died 189/804) offer several variant readings to choose
from (takhyr). Nfi b. Abd al-Ramn al-Layth accepts any reading that
he himself has once heard, and teaches his own reading (ikhtiyr)
165
only on
kull widin min al-aimma bi-riwyat man akhadha l-qira anhu tilwatan wa-add l-urf
anhu ikyatan dna man naqalah saman f l-kutub wa-riwyatan f l-uuf. Synonyms
of tilwatan and ikyatan is aran: The pupil recites, the teacher corrects (yaruddu alayhi
withaccusative of the mistakes and bi-for the correctionto be inserted). This type of teaching
is called talqn (already al-Ja, Akm al-Qurn [Constantinople, 1335] p. 445, l 4 of Ibn
Masd); one may therefore say akhadhaanhul-qirataaranwa-talqnan(Yqt, al-Irshd,
vol. 4, p. 118, l 6).
157
al-Nashr, vol. 2, p. 191, l 21.
158
Yqt, al-Irshd, vol. 2, p. 118, l 13sqq.
159
More on the subject later.
160
74/693148/765; EI
2
; Juynboll, Encyclopedia, p. 538, col. 2; Sezgin, GAS, vol. 1, p. 518.
161
al-Dhahab, abaqt (Berlin Ms., Or. fol. 3140), 16
v
.
162
al-Dn, Taysr f l-qirt al-sab on sra 89, the end; Makk b. Ab lib, al-Kashf an
wujh al-qir wa-ilalih wa-ujajih, on sra 28:60.
163
Makk b. Ab lib, al-Kashf an wujh al-qirt, on sra 1:3.
164
Makk b. Ab lib, al-Kashf, on sras 55:56, 67:11, and often.
165
Makk b. Ab lib, Ibna (BerlinMs, 578), 508, ina separate chapter onthe causes of the
deviations between different transmissions of the same qira; al-Dhahab, abaqt (Berlin
Ms, Or. fol. 3140) 15
v
.
demand. Al-Kis can only cope with the throng of pupils by reciting him-
self instead of the pupils, while they are sitting and point (e.g. vocalize) their
maif.
166
As in this case, also everywhere else the written copy must have
presented considerable competition to the theory of the solely sanctioned
oral teaching. This is evident fromthe mass of old Koranic manuscript frag-
ments, and their conscientious provision with reading signs, and even refer-
ences to variants. This is also documented by separate statements of schol-
ars: No matter how esteemed a calligrapher might be in belletristic circles,
there is the equally strong warning against resorting to the muaf as the
source of Koranic wisdom.
167
What the great imms of Koranic variants, the eponyms of the Seven, [iii/146]
Fourteen, and other Readings performed was thus instruction in the Koran,
but not instruction in variant readings. More or less tolerant towards devia-
tions, they taught Koranic recitationaccording to one way, namely everyone
his own. It occurred only to the very latest of themto teach other than their
own ikhtiyr and also pass on the reading of their teachers. Particularly the
earliest among them, Khalaf b. Hishm al-Bazzr (died 229/843), was at the
same time one of the twomaintraditionists of amza b. abbal-Taym. The
century that separates him from Ibn Mujhid (d. 324/936) developed these
approaches to the later scientific teaching of the variant readings, which
centres on the transmission of a host of compact readings side by side.
After what has been said, the purpose of the isnds leading from the
great imms down to the Prophet cannot represent the transmission of the
individual reading of the predecessors in its unchanged totality. All this
says no more than that the later authority in the chain of authorities took
lessons on the Koran from the preceding authorityand this according to
the opinion of posterity, which in some cases might indeed go back to his
own statement. At best, these isnds can thus indicate the circle fromwhich
a certain reading evolved.
166
al-Dhahab, abaqt, 18
r
. Ibn Mujhid says: kna l-ns yakhudhna anhu alfahu
bi-qiratihi alayhim(ibid.). Something similar is reportedfromAbYay AYAHIBNQAYS
al-Kilb al-im (d. 121/738): kna l-ns yulina maifahumal qiratihi (Ibn al-Jazar,
abaqt [Berlin Ms, cod. simul., 55] 135
r
).
167
Ab timal-Sijistn (d. 248/862 or 55/869) says: l takhudh al-Qurnani l-muaf//
yna; similar statements are ascribed to the jurisconsults al-Shfi and [Sufyn] AL-THAWR
[EI
2
; EQ; Juynboll, Encyclopedia, pp. 628643] (al-Jabar, commentary on the Shibiyya,
towards the end of the introduction). Elsewhere (Ibn Jinn, Mutasab on sra 37:52 and 53)
AB TIM Sahl b. Muammad AL-SIJISTN speaks of ba al-juhhl of the scribes of the
Koran.
This much is certain: there had been instruction in the Koran already in
the first century. The oldest totally unsuspicious evidence is Ibn Mujhids
statement that Ab Abd al-Ramn AL-SULAM (d. 73/692 or 74/693) was
the first man to have taught al-qira al-mujma alayh at Kfa, i.e., not
the one of Ibn Masd.
168
It was also al-Sulam who transmitted the tra-
dition sanctioning the teaching of the Koran: khayrukum man taallama
l-Qurna wa-allamahu.
169
There are reports going back still further, accord-
ing to which both Ibn Masd (d. 32/652 or 33/653), and Ab l-Dard
(Uwaymir al-Khazraj, died 32/652),
170
q of Damascus, provided instruc-
tion in the Koran. As regards Ibn Masd, we only have the frequently cited
legend of the pupil who could not pronouce th.
171
It seems to be a copy of
the identical narrative about Abl-Dard.
172
It is quite likely that IbnMasd
was indeed concerned with spreading the Koran in his own recension, and
that it won him a certain number of followers at Kfa.
173
However, we must
not imagine the governor of Kfa to have been a professional instructor in
the Koran. The acquaintance with the Koran of the Kfan qurr that we
encounter in the Battle of iffn is likely to have resulted in the first place
fromthe study of maif of the recension of Ibn Masd,
174
and then also of
Uthmn.
As far as Ab l-Dards activity as a teacher of the Koran is concerned, [iii/147]
we have a detailed report
175
apart from the legend cited above. No matter
how much this reminds us of narratives of the instruction of teachers of
the Koran in historically better documented periods,
176
it probably cannot
168
Ibn al-Jazar, abaqt (Berlin Ms, cod. simul., no. 55) 108
r
.
169
Ibid., 108
v
; also in al-Bukhr (K. al-Fail al-Qurn, bb 21) he is the only one to
transmit the traditionfromUthmn. Already in Ibn Sad (abaqt, vol. 6): Biographien der
Kufier, p. 119, l 19sqq. is he the authority for the teaching of the Koran. [Juynboll, Encyclopedia,
p. 535, col. 1, The most meritorious among you is he who studies the Qurn and instructs
others.]
170
See above, p. 463 n. 606; EI
2
; Juynboll, Encyclopedia, s.v.
171
See p. 41 n. 131, and above, p. 463 n. 608.
172
173
174
175
Ibn Askir, al-Tarkh al-kabr (Damascus, 1329/1911), vol. 1, p. 69; al-Dhahab, abaqt
(Berlin Ms. cod. Or. folio, 3140) 3
v
sq.; Ibn al-Jazar, abaqt (Berlin Ms., cod. simul. 55), 159
r
= Takprzade, Mift al-sada (Hyderabad, 1328/1910) vol. 1, p. 354, l 11 sqq. = Mevzuat
ul-ulum, vol. 1, p. 452, l 17sqq.; the number of his pupils is here given as being over 1,600!
Cf. further the report cited below, p. 517 n. 51, where, apart from Ab l-Dard, also Mudh b.
Jabal (Juynboll, Encyclopedia, s.v.) and Ubda b. al-mit (ibid.) are mentioned. According
to Ibn Sad (abaqt, vol. 2, part 2): Letzte Krankheit, Tod und Bestattung, p. , l 2, Mudh
b. Jabal was a teacher of the Koran; also Ab Ms AL-ASHAR, ibid., p. , l 10.
176
The habit to assign a pupil first to a arf or subordinate muqri is also practised by Ibn
be rejected outright because Damascus held a special place in the history
of the transmission of the Koran later as well.
177
Unlike at Kfa, where a
pre-Uthmnic recensionof the Koranmight have beenofficially introduced
at an early time, at Damascus the demand for instruction in the Koran was
likely greater than at Kfa.
Teaching the Koran does not mean the teaching of one reading (qira) [iii/148]
but rather refers to teaching the Koran from the point of view of one par-
ticular reading; to be taught is as much part of life and the purpose of qira
as is the private and ritual recital. In early times there must certainly have
been discussions about the variant readings side by side with instruction in
the Koran. The interest must have centred on the individual passage of the
Koran with its numerous possibilities of pointing and vocalization. Who-
ever advocated the different ways of pronunciation had to take a second
seat. This was in stark contrast to later when one begins with one reader
and tries to find out how he read a passage. The textual expression of this
ancient science of the variant readings we may best look for indirectly in
the late dogmas of the Seven, the Ten, the Fourteen, etc.; it is expressed
immediately inthe individual readings. For various purposes it was of course
possible to select at will representations of complete readings. Moreover, a
vast number of individual readings are cited, predominantly in commen-
taries onthe Koranas well as inworks onshawdhdh
178
andtall,
179
but alsoin
the grammatical-lexicographic literature and in books on qirt etc. which,
although partly attributed to one of the well-known readers,
180
are either
missing from the official presentation of his reading or cannot have been
derived from any type of transmission, and which partly belong to readers
outside the circle of the Fourteen Readers.
181
They show us an older schol-
arly tradition of Koranic variant readings, at home among philologists and
theologians who pursued scientific investigations in the Koran, whereas the
later science of qirt after Ibn Mujhid developed basically from synopses
of different forms of practical-cultic instruction in the Koran.
Mujhid, who employed eighty-four [deputy officials] khalfa (Ibn al-Jazar, al-Nashr, vol. 1,
p. 121, l 25).
177
See pp. 435 and p. 460.
178
Cf. p. 431 sq., and p. 445.
179
= Justification of the acceptance of a variant reading. More later.
180
Among the Seven Readers it is most frequently Ab Amr b. al-Al.
181
Cf. the index of my (Bergstrers) Nichtkanonische Koranlesearten.
Criticismof the Transmission
The above survey of the history of instruction in the Koran and variant [iii/149]
readings offers at the same time a starting point for an evaluation of the
reliability of the transmission of the variant readings. As far as the complete
readings of the Koran from approximately 300ah onwards are concerned,
much knowledge has certainly been lost, but serious changes have hardly
taken place. For the period from 100ah to 300ah Muslim transmission
deserves penetrating criticism; it tends to make more recent texts look
much older. After all, the investigation of the Koranic reading of asan
al-Bar (d. 110/728)the oldest surviving and, at the same time, one of
the weakest transmissionsresulted in the great likelihood that it reflects
an approximate textual form which was used by him or at least by his
immediate followers.
182
No single direct transmission of any of the complete
readings of Koranic variants dates from the first century.
If the complete readings could be misrepresented by the incursion of
foreign readings, in the case of individual variant readings there was the
danger that they were ascribed to the wrong authorities.
183
In addition, a
tradition which was merely taughtparticularly when this was done in
writingwas more readily subject tomisunderstanding andoccasional cor-
ruption.
184
Thus, in cases when there is a complete variant reading next to
diverse single readings from one and the same reader, in principle the lat-
ter ones must be considered more likely to be the original. In the individual
182
Bergstrer, Die Koranlesung des asan von Bara, p. 55, cf. p. 48sqq.
183
The classic example of a duplicate in which one of the two references must be wrong is
the following: It is reported fromboth afa bt. Umar and isha that she particularly asked
the scribe copying for her a muaf to insert alti (or wa-alti) al-ar in sra 2:239 after
al-wus; both already in Mlik b. Anas, Muwaa, section al-alt al-wus, and Ab Ubayd
al-Qsimb. Sallm, Faail (Berlin Ms. 451), 37
v
; cf. Goldziher, Introduction to Islamic theology
and law, p. 15, no. 8, sq. [Juynboll, Encyclopedia, p. 393, col. 1]. This is unlikely to be a simple
error, it is rather that the two versions are competing tendentious traditions, one of which is
the favourite isha, who is generally consideredthe highest authority, andthe other is afa,
who in this special case was generally better qualified because of the role she played in the
first semi-official collection of the Koran. Cf. further, the frequent fluctuation between Ibn
Masd and Ubayy b. Kab, see above, p. 433 n. 302 and n. 314, 435 notes 331 and 334, p. 439
n. 377, p. 440 n. 402, p. 441 n. 418, p. 443 n. 439, p. 447 n. 474, p. 449 n. 491, p. 450 n. 501, and
p. 453.Extraordinary frequently contradicting variant readings are transmitted from one
and the same reader; Ibn Mujhid (in the Mutasab of Ibn Jinn) usually points this out by
adding bi-khilf or a similar remark.
184
Cf., e.g., above, p. 438 n. 367 and n. 374, p. 441 n. 413, p. 442 n. 425, p. 443 n. 439, p. 448
n. 483, p. 449 n. 495, p. 450 n. 505, and p. 451 n. 518; also Bergstrer, Die Koranlesung des
asan von Bara, p. 48.
case, however, one can never be certain that the variant reading really origi-
nated fromthis particular person, and what it looked like.
185
The uncertainty
becomes particularly great in the case of variant readings that are ascribed
to authorities of the first century ah; here, the danger of tendentious fab-
rications is an additional aspect, and this the more so the closer we come
to the Prophet.
186
In any case, the critical investigation of the reports about
at least Ibn Masds
187
Koran produced a more favourable result than might
have been expected under the circumstances.
188
Apart from Ibn Masd (and Ubayy b. Kab) a special place among the
authorities of the first century ah belongs to Abd Allh IBN ABBS (died
in 68/687).
189
This is based on his importance for the case of tafsr. The
statements about his variant readings are part of the statements about his
explanations and can only be appreciated together. As we cannot assume
that the earliest authorities in particular were familiar with all of the Koran,
one would expect each of themexcept in indicated passagesto occur
only in particular parts of the Koran. An investigation as to whether, and
to what extent, this applies, might allow further conclusions regarding the
degree of the reliability of a transmission.
Now and then Muslims as well criticized the reliability of the transmis- [iii/151]
sion of the variant readings. Criticism of the transmission is always likely
to be of secondary importance because it was primarily the variant read-
ing itself that was the target. Already Mlik b. Anas (d. 179/795) qualifies,
and at the same time explains, his rejection of Ibn Masds text by speaking
of al-qira allat tunsab il Ibn Masd.
190
Al-Zamakhshar acquits himself
185
Cf. Bergstrer, Die Koranlesung des asan von Bara, p. 48sq.
186
See above, note 183.
187
They belong here to the transmission of individual variant readings since we no longer
have a scholastically perpetuated and complete transmission of Ibn Masds reading.
188
189
This is evident, for example, from al-Kashf an wujh al-qirt of Makk b. Ab lib,
where from among the early representatives of the canonical variant readings listed, Ibn
Abbs is the eldest. He is followed by Ab Abd al-Ramn AL-SULAM (see above p. 439)
and Ab Sulaymn YAY IBN YAMAR al-Adwn (died before 90/708) [EI
2
; EQ; Sezgin,
GAS, vol. 9, pp. 3334], who seems to have been of similar importance to Bara as did
al-Sulam toKfa. (See below, the chapter The historical development. p. 458). [EI
2
; Juynboll,
Encyclopedia, p. 1; Sezgin, GAS, vol. 8, pp. 2122.]
190
InMakk b. Ab lib, Ibna(BerlinMs. 578), 511. Similar, Sbawayh119, 269, 454; al-q
Isml b. Isq al-Mlik, d. 282/895 (Ibna 501;)cf. on Ubayy b. Kab above, p. 455 n. 574;
Sbawayh 276. Also in other readings that were later considered shdhdh the reliability of
their isnd has been challenged, e.g., in the case of the one of [Ab Abd Allh Muammad b.
Abd al-Ramn] IBN AL-SAMAYFA [al-Yaman, d. before 169/785] (Ibn al-Jazar, al-Nashr
f l-qirt al-ashr, vol. 1, p. 16, l 5; abaqt [Berlin Ms., cod. simul, 55] 205
v
), the one of
of the linguistically inconvenient variant reading of Ab Amr b. al-Al
baghattatan
191
in sra 47:20 (instead of baghtatan or baghatatan) by assum-
ing a mistake of the rw, although Ibn Mujhid supplies the complete isnd
in Ibn Jinn, Mutasab, s.v.
Orthodox Teaching
Ibn Mujhids (d. 324/936) crucial innovation, the great break with the past, [iii/152]
consists not so much of setting in motion the canonization of the Seven
Readings. Rather, it was evidently he
192
who finally replaced the study of
individual variant readings in the science of the variant readings with the
study of complete readings as taught in the schools of the Koran. What
persuadedhimwas not only the traditionalismthat receiveda freshimpetus
from everywhere during this time.
193
The decisive factor was that there was
a need for a tradition-based guarantee for the qirt as well as for the
Koran itself,
194
which, after all, existed only in the readings: The muaf
did not suffice when it came to oral transmission. A collection of single
readings, even with perfect isnds, was an insufficient guarantee; required
were impeccable isnds for readings that applied to the whole of the Koran.
The preference for complete readings caused a rethinking in the entire
field of the science of variant readings. In the same way that this prefer-
ence gained ground only gradually, this rethinking, too, spread only step
by step. Makk b. Ab lib (d. 437/1045) still formulated his canon of the
three criteria for the approach to the individual readings,
195
and avails him-
Ab l-Samml [EI
2
] (Nashr, ibid., Tabaqt, 170
r
) and the one of im al-Jadar (abaqt,
91
v
). The qira of Ab anfa (d. 150/767) is recognized as a simple forgery, collected by
Ab l-Fal Muammad b. Jafar AL-KHUZ (d. 408/1017) [Brockelmann, GAL, suppl. vol. 1,
p. 723; Sezgin, GAS, vol. 1, pp. 1617] (Ibn al-Jazar, al-Nashr f l-qirt al-ashr, vol. 1, p. 16, l
6sqq.).
191
Not included in the canonical reading of Ab Amr b. al-Al.
192
On the one hand, it is certain that the change must have taken place during his time,
on the other hand, his activity was epoch-making (see below), and he was the most severe
enemy of ancient licences (see above, p. 468 and 476). Thus, the innovation is to be accorded
to him. But explicit reports supporting this are hardly available; still, cf. perhaps al-Dhahabs
statement, abaqt (Berlin Ms. 9943, p. 127; missing in Ms. Or. folio, 3140, 23
v
). Ibn Mujhid
preferred Ab Muammad al-Yazd from among the transmitters of Ab Amr b. al-Al
because he transmitted exclusively from him.
193
The end of the third century ah is the time when the door of ijtihd was closed.
194
al-Naysbr, commentary on the Koran, muqaddima 3, masala 1; cf. Ibn al-Jazar,
al-Nashr, vol. 1, p. 13, l 2, citing al-Suy, al-Itqn, naw 22 (Sprenger ed., p. 179, l 8).
195
self of ikhtiyr.
196
This, however, must not be misinterpreted, because he
teaches the Seven Variant readings essentially as units. In reality, his three
criteria had already lost their meaning when he was teaching them. In the
Seven Readings not only were linguistic mistakes tolerated,
197
but indeed
also deviations from the Uthmnic text.
198
Later scholars, thus, proceeded
accordingly and abandoned the three criteria.
199
The ikhtiyr outlasted Makk b. Ab lib but in a shabby and narrow [iii/153]
form: The same way that the majority carriedweight only withina particular
variant reading,
200
a reading was now not chosen for oneself, but rather
this was done for the originator of the variant reading or the riwya of
a reading by settling the differences of the transmission,
201
and filling in
the lacunae.
202
In constantly weakening form the ikhtiyr survives until the
threshold of modernity: Ab Muammad Abd Allh b. Muammad YSUF
EFENDI-ZADE
203
(d. 1167/1753) composed an entire book, al-Itilf f wujh
al-ikhtilf
204
to establish and justify his ikhtiyr from pending differences
within the Ten Readings. Only a man like al-Zamakhshar (died 538/1143),
dogmatically heterodox and not a genuine representative of the science of
variant readings, can rescue the word from losing its complete old sense.
205
Lacunae in the transmission were not filled at discretion but, if possible, [iii/154]
by ones own method, qiys, which has little in common with the qiys
that is one of the four roots of fiqh. It nevertheless has its roots in fiqh
and corresponds exactly to what was called there qiys qawl fuln,
206
a
196
See above, p. 485.
197
198
199
200
Cf. ijm in al-Dn, Taysr, passim; jumhr, in Ibn al-Jazar, al-Nashr, vol. 1, p. 392, l 15,
and frequently.
201
Ibn Mujhid (in Makk b. Ab lib, al-Kashf an wujh al-qirt, s.v.) thus chooses
sra 18:95where from Ab Bakr al-iddq an im al-Jadar both tn and tn are
transmittedfor him the second one; similarily in the Kashf on sra 27, at the end, in the
Taysr of al-Dn, chapter madhhab Ab Amr f tark al-hamza, and frequently.
202
E.g., nothing has been transmitted about the position of Ibn Kathr al-Kinn and Ibn
mir al-Yaub regarding waqf al marsmal-kha (see above, p. 408 n. 138), and al-Dns
teachers filled up the gap with ikhtiyr (Taysr, in the respective chapter).
203
Sezgin, GAS, vol. 1, pp. 125 and 139.
204
Lithograph, Constantinople, 1312/1894, in the margin of mid b. Abd al-Fatt al-
Plaw (BLAW), Zubdat al-irfn.
205
E.g., maliki huwa l-ikhtiyr on sra 1:3, arayta laysa bi-l-ikhtiyr on sra 107:1.
206
From fiqh, e.g., al-abar, Ikhtilf, ed. by Kern, vol. 1, p. 45, l 12 and p. 48, l 3, and often;
from the variant readings, al-Dn, Taysr on sra 39:12 and 41:44, synonymous with qiysan
al madhhab ibid., qiysu madhhabi ibid., chapter al-waqf al marsm al-kha; cf.
al-qiysu ibid., chapter, al-hamzatayn f kalima.
a ruling based on the opinion of a particular authority or school; here as
there, the opposite is na,
207
an explicit ruling. Makk b. Ab lib
208
(d. 437/
1045) supplies a theory of qiys inthis sense: The content of his book is made
up of three parts (1) qism qaratu bi-hi wa-naqaltuhu wa-huwa man f
l-kutubmawjda(2) wa-qismqaratubi-hi wa-akhadhtuhulafanawsaman
wa-huwa ghayru mawjd f l-kutub, and (3) wa-qism lam aqra bi-hi wa-l
wajadtuhu f l-kutub wa-lkinqistuhu al m qaratubi-hi wa-huwa l-aqall.
Ibn al-Jazar
209
also includes within the scope of this qiys phonetically
difficult matters (ghum wajh al-ad), apart fromthe case in which na is
wanting. We are indebted to this method for the apparent completeness of
the variant readings that have come downto us. Inorder to get anidea of the
real stock one would have to eliminate first of all the entire set of variants
qiys, and then also the other ikhtiyrt.
Shdhdh is now what lies outside the recognized canonical readings.
210
[iii/155]
The only controversy remaining is whether the canonical readings are the
Seven or the Ten or still more variants: shdhdh ani l-saba wa-l-ashara
wa-ghayrihim.
211
Linguistic usage which contrasts shdhdh with the Seven
can be traced back to Ibn Mujhid (d. 324/936). He supplemented his Kitb
al-Saba with a Kitb al-Shawdhdh.
212
As its crowning event and in order to achieve a firmer framework than
that of earlier perceptions,
213
the newsystemadopts the concept of tawtur,
207
Occasionally especially athar: al-Dn, Taysr, chapter al-hamzataynminkalimatayn; or
sam; Makk b. Ab lib, al-Kashf on sra 70:1. Different from na, but in contrast to qiys,
is ad (e.g., Taysr, chapter hamzatayn min kalimatayn; Ibn al-Jazar, al-Nashr, vol. 1, p. 17, l
19), pronunciation which is transmitted by means of enunciation and repetition (mushfa,
for example, al-Taysr onsra 26:56), but not by a pronunciationthat is explicitly established;
cf. further, in the text.Qiys, incidentally, also as one of the roots of fiqh is the antithesis to
na, namely Koran and tradition.
208
Tabra (Istanbul Ms, Nuruosmaniye Ktphanesi, 55), 162
r
, cited in Ibn al-Jazar, al-
Nashr, vol. 1, p. 17, l 1 sqq.
209
al-Nashr, vol. 1, p. 17, l 18. According to Ibn al-Jazar (ibid., l 20sqq.) all these matters are
not qiys in the strict sense (al l-wajh al-iila) rather nisbat juzyin al kullyin.
210
The old meaning, somewhat reduced, is held by Ab Shma (d. 665/1266; cf. p. 446); for
him shdhdh is the opposite of mujma alayh (cf. Ibn al-Jazar, Nashr, vol. 1, p. 10, l 3, citing
al-Suy, al-Itqn, naw, 22 [Sprenger edition, p. 177, l 5]). Also in Ibn al-Jazar (died 833/1429)
himself we occasionally find this meaning; e.g., he says: shdhdh ani l-mma in abaqt
(Berlin Ms., cod. simul. 55), p. 170
r
.
211
212
Ibn Jinn (d. 382 sic) accepts this division without hesitation and speaks of arban
sammhu ahl zamnih shdhdhan ay khrijan an qirat al-qurr al-saba (al-Mutasab,
[Istanbul Ms., Ragp Paa Ktphanesi, 13] 1.) Also Ibn al-Nadm (writing in 377/987) bases
the disposition of the section on the readers of the Koran on this (al-Fihrist, p. 28sqq.).
213
See above, p. 485sq., [the dogma of tawtur].
i.e. a transmission going back to several independent sources, taken from
the terminologies of the criticism of tradition (muala) and ul al-fiqh
and applied to the science of the various readings.
214
The canonical readings
are mutawtir, and their transmission rests with each generation, not with
the attestation of individuals (khabar al-wid, al-d), but on general
agreement.
215
Mashhr, too, some men now no longer use in the general
sense of recognized but in its terminological meaning in muala, where
it indicates the second stage of tawtur, practically an equivalent, but as far
as attestation is concerned not quite conforming to its rules.
216
The last two true scholars among the representatives of the science of [iii/156]
the variant readings, Abd al-Ramn b. Isml AB SHMAH (died 665/
1266) and Ibn al-Jazar (died 833/1429) did not submit to the orthodoxy
of the dogma of tawtur. Ab Shma considers them applicable only to
those parts of the Seven Readings in which these do not deviate from one
another;
217
with this, it loses its raison d tre. At first, Ibn al-Jazar tolerated
214
This happened rather late; Ab Shma (d. 665/1266) in (Ibn al-Jazar, al-Nashr, vol. 1,
p. 13, l 9sqq., citing al-Suy, al-Itqn, naw 22 [Sprenger ed., p. 179, l 14), andIbnal-Jazar (died
in833/1429), himself (al-Nashr, vol. 1, p. 13, l 1, citedibid., [Sprenger ed., p. 179, l 7]) attribute the
dogma of tawtur to some mutaakhkhirn (modern authors). When Ibn al-Jazar (al-Nashr,
vol. 1, p. 30, l 8sqq., citing al-Itqn, naw, 2227, tanbh 2 [Sprenger ed. 189, p. 4]) says that the
well-known dogmatist al-q Ab Bakr b. al-ayyib AL-BQILLN (d. 403/1012) has (in his
Kitb al-Intir) naa al tawtur dhlika kullih (namely of the ad) this can hardly be a
direct citation. It is even less likely that the content, f l-qirt al-sab al-mutawtira, which
Haji Qalfa adds to the title, Kitb al-Sab, can originate fromthe author himself, Ibn Mujhid
(died in 324/936).
215
Ibn al-jib (d. 646/1248; Brockelmann, GAL, vol. 1, p. 303; EI
2
) in Ibn al-Jazar, al-Nashr,
vol. 1, p. 30, l 4sqq., abbreviated in al-Suy, al-Itqn, naw, 2227, tanbh, 2 (Sprenger ed.,
p. 188, l 21 sqq.); al-Naysbr (d. ca. 706/1306; see above, p. 407 n. 130) in the passage cited
on p. 443 n. 193; al-q Jall al-Dn (Muammad b. Ibrhm) b. al-jj AL-BALLAFQ
(d. 770/1368), so Ibn al-Jazar, abaqt [Berlin Ms., cod. simul., 55] 225
v
) in al-Itqn, naw
2227, tanbh 2 (188, 16);from the representatives of ul, for example, adr al-Shara
al-Thn (Ubayd Allh b. Masd, d. 747/1346, Brockelmann, GAL, vol. 2, p. 214, [EI
2
]), Taw,
qism1, rukn 1, beginning; Tj al-Dn AL-SUBK (died 711/1370; Brockelmann, GAL, vol. 2, p. 89),
Jam al-jawmi, beginning of kitb 1, and in Ibn al-Jazar, al-Nashr, vol. 1, p. 44, l 12sqq.
(Citations from this passage and from Man al-mawni, with a discussionconcluded with
al-Subks signed expertisebetween himand Ibn al-Jazar about the question whether also
the Three after the Seven are mutawtir.)
216
Cf. al-Suy, al-Itqn, naw 23 (Sprenger ed., p. 181, l 6sqq.), and naw, 2227, tanbh,
2 (Sprenger ed., p. 188, l 16sq.). The complete scaleal-Itqn, naw, 2225consists of the
four stages, mutawtir, mashhr, d, and shdhdh. It is a deterioration when al-Ballafq,
loc. cit., puts mutawtir = al-qirt al-saba (sic!) al-mashhra, d = the Three after the
Seven as well as the Companions of the Prophet, shdhdh = all the rest of them.
217
See above, p. 503 n. 215. The wording is not quite clear. It could also be interpreted
to mean that Ab Shma intended to exclude only those variant readings from tawtur on
this, but he later rejected it with the pertinent argument that tawtur dis-
penses with the criteria of both correct grammar and agreement with the
muaf or does not allow their application.
218
Both are late and impracti-
cal attempts at adhering to the individual appreciation of the variant read-
ings.
219
Some scholars enjoy somewhat more licence since they remain outside [iii/157]
the guild of the muqrin; this applies particularly to the commentators Ab
Muammad al-usayn b. Masd al-Farr AL-BAGHAW (d. 510/1117) and
al-Zamakhshar (d. 538/1143).
220
which the transmitters of the same qira were at variance. This way the restriction would
lose all importance. Already Tj al-Dn AL-SUBK (Jam al-jawmi, beginning of kitb 1) inter-
preted Ab Shma the way it was done in the afore-mentioned text. A different restriction
is the one of both Ibn al-jib (d. 646/1249) and al-Naysbr (d. ca. 706/1306), loc. cit., who
exclude the peculiarities of the pronunciation, ad, from tawtur. This reflects the proper
recognition that these matters were settled later than the actually variant readings.Tj al-
Dn AL-SUBK mentions both restrictions, loc. cit.Badr al-Dn Muammad b. Abd Allh
AL-ZARKASH (d. 794/1391; Brockelmann, GAL, vol. 2, p. 91) recognizes tawtur only back to
the Seven Readers, but not back from them down to the Prophet (al-Suy, al-Itqn, naw,
2227, tanbh, 2 [A. Sprenger ed., p. 179, l 17]).
218
Munjid (al-muqrin wa-murshid al-libn), chapter 6, and in contrast al-Nashr, vol. 1,
p. 12, l 23sqq. (citing al-Suy, al-Itqn, naw, 22 [A. Sprenger ed., p. 179, l 4sqq.]), where this
particular interpretation is considered the later one.
219
Ab Shma goes so far as to find shdhdh variant readings also within the Seven
Readings (inna l-qirt al-mansba il kull qri min al-saba wa-ghayrihim munqasima il
l-mujma alayhi wa-l-shdhdh, in Ibn al-Jazar, al-Nashr, vol. 1, p. 10, l 2sq.); similarly some-
what earlier al-usaynb. Masd AL-BAGHAW (d. 510/1117); [EI
2
; EQ; R.A. Nicholson, Literary
history of the Arabs, p. 337; Sezgin, GAS, vol. 1, p. 155 (1)]; in Ibn al-Jazar, al-Nashr, vol. 1, p. 44,
l 10.
220
See the preceding foot-note, and p. 423sq. More details below.
THE READERS AND THE VARIANT READINGS
Sources
The biographies of readers and representatives of the science of variant [iii/157]
readings can be found in many different types of biographical collections,
primarily of grammarians, littrateurs, uff, and traditionists. Isolated bits
of important informationare contained inthe historical literature. Ina class
of its own is Ibn al-Nadms (fl. 377/987) Fihrist. Its relevant sections,
1
even
if just for their antiquity, are important as an independent presentation
outside the science of the variant readings.
The introductions to the works on the variant readings, with their inev- [iii/158]
itable isnds, afford ample opportunity to supply ex professo information
about the readers. Probably the most valuable of these biographical intro-
ductions are contained in the Kitb al-Iqn f id ashra qira of Ab Al
(al-asan b. Al) AL-AHWZ
2
(d. 446/1055), and the K. al-Kmil f l-qirt
of (AB AL-QSIM Ysuf b. Al) AL-HUDHAL
3
(died 465/1072).
4
The cor-
responding sections of Ibn al-Jazars al-Nashr
5
constitute for us the most
important example of this genre; in addition, Ibn al-Jazar supplied a con-
cise but important history of the science of the variant readings
6
in that part
of his introduction dealing with basic principles.
The oldest collection of biographies of readers seems to have been com-
posed by (Ab l-asan [or al-usayn] Amad b. Jafar) IBN AL-MUND
(d. 334 or 336/945 or 7);
7
not much later is the K. al-Mujam f asm al-
1
Pp. 2731, on readers; pp. 3133, and 3839, on scholars of the variant readings.
2
Brockelmann, GAL, vol. 1, p. 407, ibid., suppl. vol. 1, p. 720; Sezgin, GAS, vol. 1, p. 603, l 5;
andalsoYqt, Irshd, vol. 3, p. 152. Yqt made use of al-Iqn, cf. G. Bergstrer, Die Quellen
von Jqts Ird, in Zeitschrift fr Semitistik, vol. 2 (19231924), p, 198, no. 98 (preceded by
some other related sources of Yqt).
3
Yqt, Irshd, vol. 7, p. 308.
4
Ibn al-Jazar used both books in his Ghyat al-nihya f abaqt al-qurr.
5
Vol. 1, pp. 53192.
6
al-Nashr f l-qirt al-ashr, vol. 1, p. 33, l 14 to p. 35, l 10.
7
Sezgin, GAS, vol. 1, p. 44. Cf. G. Bergstrer, Die Quellen von Jqts Ird, loc. cit.,
no. 95. The most reliable trace is Yqts statement in Irshd, vol. 5, p. 248, l 11 sq.: ql
Ibn al-Mund f man mt f sanat 287 The title of the presumable book that was
arranged by date of death cannot be identified. More about the author in Fihrist, p. 38sq.
(Ab l-asan, d. 334/945); Ibn al-Jazar, abaqt (Berlin Ms. cod. simul., no. 55), 12
r
sq. (Ab
506 the readers and the variant readings
qurr wa-qirtihim of Ab Bakr (Muammad b. al-asan) AL-NAQQSH
(died 351/962).
8
The first of these made little impression, and the second,
hardly any at all. Of prime importance for later scholars are only abaqt
al-qurr of al-Dn (d. 444/1052), which apparently did not survive, along
with the K. Marifat al-qurr al-kibr al l-abaqt wa-l-ar (mostly called
abaqt) of (Muammad b. Amad) AL-DHAHAB
9
(d. 748/1348), who un-
fortunately does not list his sources. These works constitute the main
source
10
of what is still the most comprehensive collectioninthe field, Ibnal-
Jazars (d. 833/1429) Nihyat al-diryt f asm rijl al-qirt,
11
which seems
to have survived in excerpts by the author, entitled Ghyat al-nihya.
12
al-
Dhahab restricts himself to the better knownreaders but treats themrather
thoroughly. Ibn al-Jazar is likely to have gone into greater detail in his main
work, but in his excerpt the meagre lists of names of pupils and instruc-
tors predominate by far. According to Ibn al-Jazars own statement, the
materialprobably meaning the number of persons listedwas approxi-
mately double in comparison with the works of both al-Dn and al-
Dhahab; yet compared with al-Dhahabs arrangement by abaqt, the
alphabetical arrangement greatly facilitates the location of individual per-
sons. The growth of the material can be explained by Ibn al-Jazars plan,
first of all, to bring the collection down to his own time, and then to present
l-asan, d. 336/947 sic). al-Suy, Bughya, s.n. (Ab l-usayn, d. before 320!/932).The
K. Afwj al-qurr of al-q Ab Ysuf Abd al-Salm al-Qazwn, referred to by Bergstrer,
in Zeitschrift fr Semitistik, loc. cit., no. 94, I cannot date; it seems to have been written not
much after the time of Ibn Shannabdh (d. 328/939), and was used by Yqt.
8
Sezgin, GAS, vol. 1, pp. 4445; cf. also Fihrist, p. 33; Yqt, Irshd, vol. 6, p. 496sqq.
(the title, p. 497, l 2sqq., three editions, al-akbar, al-awsa, al-aghar); Takprzade, Mift
al-sada (Hyderabad, 1328/1910), p. 416sq. = Mevzuat ul-ulum, vol. 1, p. 531 sq.
9
Brockelmann, GAL, vol. 2, p. 46. Apart fromthe select biographies, BerlinMs. 9943, used
earlier, there is nowavailable for the abaqt al-qurr the complete andmuchsuperior Berlin
Ms., cod. Or. folio, 3140 (Ahlwardt, Verzeichnis).
10
A qurr work composed between the time of al-Dn and al-Dhahab, the K. al-Intir
f marifat qurr al-mudun wa-l-amr of Ab l-Al (al-asan in Amad AL-AR) AL-
HAMADHN (died 569/1173); [Sezgin, GAS, vol. 1, p. 12, l 1; cf. Pretzls Verzeichnis, no. 52,
p. 241] Ibn al-Jazar did not seeaccording to his own statement(abaqt, [Berlin Ms.,
cod. simul., 55]), 54
v
sq. Cf. about him, Yqt, Irshd, vol. 3, p. 26sqq.; Ibn al-Jazar, abaqt,
loc. cit.
11
This is how he cites the title in the introduction to the abaqt. Also calledactually
inappropriatelyKitb abaqt al-qurr al-kabr.
12
Also (Mukhtaar) abaqt al-qurr. A photocopy of the Istanbul manuscript, Nuru-
osmaniye Ktphanesi, no. 85, can be found as cod. simulata orientalia, no. 55, in the Staats-
bibliothek, Berlin. An excerpthardly usableof Tarjim rijl kitb al-Nashr min naqaat
al-qirt al-ashr of al-Sayyid MUAMMAD RIF AL-IF b. al-Sayyid Ibrhm can be
found as an autograph, completed in 1202/1787, also at the Staatsbibliothek, Berlin, cod. or.
oct., no. 2192.
the readers and the variant readings 507
quite a number of important works on variant readings with references to
these authorities. An excerpt of biographies of Ibn al-Jazars shorter version
of his abaqt, supplemented by statements of other provenancepartly
in a much abbreviated versionis contained in the section on readers by
Takprzade
13
(d. 968/1560), which is still valuable as the only printed col-
lection of such biographies.
Survey of the Older Readers
The earliest list of outstanding readers that we know of is the one by Ab [iii/160]
Ubayd al-Qsim b. Sallm
14
(d. 223 or 4/838). It includes a number of Com-
panions of the Prophet, followed by some forty Followers, and finally fif-
teen actual readers, three each from the five amr, Medina, Mecca, Kfa,
Bara, and Damascus.
15
Thus, in the last group, as far as the reading is con-
cerned, the five early Muslim centres are accorded equal treatment. This is
13
See above, p. 389 n. 2; the section on readers (Mift al-Sada [Hyderabad, 1328/1910],
vol. 1, pp. 347397 = Mevzuat ul-ulum, vol. 1, pp. 444507) discusses Ab Bakr and Umar,
then the younger aba and tbin of the list of Ab Ubayd (see below), further the
Seven Readers, with their two transmitters each, the Three after the Seven, and, finally, from
among the authors of works on readings only al-Dn and al-Shib in addition to several
commentators of the Shibiyya as well as Ibn al-Jazar and his sons. Masrq (Ibn al-Ajda) is
missing from among the list of tbi [Juynboll; Encyclopedia; Sezgin, GAS, vols. 1 and 2].
14
Most complete, with authors name, in Ab Shmas commentary on the Shibiyya at
the beginning of the introduction, and in al-Suy, al-Itqn, naw 20, who merely divides the
list intotwoparts, tracing backthe secondpart toal-Dhahab (Sprenger ed., p. 169, l 15sqq. and
p. 171, l 7sqq. Without the name of Ab Ubayd al-Qsimb. Sallm: Ibn al-Jazar, Nashr, vol. 1,
p. 8, l 5sqq.) (when citing individual statements from the list he occasionally mentions Ab
Ubayd, for example, abaqt [BerlinMs., cod. simul. 55] 85
v
); andsee above, note 13. Only the
Fifteen Readers, for example, also Takprlzade, Mift al-sada (Hyderabad, 1328/1910),
vol. 1, p. 366, l 10sqq. = Mevzuat ul-ulum, vol. 1, p. 467, l 1 sqq.Half a century more recent is
the compilation of nineteen brief biographies of readers in Ibn Qutayba (d. 276/889) [Sezgin,
GAS, vol. 8, p. 161] K. al-Marif (ed. Wstenfeld), pp. 262264; it does not seem to have made
any impression on posterity.
15
Medina: Ab Jafar IBN AL-QAQ al-Makhzm, Shayba b. Ni [Sezgin, GAS, vol. 9,
p. 203], Nfi b. Abd al-Ramn al-Layth; Mecca: IBNKATHR AL-DR [EI
2
; EQ], umayd b.
Qays AL-ARAJ, Ibn Muayin; Kfa: Yay b. Waththb al-Asad [G. Junyboll, Encyclopedia,
256, col. 2], im al-Jadar, Sulaymn b. AL-AMASH; Bara: Ibn Ab Isq al-aram, s
b. Umar al-Thaqaf, Ab Amr b. al-Al; Damascus: Ibn mir al-Yaub, Yay b. al-rith
al-Dhimr (this, the second name in Ab Shma, the other lists have a different order). The
last name, the one of the third Damascene reader, is omitted in the original version of the
list because the transmitter allegedly forgot; the lacuna is filled differently. In the ordinary
version the list for Kfa as well as Bara has been enlarged by two readers each who do not
belong there, namely amza b. abb al-Taym and al-Kis for Kfa, and im al-Jadar
and YAQB b. Isq b. Zayd AL-HARAM for Bara.
misleading. More to the point is the picture provided by the distribution
of the tbin, or even the selection of the Seven Readers. From among
the remaining cities this comprises one each, but three from Kfa. In the
earliest time, Kfa was the centre of the reading of the Koran as well as
the supply centre of Kfic manuscripts. The remaining cities, furthermore,
enjoy by no means an equal status: Medina and Bara come to the fore,
whereas Damascus recedes totally. The tabular survey of the better-known
early readers,
16
arranged by locality and date of death,
17
will illustrate this.
Damascus does not even require a separate column, as the number of local
readers is so insignificant that an appendix is all they need. The table has
been continued only to the middle of the second century ah, by which
time Medinaas Mecca did previouslyrecedes totally from the picture,
whereas Baghdad begins to appear at the side of Kfa and Bara.
The Historical Development
These readers are connected with one another and with the Companions [iii/169]
of the Prophet by chains of authority stating that so-and-so received the
qira from so-and-so, and this mostly aran, namely, the pupil recited and
the instructor corrected. Although these chains of transmitters (isnds) can
never be ascertained individually, the overall picture of the history of the
16
Listed are: (1) The tbin (Followers) and readers on Ab Ubayd al-Qsim b. Sallms
list (the older Companions have been omitted). (2) All readers that I knowof fromno matter
what source to whom an individual qira (or ikhtiyr) is ascribed. (3) Some other readers
whose importance ensues from the isnds. Among the sources to no. 2 there is first of all
the list in Fihrist, p. 30sq.; it is extraordinary typical of our insufficient information that
from among the authors of individual readings mentioned there some who cannot even
be identified with the help of Ibn al-Jazars abaqt, thus, not considered by him to have
been a reader. Not identified are (Medina), if not merely a mistake for
(Mecca), possibly identical with the traditionist missing in Ibn al-Jazar,Abd
al-Ramnb. AbdAllhb. Ab Ammr [Juynboll, Encyclopedia, p. 223; col. 1] (3rdgroup, what
would fit according to al-Fihrist, that he is the teacher of Ab Amr b. al-Al), and
(Damascus), if not even= Yazd b. Quayb (see below, p. 517). Missing inIbnal-Jazar are Abn
b. Uthmn (Medina) and Khlid b. Madn al-Kal [EI
2
; Juynboll, Encyclopedia, pp. 141, 180]
(Damascus) who is at least mentioned in the biography of Ab l-Dard, abaqt (edited by
Bergstrer and Pretzl, 19331935) nos. 1850 and 2480.The main source of information is
apart from the literature of rijl the abaqt of Ibn al-Jazar.
17
The dates of death frequently differ considerably, particularly when it concerns older
or less known authorities. Whenever Ibn al-Jazar or another of the later scholars settled for
one of the respective dates, to simplify matters, I used only these. In cases of vague dates, and
others, which are not contradictory, I referred to the last one only. Here, as elsewhere in the
book, I juxtaposed the earliest and the latest of the more or less ascertained dates.
text of the Koran in the first and second half of the second century, sup-
plemented by other references from the following period to the beginning
of the science of variant readings, nevertheless fills the outlines of the pre-
ceding considerations with individual traits and is in agreement with both
internal probability and historical conditions.
The isnds most clearly
18
showthe basic, familiar fact that (1) the tradition [iii/170]
of Koranic readings of each of the amr constitutes a complete unit;
19
(2)
that the individual readers, however much they personally help establish
local custom, are not merely individuals but also exponents of this local cus-
tom; and (3) that the differences, for example, between Ab Jafar Yazd IBN
AL-QAQ al-Makhzm and Nfi b. Abd al-Ramn al-Layth disappear,
considering that both of them represent the Medinan reading.
This local limitation of the Koranic reading is partly due to the influence
of the local model copies,
20
yet their difference was too minimal and their
ambiguities too great to be the decisive factor. In this we come face to face
with one aspect of an important phenomenon: immediately after the con-
quest, the individual Islamic centres, both old and new, constitute such self-
sufficient entities that there was virtually no exchange of ideas among them.
The domination of local custom over the individual reader is most obvi- [iii/171]
ous at Damascus. The generations before and after Abd Allh IBN MIR
AL-YAUB each have only one reader known by name; and this meagre
isnd, Ibn Ab ShihbAbd Allh IBN MIR AL-YAUBYay b. al-
rith al-Dhimr, lacks connection to the top.
21
Only as the champion of
the Damascene reading, which could not be totally ignored, was Ibn mir
al-Yaub accepted into the Seven by Ibn Mujhid, who himself was one of
their transmitters, albeit against the reservations of other members.
22
18
For the isnds traced back fromthe familiar readers (the Seven, etc.) to the Prophet, we
mainly used the abaqt of Ibn al-Jazar, and for the isnds in the opposite direction it is the
large list inal-Nashr, vol. 1, pp. 98192, withsupplementary details regarding secondary trans-
missions, ibid. pp. 4142; further, al-Naysbr, commentary on the Koran, al-muqaddima
al-l, and al-Jabar, commentary (Kanz al-man) on the Shibiyya, introduction (the
transmissions used by Ibn Mujhid).
19
See above, p. 492.
20
21
Already Muslimcritics, primarily al-abar, realized the futility of trying to have IbnAb
Shihbor even Ibn mir himselflearn the Koran from Uthmn or Ibn mir from Ab
l-Dard. The long lasting polemics are still reflected in a long appendix to the biography of
Ibn mir in the Mevzuat ul-ulum, vol. 1, pp. 477478.
22
Ab tim al-Sijistn [d. 255/869] and other writers do not consider him (Makk b.
Ab lib, Ibna, [W. Ahlwardt], Ms Berlin, no. 578, p. 496). AB HIR Abd al-Wid b.
Ab Hshim AL-BAZZR (280/893349/960; Sezgin, GAS, vol. 9, pp. 167168), states that if it
had not been for Ibn Mujhid who had opted for Ibn mir, he himself would have preferred
al-Amash. (al-Dhahab, abaqt, [Ahlwardt, Verzeichnis, Ms., Berlin, Or. fol. 3140,] 2b, l 11).
Medina: Mecca:
57/8 Ab Hurayra (al-Daws)
63 Mudh (Ibn al-rith al-Anr
al-Najjr) al-qri
68 (Ab l-Abbs Abd Allh) b. Abbs
(al-Hshim)
after 70/78 (Ab l-rith Abd Allh) b.
Ayysh b. Ab Raba (al-Makhzm),
aqra ahl al-Madna f zamnih
private reading; (Fihrist)
23
ca. 70 Abd Allh b. al-Sib
(al-Makhzm), qri ahl Makka
24
74 (Ab im)UBAYD IBN UMAYR
(al-Layth), al-q [EQ]
91/3 (Ab amza) ANAS IBN MLIK
(al-Anr al-Khazraj)[EI
2
; EQ;
G. Juynboll, Encyclopedia, 131134.]
93/5 (Ab Abd Allh) URWAH IBN
AL-ZUBAYR (Ibn al-Awwm)[EI
2
; EQ]
94 (Ab Muammad) SAD IBN
AL-MUSAYYAB (al-Qurash
al-Makhzm) [EI
2
; GAS, I, 276;
G. Juynboll, Encyclopedia]
23
Where wrong: instead of .
24
Juynboll, Encyclopedia, pp. 218, 626, 661.
Kfa: Bara:
53/64 (Ab Maysara) AMR IBN
SHURABL (al-Hamdn)
61/2 (Ab Shibl) ALMAQAH (IBN QAYS
AL-NAKHA)[GAS, I, 398]
61/4 (Ab Yazd) AL-RABI IBN
KHUTAYM (al-Thawr)[GAS, I]
63 (Ab isha) MASRQ (IBN
AL-AJDA AL-HAMDN)[GAS, I]
65 al-rith b. Abd Allh al-Hamdn
al-awar
69 Ab l-Aswad (lim b. Amr)
al-Dual, q of Bara, allegedly the
founder of grammar [EI
2
]
72/3 Abda b. Amr (or b. Qays)
(al-Salmn)[GAS, I]
after 70 in Abd Allh (al-Raqsh or
al-Sads) [G. Juynboll, Encycl., 442]
73/4
25
Ab Abd al-Ramn (Abd Allh
b. abb) AL-SULAM al-arr,
muqri al-Kfa
26
[EI
2
; GAS, I, 671]
74 (Ab Muwiya) UBAYD IBN
NUAYLAH (al-Khuz), muqri ahl
al-Kfa
74/5 AMR IBN MAYMN (al-Awd)
[G. Juynboll, Encyclopedia]
74/5 al-Aswad b. Yazd (Ibn Qays al-
Nakha), nephew of Alqama [GAS, I]
82 (Ab Maryam) ZIRR IBN HUBAYSH
(al-Asad)
27
before 90 (Ab Sulaymn) YAY
IBN YAMA/uR (al-Qays al-Jadal
al-Udwn),
28
q of Merv, where he
died [EI; EQ; GAS]
94/5 Sad b. Jubayr al-Asad al-Wlib),
mawl, executed by al-ajjj [EI
2
; EQ;
GAS]
89/90 NAR IBN IM (AL-LAYTH or
al-Dual, grammarian [EI
2
; EQ; GAS,
IX]
25
According to others only 85/704.
26
See above, p. 496.
27
Juynboll, Encyclopedia, 5 s.v.
28
Medina: Mecca:
101 Umar b. Abd al-Azz, caliph,
99/717101/720 [EI
2
]
102/4 (Ab l-ajjj) MUJHID (Ibn
Jabr) al-muqri, private ikhtiyr
102/3
29
A b. Yasr (al-Hill) al-q,
mawl of Maymna
Dirbs, mawl of Ibn Abbs; private
reading (Fihrist)
105 (Ab Sad) ABN IBNUTHMN
(Ibn Affn al-Umaw;) private
reading (Fihrist)[EI
2
; GAS, I, 277]
105/7 (Ab Abd Allh) IKRIMAH
(al-Barbar) al-mufassir, mawl of
Ibn Abbs
107 (Ab Ayyb) SULAYMN IBN
YASR (al-Hill), mawl of
Maymna, brother of preceding
[G. Juynboll, Enclyclopedia, s.v.]
106 (Ab Abd al-Ramn) TWS IBN
KAYSN (al-Yamn al-Jand) [EI
2
;
EQ]
106/30 (Ab Abd Allh) MUSLIM IBN
JUNDA/UB (al-Hudhal) al-q,
mawl
114/5 (Ab Jafar) (Muammad b. Al)
AL-BQIR, the 5th Imm
114/5 (Ab Muammad) A IBN
AB RAB al-Qurash al-Yamn
al-Janad, mawl [EI
2
; EQ]
117 (Ab Dwd) Abd al-Ramn b.
Hurmuz AL-ARAJ (al-Hshim)
al-qri, mawl, d. in Alexandria
[GAS, IX, 3435]
117 (Abd Allh b. Ubayd Allh) IBN AB
MULAYKAH (al-Taym) [Juynboll,
Encyclopedia]
122 (Ab l-usayn) ZAYD IBN AL
(Zayn al-Abidn b. al-usayn)
120
30
(Ab Mabad Abd Allh) IBN
KATHR (AL-KINN) al-Dr, dealer
in spice,
31
mawl from Yemen, of Per-
sian background, one of the Seven
123/5 (Ab Bakr Muammad b.
Muslim b. Ubayd Allh) b. Shihb
(al-Qurash) AL-ZUHR; private
reading [EI; EQ; GAS]
123 (Muammad b. Abd al-Ramn)
IBN MUAYIN (al-Sahm), mawl;
one of the Four after the Ten [EI
2
]
29
According to others already 94/712 or 97/715. [EI
2
, died 104/722.]
30
According to Ibn Khallikn, no. 326, this is a mistake of Ibn Mujhid who must have
confused the reader with a Abd Allh b. Kathr al-Qurash, who died in 120/737; the reader
really died later.
31
According to others, a nisba that others then derive from the Ban Abd al-Dr. The
nisba, Qurash, occasionally applied to Ibn Kathr, although a mawl of a Kinn, follows
partly from this interpretation, and partly from the confusion mentioned in the preceding
footnote.
Kfa: Bara:
95/6 (Ab Imrn) Ibrhm (Ibn Yazd)
AL-NAKHA, nephew of Alqama [EI;
EQ; GAS, I, 403404]
95/6 Ab Amr (Sad b. Iys) al-Shaybn
[G. Juynboll, Encyclopedia, pp. 109, 247]
103 Yay b. Waththb (al-Asad) muqri,
mawl [Juynboll, Encyclopedia, p. 526]
103/721 (Ab Amr mir b. Sharl
al-imyar) AL-SHAB [GAS, I, 277;
Juynboll, Encyclopedia]
105 Ab Raj (Imrn al-Urid)
[G. Juynboll, Encyc.]
110 (Ab Sad) al-asan (Ibn Ab
l-asan Yasr) (asan al-Bar,
mawl, one of the Four after Ten
112 ala b. Muarrif (al-Hamdn
al-Iym or al-Ym); private reading;
ikhtiyr [ibid.]
110 (Ab Bakr Muammad IBN SRN
(al-Anr)[GAS] mawl of Anas b.
Mlik
(3rd level) Ab Zura b. Amr b. Jarr (al-
Bajal) [Juynboll, Encyclopedia, s.q.]
117 (Ab l-Khab) QATDAH (Ibn
Dima al-Sads) al-am; private
ikhtiyr [EI
2
; GAS, I, 31]
127/9 (Ab Bakr) SIM (IBN AB
AL-NAJD Bahdalah al-Asad)
al-ann, mawl, one of the Seven
[EI
2
]
117/29 (Abd Allh) IBN AB ISQ
(al-aram), grammarian; private
reading (Fihrist) [EQ; GAS, VIII, 36]
before/ca. 130 (Ab amza) UM-RN
IBN AYAN (al-Shaybn), mawl
128 (Ab l-Mujashshir) IM (Ibn
al-Ajjj or Maymn) AL-JADAR;
private reading
Medina: Mecca:
130 Ab Jafar (Yazd IBN AL-QAQ
al-Makhzm) al-qri, Ayysh; one
of the Seven. [GAS]
130 (Ab afwn) umayd b. Qays
(al-Asad) mawl of Ibn AL-ARAJ,
mawl, private Three after reading
[GAS; Juynboll, Encyclopedia]
130 Shayba b. Ni, mawl of Umm
Salama, q of Medina; private
reading [GAS, IX, 203]
136 (Ab Usma) ZAYD IBN ASLAM
(al-Adaw), mawl of Umar b.
al-Khab [EQ; GAS]
148 (Ab Abd Allh) JAFAR (Ibn
Muammad) AL-DIQ, 6th imm
169 Nfi (Ibn Abd al-Ramn) b. Ab
Nuaym (al-Layth), from Isfahn,
mawl, one of the Seven [EI
2
; GAS, I]
Damascus:
32
Khlid b. Sad, ib Ab l-Dard
33
91 (Ab Hshim) AL-MUGHRAH IBN AB SHIHB (Abd Allh al-Makhzm)
118 (Abd Allh) IBN MIR (AL-YASUB), q of Damascus, imm and adminis-
trator at the Umayyad Mosque; one of the Seven
121 (Ab Yay) AYAH IBN QAYS (al-Kilb al-im), later at Damascus
145 YAY IBN AL-RITH (al-Ghassn) al-Dhimr, imm at the Umayyd
Mosque, private ikhtiyr. [GAS, IX]
Inadditionim, fromwhere originates also the above-mentionedAiyya b. Qays:
34
32
I cannot identify the Damascene Isml b. Abd Allh b. al-Muhjir. Cf. Ibn Ab l-
Muhjir, Fihrist, p. 29, l 18 (Schwally); [also Sezgin, GAS, vol. 1, p. 300?]
33
He is the last tabi on Ab Ubayd al-Qsimb. Sallms list, and he is also mentioned in
the biographies of Ab l-Dard as one of his pupils. He early fell into oblivion. The name is
uncertain: instead of also is to be found; instead also .
34
Also Ibn mirs nisba points to Homs or its vicinity (cf. Samn, s.n.).
Kfa: Bara:
132 (Ab Isq Amr b. Abd Allh
al-Hamdn) AL-SAB
141/53 (Ab Sad) ABN IBN TAGHLIB
AL-RABA
35
148 (Ab Abd al-Ramn Muammad
b. Abd al-Ramn) IBN AB LAYL
(al-Anr), q of Kfa, renown
jurisconsult; private reading (Fihrist)
[EI
2
]
149 (Ab Umar) S IBN UMAR
al-Thaqaf,
36
grammarian; private
reading
148 (Ab Muammad Sulaymn b.
Mihrn) AL-AMASH (al-Asad
al-Khil), mawl; one of the Four
after the Ten
148/55 Ab Amr b. al-Al (al-Tamm
al-Mzin), born at Mecca, lived
mostly at Bara, died at Kfa,
grammarian; one of the Ten
156 (Ab Umra) AMZAH IBN ABB
(al-Taym) al-zayyt; mawl; one of
the Seven
156 (Ab Umar) S IBN UMAR
(al-Asad) AL-HAMADHN,
al-am; private reading (Fihrist).
37
77/after 86 Ab Bariyya ABD ALLH IBN QAYS (al-Kind al-Sakn
38
al-
im), ib Mudh b. Jabal, military leader under Muwiya; private ikhtiyr.
103/8 (Ab Abd Allh) KHLID IBN MADN (al-Kal al-im); private reading
(Fihrist).
39
Finally in the Yemen:
40
before 169
41
(Ab Abd Allh Muammad b. Abd al-Ramn) IBN AL-SUMAYFA
(al-Yamn), later at Bara; private reading (ikhtiyr).
35
Sezgin, GAS, vol. 1, p. 24, vol. 2, p. 131, etc.
36
37
38
This nisba has no punctuation in the consonantal text, particularly in the first letter ()
and is vowelled differently; al-Yazghim?
39
Under Sham.
40
Also the Meccan readers, ws b. Kaysn, A b. Ab Rabh, and Ibn Kathr have a
Yemenite background (see above).
41
As far as his span of life is concerned we only know that he was older than Nfi, under
whom he still studied. (Ibn al-Jazar, abaqt, 3106.)
Abd Allh IBN MIR AL-YAUBs two canonical transmitters (Ab
Amr Abd Allh b. Amad ) IBN DHAKWN (al-Qurash al-Fihr
42
al-
Dimashq, died 245/856) and (Ab l-Wald) HISHM (IBN AMMR al-
Sulam al-Dimashq, d. 245/859) belong only to the second next generation
after (Yay b. al-rith) al-Dhimr. From among the later representatives
of the Damascene reading the best known is the grammarian (Ab Abd
Allh HRN b. Ms) AL-AKHFASH (al-Dimashq, d. 291 or 2/904). Part
of Mesopotamia
43
belongs to the territory of Damascus, politically as well
as in matters of Koranic reading; for this reason the Koranic commenta-
tor, al-Naqqsh (d. 351/962) represents the reading of IBN MIR al-Yaub
(parallel to that of Ibn Kathr [al-Kinn]). At his time, however, its decline
had already set in. Al-Maqdis, writing in 375/985, and as a Syrian adhering
to it, encountered it only in Damascus proper, not in the rest of Syria.
44
In
Iraq
45
and Egypt,
46
where all Seven Readings were represented, the Dama-
scene reading was also being studied, although in Egypt less than the other
readings, so that his wish to read according to IBN MIR roused consider-
able interest among the muqrin because it was so unusual. Consequently,
when he was asked why he followed an odd reading (tajrd)
47
contrary to
the majority (mma), ABAL-AYYIB [Abd al-Munim] IBNGHALBNal-
alab (d. 389/998
48
) said to him: da hdhih al-qira fa-innah atqa.
49
It
was not long until this reading was replaced at Damascus by that of Ab
Amr b. al-Al.
50
Damascus did not become the incontestable centre of MuslimSyria until [iii/172]
the Umayyads established their residence there. Previous rival towns had
beenim (Homs) and al-Jbiya. Corresponding to the political importance
in the early period, im had its own tradition of Koranic variant read-
42
Thus Ibn al-Jazar (Nashr, vol. 1, p. 144, l 2; abaqt, 1720); al-Dhahab (abaqt, [Ms.,
Berlin, Or. fol. 3140,] 19
r
) and the rijl books, rather [AB AL-YAMN] al-Barn, [Sezgin,
GAS, vol. 1, p. 102] a nisba which again points to Homs (cf. Samn, s.n.) [Sezgin, GAS, vol. 1,
pp. 111112].
43
Ibn mirs reading at the time of Ibn Mujhid (d. 324/936) in al-Jazra: al-Dhahab
(abaqt, Ms. Berlin, Or. fol. 3140), 11
r
; Ibn al-Jazar (abaqt, vol. 1, p. 424, l 22and still at
the time of al-Maqdis: (writing in 375/985) Descriptio imperii Moselmici, ed. M.J. de Goeje,
1906, p. 142, l 13.
44
al-Maqdis, Descriptio imperii Moslemici, ed. M.J. de Goeje
2
, p. 180, l 9.
45
Ibid., p. 128, l 4.
46
Ibid., p. 202, l 18sqq.
47
Ibid., p. 144, l 10sqq.
48
Sezgin, GAS, vol. 1, p. 15, no. 20: Brockelmann mistook the author for his son, hir.
49
Ibid., p. 202, l 18.
50
See below, p. 521 n. 89. [Sezgin, GAS, vol. 1, p. 15, no. 20.]
ing
51
that can be traced back to Mudh b. Jabal.
52
In accordance with the
subsequent political development, this tradition joins the one of Damas-
cus: A pupil of Abd Allh b. Qays (al-Kind) al-Sakn is Yazd b. Quayb
(al-Sakn al-im al-Sham), who had a private ikhtiyr, his pupil, the q
of Damascus, (Ab Ibrhm) IMRNIBNUTHMN(al-Zubayd), author of
a qira shdhdha; he is the teacher of the muqri al-Sham from im, Ab
aywa (Shurayb. Yazdal-aram, d. 203/818), to whoma qirashdhdha
is also ascribed, namely very likely the qirat al-imiyyn, which he contin-
ued to teach. This line can be followed up to the third century through his
son aywa b. Shuray (al-im, d. 224/838).
After Damascus, less important was Mecca, whichpolitically andintellec- [iii/173]
tually had receded early into the background. Al-Maqdis finds Ibn Kathr
al-Kinn together with the rest of the Seven in Iraq
53
and in Egypt,
54
and
next to some other readings (very strange!) in Jibl;
55
but at this time, other
readings had begun to penetrate even Mecca.
56
Like Ibn mir al-Yaubs
reading also the one of the equally old Ibn Kathr al-Kinn can be doc-
umented only much later. Even if the isnd of the latter is weaker than
51
Upon the request of Yazd b. Ab Sufyn [EI
2
] Umar may send him to Syria, man
yuallimuhumal-Qurn wa-yufaqqihuhum, Umar sent AB AL-DARD al-Khazraj, Mudh
b. Jabal, and Ubda b. al-mit [EI
2
; Juynboll, Encyclopedia, s.v.] with the instruction to
go first to im; one of them was supposed to remain there, the second one continue to
Damascus, and the third one to Filasn. Ubda remained at ims, Ab l-Dard returned to
Damascus and Mudh to Filasn, where later also Ubda went (al-Muttaq al-Hind, Kanz
al-umml, vol. 1, no. 4773). This narrative presupposes the old division of Syria and Palestine
intojund. (Missing are al-Urdunn, andQinnasrn, whichwas not organizedas junduntil Yazd
I.) According to the role attributed to im, the narrative must date back a very long time.
52
The account mentioned above, p. 221, that the inhabitants of im traced back their
reading toAL-MIQDD(IbnAmr ) b. al-Aswad[EI
2
; cf. Juynboll, Encyclopedia, s.v.] does not
belongas one might thinkto the corpus of the transmissions about Uthmns recension
of the Koran. None of the old sources mentions al-Miqdd. We first meet him, and this
without patronymic, in Ibn al-Athr. Since al-Miqdd is not known either for any kind of
interest inthe Koranor any relationwithis, it is likely tobe a mere confusionwithMudh
(Ibn Jabal). Mudh is one of those who allegedly collected the Koran already during the
Prophets lifetime (see above, p. 217); he also has been at im, at least for some time Ibn
Sad (abaqt, vol. 2, part 2): Biographien der medinischen Kmpfer, p. 125, l 12sqq. = ibid.:
Biographien der Basrier von der dritten Klasse bis zum Ende, p. 115, l 3sqq.). Other than he, it
could only have been Ubda see preceding foot-note, and cf. above, p. 217); but apparently
the literature of the variant readings does not seem to know anything about him (he is also
missing from the abaqt of Ibn al-Jazar.)
53
al-Maqdis, Descriptio imperii Moslemici, ed. de Goeje
2
, p. 128, l 4.
54
Ibid., p. 202, l 18.
55
Ibid., p. 395, l 9.
56
Ibid., p. 97, l 3.
that of the Damascene, similarly also the two main canonical traditionists
(Ab l-asan Amad b. Muammad b. Abd Allh b. Ab Bazza) AL-BAZZ
(d. 250/864), and (Ab Amr Muammad b. Abd al-Ramn b. Muammad)
QUNBUL (d. 280/893) are from much later than their Imm. In the form
of this riwyt the variant reading is considerably more recent than the
one of Ibn Muayin (d. 123/740), who was not reckoned among the Seven
because he deviated too far from the text of the Koran,
57
and probably also
because he was considered, very likely erroneously, younger than Ibn Kathr
al-Kinn.
58
There is hardly any evidence that the Meccan reading spread
beyond its place of origin, except to Baghdad where (Muammad b. Ms
Ab Bakr) AL-ZAYNAB
59
(d. 318/930), and (Ab Muammad Umar b. Abd
al-amad) IBNBUNN
60
(d. 374/984) represent IbnKathr al-Kinn. Meccas
relation to South Arabia, on the other hand, originates from readers with
SouthArabianbackground, including IbnKathr al-Kinn himself,
61
andthe
reverse influence of Ibn al-Sumayfas (d. before 169/785) Yemenite reading
tradition was attributed to Mecca, and was from there transmitted to other
places.
62
Contrary to the Meccan reading, the Medinan counterpart asserted itself [iii/174]
and has survived to this very day. The conquered Occident followed Med-
inan fiqh and qira, which it has strictly and exclusively preserved until
today. The Medinan reading reached Egypt by way of (Uthmn b. Sad al-
Qib) WARSH(d. 197/812), a pupil of Nfi b. Abd al-Ramn al-Layth, who
had displaced the two older readings of Ab Jafar Yazd IBN AL-QAQ
63
and Shayba b. Ni.
64
The former reading is still preserved among the Ten,
but not the latter one, although both of them were equally respected in the
57
Ibn al-Jazar, abaqt, 3118.
58
Cf. above, p. 511 n. 25.
59
60
Ibid., 2430.
61
P. 454 n. 35.
62
63
Regarding Nfis connection with it, see above, p. 489. It might not be without impor-
tance that wherever Nfi is positioned against Ab Jafar IBN AL-QQ he is frequently in
agreement with Ibn mir: thus, for example, sra 2:119, wa-ttakhadh for wa-takhidh; 2:160,
tar (thus also an occasional transmission from Ab Jafar) for yar; 2:172, wa-lkin al-birr
for wa-lkinna birr. Nfi and Ibn mir in this instance have all the rest of the Seven or even
the Ten against themselves. Not considered are the agreements which followfromthe Medi-
nan and Damascene consonantal text (cf. above, p. 399sq.). Acertain influence of Damascus
upon Medina in the Umayyad period is quite logical, particularly as it is concurrent with
Nfis period of teaching.
64
Sezgin, GAS, vol. 9. p. 203.
earlier period.
65
Both scholars are contemporaries and their readings are so
close to one another that one of them suffices as a model.
66
Ab Jafar IBNAL-QAQs reading does not seemto have ever been used [iii/175]
outside of Medina,
67
whereas the trail of Nfis reading leads to Iraq. Already
AbIsq(ISMLIBNJAFARb. Ab Kathr) AL-ANR
68
(died180/796)
the next best known of Nfis two main canonical transmitterswent to
Baghdad; also those two transmissions have been brought there, the one of
Ab Ms s b. Wirdn QLN
69
(d. 220/835) by Ab Jafar Muammad
b. Hrn (known as Ab Nash,
70
d. 258/871) and the one of Warsh by AB
BAKR Muammad b. Abd al-Ram b. Shabb AL-IBAHN
71
(d. 296/908).
The transmitted qira could not gain a footing there. Al-Maqdis found it
prevalent only in the ijz,
72
in Egypt
73
and in the Maghreb.
74
Ab Muammad AL-GHZ IBNQAYS (d. 199/814
75
), an immediate pupil
of Nfi, brought the latters qira together with Mlik b. Anas Muwa
76
to
al-Andalus. That Nfis best known transmitter, Warsh, carried the qira
to his homeland Egypt, where his Medinan pupil, AB YAQB Ysuf b.
65
Thus the commentator to Mliks Muwata, Ab Bakr IBN AL-ARAB (d. 543/1148) in
Ibn al-Jazar, al-Nashr, vol. 1, p. 37, l 15 (more precisely al-Suy, al-Itqn, vol. 1, p. 82, l 23),
and also Ibn Taymiyya (d. 728/1328 [EI
2
]) in al-Nashr, vol. 1, p. 39, l 11, which refers to Amad
IBNANBAL; Makk b. Ab lib in al-Ibna (Ms. Berlin, 578, p. 509) considers both readings
ghayr matrka.
66
Ibn Jinn in his Mutasab, whenever Shayba (Ibn Ni) is mentioned, it is always
together with Ab Jafar IBN AL-QQ, with the sole exception of vol. 1, p. 28, l 28 (G. Berg-
strer, Nichtkoranische Koranlesearten imMutasab, see its index).
67
The transmitters of his variant reading in al-Nashr are either Medinans or professional
scholars and collectors of qirt, who transmit several readings side by side.
68
Ibn al-Jazar, abaqt, no. 758. However, he also studied under Shayba (Ibn Ni), and
the two main transmitters of Ab Jafar IBN AL-QQ, possibly even under him personally.
[Sezgin, GAS, vol. 1, pp. 9495.]
69
70
Ibn al-Jazar, al-Nashr, vol. 1, p. 102, l 6, and p. 112, l 22; abaqt, 3504.
71
Ibn al-Jazar, al-Nashr, vol. 1, p. 110, l 19, and p. 113, l 21, abaqt, 3129. [Ab Yaqb Ysuf
b. Amr b. Yasr, d. ca. 240/854; Sezgin, GAS, vol. 1, p. 11, no. 5, l 12] (AB YAQB) AL-AZRAQ,
the secondespecially Egyptiantransmitter after Warsh. See also al-Suy, usnal-muara
f akhbr Mir, vol. 1, p. 207, l 24. However, this arq is also found in Petzls edition of Taysr,
p. 11, l 2, and still in al-Nashr, vol. 1, p. 111, l 6 and 10; but is no longer a private arq. Al-Azraq
became known in Baghdad only after 700.
72
Descriptio imperii Moslemici, ed. de Goeje
2
, p. 39, l 11.
73
Ibid., p. 202, l 18.
74
Ibid., p. 238, l 15.
75
Sezgin, GAS, vol. 2, p. 661, l 3.
76
Ibn al-Jazar, abaqt, 2534. This would seem to contradict the statement in abaqt,
3314: huwa lladh qadima bi-qirat Nfi al tilka l-bild (i.e. Qayrawn) fa-innahu kna
l-ghlib al qiratihimarf amza wa-lamyakun yaqra li-Nfi ill khaw al-ns.
Amr [Ibn Yasr] AL-AZRAQ (d. ca. 240/854) continued to teach it,
77
and
from where it conquered the Occident, is one of the most decisive events
in the history of Koranic readings and initiated a schism that still per-
sists, the Occident following Nfi b. Abd al-Ramn al-Layth in its reading
whereas the Orient does that of IM b. al-Ajjj al-JADAR. This division
appears rather early. When Amad IBN ANBAL (d. 241/855) was asked
which qira he likes best he replies:
78
qirat ahl al-Madna fa-in lam yakun
fa-qirat im, and Makk b. Ab lib (d. 437/1045)
79
says: qirat hdhayn
al-immayn (Nfi and im) awthaq al-qirt wa-aauh sanadan wa-
afauh f la-arabiyya wa-yatlhum f l-faa khatan qirat Ab Amr,
which, like other similar statements, serves to justify a fait accompli.
im al-Jadars variant reading is the one Iraqi reading that survived [iii/176]
as a representative of this most important group of readings. Since Iraq
is the scholarly-religious centre of early Islam also in other respects, this
applies also to its Koranic reading and the rising science of the Koran as
well as the production of manuscripts of the Koran.
80
The Iraqi domination
is evident whenit is rememberedthat four of the SevenReadings are located
there (at Bara: Ab Amr b. al-Al; at Kfa: im al-Jadar, amza b.
abb al-Taym, and al-Kis), from the Three after the Seven, two more
(at Bara: YAQB (Ibn Isq b. Zayd) AL-ARAM, at Kfa: Khalaf b.
Hishm al-Bazzr), and three from the last Four of the Fourteen (at Bara:
Ab Muammad al-Yazd, al-asan al-Bar, at Kfa: Sulaymn b. Mihrn
AL-AMASH); thus from the Fourteen readings no less than nine.
The Baran Koranic reading follows Mecca; this is evident from its char- [iii/177]
acter ever since al-asan al-Bar,
81
and is reflected in Ab Amr b. al-Al as
well as in his isnds, where the Meccan authorities predominate. The oldest
77
According to abaqt, 3518, Muammad IBN WA [Sezgin, GAS, vol. 1, pp. 474
475] introducedthe riwyaof Warshtoal-Andalus according tothe traditionof Abdal-amad
[b. Abd al-Ramn AL-UTAQ d. 231/845.] This Abd al-amad, a pupil of Mlik b. Anas
(abaqt, 1660), seems to have been soon replaced by the arq of ABYQB Ysuf b. Amr
AL-AZRAQd. ca. 240/854 [Sezgin, GAS, vol. 1, p. 11, l 15]. In any case, Ab l-Fal Muammad b.
Jafar AL-KHUZ, d. 408/1017 (abaqt, 2893) says that in Egypt and the Maghreb it was read
exclusively according to ABYAQB (= AL-AZRAQ) and Warsh. See also usn al-mudara
(Cairo, 1327), vol. 1, p. 207, l 24. This arq, however, is also found in al-Taysr (ed. O. Pretzl,
p. 11, l 2) and also in al-Nashr, vol. 1, p. 111, l 6 and 10; but it is no longer a private arq.
78
Ibn al-Jazar, abaqt, vol. 1, p. 348, l 11, and vol. 2, p. 332, l 1; al-Dhahab, abaqt
al-qurr, in the unfinished print of the Constantinople periodical, al-Hidya, 4 (1331), p. 653,
l 18, and p. 709, l 24.
79
Ibna (Ahlwardt, Verzeichnis, Berlin Ms. no. 578), p. 509.
80
Cf on this already p. 449.
81
G. Bergstrer, Die Koranlesung des Hasan von Bara, p. 56.
representative of this group is at the same time the oldest of the Fourteen,
al-asan al-Bar. But already at a very early period he no longer was of
individual importance
82
and was replaced by Ab Amr b. al-Al, who, on
the other hand, was at times of great importance and apparently dominated
the Orient. According toMakk
83
b. Ab lib(died437/1045), his reading was
practically in use at Bara about 200/815. The isnds are no indication of an
early dispersion, except that the early start of the exchange Bara-Kfa and
an encroachment of both upon Baghdad is ascertained.
Al-Maqdis finds Ab Amr b. al-Als reading encroaching upon the Ye-
men,
84
Syria (except Damascus),
85
Jibl,
86
and even Egypt.
87
About 500/1106
this reading had spread also to Damascus.
88
At the time of Ibn al-Jazar
(811/1408) it dominated the entire Koranic teaching of Syria, the ijz, the
YemenandEgypt, certainly the farsh (private variant readings), eventhough
many mistakes were made in the rules of pronunciation (ul).
89
Naturally,
the great iddighm of Ab Amr b. al-Al posed great difficulties, although
not present in all the forms of the reading. Probably not much later the
reading was surpassedby af b. Sulaymn, the maintransmitter of imal-
Jadar. In any case, Burhn al-Dn AL-ALAB, d. 956/1549,
90
whose expo-
sition of the anafite rite in the Ottoman Empire became authoritative, still
recommends in prayer Ab Amr along with af. b. Sulaymn.
91
In remote
regions it has survived until the present, namely in the Sudan.
92
82
Ibid., pp. 46 and 50.
83
In al-Suy, al-Itqn, vol. 1, p. 82, l 30sqq.
84
al-Maqdis, Descriptio imperii Moslemici, ed. M.J. de Goeje, p. 97, l 3.
85
Ibid., p. 180, l 9.
86
Ibid., p. 395, l 9.
87
Ibid., p. 202, l 18.
88
Ibn al-Jazar, abaqt, no. 1285 (vol. 1, p. 292, l 6) and no. 1790 (vol. 1, p. 424, l 24.) The first
man to introduce the reading of Ab Amr b. al-Al to Damascus was Ab l-Barakt Amad
b. ws, d. 492/1098 (abaqt, vol. 1, p. 425, l 1 and no. 327) or Subay b. al-Muslim b. Qir,
d. 508/1114 (abaqt, 1319).
89
abaqt, no. 1285 (vol. 1, p. 292, l 3).Somewhat earlier, [Muammad b. Ysuf] Ibn
[i.e. Ab] ayyn al-Jayyn [i.e. Ab ayyn al-Gharn al-Andalus] (654/1256745/1345,
Brockelmann, GAL, vol. 2, p. 109) mentions that the reading had spread to Syria and Egypt
(Ibn al-Jazar, al-Nashr, vol. 1, p. 41, l 13). [Sezgin, GAS, vol. 9., pp. 24, 26, 62, 63, 219.]
90
Burhn al-Dn Ibrhm b. Muammad AL-ALAB; Brockelmann, GAL, vol. 2, p. 432;
EI
2
.
91
Ghunyat al-mutawall, commentary on Munyat al-muall [Brockelmann, GAL, vol. 2,
p. 109], bb 6, fal 2, at the end.
92
From a letter by Prof. Arthur Jeffery, Cairo, to the author.
Ab Amr b. al-Als reading was temporarily confronted within Baras [iii/178]
sphere of influence with the rivalry of im al-Jadar (d. 128/745), and
particularly with that of Ab Muammad YAQB b. Isq AL-ARAM
(died 205/821),
93
who had also studied the Kfan variant readings.
94
Already
inhis lifetime his reading was equivalent to that of Ab Amr b. al-Al.
95
The
grammarian Ab Uthmn Bakr b. Muammad AL-MZIN (d. 249/862)
had a vision that the Prophet recommended him to follow the reading of
Yaqb (al-aram).
96
In the second half of the fourth century, the imm of
the Mosque of Bara recited exclusively according to Yaqb al-arams
reading,
97
and this is how al-Maqdis found it in his time.
98
Musfir b. al-
ayyib al-Bar (d. 443/1051) carries his reading to Baghdad: kna baran
bi-qirat Yaqb hfian la-h.
99
He narrowly missed being received into the
Seven because Ibn Mujhid turned against him in the end and opted for al-
Kis.
100
Yet in spite of this he holds a special place among the Three after
the Seven since because of him, qirt books on the Eight appear not infre-
quently, and he frequently receives a special treatment in the mufradt as a
supplement tothe Seven.
101
Among the Four after the Tenthere is yet another
Baran(ABMUAMMADYay b. Mubrak) AL-YAZD (d. 202/817), who
at the same time is the main transmitter of Ab Amr b. al-Althe two
canonical transmitters of Ab Amr (Ibn al-Al) being traced to himwho,
however, departs fromhimonly in details, and did not gain an independent
status.
Corresponding to the Alid sympathies of Kfa, the Kfan reading fol- [iii/179]
lows, on the one hand, Al, but, on the other hand, Ibn Masd, who had
been an administrator at Kfa, and whose codex of the Koran had been
93
Ibn al-Jazar, abaqt, 1498 and 3891. [Sezgin, GAS, vol. 1, pp. 1112.]
94
Traces of it in his Koranic reading; wherever he deviates from Ab Amr b. al-Al, he
partly follows the Kfans. Thus in sra 2:77, asanan, with amza b. abb and al-Kis
instead usnan; verse 153, yaawwa, with the same, instead taawwaa; verse 178, muwain,
with the same, as well as the one transmission from im al-Jadar, Ab Bakr SHUBAH
IBN AYYSH [Sezgin, GAS, vol. 1, p. 10, no. 4] instead min. Some other cases exist only in a
transmission of Yaqb al-aram.
95
al-Suy, al-Itqn, vol. 1, p. 83, l 2.
96
Ibn al-Jazar, abaqt, 3891 (vol. 2, p. 388, l 18).
97
Ibn al-Jazar, Nashr, vol. 1, p. 43, l 1, and abaqt, 3891, often; cf. also Ibn Khallikn,
no. 705.
98
al-Maqdis, Descriptio imperii Moslemici, ed. M.J. de Goeje, p. 128, l 4.
99
abaqt, 3589.
100
Makk b. Ab lib, Ibna (Ms. Berlin, 578) p. 496, and fromthis al-Suy, al-Itqn, vol. 1,
p. 83, l 15. Cf also Ibn al-Jazar, Munjid al-muqrin (Cairo, 1350), p. 75, l 17.
101
See below, p. 561.
in use for a long time.
102
Both appear next to others as the authorities
of Ab Abd al-Ramn AL-SULAM,
103
who was the best known of the
older Kfan readers. An early indication of the importance of Kfa as a
Koranic centre is the report that ala b. Muarrifs
104
(d. 112/730) reading
was accepted at al-Rayy, where one of his pupils was teaching it.
105
A stu-
dent of, and the successor to, the imma f l-qira, the aforementioned Ab
Abd al-Ramn AL-SULAM, is the earliest Kfan of the Seven, im al-
Jadar.
106
Also ims second teacher, Zirr b. ubaysh,
107
had been a stu-
dent of Ibn Masd and Al. The later theory is attempting to ascribe the
two origins of im al-Jadars reading to his two canonical transmitters,
Ab Bakr SHBAH IBN AYYSH
108
(d. 193/808) and Ab Umar AF (Ibn
Sulaymn,
109
d. 180/796), andthis way at the same time explaintheir conspic-
uously great differences,
110
namely im al-Jadar allegedly transmitted
Als reading to Ab Bakr SHUBAH IBN AYYSH, but Ibn Masds read-
ing to af (Ibn Sulaymn). Incidentally, also amza b. abb al-Taym
111
is
brought in connection with Ibn Masd: His teacher (Ab amza) UM-
RN IBN AYAN (al-Shaybn) kna yaqrau qirat Ibn Masd wa-l yukh-
lif muaf Uthmnyatabir urf mani AbdAllhwa-lyakhruj minmuw-
faqat muaf Uthmn, and this was allegedly amzas ikhtiyr.
112
By the
same token, amza b. abb is also considered the champion of the
qirat Al: His teacher, Muammad b. Abd al-Ramn IBN AB LAYL
113
(d. 148/765) kna yujawwid arf Al.
114
In the struggle of the two transmitters af (Ibn Sulaymn) prevailed [iii/180]
against im al-Jadar. That in the rivalry of the Kfan readings among
each another and, beyond this, in the struggle against other readings, the
riwyat af an im was successful, must be attributed to its neutral-
ity and its nearly total congruence with the prevalent pronunciation of
102
See above, pp. 446sq., and 456.
103
See above, pp. 496 and 499 n. 189.
104
Juynboll, Encyclopedia, p. 623, col. 2624, sqq.
105
Ibn al-Jazar, abaqt, 1488; Juynboll, Encyclopedia, s.q.
106
al-Dhahab, abaqt, p. 653, l 7.
107
Juynboll, Encyclopedia, p. 58sqq.
108
EI
2
; Sezgin, GAS, vol. 1, p 1011.
109
EI
2
110
Ibn al-Jazar, abaqt, no. 1496 (vol. 1, p. 348, l 7).
111
EI
2
112
113
EI
2
; Juynboll, Encyclopedia, p. 538, col. 2; Sezgin, GAS, vol. 1, p. 518.
114
classical Arabic. The final conquest of the Orient, which was accompanied
by the spread of the anafite madhhab, seems to have taken place only
with the rise of the Ottomans. The somewhat later reading of (Sulaymn
b. Mihrn) AL-AMASH (d. 148/765), also a pupil of im al-Jadar, was
the last one to be accepted among the Fourteen, but was of hardly any
great importance, althoughplacing ABHIRAbdal-Wid(IbnUmar b.
Muammad) b. Ab Hshim AL-BAZZR
115
(d. 349/960) over IBN MIR al-
Yaub.
116
On the other hand, the two even later ones, amza (Ibn abb,
d. 156/773) and (Al b. amza) AL-KIS (d. 189/804) attained greater repu-
tation, even if far more controversial. Ibn Mujhid added both to the Five in
order to complete the Seven.
117
Al-Kis belongs here only on account of the
origin of his reading. He actually lived and taught at Baghdad and is there-
fore occasionally separated from the Kfans as being min ahl al-Irq.
118
The
fact that the reading of amza (Ibn abb) could establish itself and even
be considered meritorious is perhaps also partly based on its peculiar fea-
ture, namely taqq, i.e., the exaggerated explicit pronunciation with its far-
fetched accuracy, and its slavish scriptural conformity. We thus find amza
b. abb about 200/815 represented next to im al-Jadar at Kfa. At
Baghdad, he is particularly mentioned for the earlier period;
119
al-Maqdis
finds him together with im, al-Kis, and Ab Amr b. al-Al in Iraq;
120
in
Syria al-Kis is represented among other readers;
121
the reading of amza
is transferred to Nibn by AB AL-FAL Jafar b. Muammad AL-NAB
(d. 307/919); according to a report mentioned above on p. 462 n. 76 he is
alleged to have dominated the Occident as well for some time. On the other
hand, until the end of the seventh century the reading of al-Kis was com-
mon in Transoxiana and Ifahn
122
for the teaching of the Koran (talqn) as
well as in prayer. But at an early period opposition against the reading of
amza b. abb also asserted itself. Abd Allh b. Idrs
123
hears recitation like
amza with ifr min al-madd wa-l-hamz wa-ghayr dhlika min al-takalluf,
115
116
117
Cf. the quotation fromABAL-FAL Abd al-Ramn b. Amad AL-RZ (d. 454/1062)
[Brockelmann, GAL, suppl. vol. 1, p. 721; Sezgin, GAS, vol. 1, p. 674, no. 27] in Ibn al-Jazars
Munjid al-muqrin, p. 75, l 13.
118
Makk b. Ab lib, Ibna (Ms. Berlin, Ahlwardt, Verzeichnis, no. 578), p. 509.
119
al-Maqdis, Descriptio imperii Moslemici, ed. M.J. de Goeje
2
, p. 128, l 4.
120
Ibid., p. 39, l 11.
121
Ibid., p. 180, l 9.
122
123
Juynboll, Encyclopedia, p. 7; fl. 115/733192/808, Kfa.
and therefore rejects this reading (K R H [ ]
124
). Al-Kis, when on pilgrim-
age, recites in prayer sra 4:10, saying ifan with imla like amza; he is
beatenupandtrampled, andthe CaliphHrnal-Rashdreproaches him. As
a consequence, he abandons much of amzas reading.
125
In fact, al-Kiss
reading compared with that of amza is not very independent, particularly
inthe case of farshal-urf (the single passages); it is inmany respects noth-
ing but a watered down recension of that one,
126
even if in the ul (imla!)
al-Kis is sometimes more extravagant. Abtimal-Sijistn, whenestab-
lishing his ikhtiyr,
127
disregards bothreadings,
128
whichis tosay that bothare
lacking the third criterion, namely that ijtim al-umma alayh is wanting. In
general, however, he holds a low opinion of the Kfans. Among them it is
al-Kis who is the most competent in the Koran and Arabiyya, although
he, too, became known only through his connection with the court.
129
The
opposite opinion, however, is also an old one. Probably the first person who
explicitly sided with the future victorious reading of af an im is the
famous critic of tradition, Yay b. Man [Mun] (died in 233/847).
130
Yet another Kfan is the last among the Three after the Seven (Ab
Muammad) KHALAF
131
(IBN HISHM AL-BAZZR, d. 229/844); he is the
Kfan counterpart to the Baran (Ab Muammad) al-Yazd, and his read-
ing is in reverse ratio to that of amza b. abb, comparable to that of Ab
Muammadal-Yazd tothat of AbAmr b. al-Al.
132
Anyhow, it was of some
importance: Muammad b. Isq b. Ibrhm al-Marwaz (d. after 280/893)
reads only according to him.
133
The younger representatives of local readings that we mentioned belong [iii/182]
already to a time when that formof the science of variant readings appeared
that is characterized by the existence of several parallel readings. As an
intermediate stage between the local readings and the systematic Koranic
reading we can identify two contrasting endeavours: first, the endeavour
124
Ibn al-Jazar, abaqt, 1190 (vol. 1, p. 263, l 13); in justification it is claimed that amza
himself rejected such exaggerations. A similar account with reference to Amad b. anbal
in the same source.
125
Yqt, Irshd, vol. 5, p. 186, l 19.
126
See above, p. 486 n. 108, and p. 490.
127
128
Makk b. Ab lib, Ibna (Ms., Berlin, Ahlwardt, Verzeichnis, no. 578), p. 496.
129
Yqt, Irshd, vol. 5, p. 193, l 20.
130
al-Dhahab, abaqt [sic] (Ms. Berlin 9943), p. 90 (missing in print 724, 221!) [sic];
Juynboll, Encyclopedia, s.v.; Sezgin, GAS, vol. 1, pp. 106107.
131
EI
2
; Sezgin, GAS, vol 1, p. 12, no. 9.
132
133
to establish a compromise between the different readings in the ikhtiyr,
mainly represented by Ab Ubayd al-Qsim b. Sallm and Ab tim al-
Sijistn in the first half of the third century; and
134
second, a stronger influ-
ence of the principle of tradition
135
on the Koranic reading which, analogi-
cally to adth and fiqh, led to the collection of many variant readings.
A clear expression of this type of Koranic work and teaching we wit-
nessed also in the literature of shawdhdh; the earliest sources known to
us
136
originate, of course, from the period after Ibn Mujhid, but still did
not participate in the revolution of the instruction that he initiated. Their
characteristics are: FirstNext to a great number of sporadically appearing
authorities a greater number of frequently mentioned teachers of the Koran
can be found; among them particularly such men who are later to be found
among the Sevenuncanonical readings of the Fourteen. SecondThe bases
of pronunciation are completely wanting; also the pure differences of pro-
nunciationrecede intothe backgroundcomparedtoactual variant readings.
ThirdVery many individual cases are treated without consideration for
parallel passages. FourthThe variants are limited neither to the principles
of the canonical reading, nor are they in accordance with the Arabiyya, or
the consonantal text.
Against this historical background the revolution of the science of the [iii/183]
Koran that Ibn Mujhid initiated can be illustrated. First of all, it affected
the local readings and the authorities in power in the third century by limit-
ing the arbitrarily established number of imms to seven. It also meant a not
insignificant impoverishment withinthe individual readings of these Seven.
Secondly, Ibn Mujhid eliminated in his book the greater part of the occa-
sionally transmitted variants of the Seven, and then included only such of
them which to him seemed to be particularly well documented, and which
were in agreement with both the rules of Arabic grammar and the con-
sonantal text. His decisive innovation was the creation of the principle of
complete variant readings, which obliged him by way of qiys in each indi-
vidual case to ascribe to each of the Seven one particular readingeven
when the transmission was wantingand, by way of ul, also systematize
the pronunciation of each one of them. Under him, the individual read-
ing ikhtiyr was still in full force. This was the rub of the inconsistency
vis--vis the principle of tradition that had led to complete readings, and
134
See above, pp. 484486 and 491 sq.
135
P. 424.
136
G. Bergstrer, Nichtkanonische Koranlesearten im Mutasab des Ibn inn (1933), and
Ibn Khlawayh, Mukhtaar f l-shawdhdh, ed. by G. Bergstrer.
which would have required complete consideration of all existing chains
of transmission from the reader down. Already a century later, in the writ-
ings of Makk b. Ab lib (d. 437/1045) and al-Dn (d. 444/1052) this was
more rigidly applied in the canonical form so that from every reader two
transmitterspartly direct pupils, partly fromlater generationslead with
an uninterrupted isnd down to the reader. As a consequence, the num-
ber of the differences treated by Ibn Mujhid and subsequent men dif-
fer somewhat; much was newly included, some things even eliminated.
The roots of the system of dual transmitters had been established already
in the time of Ibn Mujhid in so far as the two transmitters of im al-
Jadar, Ab Bakr SHUBAH [IBN AYYSH
137
] and (Ab Umar) AF IBN
SULAYMN,
138
frequently appear next to one another. This was probably the
starting point of the later rule which al-Draqun (d. 385/995) allegedly ini-
tiated.
139
A consistent development was achieved by applying the system of
complete readings also to single transmissions, the final offshoot of which
was the prohibition of talfq,
140
i.e., mixing different strands of transmis-
sion.
There is unanimity that Ibn Mujhid was the first to limit himself to the [iii/184]
Seven.
141
The fact that particularly this number was chosen relates to the
tradition of the Seven aruf
142
and the alleged sevenfold recension of the
Koran produced for the Caliph Uthmn.
143
The systemof the Seven was pre-
ceded by a selection of five readers that contained one reading from each
mir;
144
there was possibly also a system of six.
145
It cannot be ascertained
whether or not these systems which antedate Ibn Mujhid already consti-
tute complete readings or are merely collections of single readings. Neither
137
EI
2
138
EI
2
139
See below, p. 551 sq.; [Sezgin, GAS, vol. 1, pp. 206209].
140
Bergstrer, Koranlesung in Kairo, p. 29.
141
Makk b. Ab lib, Ibna (Ms. Berlin, Ahlwardt, Verzeichnis, no. 578), p. 509; Ab
Shma, Ibrz al-man (commentary on the Shibiyya, printed Cairo, 1349/1930), p. 4, l 7;
142
See above, p. 38sqq. [sic].
143
See above, p. 306.Makk b. Ab lib, Ibna, and Ab Shmah, loc. cit.; Ab Shmah
says: ikhtra Ibn Mujhid fa-man badahu hdh l-adada muwfaqatan li-qawlihi inna
hdh l-Qurn unzila al saba taruf.
144
Is considered by Ab l-Fal al-Rz a bida preceding the selection of Seven in Ibn
al-Jazar, Munjid al-muqrin, p. 74, 19. See below, p. 547.
145
Such a system seems to be assumed by the statement of Ab l-Abbs Amad b. Abd
Allh AL-TANFIS (d. 2nd half of the third century) according to which the reading of
Ab Amr is asan al-qirt, Ibn Kathr al, im afa al-qirt, Ibn mir aghrab al-qirt,
amza al-athar, al-Kis araf al-qirt, and Nfi al-sunna (Ibn al-Jazar, abaqt, 337).
the limitation to seven nor its sequence asserted itself immediately. Ibn al-
Jazar collects statements of scholars opposing the limitation to seven
146
as
well as the names of Koranic scholars who went beyond this number.
147
As
regards literature dealing with more than the Seven, see below, p. 561 sqq.!
The sequence of the imms
148
as established by Ibn Mujhid,
149
and re- [iii/185]
tained by al-Dn in his Taysr, was changed in other qirt works for a
variety of reasons. The Meccan (Ab Mabad) IBN KATHR (al-Kinn) is
particularly favoured by the Iraqis over the Medinan Nfi.
150
In accordance
with Ab Amr b. al-Als reputation in the early days he frequently is head-
ing all the others;
151
it is rare to find him at the end.
152
In al-Ahwz it is
Ibn mir who stands at the top,
153
particularly with the Iraqis he precedes
Ab Amr b. al-Al al-Tamm al-Mzin.
154
Conspicuous is the separation
of the Barans and Kfans by the inserted Damascene in the regular order
146
Most outspokenis the statement of Abl-Abbs AMADIBNAMMRAL-MAHDAW
(d. 440/1048 [Sezgin, GAS, vol. 9, pp. 215216]) and AB AL-FAL Abd al-Ramn b. Amad
AL-RZ (d. 454/1062) which Ibn al-Jazar lists in the final chapter of his Munjid al-muqrin,
p. 70sqq.; excerpts of the same statements, with general remarks about the question of the
Seven as well as numbers going beyond, al-Nashr, p. 1, l 36sqq.
147
Munjid al-muqrin, pp. 2946.
148
More details on the subject can be found in the anonymous biographical work on the
Seven, Asin al-akhbr f masin al-saba al-akhyr (handwritten, my personal property);
from it the details, unless not particularly documented elsewhere.
149
jj Khalfa under Kitb al-saba.
150
Thus already AB AL-HASAN AL-RZ Al b. Jafar al-Sad al-Shrz (d. ca. 410/1019)
according to Ab Abd Allh Nar b. Al b. Muammad IBN AB MARYAM al-Fris [fl.
557/1162565/1170; Sezgin, GAS, vol. 9, pp. 105 and 249] in his Kitb al-Mi, cf. Pretzl,
Verzeichnis der Qirt-Werke, no. 19, pp. 3234; also al-Naysbr (d. after 768 [i.e.
708]) in his commentary on the Koran, in the margin of al-abars commentary (Cairo,
1321/1903), vol. 1, p. 8sqq.; further, Ibn al-Fam [Brockelmann, GAL, suppl. vol. 1, p. 722;
EI
2
] in his Tajrd, Pretzl, Verzeichnis , no. 15, p. 30; and in (Muammad b. al-usayn b.
Bundr) AL-QALNISs (d. 521/1127) Kifya (Brockelmann, GAL, suppl. vol. 1, p. 723); Pretzls
Verzeichnis , no. 27, pp. 3940.
151
For example, Ishra of Manr b. Amad al-Irq, Pretzl, Verzeichnis der Qirt-
Werke, no. 21, p. 35; Brockelmann, GAL, suppl. vol. 1, p. 721; and also al-Naysbr, loc. cit., in
both cases the sequence: Ab Amr, Ibn Kathr, Nfi.
152
In the K. al-Wajz of al-Ahwz (d. 446/1054), apparently also in his larger work, Iqn
(Ms. Damascus, hiryah, 54), also al-Kifya of al-Qalnis, see above.
153
This is allegedly the case in the larger work of al-Ahwz, but it can no longer be
ascertained from the fragment, Ms. Damascus, hiriyya, 54.
154
Thus Ibn Mujhids younger contemporary, Ibn Mihrn (d. 381/991; Sezgin, GAS, vol. 1,
p. 15) in his Ghya f l-qirt al-ashr (abridged in Ibn al-Jazar, Nashr, vol. 1, p. 38, l 1); further,
Muammad b. Jafar AL-KHUZ (d. 438/1017) in his al-Muntah f l-qirt al-ashr (Ms.
Cairo, Taymr Pasha, Tafsr, 434): in the Rawat al-uff of Ab Isml Ms AL-MUADDIL
(Brockelmann, GAL, suppl. vol. 1, p. 727; Pretzl, Verzeichnis, no. 31, pp. 4344) in the Tajrd
of Ibn al-Fam (died 516/1122), Pretzl, Verzeichnis, no. 15, p. 30, also in al-Naysbr in the
commentary on the Koran (see above), vol. 1, p. 8, etc.
(of the Taysr). amza b. abb, and the more famous al-Kis, occasion-
ally change places. Among the transmitters of the Seven the order is even
more fluctuating. As the second and third choice one later added three
more to the Seven, and then once more four, thus creating systems of Ten
and Fourteen. Particularly frequent was also the system of Eight by includ-
ing Yaqb al-aram. Here, selection and order are still less established.
Frequently others are inserted among the Seven, regardless of geographic
affiliation.
155
More details are presented below when discussing the subject
of literature, and in Otto Pretzls Verzeichnis (der handschriftlich erhalte-
nen lteren Qirt-Werke) [i.e.: inventory of early writings on qirt extant
in manuscripts],
156
numbers 17, 21, 23, 34, 27, 30, 31.
The Systems of the Seven, the Ten, and the Fourteen Readings
The system in its classical form of more or less recognized (canonical) [iii/186]
readings, each represented by two transmissions, is as follows:
I. The Seven.
1. Nfi d. 169, Medina (see above, p. 514) 144 arq.
157
(a) Qln d. 220 (see above, p. 519) 83 arq.
(b) Warsh d. 197 (see above, p. 518sqq.) 61 arq. The most impor-
tant tangles of his transmission are:
Warsh

al-Azraq d. 240 Abd al-amad d. 231 a number of pupils
(see above, p. 520) (abaqt 1660) and children of pupils
(see above p. 520 n. 77) of Warsh
al-Ibahn d. 296
(see above, p. 519)
155
Thus, [Ysuf] Ibn Mihrn, al-Qalnis, al-Muaddil.
156
Part 1, pp. 1447, of his article Wissenschaft der Koranlesung (ilm al-qira), ihre
literarischen Quellen und ihre Aussprachegrundlagen (ul) [i.e. the science of the readings
of the Koran, their sources in literature as well as their principles of pronunciation] in
Islamica 6 (19331934).
157
According to Ibn al-Jazars own enumeration, al-Nashr f l-qirt al-ashr, vol. 1, p. 98
sqq., the number of paths are given, by whichthe respective reading or transmissionreaches
him.
2. Ibn Kathr d. 120 (?) Mecca (see above p. 512) 73 arq.
(a) Qunbul d. 291 (see above, p. 518) 32 arq.
(b) al-Bazz d. 250 (see above, p. 518) 41 arq.
159
3. Ab Amr d. 148/55, Bara (see above, p. 515) 154 arq.
(a) Ab Umar AL-DR d. 246 [EQ]; 126 arq.
160
(b) Ab Shuayb al-Ss d. 261; 28 arq.
161
4. Ibn mir d. 118, Damascus (see above, p. 514) 130 arq.
(a) Ibn Dhakwn
162
d. 242; 79 arq.
(b) Hishm [Ibn Ammr] al-Sulam d. 245; 51 arq.
163
5. im [al-Jadar] d. 127/9, Kfa (see above, p. 513) 128 arq.
(a) Ab Bakr SHUBAH IBN AYYSH d. 193 (see above, p. 523) 76
arq.
(b) af d. 180 (see above, p. 523) 52 arq. Important additional
tangles of his transmission are:
af

Ubayd b. al-abb al-Nahshal Amr b. al-abb al-Baghdd [iii/188]
d. 235 d. 221

Ab l-Abbs Amad b. Sahl Ab Jafar Amad Ab l-asan
AL-USHNN d. 307/919 b. Muammad Zurn al-
b. umayd Daqqq, died

Al-Fm al-Fl ca. 290
Ab l-asan Al AB HIR died 289
b. Muammad Abd al-Wid
b. li al- b. Umar [AL-
Hshim d. 368 BAZZR] d. 34
(above, p. 524)
158
The order varies; Qunbul infirst position, e.g., inal-Taysr of al-Dn, insecondposition,
e.g., in al-Nashr of Ibn al-Jazar.Ibn Kathr and his transmitters are generations apart, one
of which is represented by only one transmitter, Ab Isq Isml b. Abd Allh b. Qusann
al-Qus d. 170/786 (or 190/805).
159
Ibid.
160
Between Ab Amr b. al-Al and these two there is Ab Muammad al-Yazd, d. 202/
817 (see above, p. 524).
161
Ibid.
162
His name is Abd Allh b. Amad not Abd al-Ramn as it appeared in al-Nashr, vol. 1,
p. 144, l 2. There (vol. 1, p. 145, l 5) also the date of death is erroneously given as 202/817.
163
The order varies; Ibn Dhakwn in first position, e.g., in al-Taysr of al-Dn, in second
position, e.g., in al-Nashr.Ibn mir al-Yaub and his transmitters are two generations
apart; cf. above, p. 516.
6. amza [Ibn abb] d. 156, Kfa (see above, p. 515) 121 arq
(a) Khalaf [Ibn Hishm al-Bazzr] d. 229 (see above, p. 525) 53
arq
(b) Khalld (Ab s al-Shaybn) d. 220 [EQ];
164
68 arq.
7. al-Kis d. 189, Kfa (see above, p. 524) 64 arq.
(a) Ab l-rith (AL-LAYTHIBNKHLIDAL-BAGHDD) d. 240;
40 arq.
(b) al-Dr d. 245 = 3 a; 24 arq.
II. The Three after the Seven.
8. Ab Jafar IBN AL-QAQ d. 130, Medina (see above, p. 514) 52
arq.
(a) Abl-riths b. Wardn(al-adhdha) diedca. 160; 40 arq.
(b) Ab l-Rab (Sulaymn b. Muslim) IBN JAMMZ (al-Zuhr),
died after 170; 12 arq. [EQ]
9. Yaqb al-aram d. 205, Bara (see above, p. 522) 85 arq.
(a) Ruways (Muammad b. al-Mutawakkil) d. 238; 41 arq [EQ].
(b) Raw (Ibn Abd al-Mumin) [GAS, I] d. 234/6; 44 arq.
10. Khalaf b. Hishm al-Bazzr d. 229, Kfa = 6a; 31 arq.
(a) Isq al-Warrq d. 286; 22 arq.
(b) Idrs al-addd d. 292; 9 arq.
III. The Four after the Ten.
165
11. Ibn Muaysin d. 123 Mecca (see above, p. 512).
12. Ab Muammad al-Yazd d. 202,
166
Bara (see above, no. 3).
13. al-asan al-Bar
167
d. 110, Bara (see above, p. 513).
14. al-Amash d. 148, Kfa (see above, p. 515).
164
Anintermediate member betweenamza andhis transmitters is Abs SULAYMIBN
S al-anaf d. 188/9 or 200.
165
I [Pretzl] no longer list the transmitters.
166
In the work on Fifteen readers, Rawat al-uff, of al-Muaddil (Pretzls Verzeichnis,
no. 31, pp. 4344), AbMuammadal-Yazd andal-asanal-Bar are absent fromamong the
others, but included instead are umayd b. Qays AL-ARAJ (d. 130/747) [Juynboll, Encyclope-
dia, p. 575, col. 2], Meccanandteacher of AbAmr (see above, p. 514), further, Ibnal-Sumayfa
(Ibn al-Jazar, abaqt, 3106, without date of death, a pupil of Nfi and ws b. Kaysn), al-
Yamn and ala b. Muarrif (d. 112/730).Ab Muammad al-Yazd and al-asan al-Bar
are also missing in K. al-Jmi of Ab Mashar [Abd al-Karm] al-abar [d. 478/1085; Brock-
elmann, GAL, suppl. vol. 1, p. 722]; (Pretzl, Verzeichnis der Qirt-Werke, no. 32, p. 45)
which, apart from the Seven, contains a large selection of ikhtiyrt.
167
Ibid.
The hierarchy of the three groups corresponds to their sequence, each [iii/189]
successive group enjoying lower reputation. There is also a hierarchy within
each group, but this is not as apparent as is the arrangement by location.
From among the Seven, Nfi b. Abd al-Ramn al-Layth and im al-
Jadar are the most respected men; among the Three after the Seven, it is
Yaqb al-aram; and among the Four after the Ten, probably al-Amash.
The arq numbers added to the Ten and their transmitters permit all
sorts of conclusions with respect to meaning and distribution of the various
readings and transmission in the older period. Ab Amr b. al-Al has the
highest number; for a long time his reading was the most wide-spread of
all. He is followed by Nfi al-Layth. im al-Jadar is preceded by Ibn
mir al-Yaub. im al-Jadars variant reading spread too late to have
any effect on the arq figures, whereas Ibn mir al-Yaub is pushed back
only later by objections to his reading
168
for matters related to the science of
tradition. Ibn Kathr al-Kinn and al-Kis have the lowest figures among
the Seven; Yaqb al-aram surpasses both of them. At the end follows
Khalaf b. Hishm, but with considerable distance from the preceding Ab
Jafar Yazd IBN AL-QAQ al-Makhzm. If you add up how many paths
accrue to the individual amr, the outstanding importance of Kfa will
become evident, even though it is not reflected in the individual figures.
Bara and Medina follow closely behind, followed at a good distance off by
Damascus, and lastly by Mecca. As far as the transmissions are concerned,
Ab Umar AL-DR an Ab Amr (Ibn al-Al) alone has nearly as many
paths as dothe twotransmitters of imal-Jadar takentogether. Warsh, as
well as af b. Sulaymn,
169
who later became the authoritative transmitters
of Nfi b. Abd al-Ramn al-Layth and simal-Jadar, remain behind the
two other transmissions of the variant reading. Their day was yet to come.
The Characteristics of the Canonical
Variant Readings and Their Distinctions
The following discussion is limited to the canonical Seven. Their reciprocal [iii/190]
relation as derived from among a larger sample among the individual vari-
ant readings is roughly as follows:
170
there are two distinct groups, Kfa on
the one hand, and the rest of the amr on the other. In this case, Medina,
168
Cf. above, p. 509 n. 21.
169
EI
2
: Sezgin, GAS, vol. 1, p. 10, no. 3.
170
Cf. G. Bergstrer, Koranlesung des Hasan von Bara, p. 55sqq.
Mecca, and Bara are closer to one another than Damascus is to the three
of them. And among those three, Medina and Mecca constitute a close unit.
It is most typical that Bara has by no means a close connection with Kfa.
Among the Kfan variant readings the one of imal-Jadar is still the one
with the closest connections with the non-Kfan group. As we have been
able tosee, the readings of amza b. abbandal-Kis are extremely Kfan
and closely related to one another. Among the differences within any single
reading, the most important within the reading of imal-Jadar are by far
those of Ab Bakr SHUBA (IBN AYYSH) and af b. Sulaymn. These dif-
ferences are equally considerable as those of Ibn Kathr al-Kinn and Nfi
b. Abd al-Ramn al-Layth or al-Kis and amza b. abb. Perhaps half
as great is the difference among the Damascene variant readings. In each of
the remaining readings, the differences between the two transmissions are
negligible.
The most basic rules of pronunciation are recognized by all the Seven [iii/191]
readings. They are subject to tajwd, namely a propaedeutic to the Koranic
variant readings proper. The great teachers of the Koran dealt with them
in separate writings;
171
occasionally they precede the works on the variant
readings as an introduction.
172
These works also contain instructions for the
presentation of the material, apart fromgeneralities of prerequisites for the
teachers and students of the Koran and an introduction to the terminol-
ogy of variant readings. Among the numerous ways of presentation referred
to, tartl, the oldest also appears in the Koran (sras 25:34, and 73:4). Origi-
nally this was likely no more than a term for recitation in general, but early
became an expression for clear recitation, and particularly slow psalmody
allowing time for meditation. In literature it is equated with the Koranic
mukth
173
(sra 17:107), which, as far as I know, was not adopted as a technical
term. adr, the presentation in normal cadence of Koranic recitation, very
early attained equality with tartl, evidently with the rise of the idea that
Koranic recitation is meritorious, but also the logical endeavour to recite a
171
On this subject see O. Pretzl, Die Wissenschaft der Koranlesung, p. 10, and further
under literature.
172
Thus in al-Mi f wujh al-qira of Nar b. Al (IBN AB MARYAM) al-Fris [fl.
557/1162565/1170] (Pretzl, Verzeichnis der Qirt-Werke, no. 19, pp. 3234) in the Iqn of
(Ab Jafar Amad b. Al al-Gharn) IBN AL-BDASH [d. 540/1145] (Pretzls Verzeichnis,
no. 11, pp. 2728), inal-Nashr of Ibnal-Jazar, etc. Brockelmann, GAL, suppl. vol. 1, p. 723, no. 7a;
Sezgin, GAS, vol. 9, p. 105, no. 19, and p. 249, l 2.
173
Cf. the quotation fromal-Mi f wujh al-qira of Ab Abd Allh Nasr b. Al al-Fris
in O. Pretzls Wissenschaft der Koranlesung, p. 33, bottom!
great deal (istikthr al-qira).
174
The traditions and statements that are used
to explain tartl offer an insight into the struggle for the recognition of adr.
There is a reminder of this still in al-Khqn (died 325/937),
175
who explic-
itly states (in[qada], verse 13) that adr is permitted(murakhkha) besides
tartl, justifying his decision with the intention inherent in Islam to lighten
religious obligations. Already in al-Khqn a new expression appears with
verse 27, namely taqq, replacing tartl as the prevalent formin the science
of the variant readings;
176
while in the system of al-Nashr, it recedes some-
what into the background. The expression, taqq, as supported by the very
meaning of the word, indicates recitation under most careful observation of
every detail of pronunciation. It is Ibn al-Jazar who defines the relationship
of taqq to tartl which is evidently already applicable in al-Khqn, and
which at the same time explains the disappearance of tartl fromthe termi-
nology of the science of the variant readings. In fact, Ibn al-Jazar interprets
taqq in such a way that it serves as exercise and instruction, whereas tartl
is meant for meditation, and that every taqq serves at the same time as
tartl, but not vice versa.
177
As time went on, the Koran became recited more rapidly. But rather [iii/192]
than devising a new technical term for the greater speed, the designation
adr, once agreed upon, was retained; and an intermediate stage between
adr and taqq was introduced. It was named tadwr in al-Nashr, and
in the respective passage of al-Mi
178
it appeared as tajwd. However,
this attempt to integrate tajwd into the system of recitation failed. The
expression is first found in an explanation of tartl ascribed to Al: Tartl is
good pronunciation (tajwd) of the letters and observation of the pauses.
A statement ascribed to Ibn Masd reads: jawwid al-Qurn, pronounce
the Koran well.
179
This referred no more to a different way of recitation
than was originally the case with tartl. But whereas the definition of tartl
became more restricted, in the case of tajwd it was retained, becoming the
synonymof linguistic purity (faa)
180
when reciting the Koran, which had
already been required by al-Khqn (verse 21). The inclusion of tajwd in
methods of recitation did not last. In al-Nashr, both tajwd and tartl are
attached only on the face of it to a three-level system, leading over to the
174
Ibid.
175
See below, p. 568; Sezgin, GAS, vol. 1, pp. 1415.
176
Cf al-Taysr (ed. Pretzl), p. 31, l 2; Itf of al-Dimy al-Bann, p. 26, l 3.
177
al-Nashr, vol. 1, p. 209, l 7; al-Suy, al-Itqn, vol. 1, p. 102, l 7.
178
Anonymous! See below, p. 568.
179
al-Itqn, vol. 1, p. 109, l 11.
180
Cf. the comments on faa in al-Mi, fol. 7
v
which are intimately related.
institution of rules of pronunciation, thus serving in reality as name of the
subject, not a way of recitation. Still, al-Mi, as well as al-Iqn of Ibn
al-Bdhash,
181
knows of yet another way of recitation, thus completing a
five-level system, namely tam, which is characterized by a particular way
of drawing out the prolongations by simultaneously exhaling (JRY [ ]) in
a way that can only be acquired by training in person (this is the essence of
the name).
In the works on variant readings several attempts were made to cate- [iii/193]
gorize the readers of the Koran, mainly the Seven, within this system. But
in this case it mostly concerns a construction that begins with the variant
readers attitude to certain peculiarities of extreme prolongation of vowels
before hamza. There are only two points at which the transmitted reports
are so uniform and precise that it inspires confidence in attributing adr to
Ibn Kathr, but, most of all, taqq to amza b. abb: Ab Amr b. al-Al
is mentioned in the case of adr as well as (ishtiqq al-)taqq. This con-
forms with the transmitted reports that claim that he had different ways of
readingwith and without the far-reaching assimilations that are typical
for himand thus permitting one to comprehend in single instances
182
the
historical relation between the rapidity of speech and phonetics as reflected
in the descriptions of the postulated levels of the system.
From among five additional ways of recitation with only limited applica-
bility, tarqq (thinning out) is an offshoot of tajwd or taqq. The remain-
ing four methods indicate different types of musical-dramatic recitation:
tard, recitation with a trembling voice, tazn, sorrowful emotion, tarb,
enrapture, namely an offshoot of tam, reciting in a singsong voice (taran-
num, tanaghghum), and finally taln, psalmody. The controversy regarding
these methods of recitation follows basically the two traditions: Embellish
the Koran with your voices, and Recite the Koran with melodies (lun)
and voices of the true Arabs, both of which purport to support musical
recitation but are interpreted differently by their opponents, who could
advance unequivocal contradictory statements of older authorities. Today,
actual singing, whichal-Mistill knewtobe controversial outside Koranic
instruction, has been recognized for a long time.
183
181
Sezgin, GAS, vol. 9, p. 218, Ibn al-Bdhish al-Gharn, d. 528/1133.
182
In al-Nashr, and elsewhere, for adr performing the permissible abridgements, reduc-
tions and vocalizations, possibly under assimilation of the affected consonants, alleviation of
hamz, avoidance of pause; for taqq the opposite, distinct division of nearby letters through
lingering, and emphasizing nasalization and gemination.
183
G. Bergstrer, Koranlesung in Kairo, p. 110sqq.
In the writings on tajwd the old pronunciation of the consonants as basi- [iii/194]
cally found in Sbawayh
184
is presupposed: as velar q, as palatal j (not
quite certain), , one-sided (mostly left) articulated lateral, emphatic spi-
rant, , non-rolling r of the tongue, , voiced emphatic dental, , voiced
emphatic dental spirant. In practice the modern classic value prevails; only
to a still applies the pronunciation as spirant, not sibilant, and has
become a voiced emphatic dental spirant so that both sounds amalga-
mate.
185
In the case of and there are two varieties, velar (mufakhkham)
and palatal (muraqqaq). The border-line is controversial. In general, velar
applies to Allh when preceded by u or a (not in -i Allh) and palatal
when close to i. A peculiarity of the pronunciation of the vowels is the
prolongation (madd) of a long vowel in a closed syllable (in particular
before gemination) and before hamz (see below). When consonants meet,
the basic rule is ihr, namely both consonants remain unchanged and
completely separate, which in practice means that particularly voiced con-
sonants and at the end of a syllable nearly constitute a separate sylla-
ble.
186
Exceptions are the assimilations lr,
187
td, t, dhth, and dh, but par-
ticularly the one with t, which is important because of the afformative t-:
t and t, the latter, however, iddighman ghayra mustakmalin bal tabq
maahu ifat al-ibq wa-l-istil,
188
retaining the emphatic-velar-like char-
acteristics, namely fusion without assimilation. Special conditions control
m and n. In the case of m, ikhf becomes effective before b; ikhf further
occurs in n (also tanwn) before all consonants except larynx-related conso-
nants as well as ghandkh. The termikhf indicates the decrease of duration,
disappearance of n or m with the formation of the closure at the place of
articulation of the following sound with ghunna (nasalization). Completely
assimilated is n to l and r (without ghunna), partially (so that ghunna is
retained) n is assimilated to m or n, and also to w and y (only in Sandhi,
thus, not in the case of duny inwn).
189
Further, n is changed to m (qalb)
before b.
184
A. Schaade, Sbawaihis Lautlehre (1911); cf. also Meir M. Bravmann, Materialien und
Untersuchungen zu den phonetischen Lehren der Araber (1934).
185
G. Bergstrer, Koranlesung in Kairo, part two, p. 133.
186
Ibid.
187
Occurs only in Sandhi; Sandhi or medially applies only in case of direct interference,
like qul rabbi/ (sra 18:21 and 28:85). Otherwise separation by sakt.
188
189
This is the official rule which, however, does not apply incontestably to all transmis-
sions.
Fundamental differences of pronunciation as far as they effect the whole [iii/195]
of the Koran are the subject of ul in the writings on the variant readings.
They comprise several large sections, including the following seven:
(1) The iddighm kabr: According to the common view of teachers of the
Koranthis refers to AbAmr b. al-Als idiosyncrasy inrapidreading (adr)
to completely assimilate and contract (iddighm a) vocalized conso-
nants inSandhi as well as medially following identical or similar consonants.
Since this idiosyncrasy contradicts the general rules of tajwd, and occurs
only in one particular manner of recitation, and, most importantly, it is not
consistently documented in all the transmissions of this imm, it is totally
ignored in a large number of writings on the variant readings. There are seri-
ous reservations against the usual view, raised quite early, that this is a case
of idghma.
190
(2) Differences inthe pronunciationof hamz take upmuchspace inthe ul. [iii/196]
In the science of qirt the question does not regard qualitative difference
of hamz existing in the Arabic dialects, rather it is only whether or not it is
pronounced at all (taqq), or as an intermediate sound (bayna bayna), i.e.,
facilitating a direct transition from one vowel to another, or being replaced
(ibdl) by a arf al-madd, a letter of prolongation. As far as the alleviation of
hamz is concerned, in cases when two hamzs meet, medially or in Sandhi,
the two ijzs, Ibn Kathr, and Nfi, venture furthest; among the transmit-
ters of the latter it is particularly Warsh as well as Ab Amr b. al-Al and
HISHM IBN AMMR al-Sulam.
The initial hamz after a vowelless consonant is totally omitted and its
vowel is pronounced together with the preceding vowelless consonant, i.e.,
in the reading of Warsh, who also in other cases largely eases vocalized
and unvocalized hamz alike, a method that is followed only by Ab Amr
b. al-Al with the greater cadence of his reading (vowelless hamz is in this
case compensated by prolongation of the vowel). One peculiarity of amza
b. abbalso in other cases a champion of taqq al-hamzis that in
pause after a word with hamz he pronounced neither medial, initial nor
final hamz, wherein he is followed only by Hishm (Ibn Ammr al-Sulam)
by alleviating the final hamz. The effect of an initial hamz upon preceding
vowelless consonants produces a sakt in the reading of amza (Ibn abb),
190
See Pretzl, Die Wissenschaft der Koranlesung, p. 293sqq.Ibid., also details to the
following.
namely a prolongation of the consonant and a brief pause after the same
respectively. According to the less likely transmission, this sakt occurs in all
consonants, but according to another transmission (in the only extant copy
of) al-ujja (f l-qirt al-sab) of Ab Al al-Fris (288/901377/987)
191
only
in the case of the l of the article. In the middle of a word the same peculiarity
is transmitted about the pronunciation of shayun. This phenomenon the
author of al-ujja appropriately places parallel to the following one, which
leads us to the field of the pronunciation of the vowels among the readers
of the Koran.
(3) Hamz produces prolongation of the preceding long vowel in the middle [iii/197]
of a word in all readings. Initial hamz in pronunciation in context has
the identical effect on preceding long vowels for Warsh, amza b. abb,
al-Kis and Ibn mir al-Yaub, however notor at least disputedfor
Ibn Kathr, Qunbul (an Nfi), Ab Amr b. al-Al and IM b. al-Ajjj
AL-JADAR. The measure of prolongation is in both cases different in the
individual readers. For the practical reading of the Koran the measure is
fixed by a most complicated system that establishes the unit of the normal
prolongation as one alif.
192
(4) The field of the pronunciation of vowels is dominated by the discus-
sion of the deflection of the pronunciation (imla) of the letter shaded
towards . As in the case of hamz, also in this instance only the fact of the
deflection in its variations is established, without determining its strength.
The following distinctions are made: FirstThe pure pronunciation (fat,
ikhl al-fat, al-fat al-mutawassi). SecondDeflection of towards
[with a horizontal stroke above the (or , Wright, A Grammar)] (imla
maa or shadda, ba, ij).
193
ThirdAnintermediate stage betweenthe
191
Sezgin, GAS, vol. 9, pp. 101110. Includes a complete list of manuscripts; Twentieth
century editions, Cairo, 1965, and Damascus, 1984.
192
As a general principle in qirt literature madd means prolongation, while normal
lengthening as in qla, and mlik is called qar. Except in the above-mentioned case where
prolongation is occasioned by hamz, there is unanimity that it also occurs when a long vowel
is followed by a vowelless simple consonant, e.g., mayy, or a double consonant. The same
applies to urf al-hij at the beginning of sras. These cases, however, are not included in
most of the books on variant readings because they are not controversial.
193
Alsonamedkasr, but it must not be concludedthat the deflectionbecomes a pure i. The
most extreme contrast (outside the Seven) is encounteredinthe expressionqaraabi-l-amm,
indicating the pronunciation from to , a sound that al-Dn calls fat shadd in his Tajwd.
Cf. to this Ab ayyns commentary, al-Bar al-mu, vol. 6, p. 172, l 11.
first sounds mentioned(baynabayna, imlaqallaor mutawassia, alsotaqll
or talf). According to the two Kfans, amza b. abb and al-Kis, imla
occurs in the case of final long written with y []; conversely, the pro-
nunciation of imla of this final in Ab Amr b. al-Al is conditioned
by the immediately preceding or by the rhyme. The reading of Nfi is
very controversial. It is certain that Warsh, his transmitter, pronounced the
rhyming ends with small imla (bayna bayna); on the other hand, many
authors make the pronunciation bayna bayna of other words (as in Ab
Amr) dependent on a preceding . In the case of medial imla occurs
on two occasions: FirstIn the ten verbs, ja, sha, zda, rna, khfa,
ba, khba, qa, qa, and zgha. These are hollow verbs of the basic
form faila. amza b. abb pronounces them with full imla; Nfi, bayna
bayna; whereas al-Kis and Shuba b. Ayysh an im al-Jadar, only say
bal rna; Ibn Dhakwn only says ja and sha with imla. Second in
the penultimate syllable followed by -ri is read with imla by Ab Amr b.
al-Al in the transmission of al-Dr (in the transmission of Ab l-rith
AL-LAYTH IBN KHLID al-Baghdd (d. 240/854) only when yet another
precedes ), by Warsh, however, bayna bayna, likewise amza b. abb,
but only when preceded by (the only case where the pronunciation bayna
bayna occurs).
Apart fromthese basic rules inthe ul andfarshal-urf still a number of
individual cases are treated. The imla of the short a is discussed only when
related to the tarqq of in Warsh (see above), and in the pausal feminine
ending -ah which, according to older sources, is pronounced uncondition-
ally with imla by al-Kis, according to other sources only if not preceded
by urf mustaliya ( , , , , , , ) or laryngeal consonants.
A deflection from toward is to be found in al-Kis in words like qla, [iii/198]
gha, sa, sat, la, sqa, i.e., hollow verbs of the form fuila.
(5) Finally, differences of the pronunciation of vowels are usually still found
in the suffixes of the 2. and 3. plural masc. as well as in the 3. sing. masc.
suffix. The assimilation of -humand -hu after preceding i or y is consistently
applied. But al-Kis always reads alayhum, ilayhum, ladayhum; further,
amza b. abb and al-Kis do not assimilate when -hum is followed by
hamzat al-wal, but pronounce in this case -humu, whereas Ab Amr b.
al-Al says -himi and all the others, -himu. Ibn Kathr al-Kinn pronounces
the -m of the plural suffixes unconditionally with a long , Warsh does it
only when followed by hamzat al-qa. In the same way, Ibn Kathr al-Kinn
lengthens the suffix of the 3. sing. -h, if preceded by either , or sukn,
and -h, when preceded by .
In farsh al-urf a partial (ikhtils, so mostly Qln) or complete reduc-
tion of the vowel of these suffixes is frequently documented; this is the
case in yuaddih, nutih, nuwallih, nulih; further, sra 6:90, fa-bi-hudhumu
iqtadih (or iqtadi) instead of iqtadihi; sra 20:77, yatihi, yatih; sra 27:28,
fa-alqihi, fa-alqih ilayhim; sra 39:9, yarahu, yarah; sra 99:7 and 8, yarah;
sras 7:108 and 26:35 next to eachother are to be found arjih, arjih, arjih,
arjihi, arjih, and arjih.
(6) Very early differences of contextual and pausal pronunciation found [iii/199]
entry into the ul of the Koranic variant readings: (a) The most general
rule, namely that short final vowels are not pronounced in pause, finds
an exception in the peculiarity of the Baran and Kfan readers who hint
at the final vowel either by its barely audible remainder (only in the case
of and ) or by merely exhibiting the position of the lips. In the Kfan
terminology the former is called rawm, and the latter, ishmm; in the Baran
terminology it is vice versa. (b) Ab l-asan AL-BAZZ (an Ibn Kathr)
pronounces the m of question, when connected with prepositions mah
(with h al-sakt; other cases of h al-sakt are graphically established).
194
(c)
Most of the details discussed are occasioned by the peculiarities of the
older Koranic orthography, and partly of only theoretical importance for
the reading of the Koran. Here belong primarily those passages already
discussed above, on p. 358sqq., where words are written in contextual form
instead of pausal form, or where the combination and division of individual
particles is irregular, particularly the omission of the final (rarer ). Amore
or less slavish imitation of the peculiarities of the Koranic orthography also
led to different pronunciations of these words in pausal form.
The deviations of the Seven readings as far as they are dealt with in farsh [iii/200]
al-urf of the writings on the variant readings go beyond phonetics in the
narrowsense, i.e., variants of functionally uniformsounds; apart froma few
exceptions they remain within the scope of differences limited to vocaliza-
tion (including tashdd). This concerns primarily the not infrequent differ-
ent use of cases and moods; the difference of identically written endings of
the perfect tense, -tu, -ta, -at. Alarger group consists of the differences in the
use of verbal radicals, withand without difference of meaning andconstruc-
tion: active voice and passive voice change in Ist Formof DD, RF, QY; Ist
194
Sra 69, verses 19, 20, 25, 26, 28, 29 kitbiyah, isbiyah, mliyah, sulniyah, sra 101:10,
m hiyah. Additionally sra 2:261, yatasannah, and 6:90, iqtadih, besides -h. See also above, l
18.
Form and IInd Form of the verbs BShR, RF, FT, QTL, QDR, KDhB, MYZ;
Ist active with IInd passive of NZL; Ist Form with IVth Form (in so far as
the consonants are not different) of TY, RY, LL, FQH, LD, NR; Ist Form
active with IVth Formpassive of BLGh, RJ, GhShY, MDD, MSK, NJW, NZL; Ist
Form with Vth Form (dialectic forms with assimilation of t) of DhKR, HR,
LQF; IVth Formwith Vth Form(identical) of L; Ist Formwith VIIIth Form
in TB, but also (dialectically, in sras 10:36, and 36:49) KH HDY.
Yet another group represents the differences of the characteristic vowel
in Ist Form perfect and imperfect, e.g., a:u of MKTh; imperfect i:u, MTh,
RSh, KF, ZB; i:a SB, QN; further, yair:yaurru; ir:ur; asaytum:astum.
More remote ukhf (imperf.): ukhfiya (perf. pass.). Also in the nominal roots
considerable differences appear. Particularly frequent is the change from
monosyllablic to disyllabled roots: jubl:jubul and jibill, further, ma(a)z,
kis(a)f, nashr:nushur; may(yi)t, ay(yi)q, etc. (Numerous additional exam-
ples listed among the Kfan peculiarities.) Vowel change in araj:arij,
na:nu, shiw:shuw, maskan:maskin, mansak:mansik. Some IVth
Form infinitives interchange with afl plurals, for example, idbr:adbr,
isrr:asrr, mn:aymn. The ful plural of nouns with y as the second
root consonant becomes in many readers fil: buyt:biyt, ghuyb:ghiyb,
khuyb:khiyb, shuykh:shiykh, uyn:iyn. Now and then verb and noun
interchange: khalaqah:khalqih. Particularly frequently we find differences
in the case of identically written particles: la:li, la-ma:li-ma:lamma, anna:
inna, an:in, anna:an, lkinna:lkin, aw:a-wa.
There is much leeway for variant readings because of the inaccurate writ- [iii/201]
ing of . Where no alif appears, it might still be read, but also vice versa,
where there is an alif, it might be considered secondary or a variant, and
may still be read . This way an additional number of verbal roots might be
confused: Ist Form and IIIrd Form of the verbs KhD, DRS, DF, QTL, WD;
IInd Form with IIIrd Form, BD, F, QD, FRQ; Vth Form with VIth Form,
D, HR; IVth Formperfect with Infinitive in aama:imun. The same way
also jaan:jn, further, qul:qla,
195
takhaf:takhfu.
196
Also in the case of
the nominal roots does this defective writing show an effect: For example,
mahd:mihd; silm:salm; khalf:khilf; khi:khaa:khi; sir:sir:sar;
ghashwa:ghishwa; salam:slim; asin:sin; lim:allm; likewise in several
Plural forms: sakr:sakr:sukr. Further, the frequent confusion of Sin-
gular and Plural must be attributed to this, in the Sound (Fem.) Plural, for
195
196
example, rislati : rislt as well as in the Broken Plural, for example,
abd:ibd; am:im; r:riy; ruhn:rihn; ir:r; jidr:judur; sirgh:
surugh; kitb:kutub; khshi:khushsh; kfir:kuffr; maskan:maskin; and
masjid:masjid.
Also other orthographic inaccuracies
197
permit variant readings. Since the
ending of Fem. Singular can be written with -t,
198
it might be confused with
the Sound Plural: ghaybat:ghaybt, kalimat:kalimt, etc.; also in alt,
which is written with w, and can therefore be read alawt. Similarly, the
writing of GhDWH produces the variants ghadh and ghudwah.
199
Since
in the dhawt al-y the is written with y,
can be interpreted as
yughshkum as well as yaghshkum. Since the final was frequently not
written, confusions like tasalanna:tasalni (= -n), also yarta:yartai (= -)
are possible. The absence of word divisions produce, for example, anra
Allhi:anran lil-Allhi, idh adbara:idh dabbara. The imperfections of the
orthography of hamz permit additional leeway, for example, in na: na;
ayr: ir; ayf : if ; rauf : raf ; kabr: kabir; shirk: shurak ; khaa:
khat : khay; durryun: durrun (dirrun) ; bs: bis: bayas: bas also
yasa:yas (because two ws one after another are not expressed in writ-
ing); layka:l-ayka; particularly in personal names: Jabrl, Jibrl, Jabril,
Jabrl; Mkl, Mkil, Mkl.
200
There is in addition still a very large num-
ber of instances where it is controversial whether or not the interrogative
particle, a, occurs before a word beginning with hamz.
The great licence afforded by the ambiguity of the Kfic script through [iii/202]
the absence of diacritic points is rarely made use of by the Seven readings.
This has already been discussed above on p. 434. Still rarer are deviations
fromthe consonantal text, namely the manuscripts without diacritic points
that have been discussed above, pp. 413414.
201
197
198
See above, p. 400.
199
200
201
Cf. also above, p. 395 n. 46, and the cases of inherited consonantal variants, pp. 347349.
Apparent consonantal deviations: sra 7:120, where Qunbul reads firawnu wa-mantum
instead firawnu amantum, but in wal only; for this reason the form of the writing based
on the waqf form remains unchanged. It is similar in some isolated instances of the great
iddighm in sra 27:36, where amza b. abb reads atumiddnn instead atumiddnan;
sra 46:16, ataidnn, Ibn Kathr, Nfi, Ab Amr b. al-Al, and Hishm b. Ammr instead
of ataidnin; sra 18:94, makkann: thus all except Ibn Kathr instead makkanan. In all such
instances it is always written separately and still pronounced together.
The characterization of each individual reading is made difficult in so
far as they are inconsistent in themselves. This can easily be accepted con-
sidering the fact that none of the readings is based on a uniform linguistic
or dialectic interpretation of the consonantal text, nor has its origin in a
uniform understanding. Rather it represents the result of a most compli-
cated balancing act that has already been discussed above. Even the kind
of instruction in the Koran, the original fixation of this instruction, and also
the applicationof the critical principles, whichhave already beendiscussed,
was bound to lead to the isolation of passages and then to treat similar
instances in different ways. In such a case, particularly the adherence to
the principle of tradition was an obstruction: Whoever received a Koranic
passage transmitted from an old authority in a particular form (based on a
sound isnd) couldand onspecial conditions had toinclude this partic-
ular passage inthis forminhis ownreading. This principle of traditionmade
it impossible to correct parallel passages. Naturally, the internal uniformity
of the individual readings ran counter to the above-mentioned standardiza-
tion, excluding any consideration for other readings.
In fact, we can identify inconsistencies among the readers themselves for [iii/203]
independent purposes. Al-Kis (d. 189/804) inflects thamd as accusative
triptote noun because it appears on several occasions in Koranic orthog-
raphy in the accusative with alif (sras 11:71, 25:40, 29:37, 53:52), but other-
wise as a diptote because it is a personal name; the only exception being
sra 11:71, where the genitive reads thamdin. (Yay b. Ziyd) AL-FARR
(d. 207/822) asks himfor the reason, andreceives the reply: qarubat f l-adhf
min al-mujr wa-qabun an yajtamia l-arf marratayn f mawiin thumma
yakhtalif fa-ajraytuh li-qurbih minhu. Because at this place a-l inna Tham-
dan kafar rabbahum a-l budan li-Thamd is preceded by the triptote
accusative required by orthography, al-Kis here, and only here, applies
also the genitive triptote. Here, for once, we obtain a direct insight into the
mechanics of the old readers of the Koran. The passage is from al-Farrs
extant commentary on the Koran.
202
The extent of these inconsistencies is unbelievable; still, it must not [iii/204]
be exaggerated. In the fundamentals of pronunciation certain main lines
are indeed established; there is nothing entirely contradictory; there are,
however, smaller deviations in isolated instances. The straightforward rules
of assimilation are completely observed, to some extent also the rules
202
Man l-Qurn, Ms. Bagdatl Vehbi Efendi, Istanbul, no. 66, photocopy, Berlin, Cod. sim.
or., no. 37, s.v.
governing hamza. Most numerous are conspicuous details of imla, but
also here there are numerous ul muarida, and likewise in individual
instances the expression aythu waqaa is quite often to be found.
More difficult than in the Seven is a general typology of subsequent read-
ings of the systems of the Ten and Fourteen, with the exception of the one
of Yaqb al-aram who, on account of his great importance, has a very
good tradition. The Kfan Khalaf b. Hishm (al-Bazzr) and the Baran
Ab Muammad al-Yazd, easily fit into the framework of their amr as
far as pronunciation is concerned, corresponding to their individual cases,
where they display few idiosyncrasies. Conversely, in the case of the older
scholarsal-asan al-Bar (d. 110/722), Ibn Muayin (d. 123/740), and Ab
Jafar (Yazd IBN AL-QAQ) as well as al-Amash (d. 148/765)the presen-
tationof the basics of their pronunciationis made superfluous by the simple
fact that in teaching they attached no great importance to the phonetic imi-
tation of their own reading. Nevertheless, the individual readings contain
a large number of purely phonetical details that presuppose considerably
greater difference in the rules of pronunciation. Already Bergstrer
203
drew
attentionto the fact that al-asanal-Bar was reckonedamong the groupof
readers who, as far as the rules of pronunciationare concerned, pronounced
hamz, while in farsh al-urf a far-reaching lightening of hamz is explic-
itly documented. As we know from writings on shawdhdh,
204
for him the
chapter imla ought to be re-written, and the imla of a to u included. The
framework of the ul, originally created for the Seven, was retained also
later on, and the newly added readers of the Fourteen included only super-
ficially, mostly by merely listing them under the rubric of miscellaneous,
the bqn-system.
203
Die Koranlesung des Hasan von Bara, p. 17.
204
The most highly esteemed of them (see below, p. 566) were still unknown to Bergstr-
er. For this reason his conjectures regarding the imla of al-asan von Bara on p. 351 (Seite
18 [sic]) are not quite pertinant; certainly quite accurate are his fundamental deliberations
(above, p. 421 sqq.).
LITERATURE OF THE VARIANT READINGS
1
The Earliest Period
Whereas inthe case of adth, as Goldziher has shown,
2
writtenrecords were [iii/205]
used quite early without hesitation, and it was not until a later period that
the habitor at least the fictionof oral transmission became established,
in the narrow case of the readings of the Koran, oral transmission seems
likely to have been the original mode of transmission, at least insofar as this
already presupposes an authentic consonantal text. The task consisted of
fixing the pronunciation of an extant consonantal text into a written form,
not only by reading signs in the text itself but also by explicit description
or stipulation (na), which can hardly be anything but a later develop-
ment.
The first written records of this kindwhich are neither public nor
textual but entirely private and thus, strictly speaking, not a literature of
variant readings but rather a precursor to itgo back to the middle of the
second century, to the time of the younger of the recognized readers and the
elder pupils of the eldest among them. The technical term of these records
is nuskha; the standard expression, la-hu (pupil) an-hu (teacher) nuskha
is synonymous with the less frequent kataba l-qira an
3
This does not
refer to a work authored by the respective pupil but to a kind of notebook.
To the earliest among the myriad examples listed in the Ghyat al-nihya f
abaqt al-qurr of Ibn al-Jazar belong the following: Sulaymn b. Mihrn
AL-AMASH (d. 148/765), abaqt, no. 874; amza b. abb (d. 156/773),
1
Sources: Ibn al-Nadm, al-Fihrist, the sections from Ibn Mujhid to al-Naqqsh (odd
references already previously) and al-kutub al-muallafa f l-qirt, in addition to the fol-
lowing: Ibn al-Jazar, whose survey of the literature of the variant readings in al-Nashr f
l-qirt al-ashr has already (see above, p. 505) been mentioned; also his Ghyat al-nihya
f abaqt al-qurr (printed, Cairo, 1933 and 1935), including the entire biographical litera-
ture (see above, p. 505sqq.); Pretzls Die Wissenschaft der Koranauslegung contains a list of
the extant manuscripts of the early qirt works down to the Shibiyya.
2
Muslim studies, vol. 2, p. 181 sq.; his The hirs (1971, repr. 2009), p. 90; F. Sezgin,
Goldziher and Hadith, in: I. Goldziher, Schools of Koranic commentators, pp. xxiixxiv.
3
For example, Ibnal-Jazar, al-Nashr, vol. 1, p. 180, l 12; similarly Qln[Sezgin, GAS, vol. 1,
p. 12, no. 8] says: qaratu al Nfi qirata-hu wa-katabtu-h f kitb (al-Dhahab, abaqt
[Berlin Ms., Ahlwardt, Verzeichnis, no. 9943], 62, missing in the printed edition [p. 732, l 2]);
Ibn al-Jazar, abaqt, no. 2509 (vol. 1, p. 615, l 17).
546 literature of the variant readings
no. 22; Nfi (d. 169/785), nos. 755, 1581,
4
1965, 2509, 3322; Ab Amr YAY
IBN AL-RITH al-Dhimr (d. 148 [i.e., 145/762]),
5
no. 1965; ISML IBN
JAFAR b. Ab Kathr
6
(d. 180/796), a pupil of Ab l-Rab Sulaymn IBN
JAMMZ al-Zuhr,
7
no. 1377; al-Kis (d. 189/804), nos. 3037, 3742, 3764; Ab
s SULAYM IBN S al-anaf (d. 188/803 or 200/815), no. 3122; Uthmn
b. Sad WARSH (d. 197/812), no. 1660; the above-mentioned (p. 472) Abd
al-amad b. Abd al-Ramn AL-UTAQ;
8
and AB MUAMMAD Yay b.
al-Mubrak AL-YAZD
9
(d. 202/817), nos. 1504 and 1929.
These records do not give us much of a clue, but we may certainly assume [iii/206]
that they containedonly brief details about the controversial passages of the
particular imm.
Since in earlier days the borders between private notes and the public
book were fluid, a number of writings can be attributed to the type that
is chronologically identical with the nusakh, and appear under the title
kitb al-qirt (possibly more appropriately al-qira) or a similar title.
10
We
are probably dealing with another stage of this development when Abd
al-amad b. Abd al-Ramn (AL-UTAQ), who died in 231/845 (see above,
p. 520 n. 77), treats both Nfi b. Abd al-Ramn al-Layth and amza (Ibn
abb) in a book entitled Ikhtilf Nfi wa-amza.
11
This corresponds pre-
cisely to the fact that [Ab Sad Abd al-Malik] AL-AMA, who died [in
213/828]
12
(abaqt, no. 1965), possesseda nuskha eachof AbAmr b. al-Al
and of Nfi,
13
and another younger man, such a nuskha of Ibn Dhakwn
4
Abd al-Ramn IBNAB AL-ZIND[EI
2
,] according to the rijl books; the year of death
here indicated as 164 ought to be changed to 174/790; (EI
2
).
5
6
Sezgin, ibid., vol. 1, pp. 9495.
7
See above, p. 530; EQ.
8
The Rislat Warsh, which Brockelmann (GAL, vol. 1, p. 189, note) mentions with reser-
vations, and which comes to mind in this connection, is in reality composed by the commen-
tator al-Mutawall himself (see Yusuf E. Sarkis, Dictionnaire encyclopdique de bibliographie
arabe, col. 1617, no. 3, and cf. G. Bergstrer, Die Koranlesung in Kairo, p. 28).
9
Sezgin, GAS, vol. 2, p. 610, vol. 9, pp. 6364.
10
Primarily Ibn al-Nadm, Fihrist, section al-kutub al-muallafa f l-qirt; among the
authors are Ab Amr b. al-Al and Khalaf b. Hishm al-Bazzr; in addition the work later
called kitb al-qirt (-a?) of Abn b. Taghrb al-Raba, d. 141/752 (see above, p. 515); [Goldz-
iher, Schools, p. 45, note 69] and the writings on readings of al-Kis, the authors of which,
as far as can be documented, were all his pupils. Also al-Kiss own Kitb al-qirt; see al-
Dhahab, abaqt, p. 718, l 13; Ibn al-Jazar, abaqt, no. 2212 (vol. 1, p. 539, l 18); Yqt, Irshd,
vol. 5, p. 200, l 4; G. Flgel, Die grammatischen Schulen der Araber, p. 125, no. 3.
11
Ibn al-Jazar, abaqt, no. 97 and 3518.
12
EI
2
; EQ; Sezgin, GAS, vol. 2, p. 613, vol. 3.pp. 364365, vol. 4, pp. 333334, vol. 7, pp. 344
345, vol. 8, pp. 7176, vol. 9, pp. 6667.
13
Ibn al-Jazar, abaqt, no. 1965.
literature of the variant readings 547
and Hishmb. Ammr al-Sulam.
14
More comprehensible collections of this
kind, namely the combination of complete readings, existing for practical
purposes either at the place of their origin or in their immediate zone of
influence, presuppose travel: if even the collection of adths, which are
much less locally confined, requires extensive travel,
15
so much more does
the collection of variant readings of the Koran. Among the first men known
to have done this are al-ulwn,
16
who died after 230/844, and Ab Jafar
MUAMMAD IBN SADN,
17
d. 231/845; among the canonical transmitters
it is af b. Umar AL-DR al-Baghdd,
18
d. 246/860. The pathof collections
of comprehensive readings here leads to Ab Jafar Amad IBN JUBAYR al-
Ank AL-MUQRI
19
(died258/871), whocollectedfive of the canonical read-
ers,
20
and to al-Djn (d. 324/935), who collected eight of them (the Seven
plus Ab Jafar Yazd IBN AL-QAQ al-Makhzm
21
), both being predeces-
sors of Ibn Mujhid, who collected seven. In the case of other collections,
from the previous period, which comprise a larger number of authorities, it
is doubtful to what extent they claim comprehensive readings. Ibn al-Jazar
mentions Ab Ubayd al-Qsim b. Sallm (d. 224/838)
22
as the first author
of a considerable collection. His book comprises twenty-five readers apart
from the Seven. At approximately the same time, Ab tim al-Sijistn
(d. 250/864) composed a large and a small work on qirt.
23
The book al-
Jmi by YAQB b. Isq b. Zayd AL-ARAM (d. 205/821) is clearly on
14
Ibid., 165.
15
I. Goldziher, Muslimstudies, vol. 2, pp. 42 and 164sqq.
16
17
See above, pp. 484sq., and 486 n. 108.
18
abaqt, no. 1159 (vol. 1, p. 255, l 12: raala f alab al-qirt; at the same place it is
also said of him: awwal man jamaa l-qirt). [EQ; Sezgin, GAS, vol. 1, p. 13, died 240/854 or,
according to others, 246/860].
19
20
Ibn al-Jazar, Nashr, vol. 1, p. 33, l 19; abaqt, no. 176; Makk b. Ab lib, Ibna (Berlin
Ms., Ahlwardt, Verzeichnis, no. 578), p. 509. Apparently one each of every mir! [Sezgin, GAS,
vol. 9, pp. 138139].
21
Ibn al-Jazar, Nashr, vol. 1, p. 33, l 24sqq., where it is said that Ibn Mujhid transmitted
from him, which likely means that he used his book.
22
Nashr, vol. 1, p. 33, l 17. His book of types of reading also in al-Fihrist of Ibn al-Nadm as
well as in al-Dhahab, abaqt, p. 615, l 24. Regarding his list of the types of reading see his
Fail; cf. above, p. 430sq.
23
In the Ms., Damascus Dr al-Kutub al-hiriyya, qirt, no. 54 (see below!) it is said
of him: wa-annafa Kitbah al-kabr f l-qirt f arbana sanatan wa-yuql inna muannaft
al-Islmarbaahuwaaaduhumthummaannafakitbahal-aghr f marifat urf al-Qurn
al-mukhtalaf f-h.
the subject of individual variant readings:
24
dhakara f-hi ikhtilf wujh al-
qirt wa-nasabakull arf ilmanqaraabi-hi. Similar statements, like those
about Ab Ubayd al-Qsim b. Sallm and Ab tim al-Sijistn, can also
be found concerning some later men, for example, al-q Isml b. Isq
al-Mlik al-Azd (d. 282/895), who dealt with twenty imms, including the
Seven
25
andbased on Ab Ubayd Abd al-Qsim b. Sallm
26
the well-
known al-abar (d. 310/923).
27
Al-Tabars work, also entitled Jmi,
28
con-
sisted of eighteen volumes, albeit in large letters,
29
Jam al-qirt min al-
mashhr wa-l-shawdhdh wa-ilal dhlika wa-sharuh,
30
and his own
ikhtiyr. Its importance rests on the fact that Ibn Mujhid made use of
it.
31
The edition of large collections of individual variant readings extends
to the time of Ibn Mujhid and beyond. Ibn Mujhids great adversary, Ibn
Shannabdh (d. 328/939),
32
is mentioned as the author of a Kitb ikhtilf
al-qurr.
33
Ab Bakr Muammad AL-NAQQSH (d. 351/962) composed al-
Mujam al-kabr f asm al-qurr wa-qirtihim.
34
The last works of this
genre we might reckon to be the great collective works of the fourth and
fifth centuries, including the Iqn by Ab Al AL-AHWZ (d. 446/1055)
35
which, despite being cited repeatedly (above, note 29), has survived
24
See above, p. 530sq.; Yqt, Irshd, vol. 7, p. 302, l 16; Ibn Khallikn, no. 835; [EQ; Sezgin,
GAS, vol. 1, p. 11, no. 6].
25
Ibn al-Jazar, abaqt, no. 754, Nashr, vol. 1, p. 33, l 20; Makk, Ibna (Berlin Ms., no. 578),
p. 496. Yqt, Irshd, vol. 2, p. 258, l 3, mentions an approving remark of Ibn Mujhid.
26
Ibn Mujhid in Yqt, Irshd, vol. 6, p. 443, l 1; in the list of writings Yqt even says:
kitbuh f l-qirt yashtamil al kitb Ab Ubayd. The number of the imms seems to have
been identical in both works.
27
Sezgin, GAS, vol. 1, pp. 323328. See above, p. 485sqq.
28
Ibn al-Jazar, Nashr, vol. 1, p. 33, l 23; otherwise mostly called kitbuh f l-qirt, or
similarly; for example, Makk b. Ab lib, Ibna (Berlin Ms., Ahlwardt, Verzeichnis, no. 578),
pp. 496 and 500; Yqt, Irshd, vol. 6, p. 427.
29
Ab Al al-asan AL-AHWZ, in the Iqn, Ms., Damascus, Dr al-Kutub al-hiriyya,
no. 54: raaytuhu f thamniya asharu majalladatin ill annahu kna bi-khuin kibrin. Also
Yqt, Irshd, vol. 6, p. 427, l 7sqq. (abbreviated). Here, Yqt explicitly mentions al-Iqn of
al-Ahwz as the source. Thus, the afore-mentioned Damascus manuscript can be identified
beyond doubt.
30
Its authorship is very doubtful; Sezgin, GAS, vol. 1, p. 328, no. 9.
31
Ibn al-Jazar, Nashr, vol. 1, p. 34, l 2 (raw an probably meant by this); cf. Yqt, Irshd,
vol. 6, p. 442, l 18, where Ibn Mujhid attests to the excellence of the book, attributing some
of its mistakes to Ab Ubayd al-Qsim b. Sallm.
32
33
Yqt, Irshd, vol. 6, p. 302, l 2.
34
Fihrist, p. 33, and from this Yqt, Irshd, vol. 6, p. 497, l 3 (akbar instead kabr); see
above, p. 505.
35
al-Jazar, abaqt, vol. 1, p. 222, l 3: raw an-hu al-imma wa-rimma Ab Mashar
only as a very small fragmentand another fragmentary Kitb al-Jmi or
Sq al-ars of his pupil Ab Mashar Abd al-Karmal-abar (d. 478/1085).
36
Both works contain a great number of complete ikhtiyrt apart from the
Seven canonical readings. Further, the K. al-Kmil by ABAL-QSIMYsuf
(Ibn) Al AL-HUDHAL (d. 465/1072),
37
contains no fewer than 1,500 riwyt
and uruq, apart from the fourteen other ikhtiyrt. Likewise, the K. al-
Rawat al-uff of Ab Isml Ms AL-MUADDIL (roughly a contem-
porary)
38
contains fifteen readings, among whom we find umayd b. Qays
AL-ARAJ,
39
Ibn al-Sumayfa, and ala b. Muarrif al-Hamdn (d. 112/730).
40
The collection and citation of variant readings in the time before Ibn [iii/209]
Mujhid is not limited to the muqris themselves but is rather the work
of traditionists, Koranic commentators, and philologists alike, particularly
the latter. They appear either in the framework of grammatical and lexico-
graphical works or frequently independently; in most cases we only know
them by title. In the Mutasab of Ibn Jinn (died 392/1002),
41
however, more
details have survived through references to the famous Muammad b. al-
Mustanr QURUB (d. 206/821),
42
and also to the book of Ab tim al-
Sijistn (died 255/869).
43
The highlight of philological studies onthe variant
readings is precisely this work of Ibn Jinn, as well as his commentary on the
K. al-Shawdhdh of Ibn Mujhid, and his parallel work and model, the com-
mentary of his teacher, Ab Al al-Fris (d. 377/987)
44
on the K. al-Saba of
Ibn Mujhid, entitled K. al-ujja (see below, p. 552).
al-abar f kitb sq al-ars. [Ab Al al-asan b. Al b. Ibrahm AL-AHWZ; Sezgin, GAS,
vol. 1, p. 603].
36
O. Pretzl, Verzeichnis der lteren Qirtwerke, no. 32, p. 45.
37
38
Ibid., no. 3679; O. Pretzl, Verzeichnis der lteren Qirtwerke, no. 31, pp. 4344.
39
Juynboll, Encyclopedia, p. 575, col. 2; Sezgin, GAS, vol. 8, p. 50.
40
Juynboll, Encyclopedia, p. xxviii, col. 2, n. 4; p. 62, col. 1, p. 404, col. 2.
41
G. Bergstrer, Nichtkanonische Lesearten imMutasab des Ibn inn.
42
Loc. cit., p. 18, l 93. IbnJinn does not supply the title; it might have beenMan l-Qurn
(Flgel, Die grammatischen Schulen der Araber, p. 66, no. 1) or Irb al-Qurn (p. 67 no. 18).
[Sezgin, GAS, vol. 9, pp. 6465].
43
Loc. cit., p. 18, ll 90 and 94; Ab tim al-Sijistn is one of the main authorities also
for the author of the K. al-ujja (see below, p. 552) and is frequently cited in the Kashf of
Makk b. Ab lib. [According to Sezgin, GAS, v. 3, pp. 367368, Ab tim died more likely
in 250/864.]
44
Ibn al-Jazar, abaqt, p. 951; [Sezgin, GAS, vol. 9, pp. 101110].
The Rise of the Canonical Variant Readings [iii/210]
AbBakr (Amadb. Ms) IBNMUJHID(d. 324/936) rankedsecondinthe
science of the variant readings behind his adversary, Ibn Shannabdh,
45
but
in the opinion of his pupil, AB HIR Abd al-Wid b. Ab Hshim AL-
BAZZR
46
(d. 349/960) greatly superior to him as far as discernment is con-
cerned.
47
But most of all, he was a successful teacher.
48
His work has become
the standard reference book to the science of the Koran, although it was sur-
passed by later literature, particularly by al-Dns Taysr and by subsequent
writings. Nevertheless, his book continued to be transmitted for quite some
time.
49
Ibn al-Jazar studied it and read the Koran according to its content. It
has come down to us, at least the greater part of it, in the above-mentioned
commentary of Ab Al al-Fris.
50
There, it is most certainly not complete,
because after some introductory words al-Fris begins with the commen-
tary on the first variant reading mlik : malik, verse 1:3; but we may assume
that Ibn Mujhid first had mentioned details about the readers and trans-
mitters. Apart from the introduction, which is no longer extant, the book
was arranged according to Koranic passages relating to their differences.
There is no trace of a first part on the rules of pronunciation arranged by
subject (ul), as appears in later such writings. But there are approxima-
tions to a synopsis when at the first occurrence of a subject, similar matters
arising later are all dealt with together. In this instance he even goes fur-
45
Ibn Shannabdh prides himself with this superiority which to a large extent is due to
his extensive travel (Ibn al-Jazar, abaqt, no. 2707: vol. 2, p. 55, n. 9). His date of death is
variously given as 325/936 and 328/939.
46
Sezgin, GAS, vol. 9, pp. 167168: Ab hir al-Bazzz, 280/893349/960.
47
He says about Ibn Shannabdh: ilmihu fawqa aqlih, but about Ibn Mujhid, the
opposite. (al-abar, loc. cit.)
48
His alqa (lectures) were attended by three hundred professional (muaddar or muta-
addir) readers; he had eighty-four assistant teachers (khalfa, see above, p. 496 n. 176); al-
Nashr, vol. 1, p. 121, l 25. Ibn al-Jazar does not known of any teacher with more students than
he had (abaqt, no. 663: vol. 1, p. 142, l 5).
49
Usually entitled Kitb al-Saba, fully, Marifat qirt ahl al-amr bi-l-ijz wa-l-Irq
wa-l-Sham (thus al-Fris in the introduction to his commentary; see O. Pretzl, Die Wis-
senschaft der Koranlesung, p. 18, l 7). There is the solidary statement of al-Jabar (d. 732/1331)
in his commentary on al-Shibiyya (Ms. Istanbul, Fatih Camii Ktphanesi, 52, 550
v
.) that he
used the Kitb al-Saba al-aghr of Ibn Mujhid as the only of his writings that he was able to
find. If this was correct, it might fit the passage in the Fihrist, where the bibliography lists a
Kitb al-Qirt al-aghr next to a kabralthough next to both of them also Kitb al-Saba.
50
A second commentary is from Ibn Khlawayh (d. 370/980; Flgel, Die grammatischen
Schulender Araber, p. 232, no. 19); further injj Khalfa (under Kitbal-Saba) whoexplicitly
mentions that both the commentaries and the text are in his possession. [Sezgin, GAS, vol. 9,
pp. 167168.]
ther than al-Dn. Furthermore, there are also traces of the bqn-system
already mentioned on p. 428. Al-Fris lacks the solid system of two trans-
mitters. We therefore often find fewer transmissions. Hishm(Ibn Ammr)
an Ibn mir, for example, is hardly considered. But in most instances there
are still more, since apart from the canonical transmitters also other trans-
mitters are dealt with. For this reason, the book frequently contains variants
that are not at all mentionedby al-Dn. Many peculiarities display a smaller
measure of detail and comprehensiveness of phonetic observation. Entirely
wanting are references to the difference between context (wal) and pausal
(waqf) pronunciation of the later period.
We cannot determine who established the additional limitations and [iii/211]
impoverishments of the systemof dual transmitters since too little is known
about the writings on the Seven in the period between Ibn Mujhid and
al-Dn.
51
We are in a somewhat better position as far as the next two
advances that occurred in the immediate period after Ibn Mujhid, namely
the presentation of a systematic part about ul and a linguistic exegesis of
the variant readings are concerned.
As far as the ul are concerned, we learn that the first man to have [iii/212]
initiated the discussion of the individual readings (called farsh) was Ibn
Mujhids pupil (Ab l-asan Al b. Umar) AL-DRAQUN (d. 385).
52
Al-
Dns Jmi owes its excellence mainly to the fact that the work was mod-
elled on al-Draquns book.
53
Two books entitled Itijj al-qurr wa-irb
al-Qurn
54
have beentransmitted, one by the renownedgrammarianal-Mu-
barrad (d. 285/898), and the other by Ab Bakr IBN AL-SARRJ (d. 316/
928).
55
Still older is the second advance. Already al-Akhfash (d. ca. 292/
904, see above, p. 516) is said to have composed a book (bi-l-ilal)
56
on
the reading of Ibn mir al-Yaub, and AB AL-QSIM Ubayd Allh b.
51
The most important writings: al-Irshd of AB AYYIB (Abd al-Munim b. Ubayd
Allh) IBN GHALBN, died 389/999 [Sezgin, GAS, vol. 1, p. 15] in Egypt (used in al-Nashr of
Ibn al-Jazar, see vol. 1, p. 78) the teacher of Makk b. Ab lib (see below); the K. al-Hd
of AB ABD ALLH Muammad b. Sufyn AL-QAYRAWN, d. 415/1024 in Medina, for
whom Ibn al-Jazar still has rich transmissions (al-Nashr f l-qirt al-ashr, vol. 1, p. 65), and
which we, too, still possess (Pretzl, Verzeichnis, Istanbul, Fatih Camii Ktphanesi, no. 62);
and al-Hidya of Ab l-Abbs (AMAD IBN AMMR) AL-MAHDAW, d. 440/1048 [Sezgin,
GAS, vol. 9, pp. 215216], also used by Ibn al-Jazar (al-Nashr, vol. 1, 68), as well as the extant
corresponding commentary by the same author (Pretzl, Verzeichnis, no. 6, pp. 2425; Istanbul,
Kprl [not specified whether Kprl Fazl Ahmet Paa or Kprl Mehmet Paa], no. 20).
52
53
Ibn al-Jazar, abaqt, 2281: vol. 1, p. 559, l 4; the title of the book is unknown.
54
Yqt, al-Irshd, vol. 7, p. 143, l 20; Sezgin, GAS, vol. 8, p. 98.
55
Ibid., vol. 7, p. 11, l 11; see below, note 60; Sezgin, GAS, vol. 9, pp. 8285.
56
Ibrhm AL-AMR (d. 307/919) a book entitled Muannaf muallal
57
about
the reading of Ab Amr b. al-Al. Al-abar (d. 310/923) in his great work
on qirt
58
(see above, p. 548) refers in every reading to ilal and shar.
IBN MIQSAM al-Ar (d. 354/965), Ibn Mujhids adversary (see above,
p. 475sq.), composed a book entitled Itijj lil-qurr, where he is trying to
advocate unusual readings wujhan min al-lugha wa-l-man.
59
We also find
this latter expressioninthe title of the oldest extant workdefending this pro-
cedure of substantiation and justification of the variant readings, entitled
K. al-ujja of AB AL (al-asan b. Amad b. Abd al-Ghaffr) AL-FRIS
(al-Fasaw) who died in 377/987, which constitutes a complete commentary
on the Koran, including tall to Kitb al-Saba of Ibn Mujhid.
60
The influ-
ence of this work must not be underestimated; Makk b. Ab lib, who died
in 437/1045,
61
and Ab hir (Isml) IBN KHALAF (al-iqill), who died in
455/1063,
62
left us excerpts from it. See also below, p. 582sq. (Seite 247, Nr.
18!) [sic, where there is no such number].
The Rise of the Classic Systemof the Seven
Of decisive importance for the subsequent development of the literature of [iii/213]
the variant readings was the transfer of the new science of variant readings
to Spain. Upon the initiative of the Spanish Umayyad Caliph, al-akam II
al-Mustanir Billh, the Egyptian reader of the Koran AB L-ASAN (Al b.
Muammad b. Isml) AL-ANK (d. 377/987) moved in 352/963 to Cor-
doba.
63
But it was far more important that conversely Spaniards, namely
the two contemporaries, Makk b. Ab lib al-Qays (died 437/1045) and,
57
Ibid., 2010.
58
According to the above-mentioned work of (Ab Al al-asan b. Al) AL-AHWZ (Ms.
Damascus, al-hiriyya, 54); further, Yqt, Irshd, vol. 6, p. 427, l 8, and vol. 6, p. 442, l 2.
59
Yqt, Irshd al-arb il marifat al-adb, vol. 6, p. 499, l 1.
60
Pretzls Verzeichnis, no. 1, pp. 1721. About the author see Flgel, Die grammatischen
Schulen, p. 110; Sezgin, GAS, vol. 1, p. 18, no. 11. According to the author, the work on the
book was begun in 315/927thus, still before Ibn Mujhidby Ab Al al-Friss teacher,
Ibn al-Sarrj (Flgel, op. cit., p. 103; Sezgin, GAS, vol. 9, pp. 8285) in whose list of writings
it appears as K. al-Itijj f l-qira (Flgel, ibid., no. 6; see also jj Khalfa under Itijj
al-qurr). Ab Al al-Fris quotes in full his teachers explanationsas far as it went (only
part of the second sra!)and then adds his own remarks. Ibn Jinn reproaches him in
the introduction to his Mutasab (G. Bergstrer, Nichtkoranische Lesearten, p. 17, l 65) for
verbosity and arduousness. The first reproach applies probably more so than to Ibn Jinns
own work; the latter reproach, hardly.
61
Yqt, Irshd al-arb il marifat al-adb, vol. 7, p. 174, l 14.
62
Ibn al-Jazar, abaqt, 763; Sezgin, GAS, vol. 9, p. 244, no. III (c).
63
Ibn al-Jazar, abaqt, 2308 (vol. 1, p. 565, l 2). Sezgin, GAS, vol. 9, p. 200, n. 1.
more successfully and influentially, Ab Amr (Uthmn b. Sad) AL-DN
d. 444/1053 studied in the East.
64
Both of themcomposed two works each on
the Seven, a brief one, containing nothing but the facts (including the most
important isnds) and another, comprehensive one, whichoffers apart from
extensive isnds also the tall. Makk b. Ab libs Tabiraaccording to
his own words in the epiloguewas intended as a guide for beginners to be
memorized by heart and as replacement for a still shorter presentation, al-
Mjiz, from385/995, intended for his private use only but published against
his will. It was not until 424/1032 that he could realize his long-standing
project of writing a commentary on it, containing ilal and ujaj, which he
entitled Kashf.
65
The Tabira was a kitb naql wa-dirya, the Kashf is a kitb
fahm wa-ilm wa-dirya.
66
The Tabira retained a certain importance; AB
L-ASAN (Al b. Umar) AL-QAYJ (d. 723/1323) referred to it in his sup-
plement to the Shibiyya,
67
entitled al-Takmila al-mufda li-fi al-qada,
andIbnal-Jazar still studied it;
68
the Kashf, onthe other hand, seems to have
been of less influence upon later writers. The interest in a deeper under-
standing of the variant readings of the Koran that began in the third century
is certainly on the way out from the sixth century onwards (still one rep-
resentative, Ibn Ab Ibrhm, in al-Mi).
69
In Ibn al-Jazars life-work, K.
al-Nashr, which in other respects constitutes a certain highlight, hardly the
least traces remain.
70
Far more successful than Makk b. Ab lib was al-Dn. Still today he [iii/215]
indirectly dominates the teaching of the science of qirt because of the
literature which, in the final analysis, is based on his Taysr f l-qirt al-sab.
The Taysr,
71
as the more concise presentation, has as its counterpart the
64
As the precursor of the two men Ibn al-Jazar mentions Ab Uthmn [Uthmn]
(Amad b. Muammad b. Abd Allh) AL-ALAMANK d. 429/1037; [EI
2
;] when he mentions
him as the first man to bring the science of the variant readings to Spain (abaqt, 554,
similarly Nashr, vol. 1, p. 34, l 10) he means to say that he was the first Spaniard to have studied
in the East. His work, K. al-Rawa, Ibn al-Jazar lists among his own sources (Nashr, vol. 1,
p. 70) but without having an isnd. The work was obviously of little importance.
65
Apart fromthe very goodBerlinMs. (Ahlwardt, Verzeichnis, no. 578 =Petermann, no. 17).
an excellent Ms. in [H. Derenbourg, Les manuscrits arabes de l] Escorial, 1325. Photographs
at the Korankommision der Bayerischen Akademie der Wissenschaften, [Mnchen].
66
These details from the preface to al-Kashf.
67
68
Ibn al-Jazar, Nashr, vol. 1, p. 69.
69
Ca. 560/1164; Pretzls Verzeichnis, no. 19, pp. 3234.
70
See below, p. 563sq.
71
Edited by O. Pretzl, with the sub-title, Das Lehrbuch der sieben Koranlesungen [manual
of the seven readings of the Koran], von al-Dn (1930).
Jmi al-bayn,
72
as the more comprehensive presentation of the two. The
relation of the two works is different in Makk b. Ab lib: The prime value
of Jmi is that it is not limited to the Fourteen canonical transmissions;
rather, it considers forty of them, supplying isnds with very detailed and
biographical data. It is because of such material that the Jmi constitutes
the basis of the science of the Koranic readings. Ibn al-Jazar resorts to
logograms in his abaqt to indicate the readers who are included. Other
than Ibn al-Jazar,
73
Abd al-Ramn al-afrw,
74
author of several works on
variant readings (d. 636/1238), also studied it (see below, p. 566 n. 161
75
).
The Taysrthe classic guide to the Seven Koranic variant readings, like
the Tabira, and intended for memorization by heartbegins after a brief
introduction by supplying as the essential element some information about
the method of recitation (al-aramiyyn = Nfi, Ibn Kathr al-Kinn, etc.),
accompanied by short biographies of the Seven readers and their fourteen
transmitters, supplemented with lists of the authorities (rijl) of the Seven,
and the isnds that connect al-Dn with the fourteen transmitters. These
isnds are always supplied twice, once for the theoretical instruction (ad-
dathan bi-h , etc.), and then for the practical recitation of the Koran
afterwards (qaratu bi-h al-Qurn al ).
The first main part of the book deals with ul and the general rules of
pronunciation, which is preceded by short sections on istidha and tasmiya
as well as the different variant readings of the first sra. The ul com-
prise: First.The great assimilation (of consonants separated by a vowel,
see above, p. 537). Second.h al-kinya, suffix of 3. Pers. Sing. Masc.
Third.Prolongation (madd). Fourth.Rules of hamz (including the spe-
cial cases of hamz in words in pause in amza b. abb and Hishm b.
Ammr (al-Sulam) an Ibn mir (al-Yaub)). Fifth.The so-called minor
assimilation
76
(of adjacent consonants). Sixth.imla (including in par-
ticular al-Kiss pronunciation of the pausal feminine ending -ah). Sev-
enth.The pronunciation of r and l. Eighth.Pause (here also rawm and
72
Pretzls Verzeichnis, no. 4, pp. 2223. Of the two known Mss, Nuruosmaniye Ktpha-
nesi, 62, and the Egyptian Library, Cairo, qirt m 3, a photographic copy is with the Koran-
kommision der Bayerischen Akademie der Wissenschaften, Mnchen (dated, 1146, 375 folios,
beautiful, careful hand).
73
Ibn al-Jazar, al-Nashr, vol. 1, p. 60.
74
Brockelmann, GAL, suppl. vol. 1, p. 727, no. 13.
75
One of the sources of his Taqrb (see below, p. 566) in his Iln that Ibn al-Jazar studied
(al-Nashr, vol. 1, p. 78)about which I have not been able to find anything specificthe
content of al-Jmi al-bayn has been partially incorporated (al-Nashr, vol. 1, p. 60, l 8).
76
al-Dn does not use this expression; he rather speaks of al-ihr wa-l-idghm lil-urf
al-sawkin.
ishmm). Ninth.amza b. abbs short pause (sukt, sakt) after a vow-
elless consonant before hamz. Tenth.The pronunciation of the suffix of 1.
pers. sing. as - or iyah. Eleventh.The treatment of defective written final .
Inserted after the great assimilation is the mere rubric bb srat al-baraqa.
The second main part of the book deals with the individual variant
readings, farsh al-urf; the end of every sra consists of a compilation of
the occurring instances of a suffix of the 1. pers. sing. in the sra as well as
the defective orthography of the final , intended as a complement to the
general rules in the ul applicable in both instances.The conclusion is
made up of a section on takbr.
It is obvious that this outline is not based on a systematic structure, but [iii/216]
it also clearly carries the positive aspects of its origin by interlacing the ul
part with the treatment of the individual variant readings of sras one and
two. The chapters on ul result from the habit, evident from Ibn Mujhid,
to combine identical phenomena at the first occurring instance. Moreover,
as indeed at the beginning of the second srabefore the first individual
variant of the Seven (yukhdina : yakhdana) in verse 8examples of
the ul can be found,
77
it was logical to deal with the ul first. Al-Dns
presentation is very carefully balanced and very brief. No vowel is supplied
unnecessarily, only when the other readings have a different vowel.
78
The arrangement and presentation of the work are better and more [iii/217]
precisely structured in al-Dn; but as is evident from the close connection
with the Tabira, they were not created by him.
79
This would suggest that
one look for an older, common source. The most obvious person, apart from
al-Draqun, who died in 385/995 (see above, p. 551), is Makk b. Ab libs
teacher, Ab l-ayyib b. Ghalbn, who died in 389/999 (see above, p. 551
n. 51), who was also the father of al-Dns teacher, Ab l-asan hir b.
77
Sra 2:2, f-hi hudanthe h al-kinya, at the same place, thoughunjustly, but historically
justified, also the great idghm, which first occurs already in sra 1:23, al-rami malik; sra
2:4, bi-m unsila of the madd; 2:6, aandhartahum two hamzs in one word, etc.
78
Quite different the later Mss. of the Taysr, which not infrequently display unneccesary
additions fromIbnal-Jazars al-Tabr, and containreferences that do not go back to al-Dn.
Also the two existing Indian lithographs of the Taysr (Hyderabad, 1316/1898, and Delhi,
1328/1910), are based on such corrupted Mss.
79
This can reach literal particulars and identical details of the arrangement. The main
difference is that the presentation of the Tabira is less firm and that the arrangement is less
clear and more superficial. There are also some terminological differences to be found, for
example, m qalla dawruhu min al-urf (Tabira, thus, incidentally also K. al-Hd of Ab
Abd Allh Muammad b. Sufyn al-Qayrawn, Pretzls Verzeichnis, no. 2, p. 21) instead of
farsh al-urf (al-Taysr).
Ghalbn (d. 399/1008), the author of K. Tadhkira,
80
which is largely identical
with the two works (see below, p. 561 sq.). He is also the most likely person
to whom the introduction of the Fourteen transmitters might be ascribed.
The most weighty difference between the Tabira and the Taysr is that
in the Tabira the great iddighm is missing. This is still feasible despite the
mutual identity of the transmission from Ab Amr b. al-Al because the
idghm is only partly represented in the branches of both of them. This is
not unique to the Tabira, because Ibn al-Jazar
81
mentions a great number
of writings on variant readings that agree on this point, among them Ibn
Mujhids al-Saba; he also mentions that the chapter from the same trans-
mission found in the Taysr is to be found already in Ab l-asan hir b.
Ghalbn (d. 399/1009). Both of the last of the ul, sakt and yt, are miss-
ing fromthe Tabira. Regarding the yt, the workrepresents anearlier stage
when, influenced by orthographic contingencies, the systematic treatment
of these uneven points had not yet been attempted. From Ibn al-Jazars
details it is evident that the phonetic subtleties of sakt had not been firmly
establishedby Makk b. Ab libs time, andthis remainedso for quite some
time to come.
82
Insignificant differences, particularly with respect to the order of the [iii/218]
chapters on ul, are not wanting from the writings on variant readings
in the group represented by the Tabira and the Taysr; they nevertheless
form one entity compared with the works on variant readings which are
much more at variance with each other, and which place the ul before
the discussion of the first sra. Also, those works that deal with more than
seven readings can be divided into two groups. To the first group belongs
their most important representative work, the Nashr of Ibn al-Jazar, which,
with its systematic perfection and carefully thought-out plan in the chap-
ters on ul, considerably outdistances the Taysr, but it is still surpassed
by an older work of the second group, Rawat al-uff of Ab Isml
(Ms) AL-MUADDIL (see above, p. 549 n. 38; Pretzls Verzeichnis, no. 31,
pp. 4344).
The accumulation of homogeneous passages and the development of
the chapters on ul do not constitute the apex of the systematic arrange-
80
81
This goes up to literal conformity and identity of details of the arrangement, the
main difference being the less strict presentation of the Tabira, which in its disposition
is superficial and less logical. Also some terminological differences are to be met with, for
example, m qalla dawruhu min al-urf (Tabira, thus, incidentally, also al-Hd of Ab
Abd Allh Muammad b. Sufyn al-Qayrawn, Pretzls Verzeichnis, no. 2) instead of farsh
al-urf (Taysr).
82
al-Nashr, vol. 1, 240 and 412.
ment of the content of the variant readings of the literature of the variant
readings. It was ([Sad al-Dn] Sharaf al-Dn Hibat Allh b. Abd al-Ram)
IBN AL-BRIZ
83
(d. 738/1338), who also provided a topical arrangement
of the individual variant readings in his Shira on the Seven.
84
He did not
find a successor; the work remains a curiosity. Conversely, the detailed sub-
ject arrangement, which actually represented progress in the presentation
of the subject and, at the same time, facilitated memorization by heart,
was very likely considered a barrier against quick orientation with regard
to individual passages, so that writers revertedexcluding the most gen-
eral foundations of pronunciationto listing the particular reference in
each and every instance. Printed books of this kind are al-Mukarrar f m
tawtar min al-qirt al-sab wa-takarrar of Sirj al-Dn (Ab af Uth-
mn b. Qsim) al-Anr
85
AL-NASHSHR (ca. 900); further, Ghayth al-naf
of Al [b. Muammad] AL-NR AL-AFQUS [1053/16431118/1706];
86
Zub-
dat al-irfn of mid b. Abd al-Fatt al-Plaw [BLAW
87
]. The repetition
(hence the title Mukarrar) cannot be totally complete; each of the most fre-
quent references are mentioned once only at the first occurrence.
The Taysr, to whichwe nowreturn, was initially studied a great deal. This [iii/219]
is evident from the two commentaries,
88
one by AB MUAMMAD Abd
al-Wid b. Muammad AL-BHIL (d. 705/1305), entitled al-Durr al-nathr
wa-l-adhb al-namr, and the other by Ibn al-Jazar (d. 833/1429), entitled al-
Tabr, where he redoes the Taysr, complete with corrections and additions
as well as the reading of the Three after the Seven, but, most of all, the versi-
fication of the Taysr modelled after the Shibiyyawhich from then on
dominated the teaching of the variant readings. The commentaries were
enormously conducive to the spread of the ideas of the Taysr, although, at
the same time, pushing the commentary itself into the background.
The Shibiyya was not the first versification of the Seven. According
to Ibn al-Jazars abaqt,
89
al-usayn b. Uthmn b. Thbit al-Baghdd
83
Sezgin, GAS, vol. 8, p. 64, no. 4, l. 34.
84
Ibn al-Jazar, Nashr, vol. 1, p. 95, abaqt, 3772; Sezgin, GAS, vol. 8, p. 64, no. 4, l 34.
85
Cairo, 1326/1908; Brockelmann, GAL, vol. 2, p. 115.
86
Cairo, 1293/1876, 1304/1886, 1321/1903, all in the margin of the commentary on the
Shibiyya of Ibn al-Qi. (Sezgin, GAS, vol. 2, p. 698).
87
Lithograph, Constantinople, 1312/1894, with the commentary Umdat al-khulln f
of Muammad al-Amn b. Abd Allh (composed after 1252/1842), lithograph, Constantino-
ple, 1287/1870. In Turkey much used in teaching; deals with the Ten.
88
Cf. the introduction by the editor to the Taysr, p. . There is a very beautiful manuscript
of the first commentary at Cairo, Taymr Pasha, 235.
89
No. 1110.
(d. 378/988) had already composed such a work. Also al-Dn
90
himself had
already composed a rajaz poem, al-Iqtid, on the Seven, which apparently
disappeared at an early date. The author of the most famous versification,
Abl-Qsimal-Qsimb. Firro[Firruh] AL-SHIB (d. 590/1193)
91
represents
the repercussions of the Spanish science of the variant readings on the East.
He was born in Spain, where he studied variant readings, particularly Taysr,
continuing these studies on his pilgrimage to the East; he was subsequently
appointed by al-Q al-Fil
92
to the madrasa which he had founded, and
there composed his two poems based on al-Dn, the above-mentioned
Aqla (above, p. 406), and the other, simply called al-Shibiyya (with the
complete title, irz al-amn wa-wajh al-tahn). It is a awl-type poem
of 1,173 verses, rhyming with -l. Considering al-Dns rajaz, the latters
more rigid form created difficulties for the purpose. This was overcome
by exhausting poetic licence by means of padding, but particularly by a
system of logograms for readers, transmitters, or groups of such people.
The logograms are letters that do not stand for themselves at all; they
rather occur only inthe accompanying text of the respective variant reading.
Which of the letters are to serve as logograms is indicated by red ink in the
manuscripts, and by brackets in printed copies. Apart from the text, the
student must thus also memorize which of the letters are logograms. In the
same way that the Aqla goes beyondthe original, so also the Shibiyya goes
a little beyond the Taysr as far as content is concerned: For once, by means
of details and then also by inserting a propaedeutic chapter on phonetics
(bb makhrij al-urf wa-iftih), which had gained a place in the later
writings on variant readings in general.
The Shibiyya became famous
93
by its first commentator (ALAM AL- [iii/220]
DN Ab l-asan Al b. Muammad b. Abd al-amad) AL-SAKHW (d.
643/1245).
94
His work, Fat al-wadtogether with some other early com-
mentaries, for example, Ibrz al-man of (Ab l-Qsim Abd al-Ramn
b. Isml al-Dimashq) AB SHMA (d. 665/1266),
95
al-Lal al-farda of
Ab Abd Allh (Muammad b. asan b. Muammad) AL-FS (d. 656/
1258)
96
is part of a group of works in which, once more, now following the
90
Ibn al-Jazar, abaqt, 2091 (vol. 1, p. 505, 2); in addition, Yqt, Irshd, vol. 5, p. 36, l 6.
91
Sezgin, GAS, vol. 9, p. 41, no. II; Ibn al-Jazar, abaqt, 2600.
92
EI
2
: Ab Al Abd al-Ramn AL-Q AL-FIL, d. 1200.
93
Ab Shma, commentary on the Shibiyya, entitled Ibrz al-man (Cairo, 1349/1930),
p. 7; Ibn al-Jazar, abaqt, 2318 (vol. 1, p. 570, l 9).
94
95
Ibid., vol. 1, p. 317; printed, Cairo, 1349/1930.
96
Ibid., vol. 1, p. 409; at this place are listed also a number of commentaries, ibid., suppl.
vol. 1, p. 728. Cf. also Ahlwardt, Verzeichnis, vol. 1, pp. 337339.
Shibiyya, the old science is spread out in its fullness. Added is the most
successful of all the commentaries on the Shibiyya, the Kanz al-man
97
of
(Ab Isq Ibrhm b. Uthmn) AL-JABAR (d. 732/1331). The great major-
ity is mainly limited to the minimum necessary for the understanding of
the poem as well as the supplementary material. In addition to the com-
mentaries already mentioned, there are also the following printed com-
mentaries: Sirj al-qri al-mubtadi wa-tadhkr al-muqri al-muntah of (Ab
l-QsimAl b. Uthmn) IBNAL-QI,
98
andIrshdal-murdilmaqdal-
Qad of Al b. Muammad AL-ABB (living at Cairo).
99
The logograms became later also used in prose texts as a very practical [iii/221]
means of illustration. The first work of this kind is the M of [Ab] Abd
Allh Nar b. Al AL-FRIS;
100
the same applies to the above-mentioned
Zubdat al-irfn of al-Blaw (see above, p. 557). But as this was an imped-
iment to memorization, there was an attempt to avoid it in poems that
were in competition with the Shibiyya; and this already by al-Shibs
contemporary, Ab Abd Allh (Muammad b. Amad b. Muammad) AL-
MAFIR (d. 591/1194),
101
and later Mlik b. Abd al-Ramn IBN AL-
MURAIL (d. 699/1299),
102
and also, ABAYYNAL-GHARN (d. 745/
1344), who composed his poem, Iqd al-lal f l-qirt al-sab al-awl, in
the rhyme and metre of the Shibiyya,
103
as well as (Fakhr al-Dn Amad
b. Al) IBN AL-FA al-Hamadhn (d. 755/1354), who indicates his pur-
pose already by its title, all al-rumz.
104
On the other hand, there was an
attempt to surpass al-Shibs conciseness. Such is the case with the above-
mentioned commentator on the Shibiyya, Shula (d. 656/1258) (see above,
p. 559 n. 97) who in his Shama reduced the Shibiyya to approximately half
its size, this way formally surpassing it.
105
Such and similar attempts did not
succeed in displacing the Shibiyya.
97
This title is identical with the older commentary of Muammad b. Amad b. Muam-
mad SHULAH (d. 656/1258); al-Jabar apologizes in the epilogue of his commentary that
without knowing the predecessor he chose the identical title. (Istanbul Ms., Fatih Camii
Ktphanesi 52:550
v
f, Vakf Ibrahim, 51, Velyeddin Carullah, 15.)
98
Completed in 759/1357, printed 1321/1903, etc., recently in 1954. Yusuf E. Sarks, Diction-
naire encyclopdique de bibliographie arabe, 209.
99
G. Bergstrer, Koranlesung in Kairo, p. 27, also printed in the margin of Ab Shma,
see above, p. 558 n. 95.
100
Pretzls Verzeichnis der handschriftlicherhaltenenlterenQirtwerke, no. 19, pp. 32
34; different from those of the Shibiyya.
101
jj Khalfa, see below, Qada f l-qirt.
102
103
Ibn al-Jazar, al-Nashr, vol. 1, p. 94, l 19; abaqt, 3555.
104
Ibn al-Jazar, abaqt, 380; jj Khalfa, loc. cit.
105
Ibn al-Jazar, al-Nashr, vol. 1, p. 94, l 1; abaqt, 2780. [Recently published, Shar Shula
al l-Shibiyya (1954).]
The Taysr and the Shibiyya displaced a number of other teaching [iii/222]
guides on the Seven that had temporarily dominated the teaching institu-
tions in certain areas. The most important of these was the Unwn
106
of
Ab hir IBN KHALAF AL-ANR (d. 455/1063); although more recent,
the book has a better isnd than the Taysr. It is an excerpt from the same
authors larger work entitled al-Iktif.
107
In Egypt it had been the most com-
mon school-book for a long time. It was there also provided with a com-
mentary. Ibn al-Jazar (d. 833/1429) composed a book on the comparison
of the Unwn to the Shibiyya entitled Tufat al-ikhwn f l-khilf bayn al-
Shibiyya wa-l-Unwn.
108
Together with the Taysr and the Shibiyya, the
Unwn formed the basis of Mun al-muqri al-nirr al m akhta bih
al-Unwn wa-l-Shibiyya wa-l-Taysr of Shihb al-Dn Amad b. Al b. Abd
al-Ramn AL-BILBAYS (d. 779/1377),
109
as well as of a work of Umar b.
Qsim al-Anr entitled al-Badr al-munr (about Nfi b. Abd al-Ramn
al-Layth, Ab Amr b. al-Al and Ibn Kathr (al-Kinn)) that the author of
Kashf al-unn [jj Khalfah] erroneously considers a commentary on the
Taysr.
110
Somewhat later than the Unwn appeared the Tajrd of Ab l-Qsim [iii/223]
(Abd al-Ramn b. Afq b. Ab Bakr) al-iqill IBNAL-FAM
111
(died 516/
1122), reckoned by Ibn al-Jazar to be one of the best books on variant read-
ings,
112
and becoming the subject of a similar comparative study entitled al-
Taqydf l-khilf baynal-Shibiyyawa-l-Tajrd.
113
ABAL-ASANAl b. Umar
(not Uthmn) al-Kattn AL-QAYJ (d. 723/1323) in his work, al-Takmila
al-mufda li-fi al-Qadaa poem in the same metre and rhyme of the
Shibiyya
114
made a comparison with the Shibiyya and drew upon it for
its completion, namely Makk b. Ab libs Tabira (see above, p. 553) (Ab
Abd Allh Muammad) IBN SHURAYHs (al-Ruayn, died 476/1083)
115
al-
Kf as well as the Taysrthe oldest available book on variant readings in
106
Pretzls Verzeichnis, no. 9, p. 27.
107
Pretzls Verzeichnis, no. 8, pp. 2627.
108
Pretzl, Die Wissenschaft der Koranlesung,, p. 27. As it is evident fromthe isnds of the
Unwn, the author is indeed the famous Ibn al-Jazar, although the manuscript gives as his
kunya Ab Abd Allh instead of Ab l-Khayr, and although the work is not mentioned in the
lists of his works I know of.
109
Brockelmann, GAL, vol. 2, p. 111, no. 11.
110
See introduction to al-Taysr, p. , foot-note.
111
Pretzls Verzeichnis der Qirtwerke, no. 15.
112
Ibn al-Jazar, Nashr, vol. 1, p. 74sqq.
113
114
Ibn al-Jazar, Nashr, vol. 1. p. 96.
115
Ibid., vol. 1, p. 66sqq.; Sezgin, GAS, vol. 1, p. 11, bottom.
print at the moment
116
and (Ab Muammad Abd Allh b. Al b. Amad)
SIB AL-KHAYYs (d. 541/1146)
117
jz.
From among the flood of works on the Seven there remains only one
scholarly work to be mentioned, Ab Jafar (Amad b. Al) IBN AL-B-
DHASH
118
al-Gharns (d. 540/1145) al-Iqn, which represents three hun-
dred uruq.
119
This combination of limitation to the Seven, accompanied by
the greatest possible richness within this framework, is the exception; most
of them that found scholastic teaching insufficient went beyond the Seven.
The arrangement of the Iqn departs not inconsiderably fromwhat was cus-
tom (including tajwd). Only this work includes chapters on ikhtilf madh-
hibihim f kayfiyyat al-tilwa wa-tajwd al-ad as well as m khlafa fh
al-ruwt aimmatahum. Ab ayyn al-Andalus (d. 745/1344
120
), the impor-
tant commentator of the Koran, considers it the best work on the Seven (see
below, p. 578).
The Extension of the Systemof the Seven
The obvious next step beyond the Seven is the inclusion of Yaqb al-ara- [iii/224]
m who, next to them, is the most renowned writer. Various writers on the
Seven considered him worth a monograph, a mufrada; they were al-Dn,
121
Ibn Shuray,
122
and Ibn al-Fam;
123
the latter work was versified by Ab
ayyn al-Andalus (d. 745/1344)
124
in his Ghyat al-malb f qirat Yaqb
(corresponding to his Iqd al-lal (see above, p. 560)).
The oldest surviving work on the Eight that we have is the above-men-
tioned Tadhkira (p. 496) of Ab l-asan hir b. Ghalbn (d. 399/1009),
125
modelled on the Taysr; from about the same time originate the K. al-Wajz
of Ab Al al-asan AL-AHWZ (d. 446/1055)
126
and a work of AB AL-
116
Printed in the margin of the Mukarrar (see above, p. 557sq.).
117
Ibn al-Jazar, Nashr, vol. 1, p. 82.; Brockelmann, GAL, suppl. vol. 1, p. 723.
118
Pretzls Verzeichnis, no. 11, pp. 2829; Sezgin, GAS, vol. 9, p. 60, no. 45, etc.
119
Ibn al-Jazar, abaqt, no. 376 (vol. 1, p. 83, l 11), cf. Nashr, vol. 1, p. 87. As a comparison to
the number of uruq (singular arq) might serve the eight hundred uruq for the Sevenbut
it is doubtful that the method of his count is identical.
120
121
Pretzls Verzeichnis, no. 34, pp. 4546; Ms. Istanbul, Nuruosmaniye Ktphanesi, no. 62.
122
Ms Damascus, Dr al-Kutub al-hiriyya, qirt, 67.
123
Pretzls Verzeichnis, no. 35, p. 46: Mufradat Yaqb or Mufradat Ibn al-Fam.
124
Ibn al-Jazar, Nashr vol. 1, p. 95, l 3.
125
Pretzls Verzeichnis, no. 16, pp. 3031; Sezgin, GAS, vol. 1, p. 16, no. 23.
126
Ibid., no. 18.
ASAN AL-RZ Al b. Jafar AL-SAD al-Shrz (with ilal) utilized in the
M f wujh al-qira of Ab Abd Allh Nar b. Al AL-FRIS.
127
Some-
what later appeared the Talkh of [AB] MASHAR (Abd al-Karm) AL-
ABAR (d. 478/1085).
128
It must be mentioned here that the Fihrist knows
of an earlier qirt work of this type where not Yaqb al-aram
129
is the
eighth reader but Khalaf b. Hishm (al-Bazzr).
Whereas in the Maghreb, in Egypt, and also in Syria, the Seven or Eight [iii/225]
readings were widespread, the East produced a greater number of works
on the qirt of the Ten. The earliest mentioned in Ibn al-Jazars Nashr,
vol. 1, p. 88, is the K. al-Ghya f l-qirt al-ashr of Ab Bakr Amad b.
al-usayn IBN MIHRN (d. 381/991) on which Ab l-asan Al b. Muam-
mad AL-FRIS
130
(d. 461/1069) wrote a commentary.
131
About the same time
appeared the Ishrat al-mubtad wa-tadhkirat al-muntah
132
of Ab Nar
MANR IBN AMAD AL-IRQ.
133
According to Ibn al-Jazar, the Irshd
al-mubtad wa-tadhkirat al-muntah
134
of Ab l-Izz Muammad b. Bundr
AL-QALNIS (d. 521/1127) was equally widespread in the East as was the
Taysr inthe West.
135
It was anexcerpt fromthe authors muchlarger work, al-
Kifya al-kubr f l-qirt al-ashr,
136
and (like the Shibiyya) it was repeat-
edly versified.
137
Of less importance was the Jmi of Ab l-usayn Nar b.
Abd al-Azz AL-FRIS al-Shrz (d. 461/1068).
138
On the other hand, the
following three works must be recognized as belonging to the most note-
worthy texts of the Koranic variant readings: (1) Ghyat al-ikhtir of Ab
l-Al al-asan AL-AR AL-HAMADHN (d. 569/1173), the al-Dn of
127
Ibid., no. 19, introduction.
128
Ahlwardt, Verzeichnis.
129
Flgels edition, p. 39, l 6. Here the author is named Ab l-asan b. Murra al-Naqqsh,
this one, however, is likely tobe identical withthe one of Ibnal-Jazar, abaqt, no. 3181, called
Muammad b. Abd Allh b. Ab Murra d. 352/963, who in any case specialized in the riwya
of Khalaf b. Hishm.
130
EQ; EI
2
.
131
Incomplete Ms., Cairo, Egyptian Library, Taymr Pasha, no. 344. Brockelmann, GAL,
suppl. vol. 1, p. 722.
132
Pretzls Verzeichnis der handschriftlich erhaltenen lteren Qirtwerke, no. 21, p. 35.
Excerpts from it are mentioned by Ibn al-Jazar, abaqt, no. 1545.
133
Brockelmann, GAL, suppl. vol. 2, p. 721.
134
Pretzls Verzeichnis, no. 28, p. 40. Another Ms. of the work is located in Istanbul, Serail,
Sultan Ahmet III, no. 169; further, Damascus, Dr al-Kutub al-hiriyya, Qirt, no. 27.
135
abaqt, no. 2958.
136
Pretzls Verzeichnis, no. 27, pp. 3940.
137
See Ibn al-Jazar, abaqt, nos. 774 and 1805 (vol. 1, p. 430, l 6), and no. 2352.
138
Brockelmann, GAL, suppl. 1, p. 722: Jmi bi-qirt al-aimma al-ashara.
the East;
139
(2) al-Ikhtiyr of Ab Amr Abd Allh b. Al SIB AL-KHAYY
(d. 541/1146),
140
and (3) al-Mib al-zhir of Ab l-Karam al-Mubrak AL-
SHAHRAZR al-Baghdd (d. 550/1155).
141
The greatest dissemination after
the Irshd belonged probably to the K. al-Mustanr of Ab hir Amad b.
Al IBN SIWR AL-BAGHDD (d. 496/1102), at least until it was eclipsed
142
by Ibn al-Jazars Nashr,
143
which in its form represents a high point in the
science of qirt. In contrast to his great classic predecessors, Ibn al-Jazar
is no longer interested inthe inner justificationof the differences of the vari-
ant readings (tall), the same way that he has almost no connectionwiththe
grammarians. He is concerned only with the fact (ad) which he is trying
critically to crystallize from a large number of earlier workshe mentions
some sixty of themand then presents themwith an unsurpassed clarity of
disposition and precise expression. The fact that he dealt with Ten readings
vis--vis the almost generally accepted Seven obliged him to introduce the
development of the Koranic variant readings with a historical survey and to
take positions on principles. Drawing largely on older sources, particularly
the Ibna of Makk b. Ab lib (d. 437/1045), he develops the conceptions of
the canonical (aa and mutawtira) as well as the uncanonical (shdhdh,
pl. shawdhdh) readings to a certain universality.
144
Ibn al-Jazar, too, limits
himself in his work in general to two transmitters of each imm. His book,
however, is not unessentially enriched by considering far more uruq.
145
This
becomes quite evident in the case of the reading of Warsh, where the arq
of AB BAKR Muammad b. Abd al-Ram b. Shabb AL-IBAHN devi-
ates considerably from the arq of AB YAQB Ysuf b. Amr AL-AZRAQ
(alone considered in the Taysr). Modelled on earlier works,
146
his chapter
on ul is preceded by a brief survey on tajwd. In other ways he follows
the main lines of the presentation of the ul as known from works on the
139
Ibnal-Jazar, abaqt, 945: wa-ind annahuf l-mashriqka-Ab Amr al-Dn f l-magh-
rib.
140
Ibn al-Jazar, abaqt, vol. 1, p. 435, l 10, he mentions a versification of the systemof the
Ten.
141
142
Pretzls Verzeichnis, no. 22, pp. 3536; no. 24, pp. 3738; no. 25, p. 38; andno. 26, pp. 3839;
143
The full title reads: Kitb al-Nashr f l-qirt al-ashr. Cf. Brockelmann, GAL, vol. 2,
p. 201. The work was very carefully edited by Muammad Amad Duhmn of Damascus, who
produced a very accurate edition (Damascus, 1345/1927, 2 v., 504, 458 p.), fromexcellent Mss.,
one of which was used by the author himself, and supplied with his corrections.
144
See above, p. 471 sqq.
145
146
Seven. A novelty is a large chapter on the imla in the aruf al-hij at the
opening of the sraswhich is otherwise treated as farsh al-urfas well
as a very important and practical chapter entitled bayn irfd al-qirt wa-
jamih, where the insufficiency of mixing riwyt and uruq in the course
of the reading is treated (see above, p. 527sq.). Much of farsh al-urf was
already discussed in the ul, so that it could be considerably abbreviated.
147
Next to the Ten, al-Amash is to be found in K. al-Rawa f l-qirt al-id [iii/227]
ashra of AB AL AL-ASAN (Ibn Muammad b. Ibrhm al-Baghdd)
AL-MLIK
148
(d. 438/1046), whereas the above-mentioned Sib al-Khayy
(d. 541/1146) discusses the Seveninhis book K. al-Mubhij
149
with the excep-
tion of Ab Jafar (IBN AL-QAQ AL-MAKHZM)all together with Ibn
Muayin, al-Amash as well as Khalaf b. Hishm (al-Bazzr) and Ab Mu-
ammad al-Yazd.
It was only much later that the collections of Fourteen readings attained
greater importance. The oldest one that we know of is al-rumz wa-
mafti al-kunz of Muammad b. Khall AL-QUBQIB (Qabqib).
150
His
source for the Ten is Ibn al-Jazars al-Nashr, and for the remaining Four a
mufradt work of al-Ahwz (d. 446/1054). The largest work is K. Laif al-
ishrt li-funn al-qirt of the renowned Koranic commentator Amad b.
Muammad AL-QASALLN (d. 923/1517),
151
a work which as far as com-
prehensiveness is concerned nearly surpasses Ibn al-Jazars al-Nashr. An
excerpt of this is Itf fual al-bashar f qirt al-arbaat ashar by Amad
b. Muammad AL-DIMY AL-BANN (d. 1117/1705).
152
Still larger and ear-
lier collections have been discussed in different connections (p. 487sqq.).
147
The editor unfortunately failed to facilitate the location of many of the authors incom-
plete references by supplying page numbers.Later literature on al-Nashr in Brockelmann,
GAL, vol. 2, p. 201Apart fromthe versifications of the systemof the Ten (mentioned, above,
p. 562 n. 137) the likely most widespread system is Zubdat al-irfn by al-Blaw (see above,
p. 557 n. 87).
148
Pretzls Verzeichnis der handschriftlich erhaltenen lteren Qirtwerke, no. 29, pp.
4041.
149
Ibid., no. 30, pp. 4143.
150
Brockelmann, GAL, vol. 2, p. 113; suppl. vol. 2, p. 139; Ahlwardt, Verzeichnis, no. 669;
apparently the author hadalready supplieda short survey of the Fourteeninhis book, Majma
al-surr.
151
Brockelmann, GAL, vol. 2, p. 73; in addition the excellent Ms: Istanbul, Fatih Camii
Ktphanesi, nos. 32 and 33; Damascus, Dr al-Kutub al-hiriyya, Qirt, no. 6, further,
Cairo, Maktabat Qawala, Qirt, 1.
152
Printed at Constantinople, 1285/1868, and at Cairo, 1317/1899, and in 1407/1987.
The Literature of the Uncanonical Readings (shawdhdh)
Parallel with the presentation of complete readings of the canonically rec- [iii/228]
ognized imms, the transmission of variant readings that were eliminated
as uncanonical were studied until quite late. This is based on the fact that
the differentiation between mashhra and shdhdh
153
did not constitute an
absolute rejection of the latter. For practical purposes it was excluded from
the Koranic reading, while it was unconditionally retained in tafsr as the
discussible heritage of tradition. In reality also the literature of shawdhdh
begins withIbnMujhid(d. 324/936),
154
the manwho establishedthe system
of the Seven. He contrasted his K. al-Saba with a K. al-Shawdhdh, which
is no longer extant. Ibn Jinn (d. 324/1002) added a grammatic-lexical com-
mentary,
155
analogous to Ab Al al-Friss ujja, which purports to serve
as an equivalent work to K. al-Saba (see above, p. 550) but with the excep-
tion that Ibn Jinn did not stick as closely to the original as did Ab Al
al-Fris. Rather, as he himself states in the introduction, he selected from
Ibn Mujhids book whatever appealed to him linguistically, and probably
also resorted to information from other sources.
To this period belongs also another work on shawdhdh, the Mukhtaar
f shawdhdh al-Qurn min kitb al-bad of Ibn Khlawayh (d. 370/980).
156
Subsequently, nearly all of the important scholars of the Koran composed
alsobooks onshawdhdh, apart frombooks onthe canonical readings. Thus,
al-Dn didsoinhis Mutaw.
157
Abl-Fal al-Rzs (d. 454/1062) al-Lawmi
153
Regarding the technical terms see above, pp. 487 and 502.
154
If a Kitb al-Shawdhdh has been transmitted already from his teacher, Amad b.
Yay THALAB d. 291/903 (Yqt, Irshd, vol. 2, p. 152, l 19) [Sezgin, GAS, vol. 9, pp. 140142]
shdhdha cannot have been used in the precise meaning of being outside the Seven rather
in the meaning of a reading that is contrary to muaf and the Arabiyya. The same applies to
kitb al-Shawdhdh that his opponent Ibn Shannabdh d. 328/939 composed (Yqt, Irshd,
vol. 6, p. 302, l 2).
155
G. Bergstrer, Nichtkanonische Koranlesearten, The Korankommission of the Bay-
erischen Akademie der Wissenschaften recently added to its collection a valuable pho-
graphic copy of a very long Ms. from Medina. A complete edition of the entire work is in
preparation in Egypt.
156
There is an explicit reference to kitb al-Bad of Ibn Khlawayh in Yqt, Irshd, vol. 4,
p. 5, l 16. It cannot be determined whether here bad is used in the sense of rhetorical
figure of speech or metaphors after the occurrence of the identical title of (Abd Allh) IBN
AL-MUTAZZ d. 296/908. [EI
2
; Sezgin, GAS, vol. 2, pp. 569571,] edited by G. Bergstrer in
1934.
157
I could not make out whether this book is identical with Brockelmanns (GAL, vol. 1,
p. 407) kitb al-Tarf, Ms. Alger, 3672.
is a book that was frequently quoted even later.
158
Al-Ahwz (d. 446/1055)
discusses in his Jmi al-mashhr wa-l-shdhdh
159
both canonical and un-
canonical readings together; shawdhdh, alone, is treated in his al-M.
160
A fragment of the al-Taqrb wa-l-bayn of Abd al-Ramn AL-AFRW
(d. 636/1238) has survived;
161
as well as the entire work of one of the best
scholars of the Maghreb, Ab l-Baq AL-UKBAR, d. 616/1219,
162
Irb al-
qirt al-shdhdhah, and the Qurrat ayn al-qurr of AB ISQ Ibrhm
b. Muammad b. Al AL-QAWWS AL-MARAND, a pupil of a pupil of Ab
l-Al L-AR AL-HAMADHN (d. 569/1173).
163
The latter two works
contain infinitely more material than the two older works edited by Berg-
strer, Ibn Jinns Mutasab, and Ibn Khlawayhs Mukhtaar. In Qurrat
ayn al-qurr the author mentions, apart fromfamiliar sources, also the fol-
lowing four unknown works: al-Iqn f l-shawdhdh wa-l-ikhtiyrt of Ab
Al al-asan b. Al b. Ibrhmal-Hudhal al-Mir;
164
al-Muntah of Muam-
mad b. al-asan/al-usayn b. Bundr AL-QALNIS (d. 521/1127); al-Kf
of Al b. al-usayn AL-URAYTHTH, a pupil of al-Ahwz; and al-Minhj
of Umar b. afar. For the time being, however, the commentary on the
Koran, al-Bar al-mu of Abd Allh Muammad b. Ysuf b. ayyn, called
ABAYYN, must remain the principal source for shawdhdh (see below,
p. 578
165
).
Writings on the Individual Variant Readings (mufradt) [iii/230]
In contrast to the writings discussed up to now, which treated the variants
of different readers in the order of the Koran side by side, those are works
158
Ibn al-Jazar, abaqt, 1549; it is one of the main sources regarding details of shawdhdh
in the commentary of AB AYYN al-Andalus (see below, p. 578sq.). A specimen of its
richness is supplied by the author of al-Nashr, Ibn al-Jazar (vol. 1, p. 47, l 8) for the first sra.
159
abaqt, 1006, Ibn al-Jazar did not here mention the work but did so in his Nashr, vol. 1,
p. 34, l 19.
160
According tothe Qurrat aynal-qurr (see below) where it servedas a source. According
to this the two books, and tti, of al-Ahwz are works on the Seven.
161
162
Arthur Jeffery, Cairo, is preparing an edition of the work.
163
Ms. Escorial, 1337, according to the title page an authograph. The first pages in wrong
order and probably incomplete.
164
Not in the abaqt of Ibn al-Jazar. jj Khalfa mentions an Iqn f l-qirt al-
shdhdha of Ab l-asan AL-AHWZ d. 446/1055, adding wa-dhakaral-Jabar annahuli-Ab
l-Izz al-Qalnis. Yqt, Irshd (vol. 6, p. 427, l 1) knows of an iqn of al-Ahwz f id ashra
qira; regarding this, see above, p. 547 n. 23.
165
Published in Cairo (1328/1910) and, more recently, edited by .I. Sayyid in 1989.
that present the variants of a single reader separately. This kind of treat-
ment has as its origin the form of the nuskha, which has been frequently
mentioned previously, and for this reason is very well documented at the
earliest date.
166
The need to memorize widely differing strands of transmis-
sion later led to the composition of the special descriptions of individual
readers. For example, al-Dn (d. 444/1052) in his Tamhd,
167
Taqrb, and
jz
168
thus repeatedly discussed Nfi b. Abd al-Ramn al-Layths reading
under various aspects. Already Ibn Mujhid composed separate mufradt
for each of the Seven.
169
Many later writers of the science of qirt, like
Makk b. Ab lib (d. 437/1045),
170
Ab l-Al L-AR AL-HAMADHN
(d. 569/1193),
171
and Ab Shma (d. 665/1266) also did so.
172
Only two books
on the Seven readers have survived, al-Tahdhb of al-Dn
173
and the consid-
erably more comprehensive al-Kmil al-fard of AB MS Jafar b. Makk
b. Jafar AL-MAWIL (d. 717/1313).
174
From among the readers beyond the
Seven Yaqb al-aram in particular was discussed most frequently (see
above, p. 561). Ibn Mujhid produced special studies of the readings of both
the Prophet and Al b. Ab lib.
175
Ibn Shannabdh, too, wrote about the
latter two.
176
The second Seven were included in the Mufradt of al-Ahwz
(d. 446/1055).
177
jj Khalfa still knows of a mufradt (of unknown content
and size) of Muammad b. al-asan IBN MIQSAM (d. 354/965).
166
Fihrist, p. 31 sqq.
167
Ibn al-Jazar, abaqt, vol. 1, p. 505, l 7.
168
Pretzls Verzeichnis der handschriftlich erhaltenen lteren Qirt-Werke, no. 36, pp.
4647. Prof. A. Jeffery, Cairo, kindly made available to me a Maghrebi Ms. of another writing
of this author (of unknown title). Such works were common mainly in the Maghreb where
the reading of Nfi b. Abd al-Ramn al-Layth prevailed. It was also there that originated a
muchused, andfrequently commentedrajaz poem, al-Durar al-lawmi f al maqra al-imm
Nfi, of Al b. Muammad al-Barr (d. 730/1329). See Brockelmann, GAL, vol. 2, p. 248;
EI
2
.
169
Fihrist, p. 31.
170
Yqt, Irshd, vol. 7, p. 175, l 2 (Nfi!).
171
jj Khalfa see under mufradt.
172
jj Khalfa see under mufradt.
173
Pretzls Verzeichnis der Qirt-Werke, no. 33 (p. 45).
174
Ibid., no 37 (p. 47).
175
Yqt, Irshd, vol. 2, p. 118, l 6.
176
Ibid., vol. 6, p. 302, l 1.
177
See above, p. 564.
The Writings on Tajwd
The oldest composition on tajwdaccording to its own testimonyis the [iii/231]
poemof (Ab Muzimb. Khqn) AL-KHQN (d. 325/937).
178
In a par-
netic vein he exhorts to pleasant pronunciation (usn al-adthe expres-
sion tajwd is not yet in use) and lists its elements, mentioning at the same
time different ways of articulation, taqq, tartl, and adr. The entire atti-
tude of the poem is clear evidence that the teaching of the Koranic recita-
tion is derived fromthe soil of the cult of reciters (db amalat al-Qurn).
On the other hand, the abundance of the orthoepic-phonetic terminology
and the appearance of some rules emanating from grammar show that
for the concrete formation of the abstract requirement of pleasing pro-
nunciation the respective prescriptions of philology have been instituted.
The fusion of the two elements, fiqh and lugha, becomes clearly visible
also in the Riya of Ab Muammad MAKK IBN AB LIB (d. 437/1045)
who considers himself to be the first author of a work on tajwd.
179
Its first
part deals with the rules of order, the merit of the recitation of the Koran,
etc., while the second part is purely linguistic, concerned with the con-
sonants, their articulation and qualities, consonantal connections, dou-
bling of the consonants, and, in the final chapter, nn and tanwn in con-
textual pronunciation. But in the nearly contemporary work of AL-DN
(d. 444/1052), al-Tadd f ilm al-tajwd,
180
the parnetic part is omitted
it continues its own development
181
whereas the linguistic and phonetic
parts are somewhat enlarged. Here the propdeutic character of tajwd
becomes even more evident. The technical terms are explained; apart from
the consonants also the variations of the pronunciations of vowels, imla,
sukn of the pause, ishmm, and rawm are fully discussed. From a simi-
lar work of al-Ahwz (d. 446/1055) only remnants have survived.
182
From
another point of view, the same subject has already been studied and pub-
lished by Silvestre de Sacy
183
as F l-lan al-khaf, on hidden grammatical
178
Commentary by al-Dn (abaqt, vol. 1, p. 505, l 11); Brockelmann, GAL, vol. 1, p. 189;
179
See the introduction to Pretzls Verzeichnis der Qirt-Werke, no. 38 (p. 230).
180
Pretzls Verzeichnis, no. 39 (p. 233).
181
A number of works on the subject are listed by Ahlwardt in his Verzeichnis, vol. 1,
p. 244; best known, and in a printed edition, is Tibyn f db amalat al-Qurn by Yay
AL-NAWAW (d. 676/1277).
182
In quotations from Iqn of Ibn al-Bdhash (see above, p. 561) as well as of the M
(see below).
183
Silvestre de Sacy, Trait de la prononciation des lettres arabes, extrait du manuscrit
mistakes
184
of Ab l-Fal al-Rz (d. 454/1062),
185
the teacher of the Koran.
Arranged by articulation, the work deals with mistakes that ought to be
avoided in the individual consonants, both when standing alone (intervo-
calic) and whenincontact withone another. The arrangement of the matter
under the concept lan reminds one of al-Khqan, who requires famil-
iarity with lan so that it can be avoided. The origin of this requirement,
however, is to be found in a saying ascribed to Umar: taallam al-lan f
l-Qurn.
186
The anonymous sixth century Mi f l-tajwd at the Staatsbibliothek,
Berlin, no. 499 (written in 785/1383), is likely dependent upon the afore-
mentioned work as well as the presentation of al-Ahwz. Frequently based
on older sources, the work supplies a very thorough introduction to the
teaching of lan and then proceeds to the three chapters on consonants,
consonantal connections, and, much shorter, to vowels and the vowelless
state. The work concludes with an appendix of the types of recitation which
apparently derives entirely from al-Ahwz.
187
Somewhat more thanthree centuries later, andclearly as competiontoal- [iii/233]
Khqn, Alam al-Dn AL-SAKHW (d. 643/1245) treated tajwd in a poem
comprising sixty-four verses in kmil meter (rhyming with -n).
188
However,
the greatest influential editionof the subject comes fromthe last great repre-
sentative of the science of variant readings, Ibn al-Jazar (d. 833/1429)
189
with
his al-Muqaddima al-Jazariyya, a didactic poem of one hundred and seven
verses in rajab 2 meter. The author later returned twice to the subject,
namely in a special work, al-Tamhd, and in the section of al-Nashr based on
al-Tamhd. The arrangement of the Jazariyya is strongly reminiscent of that
arabe n 260 de la Bibliothque impriale, pp. 1058. The value of the text suffers from the
continual confusion of khff and qq.
184
Meant is inaccuracy of the pronunciation, compared to lan jal obvious linguistic
mistake, a true grammatical mistake. Lan also means way of speaking and melody.
185
186
Unfortunately I was unable to check the content of a similar work by (Al b. Jafar) AB
AL-ASANAL-RZ al-Sad entitled Kitb al-Tanbhall-lanal-jal wa-l-lanal-khaf. Ms.,
Istanbul, Vehbi Efendi, no. 40, fol. 44
r
51
v
.; [Sezgin, GAS, vol. 1, p. 17, no. 27].
187
He mentions as authorities: al-Khall b. Amad b. Umar [d. ca. 175/791; Sezgin, GAS,
vol. 2, p. 613,] Sbawayh, and his commentators, Mabramn al-Naw, [Fuat Sezgin, GAS,
vol. 9, pp. 8687,] and [Ab Sad al-asan b. Abd Allh] AL-SRF [EI
2
], Muammad b.
al-Mustanr QURUB, al-Mazn, [Ab Umar li b. Isq] AL-JARM [Sezgin, GAS, vol. 9,
pp. 7273], Ibn Durayd; al-Farr; Ibn Kaysn; Ibn Mujhid as well as individual readers of the
Seven.
188
189
Ibid., vol. 2, p. 201 sqq.
of al-Nashr. The system of the location and types of articulation is derived
from it, but from the section on the methods of recitation only some gen-
eralities on tajwd as transition to the rather badly arranged treatment of
consonants and connections of consonant, preceded by those sections that
consider tarqq. Under and are listed all the words with that occur in
the Koran: an integration of unrelated tajwd material from one of the spe-
cial works on the difference of the two sounds. It is followed by some verses
about madd, which was introduced to elementary teaching because of the
increasing interest in the more recent science of the variant readings, the
cases of prolongation, and the treatment of waqf. To tajwd belong only the
rules of the sound effect of the absolute initial sound (in double consonants
extensionalif) and final sound; they formthe end of the poemwhich, how-
ever, is preceded by the elements of the rules where in the Koran a pause
is obligatory, and where one must, may, or may not pause; and a detailed
study of two respective chapters of the rule of rasm, regarding cases when
two words are written together, and about such when the script in the fem-
inine reproduces the contextual form with -at instead of -ah.
Next to (al-Muqaddima) al-Jazariyya there still remains a mass of even [iii/234]
shorter and more elementary school booklets that have been composed in
Arabic andinother Islamic languages until the present. The most popular of
all is probably the Tufat al-afl (sixty-one verses) of Sulaymn b. usayn
AL-JUMZR composed in 1198/1783.
190
For the most modest needs people
were satisfied to present the rules of tajwd as required for the recitation
of the first sra separately. Already al-Jabar (d. 732/1331) thus composed
his al-Wia f tajwd al-Ftia. But also the more scholarly occupation
with tajwd beyond the numerous commentaries on the Jazariyya did not
stop with Ibn al-Jazar. Best known from later literature is al-Durr al-yatm
of Muammad b. Pr Al L-BIRKW (d. 981/1573)
191
which goes far beyond
al-Nashr as far as casuistry and consideration of differences of opinions is
concerned, although without references to the authorities.
Writings on Pause in Koranic Recitation (waqf)
Koranic recitation required a meaningful connection or separation of the
parts of speech which initially could have been learned from oral teaching,
but as a consequence of the many syntactic obscurities, there still remained
190
Y.E. Sarkis, Dictionnaire encyclopdique de bibliographie arabe, p. 708.
191
many controversial instances. Froma very early date they were thus treated
in special writings. Already the Fihrist
192
mentions amza b. abb al-Taym
and other older authorities.
193
The earliest of such extant writings is a work
from the second part of the third century by one Ab l-Abbs, attacking
al-Maqi wa-l-mabdi of Ab tim al-Sijistn (d. 250/864).
194
The most
important book of the early period, however, is al-waqf wa-l-ibtid of
Ab Bakr IBN AL-ANBR (d. 327 or 8/938 or 9).
195
It consists of two parts,
the first one dealing with the rules of the absolute pause, which otherwise
constitutes a subject of the ul in the general works on qirt (see above,
p. 540sq.). The second part deals with the relative pause seen from the syn-
tactic point of viewand contains extremely valuable considerations regard-
ing possible syntactic interpretations of readers and authors. He knows
two types of permissible pause: (1) al-Tmm huwa lladh yasun al-waqf
alayhi wa-l-ibtid bi-m badah wa-l yakn badah m yataallaq bih; and
(2) al-asan huwa lladh yasun al-waqf alayh wa-l yasun al-ibtid bi-m
badah. Awaqf that is neither tmmnor asan is called qab; this, for exam-
ple, includes the separation of the muf from the muf ilayh, and of the
mant from the nat.
Another important work of this type on al-Waqf wa-l-itinf originates [iii/235]
from the grammarian Ab Jafar Amad b. Muammad AL-NAS (d.
338/950);
196
like most of the following works, it is without a section on the
absolute pause, but instead is richer with respect to syntactic and exegetic
discussions. From this, and the booklet of Amad b. Muammad IBN AWS
(d. ca. 340/951),
197
another and older classification of waqf becomes appar-
ent: tmm, kf, and asan, which is already documented in a saying of a
contemporary of Ab anfa (d. 150/767) (in Manr al-hud [see below]
p. 4, l 14 [sic]), according to which this opinion is considered bida. It is still
maintained in later works, such as al-Muktaf of al-Dn (d. 444/1052), and
192
Flgels edition, p. 36; in Manr al-hud (printed edition, 1307/1889, see below), p. 4, l 5,
already even Nfi and Yaqb al-aram.
193
Ibid.
194
British Museum, Arabic, 1589, statement wrong. See on this, Pretzls Verzeichnis,
pp. 236237: The anonymous work on waqf in the Ms. Ar., 1589 of the British Museum
Catalogue (Cod. Orient. Mus. Brit. ii, partis suppl. MDLXXXIX, p. 718) identified as the work of
Ibn al-Anbr cannot be thus identified on the basis of the introduction. The author attacks
Ab l-tim al-Sijistn, the author of a Kitb al-Maqi wa-l-mabdi. He seems to be one
Ab l-Abbs, a pupil of Khalaf b. Hishm, Shuray b. Ynus, and Ibn Muammad al-r,
thus still belonging to the second half of the third century.
195
Pretzls Verzeichnis, no. 45, pp. 234237; Sezgin, GAS, vol. 1, p. 15, no. 18.
196
Pretzls Verzeichnis, no. 46, p. 237; EI
2
197
Ibid., no. 47, pp. 237238: Waqf wa-l-ibtid; Sezgin, GAS, vol. 1, p. 15, no. 18.
in Rawat al-nir of Amad b. Ysuf al-Kawwsh (d. 680/1281).
198
In this
classification kf is defined as munqai f l-laf mutaalliq f l-man. [Ab
Muammad] al-asan b. Al b. Sad al-Ummn [AL-AMMN]
199
(lived
after 550/1155 in Egypt) composed two books on waqf and ibtid,
200
one of
which, al-Murshid, was later expanded by Ab Yay Zakariyy AL-ANR
(d. 926/1519) in his al-Muqidli-talkh mf l-Murshidf l-waqf wa-l-ibtid.
201
Like Ibn al-Anbrs al-waqf wa-l-ibtid it includes also the rules of the
absolute pause, and, following Ab tim al-Sijistn, divides the relative
pause into: tmm, asan, kf, li, and mafhm.
202
We find nearly the iden-
tical arrangement in a larger but later work of Amad b. Muammad b. Abd
al-Karm AL-USHMN,
203
entitled Manr al-hud, with the exception that
here mafhm has been omitted, and the remaining parts subdivided into
tmm atamm, asan asan, etc. The arrangement of Muammad b. ayfr
AL-SAJWAND (died about the middle of the sixth century),
204
who was
the author of a larger and a smaller work on waqf and ibtid found more
widespread acceptance.
205
He differentiates:
1. lzim (indicated by the logogram ): m law wuila arafh ghuyyira
al-murd.
206
2. al-mulaq (): m yasun al-ibtid bi-m badah.
3. al-jiz (): m yajzu fhi l-wal wa-l-fal li-tajdhub al-mujbiayn min
[iii/237]
al-arafayn.
4. al-mujawwaz li-wajh (): still to justify.
5. al-murakhkha arratan li-nqi al-nafas wa-l al-kalm ( =
arr.)
He needs two additional signs: = qad qla it is also claimed (that here
pause is required) and at the end of a verse, which purports to forbid the
198
Ms. Berlin, 563; Ibn al-Jazar, abaqt, 701; Brockelmann, GAL, vol. 1, p. 416, suppl. vol. 1,
p. 737; EI
2
.
199
200
201
Ibid.: wa-zaama annahu tabia Ab timal-Sijistn.
202
The meaning of al-mafhm is defined as the last stage before qab: alladh l yufham
min-hu l-murd, approximately still intelligible.
203
Repeatedly printed, most recently in 1307/1889 at Cairo; cf. Y.I. Sarks, Dictionnaire de
bibliographie arabe, 452.
204
Brockelmann, GAL, suppl. volume 2, p. 724.
205
206
For example, m bi-muminna | yukhdina Allh. If pause is ignored, yukhdina
would be considered ifa to muminna.
obvious pause. This system, too, was subject to theoretical and practical
extensions, although in later times it received a kind of universal accep-
tance. Basically also the waqf indication of the official Cairene edition of
the Koran follows suit.
207
Retained are the signs , , ; omitted are , ,
, but instead, sub-sections of jiz are newly introduced: to indicate
the position where waqf is permissible, but wal is preferable, to indicate
positions where waqf is permissible and preferable to wal. Further, a pair of
double signs was introduced ^^ for the very few cases where the syntactic
affiliation of a word is doubtful and where waqf on the one word excludes
the same on the other, for example, in l rayba fihi hudan, sra 2:2; fhi can
belong to rayba and have waqf, but also to hudan and then the waqf is on
rayba.
Writings on the Enumeration of Verses
The numbering of the verses of the Koran was actually subject to few varia-
tions. Already inthe second century the following sevensystems of enumer-
ation were established:
208
al-madan l-awwal, with 6217 verses, al-madan
al-akhr, with 6214 verses, al-Makk, 6219, al-Bar, 6204 (5), al-Kf, 6236,
al-Shm, 6226 (7), al-im, 6232. The differences among these tallies are
unevenly distributed over the sras. Whereas twenty-eight sras show no
differences, for sra twenty alone there are no less than twenty-four differ-
ences, andfor sra fifty-six there are seventeenor sixteen. Since the enumer-
ationof the verses is theoretically not connected to the teaching of waqf, but
frequently of practical importance, and also in the teaching of the imla the
end of the verse is occasionally of practical importance, interest in the the-
oretical aspect of verse-counting has always remained alive in the teaching
of the Koran. The introductory chapters of two extant works make it quite
clear that this originates from the requirements of the Koranic orthogra-
phy but not from those of the Koranic readings. Substantially, they at times
remarkably strongly resemble the minor Masorah of the Hebrew Bible.
Many authorities of the second century are mentioned as the first authors of
207
Cf. G. Bergstrer, Koranlesung in Kairo, p. 9.
208
The data are derived froma special investigation by Anton Spitaler, Mnchen, entitled
Die Verszhlung [verse-numbering] des Koran nach islamischer berlieferung (1935), a work
that he published under the auspices of the Korankommission of the Bayerische Akademie
der Wissenschaften. Numbers in brackets indicate a different enumeration in the transmis-
sion.
suchwritings.
209
Surviving works include al-Bayn of al-Dn (d. 444/1052),
210
Kitb f adad suwar wa-y al-Qurn of AB L-QSIM Umar b. Muam-
mad IBN ABD AL-KF
211
(an approximate contemporary of the preceding
writer), K. Adady al-Qurn of Ab af Umar b. Al b. Manr,
212
a pupil of
Ab Bakr AL-NAQQSH, and also K. Mubhij al-asrr of al-asan b. Amad
L-AR AL-HAMADHN (d. 569/1173).
213
There are other sources to be
found in general works on qirt that are useful for the investigation of
this subject, for example, Rawat al-uff of al-Muaddil (fl. 5th/11th cent.)
and Laif al-ishrt of al-Qasalln (d. 923/1517), as well as Itf fual al-
bashar of AL-DIMY AL-BANN.
Writings on Koranic Orthography
Familiarity with the old Koranic orthography was a far more important pre- [iii/238]
requisite for the teaching of the Koran than the verse-counting. Not only
actual textual variants, but also purely orthographic peculiarities thus deter-
mined the pronunciation, the imla in the absolute waqf and particularly
in the tashl of hamz in pause. For this reason, writings on the orthography
of the Uthmnic model codices were in use at all periods. The best known
of these teaching aids was al-Muqni f marifat rasam maif al-amr of
Uthmn b. Sad AL-DN (d. 444/1052).
214
In his introductory chapter he
deals with the history of the establishment of the text of the Koran, fol-
lowed by the peculiarities of the Uthmnic orthography compared with the
orthography current at the time of a particular author, and also older trans-
missions regarding the particularities of the amr, which are not of purely
orthographic nature but are rather textual variants. The main sources of the
book are older writings, including Hij al-sunna of AL-GHZ IBN QAYS
al-Andalus (d. 199/815)from which derive mainly the details regarding
the Medinan codicesand also Kitb f l-hij al-maif of Muammad b.
s al-Ibahn (d. 253/867), as well as the above-mentioned al-waqf
of Ab Bakr IBN AL-ANBR (d. 327/938). Al-Dns other authorities are
209
Fihrist, p. 37; on the later literature also Ahlwardt, Verzeichnis, vol. 1, p. 174.
210
Pretzls Verzeichnis, no. 50, pp. 239240: K. al-Bayn f add y al-Qurn.
211
Ibid., no. 51, pp. 240241; Sezgin, GAS, vol. 1, p. 16, no. 24.
212
Recently acquired Ms. by Staatsbibliothek, Berlin, Or. quart. 1386, in R. Sellheim, Mate-
rialien zur arabischen Literaturgeschichte, Teil 1. Wiesbaden, 1976.
213
Pretzls Verzeichnis, no. 52, p. 241.
214
Edited by O. Pretzl in 1932.
NUAYRb. Ysuf [AL-RZ
215
] andAbUbaydal-Qsimb. Sallm.
216
The for-
mer is considered one of the most important scholars in this field. He does
not seem to have been familiar with the older and still extant work, Kitb
al-Maif of the renowned traditionist (Abd Allh b. Sulaymn) IBN AB
DWD AL-SIJISTN (died 316/929).
217
It is distinguished from al-Muqni
f marifat rasm maif al-amr by extraordinarily numerous facts about
the textual variants of older Koranic authorities; it is very important for
the teaching of the shawdhdh. The work essentially produces the same
lists of the differences of the amr but adds valuable details about ortho-
graphic peculiarities, about the treatment of the codices, their sale, etc. Like
the Taysr, al-Shibs (d. 590/1193) Muqni was also versified in a poem of
three hundredandtenverses inawl meter entitledAqlat atrbal-qaibf
asm al-maqid, andfrequently commentedon.
218
Inlater times very wide-
spread, particularly inthe Maghreb, was the rajaz poemMawridal-amnof
Ab Abd Allh Muammad b. Muammad AL-KHARRZ() (d. 711/1311
219
).
Further literature is presented above (p. 408) and in the introduction to al-
Muqni. On the pointing of the Koran see below, p. 590sqq.
The Commentaries on the Koran as a
Source of the Science of Variant Readings
Apart from the works on qirt in the narrow sense, the commentaries [iii/240]
on the Koranunless they are exclusively devoted to grammatical, lexico-
graphic, and exegetic mattersrepresent the main source of the science
of qirt. This is particularly true in the case of the investigation of the
uncanonical variant readings. Since they were not treated in the second
part of the present work, and a larger number of them have been reprinted
since then, or manuscript versions been discovered, the following com-
pilation will be greatly appreciated. The details about printed commen-
taries have been kindly made available by Prof. Arthur Jeffery, Cairo, who
215
216
His book, Ikhtilf al-maif, is not mentioned.
217
Damascus, Dr al-Kutub al-hiriyya, adth, 407. Photocopies are with the Koran-
kommission of the Bayerische Akademie der Wissenschaften, Mnchen. The work has been
edited by A. Jeffery, and entitled Materials for the history of the text of the Qorn, at Leiden in
1937. [Sezgin, GAS vol. 1, p. 175.]
218
Brockelmann, GAL, vol. 1, p. 410; Ahlwardt, Verzeichnis, vol. 1, p. 192b; Y.I. Sarks, Dictio-
nnaire de bibliographie arabe, 1092. A more recent printed edition appeared in 1908 at Kazan
(accompanied by a more recent commentary).
219
Brockelmann, GAL, vol. 2, p. 248; Ibn al-Jazar, abaqt, 3394.
himself checked themsystematically in order later to use themfor a critical
edition of the Koran.
1. Jmi al-bayn f tafsr [or an tawl] al-Qurn of Ab Jafar Muam-
mad b. Jarr AL-ABAR (d. 310/923); available in two Egyptian printings,
Maymniyya, 1321/1893, and Amriyya, 1330/1911. The latter edition is printed
somewhat more carefully than the former; both, however, contain plenty of
mistakes, so that a new edition would seem to be a great desideratum. Her-
mann Hauleiter provided an index to the first edition.
220
Al-abar pretty much lists all the more important differences of the
Seven, but rarely provides names. He himself mostly prefers the reading
of IM b. al-Ajjj AL-JADAR, and rather more the riwya of Ab Bakr
SHUBAH IBN AYYSH
221
than the riwya of Ab Umar AF IBN SULAY-
MN.
222
Fromamong the uncanonical readings he frequently cites the texts
of Ubayy b. Kab, Abd Allh IBN MASD, and some of the first caliphs, as
well as Al and Ibn Abbs. As a rule, his uncanonical references are listed
without any names. Although his collection of such variants is interesting,
it is something less than a mine.
2. Malim al-tanzl, of Ab Muammad al-usayn al-Farr AL-BAGHA- [iii/241]
W (d. 516/1122).
223
This is available inthe Bombay editionof 1296/1878, andin
twoCairene editions, one inthe marginof al-Khzinal-Baghdds commen-
tary, printedinsevenvolumes by al-b, 13311332/1912, andthe other inthe
lower part of the Manr edition of (Isml b. Umar) IBN KATHRs
224
com-
mentary (see below, no. 9). The better edition is the one in the margin of
al-Khzin al-Baghdd.
Al-Baghaw nearly always supplies the main deviations of the canonical
Seven, and usually by name. He regularly also adds the names of Yaqb al-
aram andAbJafar IBNAL-QAQ AL-MAKHZM. Inaddition, he lists
the better known variants of Ibn Masd, Ubayy b. Kab, etc. Occasionally, he
also cites the variant readings of umaydb. Qays AL-ARAJ, AbRaj Imrn
al-Urid,
225
al-asan(al-Bar), IbnAb Isqal-aram,
226
andothers. He
does not supply the sources of his variants.
220
Register zumQorankommentar des abari (1912).
221
EI
2
222
EI
2
223
Or 510/1117 according to Sezgin, GAS, vol. 1, p. 1555, no. 2.
224
Tafsr al-Qurn al-am, printed in Mir, ca. 1356/1937, Cairo, 1952, Beirut, 1966 and
1386/1970. Cairo, 1978, Beirut, 1904/1986. Brockelmann, GAL, suppl. vol. 2, p. 49; Sezgin, GAS,
vol. 1, p. 49 and vol. 6, p. 303.
225
226
EQ; Sezgin, GAS, vol. 8, pp. 3637.
3. al-Kashshf an aqiq al-tanzl of Jr Allh (Mamd b. Umar) AL-
ZAMAKHSHAR (d. 538/1143) edited by Nassau Lees in two volumes, and
published at Calcutta, 18561859. There are also several Cairene editions.
The Calcutta edition is by far the best, although not always accurate.
Al-Zamakhshar is very arbitrary in listing variants. He by no means lists
all of the Seven, but he has very many uncanonical ones. He frequently lists
the representatives of the uncanonical readings, but rather frequently fails
to supply the source. From among his sources he mentions Ibn Jinn, Ibn
Khlawayh, and Ibn Mujhid.
4. Maft al-ghayb of Muammad FAKHR AL-DN AL-RZ (d. 606/
1209). There are three Cairene printings, Blq, 12791289/18621872 in six
volumes, al-mira, 1310/1892 in eight volumes (reprinted, 13241327/1906
1909), al-usayniyya, 1327/1909, in eight volumes, and Constantinople, 1307/
1889, in eight volumes, with Irshd al-aql al-salm il mazy al-kitb al-
karm of AB AL-SUD (Muammad b. Muammad) AL-IMD (Khoja
elebi) (d. 982/1574) in the margin (see below, item no. 14).
227
Fakhr al-Dn al-Rz is very inconsistent in his treatment of the variant
readings. He sometimes lists pretty completely canonical and uncanonical
variants, then passes entire sections without even once mentioning a sin-
gle variant. He largely copies from al-Zamakhshar (or possibly the latters
sources), but nowand then he does list important variants that are not to be
foundinal-Zamakhshar. He does not supply the sources of his readings, but
when discussing them he presents the views of al-Zamakhshar and other
authorities.
5. Iml m manna bi-hi l-Ramn min wujh al-irb wa-l-qirt f jam [iii/242]
al-Qurn of Ab l-Baq AL-UKBAR (d. 616/1219). There are several printed
editions: Cairo, Sharaf, 1303/1885, in two parts; and Maymniyya, 1306/1888,
in two volumes; in the margin of the super-commentary of al-Jamal (on the
Jallayn), Tehran, 1860, and from it printed at Cairo, Taqaddum, 1348/1929,
in four volumes.
Al-Ukbar is rich in uncanonical readings, but unfortunately he seldom
identifies the reader and never the sources fromwhich he derived his infor-
mation.
6. Anwr al-tanzl wa-asrr al-tawl of Ab Sad (Abd Allh) b. Umar
AL-BAYW (d. 685/1282). Among the numerous printings, the best edition
is the one by H.L. Fleischer in two volumes, Leipzig, 18461848. Fromamong
227
EI
2
.
the Oriental editions it is the five-volume al-alab edition of 1330/1911, with
the supercommentary of al-Kzarn.
228
Al-Bayw is of course greatly dependent on al-Zamakhshar, even if
he does not supply all the variants of the latter. Not infrequently he cites
variant readings found in al-Ukbar but not in either al-Zamakhshar or
Fakhr al-Dn al-Rz. Occasionally he produces a variant that cannot even
be found in al-Ukbar. Still, his material for the qirt cannot exactly be
called rich. Apart from the Seven he likes to refer to Yaqb al-aram.
7. Madrik al-tanzl wa-aqiq al-tawl of (fi al-Dn) Ab l-Barakt
AL-NASAF (d. 710/1311).
229
There is a Bombay edition of 1279/1862, several
two-volume editions fromEgypt, 1306/1888 and 1326/1908 (Sada); the work
further appears in the margin of some editions of the commentary of al-
Khzin al-Baghdd. Reprint of the Sada edition in four parts, in 1333/1914.
This is a very brief commentary, and generally lists only the main variants
of the Seven, occasionally also an uncanonical variant, at times with the
name of the reader.
8. al-Bar al-mu of AB AYYN al-Naw L-ANDALUS (d. 745/ [iii/243]
1344), printed in eight volumes at Cairo, 1328/1910, and paid for by the Sultan
of Morocco. In the margin two smaller commentaries are printed, al-Nahr
al-mdd of Ab ayyn himself, and al-Durr al-laq min al-bar al-mu of
his pupil al-Qays.
230
Ab ayyn represents the western tradition of al-Andalus even though
he had also studied in Egypt and Mecca. His work is extraordinarily rich
in uncanonical readings, which in most cases he discusses in detail. A
considerable number of the variants that he lists is totally unknown to
the aforementioned commentaries. He is unique in the way he cites his
sources. In his introduction he refers to the Iqn al-zhir of Ab Jafar
IBN AL-BDHASH as the best authority on the Seven, and to al-Mib
of al-Shahrazr as the best on the Ten. Throughout he cites al-Kmil of
(AB AL-QSIM) AL-HUDHAL, Kitb al-Tarr, K. al-Rawa f l-qirt al-
id ashra of Ab Al al-asan b. Muammad b. Ibrhm AL-BAGHDD
(d. 458/1066),
231
al-Kashshf of al-Zamakhshar, Irb al-shawdhdh of IBN
KHLAWAYH, K. al-Tibyn of Ab l-Fat al-Hamadhn, al-Ayn of al-Khall
b. Amad al-Farhd, K. al-Lawmi f shawdhdh al-qirt of Fakhr al-Dn
228
EI
2
.
229
EI
2
.
230
The text is badly printed, particularly the seventh and eigth volumes which were obvi-
ously completed in haste. There are not only many printing mistakes but obvious omissions.
231
Brockelmann, GAL, suppl. vol. 1, p. 721, no. 4c.
al-Rz, al-Kmil of Ab l-Qsim Ysuf b. Al b. Jabbra al-HUDHAYL
(d. 465/1072), and K. al-Mukam wa-l-mu al-aam f l-lugha of (Al b.
Isml) IBNSDAH(d. 458/1066).
232
In addition, he refers to source material
of Ab Ubayd al-Qsim b. Sallm, Ab tim al-Sijistn, Ab Al al-Fris,
al-abar, Ab l-Baq AL-UKBAR, [Ab l-Abbs] AMAD IBN MMR
AL-MAHDAW, Ibn Aiyya (al-Muhrib), al-Dn (Muammad b. Amad)
AL-QURUB, Makk b. Ab lib, Ab Isq Ibrhm b. al-Sar AL-ZAJJJ
(d. 311/923),
233
al-Qsim b. Firruh AL-SHIB, Ab Al AL-AHWZ, al-
Mubarrad, Ibn Qutayba,
234
Ab l-Qsim Abd al-Karm AL-QUSHAYR,
235
et
al.
In the marginal edition of al-Nahr occasionally an obscurity of the main
text is eliminated, at times also printing mistakes are corrected. In this
respect al-Durr is of little use.
9. Tafsr [al-Qurn al-am] of Ab l-Fid Isml b. Umar IBN KATHR [iii/244]
(d. 774/1372). It has beenprintedinthe marginof Fatal-baynf maqidal-
Qurn of [Muammad Bashr al-Dn Uthmn] AL-QANNJ (Bulaq,
1300/18821302/1884 in 10 volumes); also, in the margin of al-Baghaw (Cairo,
Manr, 1347/1928). The former edition is by far better.
Isml b. Umar IBNKATHRis very arbitrary whenit comes to references
to the variants. On one occasion he supplies a great number of uncanonical
variants, whereas, to ones great surprise, in other cases even the readings of
the Seven are totally missing. It is rare that he lists a variant that cannot
also be found in Ab ayyn (al-Andalus) or al-Ukbar, yet at times he
cites a reader on a variant that cannot be found in other commentaries.
The variants are usually without substantiation. Still, now and then he
refers to sources such as al-Zamakhshar, al-Dn, al-Qurub, Ibn Aiyya
(al-Murib), Ab Bakr Abd Allh IBN AB DWD AL-SIJISTN (Ab
Bakr Amadb. Ms) IBNMARDAWAYH,
236
AbUbaydal-Qsimb. Sallms
Fail al-Qurn, etc.
10. Gharib al-Qurnwa-raghib al-furqn of al-Qumm AL-NAYSBR
(died ca. 706/1306). It is printed in the margin of al-abar (no. 1). It was also
printed twice at Tehran, 1280/1863 and 1313/1895; cf. above, p. 407 n. 130, for
a more recent edition.
The author supplies his own introduction to the qirt. In the introduc-
tionhe makes it clear that as far as the variants are concernedhe is following
232
EI
2
; EQ; Sezgin, GAS, vol. 8, p. 5; Cairo edition, 1377/1958.
233
Ibid., vol. 9, pp. 8182.
234
EI
2
.
235
EI
2
.
236
Died in 410/1019; Sezgin, GAS, vol. 1, p. 225.
the canonical ones, of which he recognizes the Ten, except in particular
instances. It is a useful work in so far as the author occasionally refers to
minor deviations among the Ten (from unknown riwyt). He sometimes
quotes uncanonical readers whenthey conformwiththe canonical ones. He
himself mentions as his main source the great tafsr of Fakhr al-Dn al-Rz,
Maft al-ghayb (see no. 4).
11. Tafsr al-Jallayn, begun by Jall al-Dn al-Maall (d. 864/1459), and
completed by Jall al-Dn AL-SUY (d. 911/1505). There are innumerable
editions, with and without super-commentary. The work is not rich as far as
variants are concerned, since it mentions only the most common ones, and
without supplying sources.
12. al-Durr al-manthr f l-tafsr bi-l-mathr of al-Suy, printed in six
volumes, Cairo, 1314/1896.
The author supplies a wealth of variants but rarely those that are not [iii/245]
already known fromolder works. The great advantage of the work is that he
lists the isnds of the variants and refers to a large number of earlier author-
ities, such as Ibn al-Anbr, Ab Dwd (al-Sijistn), al-Tirmidh, al-Wak
b. al-Jarr, Ab Ubayd al-Qsim b. Sallm, Ab Abd Allh AL-FIRYB,
d. 212/827,
237
al-Khab [al-Tibrz?], Abd b. umayd (or amd) b. Nar al-
Kiss (d. 249/863),
238
Amad b. Muammad al-THALAB, et al. al-Suy is
sometimes quite useful when it comes to ascertaining vague statements in
older commentaries.
13. al-Sirj al-munr f l-ina al marifat ba man kalm rabbin al-
akm of [Ysuf b. Muammad] al-Khab AL-SHIRBN
239
(d. 977/1569),
published Cairo, 1311/1893 in four volumes, together with al-Bayw in the
margin; previous editions in 1285/1865 and 1299/1881.
The work seldom ventures beyond the more important variants of the
Seven; the quotations are supplied always without the source.
14. Irshd al-aql al-salm il mazy al-kitb al-karm of AB AL-SUD
(Muammad b. Muy l-Dn [sic]) AL-IMD (d. 982/1574), Bulaq editions,
1275/1858 and 1285/1868, in two volumes, also in the margin of the two Cairo
editions of Fakhr al-Dn al-Rz.
The work is mainly based on al-Zamakhshar and al-Bayw, and quotes
variant readings, but without specifying the readers.
237
Muammad b. Ysuf AL-FIRYB or al-Faryb, 120/738212/827; EI
2
; EQ; Sezgin, GAS,
vol. 1, p. 40.
238
239
EI
2
.
15. Inyat al-q wa-kifyat al-r al tafsr al-Bayw, known as -
shiyat al-Shihb of Shihbal-DnAL-KHAFJ (d. 1069/1658),
240
eight parts in
four volumes, with the tafsr of al-Bayw (Anwr al-tanzl) in the margin,
Blq, 1283/1866, a work that was reprinted in about 1976 in Beirut.
It is rich in uncanonical variants. Although it is a compilation, it is useful
because of the great care with which the author presents his material, thus
facilitating the verification of doubtful passages in other works. Occasion-
ally he supplies his sources, al-Zamakhshar, Ibn Jinn (Mutasab), al-Dn,
Ibn al-Jazar (al-Nashr), Ab ayyn al-Andalus, al-Sajwand, Ab tim
al-Sijistn, etc.
16. Fat al-qadr al-jmi bayn fannay al-riwya wa-l-dirya min ilm al-
tafsr of Muammad b. Al AL-SHAWKN al-Yaman (d. 1250/1834), Cairo,
1349/1930, in five volumes.
241
This commentary, too, is a compilation from printed editions. The South
Arabian author obviously had access to a great deal of material that was
no longer accessible to the occidental writers. In his discussion of the vari-
ants he constantly quotes a number of authorities including Ab Ubayd al-
Qsim b. Sallm, Ab tim al-Sijistn, al-Zamakhshar, Ibn al-Anbr, al-
Tirmidh, al-Wak b. al-Jarr (his tafsr), Sad b. Manr,
242
Ibn Ab Dwd
al-Sijistn, Abdal-Razzqb. Hammm
243
(tafsr), al-Qurub, al-abar, Abd
b. umayd (or amd) b. Nar, etc.
17. R al-man f tafsr al-Qurn al-am wa-l-sab al-mathn of Ma- [iii/246]
md AL-ALS AL-BAGHDD (Bulaq, 1301/18831310/1892, in nine vol-
umes; new edition, Cairo, Munriyya, n.d., in 30 parts).
244
Acompilationof printed and extant manuscript commentaries. Very rich
in variants. His contributions from unusual sources are not very numerous.
18. I found the following Shite commentaries on the Koran still worth
noting: Majma al-bayn f tafsr al-Qurn (thus the title of the book in the
introduction of the lithograph is given thus: K. al-Majma al-bayn li-ulm
al-Qurn) of Ab Al al-Fal b. al-asan AL-TABARS (d. 548/1153, the work
was composed in 536/1141).
245
A lithograph, Tehran, 1275/1859, in two vol-
240
EI
2
.
241
EI
2
; Brockelmann, GAL, suppl. vol. 2, p. 819.
242
Ab Uthmn SD IBN MANR b. Shuba al-Khursn, d. 227/842. EI
2
; Sezgin, GAS,
vol. 1, p. 104, no. 38.
243
EI
2
; Juynboll, Encyclopedia, pp. xxxii, 2438; Sezgin, GAS, vol. 1, p. 99.
244
Other editions, Cairo, 1345/1926, 1353/1934 (repr. about 1970), and Cairo, 1964. EI
2
.
245
Brockelmann, GAL, vol. 1, p. 406; Sezgin, GAS, vols. 2 and 9. There are numerous modern
editions, the latest, Qum, 1403/1983; EI
2
.
umes of 560 and 536 unnumbered pages. Y.E. Sarks, Dictionnaire de bibli-
ographie arabe, column 1227, another lithograph 1314.
246
The commentary evidently follows the model of the ujja of Ab Al al-
Fris (see above, p. 552) which is frequently explicitly quoted. Following
this example, he clearly shows the subdivisions of the Koranic commen-
tary: qirt, ujaj, lugha, irb, man. But, in contrast, he considers apart
fromthe Seven (these partly also with more than the two familiar transmit-
ters) the readings of Ab Jafar IBN AL-QAQ AL-MAKHZM and Yaqb
al-aram, the Ikhtiyr of Khalaf b. Hishm and the one of Ab tim al-
Sijistn, without differentiating the riwyt. Among the qirt very many
shawdhdh are supplied. For explicit reasons, the work of Ibn Jinn (see
above, p. 565) is repeatedly quoted as well as al-Zajjj (see below, p. 584).
Al-Zajjj does not mention several of the shawdhdh readings found in Ibn
Jinn, but often still has considerably more than Ibn Jinn.
Other than al-ujja already mentioned above on p. 493, the following [iii/247]
manuscript commentaries constitute important source material for the his-
tory of the Koran.
1. K. Man al-Qurn of Ab Bakr Yay b. Ziyd b. Abd al-Ramn AL-
FARR (d. 207/822). The work is available intwo manuscripts that differ not
inconsiderably from one another: Istanbul (Bagdatl) Vehbi Effendi, no. 66
(some leaves are missing at the end), and Nuruosmaniye Ktphanesi,
459.
247
It might be a case of two different recensions of this work, pro-
duced by the author himself in order to frustrate the avarice of the book-
sellers.
248
But the difference between the two is not as great as the story
would have it.
The commentary is the most important source of information on the
Kfic recitationof the Koraninso far as it originates directly fromthe school
of al-Kis, and the tall text mostly offers, so to speak, a recension untrou-
bled by oral transmission. Other readings are unfortunately quoted fre-
quently without specific mention of the authorities. Conversely, Ibn Masd
and Ubayy b. Kab are frequently mentioned, so that this commentary con-
stitutes a considerably safer base for the establishment of their texts of the
Koranthanis the case withthe later sources that are frequently far removed.
246
It must be emphasized that here not all the printed or lithographed commentaries
listed by Sarks have been mentioned.
247
O. Pretzl, Die Wissenschaft der Koranlesung, p. 16.The manuscript of the Egyptian
National Library, Tafsr no. 10 is according to the fargh note a copy of the Nurosomaniye
manuscript.
248
See G. Flgel, Die grammatischen Schulen der Araber, p. 131.
The grammatical explanations of Yay b. Ziyd AL-FARR are highly
esteemed by the Arabs themselves as the most perfect example of what has
been achieved in the science of the Koran.
2. K. al-Man l-Qurn of Ab Isq Ibrhm b. Muammad b. al-Sar [iii/248]
AL-ZAJJJ (d. 316/928
249
), a pupil of the renowned al-Mubarrad. The work is
partly extant in a very poor manuscript at Istanbul, Umumiye, no. 247; the
Ms. VeliyeddinEfendi, 43, constitutes its continuation(ina better condition,
dated 368/978). The Egyptian National Library has a work of the same title
and author, al-Zajjj, signature Tafsr, 632, which I have not been able to see.
The work is rich in shawdhdh details, mostly without the names of the
readers.
3. K. Irb al-Qurn wa-tabyn m f-hi min al-naw wa-dhikr al-qirt of
Ab Jafar Amad b. Muammad b. Isml, known as AL-NAS (d. 338/
949). A very good manuscript at Istanbul, Umumiye, no. 245.
The work, like the above-mentioned (p. 510) book (al-Waqf wa-l-itinf)
by the same author, is on waqf and represents a very rich collection of older
grammarians of all schools onthe ujaj al-qira. It considers the shawdhdh
not always equally, but not infrequently it lists variant readings that are
otherwise totally unknown. It is particularly concerned with the differences
of the BaranandKfanschools. The author is a pupil of the aforementioned
[Ab Isq Ibrhm] AL-ZAJJJ and is very often quoted.
4. In the Egyptian National Library, Tafsr no. 385, there is a work by the
same author entitled Man l-Qurn.
249
EI
2
MANUSCRIPTS OF THE KORAN
The State of Manuscript Research [iii/249]
The close connection of the variant readings of the Koran with the Uth-
mnic text wouldlogically leadtothe conclusionthat a study of manuscripts
beyond the qirt literature is unlikely to produce anything new. How-
ever, as far as non-Uthmnic manuscripts are concerned, one would think
that we have sufficient information on the subject through the shawdhdh
works. As a matter of fact, at least from the fourth century ah onwards,
the manuscripts of the Koran as such no longer played an important role.
The expertise in orthographic peculiarities of the solely recognized Uth-
mnic recension required for the practice of Koranic variant readings, as
well as the production of manuscripts of the Koran, could be derived from
the above-mentioned secondary sources (p. 512sq.). For a quick orienta-
tion even Occidental scholars could dispense with the trouble of collating
manuscripts. Only the recognition of the relative application of the sci-
ence of qirt makes a renewed investigation of the earliest manuscripts
of the Koran rewarding, and promisesin case non-Uthmnic copies of
the Koran should indeed not come to light
1
at least a control of the nar-
row Muslim tradition and an augmentation of our knowledge of the period
before the systematization of the science of the Koran. At the suggestion
of G. Bergstrer, the Bayerische Akademie der Wissenschaften, Mnchen,
started to assemble, as far as possible, a complete collection of photographs
of our extant older manuscripts of the Koran, thus initiating for the first time
the investigationof important researchmaterial. As this project of the Bavar-
ianacademy is only inits beginning stages, the following expositiondoes not
constitute conclusions; rather, it merely represents a general introductionto
the problems and the methodology of manuscript research.
2
1
Rumours about such things have not infrequently been heard from scholars in the
Orient and the Maghreb, but I did not get a chance to follow them up. Important seems to
be the news from several quarters that before the First World War there was at Damascus
a non-Uthmnic codex from Homs. Muammad KURD AL writes in his book, Khia
al-Sham, vol. 6, p. 199, that during the War some twelve boxes with precious books, including
very oldmanuscripts of the Koran, were removedfromDamascus. It is not true that they came
to Germany. Until now they are unfortunately not available to scholarship.
2
See G. Bergstrer, Plan eines Apparatus criticus zum Koran; further O. Pretzl, Die
Fortfhrung des Apparatus Criticus zumKoran (1934).
586 manuscripts of the koran
Whereas investigation into Arabic palaeography with reference to papy- [iii/250]
rology made enormous advances inthe 1920s and1930s, particularly through
the monumental work of Adolf Grohmann,
3
in the case of palaeographic
inquiries into the earliest manuscripts of the Koran the most basic pre-
liminaries are still wanting.
4
Only a very small part of the older Korans
has been catalogued. The richest collection of this kind I found at Istan-
bul. Although most of it is located at the Topkap Saray, where now nearly
all the collections of Kfic Korans formerly held by municipal libraries are
brought together, there are still other old Korans in the Evkaf (Mzesi),
which originally owned some sixteen such manuscripts, and also recently
received some more from municipal libraries. A very valuable collection,
particularly because of the great variety of scripts, is the collection of the
earliest fragments of the Koran at the Bibliothque Nationale de Paris. Sev-
eral very old codices are held by the Egyptian Library at Cairo, and the
Azhar Mosque. During my study tour of Morocco in the spring of 1934,
surprisingly valuable copies were discovered there. Larger and smaller col-
lections (mostly fragments) are preserved in various libraries in the Occi-
dent.
The Script of the Older Korans
(1) The Korans of the first four centuries were mostly written in a script dif- [iii/251]
ferent from the common Arabic cursive script. From a very early period on,
they were lumped together under the misleading name of Kfic. The ori-
gin of this name cannot be determined with certainty (see below, p. 590sq.).
3
Cf. first of all Ad. Grohmann, Corpus Papyrorum Raineri archiduci Austriae, series ara-
bica, 3 (1924).
4
Valuable help is at least provided by B. Moritzs article, s.v. Arabia (a) Arabic writ-
ing in: Encyclopaedia of Islam, vol. 1 (19081913), pp. 381393. Further G. Bergstrer, Zur
ltesten Geschichte der kufischen Schrift (1919).Reproductions of Korans are to be found
in B. Moritz, Arabic palaeography (1905). Unfortunately the work does not list either library
shelf numbers or size. He also fails to supply references to his datings.Very valuable is also
the reproduction of the so-called Samarqand codex (see above, p. 393 n. 37). From among
the relevant literature I know: Henry Lansdell, Russian Central Asia (1885), Chauvin X, no. 94.
Bericht ber die Kgl. Bibliothek Petersburg [St. Petersburg Library], p. 346. Aleksandr L. Kun,
in: : , ed. Nikolai]
A. Maev, p. 401. A larger number of reproductions also in L.H. Mller, Palologische Beitrge
aus den herzoglichen Sammlungen in Gotha (1844). Some samples also in the Oriental series
of the Palaeographical Society, London, 18751883, and Palographie universelle, ed. Joseph
B. Silvestre, vol. 1 (Paris, 1839). On illuminated Korans see E. Khnel, Islamische Kleinkunst
(1925), pp. 2637!
manuscripts of the koran 587
This designation undoubtedly indicates the lapidary script. It is character-
ized by reducing letters that originally displayed a greater variety of shapes
down to a few basic artistic forms, namely to the simple hook , for the
letters b, t, th, n, y (only in the initial or the medial forms; the final forms,
as is the case also in the following groups, differ very considerably fromone
another, being not uniform even in the case of one and the same letter!), to
the group for the letters m, w, f, q, and to the very much stretched parallels
for the letters d, dh, s, , , k.
Some letters do not appear in an identical shape even in the earliest time,
so that , whichappears as a line crossedfromtopleft to right belowthe line
\ or in the shape still currently used, an acute angle >. Alif appears with or
without horizontal line; as an angle open at the top and standing on the
line , or as a semi-circle supported by a stem|. The initial has the shape
of a semi-circle open to the right which, at timesin the one ductus of the
Samarqand codex as well as in other casesattains the length of an alif,
but mostly retains the height of the smaller letters. As this style of the script
commands high artistic ability, the script fluctuates at all times between
calligraphic perfection and clumsy imitation whenever used for the Koran
in general. The fact that the perfect form can already be documented in
older numismatic inscriptions and monuments speaks against the attempt
to date this artistic perfection to the end of a development. The main forms
recur also in later documents of this genre, so it cannot be a question of
development.
5
Still, on account of minor changes of the forms,
6
and partic-
ularly because of more diverse final forms, a variety of manuscripts can be
combined into smaller stylistic groups that then display a certain continu-
ity of form, both in the number of lines and in the orthography. Only later
do we find excessesafter the third century, when Korans generally are
more frequently datedinthe so-called ornamental or floral Kfic; thus,
since the beginning of the fifth century we see, for example, a very strong
5
I would like to emphasize that the only remarkable development is the m, which
originally reached half below the line (unless a connection made this impossible, cf. plate
no. 1), but in subsequent times was always put on the line.
6
This includes most of all the change of the circle for m and w, rarely for f and q, into
semi-circles, semi-pear-shaped or egg-shaped forms (cf. plate 2!). It seems to me that the
stylistic development of the circle into triangles or squares appears only in very late Korans.
The oldest dated document of the latter type I know of is the vakfiye on the Koran, Evkaf
Mzesi, no. 1474, of 337/948! The Koran, no. 114, of the EgyptianLibrary is a very good example
of howa different style evolves in one and the same hand. Whereas the mis at the beginning
pretty much round, it gradually appears with a point at the upper left, changing eventually
to a pronounced pear-shaped form.
emphasis on the varied strokes vis--vis the script, and the decoration of the
upper lengths as well as the straight line with floral ornaments.
7
The overwhelming majority of Korans writteninlapidary script onparch- [iii/253]
ment are in horizontal format. Some of the oldest copies we know of are
nearly square, as, for example, Evkaf, no. 3733 (with the gargantuan size
of 56 by 63cm, twelve lines to the page) which in script and appearance
equals the Koran of the Egyptian Library, reproduced by Bernhard Moritz
in plate no. 1. Nearly equally large are both the Samarqand codex and the
Paris Ms. no. 324, twelve lines each, but in a different hand. We do have evi-
dence from a very early time that small sizes, and the form of notebooks
that were used for other types of books, were frowned upon. It is reported
from Ibrhm AL-NAKHA (see above, p. 513) and others:
8
kn yakrahna
an yaktub l-maifa f l-shay al-aghr [kna] yaql aim al-Qurn. Al-
ak says:
9
kna yakrah al-karrs yan al-maif tuktab fh and: l
tattakhidh lil-adth kurrssatan ka-kurrs al-muaf. Indeed, we find only
oversize copies among old Korans. However, the huge sizes of the above-
mentioned copies are likely to have been rare. They have survived probably
only because they are precious.
(2) The script of a smaller group of manuscripts takes the middle place [iii/254]
between the lapidary and the cursive forms known from papyri. Compared
with the length of vertical strokeswhich in most cases extends to the pre-
ceding lineits script is quite compact. The vertical strokes throughout
lean toward the right. Apart from certain subsequent intermediate forms,
these manuscripts are always in vertical format, and written in deep-black
ink (made of soot) which was extremely unevenly flowing. Based on a pas-
sage in Ibn al-Nadms Fihrist,
10
Joseph von Karabaek already considered
this script ijz. Their division of the verses, as far as I could ascertain, and
their other characteristics, do indeed point to the narrow group of Medi-
nan and Damascene manuscripts. As far as their orthography is concerned,
7
One of the most magnificent codices of this genre, dated 909, is to be found in the Saray,
Revan Kk, no. 18.
8
Abd Allh IBN AB DWD AL-SIJISTN, K. al-Maif (see above, p. 575) in juz 4,
beginning of bb tamwa-taghr al-maif.
9
Ibid., somewhat earlier!
10
Flgels edition, vol. 1, p. 6, l 3:
] [

. J. Karabeek in Julius Eutings Sinatische Inschriften, p. 323, translates
inaccurately, as far as the Meccan and Medinan scripts are concerned, in their alifs there is a
bendto the right side of the hand and inthe body of the text a slight inclinationto the side.
Omitted is wa-al al-abi, for which I, too, cannot supply a translation; possibly wa-huwa
al al-abi has the greatest vertical length?
they represent an older stage of development than the earliest lapidary
manuscripts we know of. On the other hand, they are connected with the
oldest lapidary codices by certain peculiarities of style; for example, the
leaning towards the right of the vertical strokes is found also in the two
oversize manuscripts, Evkaf, no. 3733, and the Egyptian Library, mentioned
above, whereas the final y drawn far backwards and underlining several
words (see plate 8, l 10) is also frequently met in the Paris codex no. 324.
Almost everywhere shay is found as .
11
Typical for all is the largely defec-
tive writing of the .
As a demonstration of the extent of the defective writing of this group [iii/255]
of manuscripts the collation of two codices on sra 3, verses 32 to 37: Biblio-
thque Nationale de Paris, no. 328, andIstanbul Saray Medina 1 might serve.
It must be observed that in this passage the group of the lapidary Korans is
generally in agreement with the Cairo edition as far as the defective writ-
ing is concerned, so that passages which already appear there defectively
are not listed again. The two manuscripts have an additional fifteen defec-
tive writings, namely eight times and for ql and qlat, verse 32,
for nabtan verses 32 and 33, for mirb, verse 35, for hunlika,
for fa-ndathu, verse 35, for imraati, for qirun.
12
Consonantal
script of the occurs in this section only in the following words: verse 32,
isban, verse 33, da, al-dui, qimun, verse 35, yashu, verse 36, al-
nsa, ayymin.
13
This amount of defective writing is disproportionally large
when compared with the other codices. The introduction of yet another
Kfic consonantal script going even further is ascribed by the author of the
Kitb al-Maif, Abd Allh IBN AB DWD AL-SIJISTN,
14
to the gov-
ernor Ubayd Allh b. Ziyd (d. 69/688), who was prompted by his secre-
tary, Yazd al-Fris.
15
There it is reported that he enlarged the muaf by
two thousand letters. Wa-kna lladh zda Ubayd Allh f l-muaf kna
11
Shay is also found in the Samarqand codex, where it is written with alif on several
occasions, whereas according to Muqni (edited by Pretzl, p. 45, l 2) this could only be the
case in the passage, sra 18:23. Conversely, also in the muaf of Ubayy b. Kab it is always
written with alif.
12
The two last passages are defective also in the Samarqand codex, which has a gap to
verse thirty-two.
13
Among them three passages where alif is followed by hamz.
14
See above, p. 574sq.; the passage occurs at the endof juz 3 towards the endof bbikhtilf
khu al-maif.
15
For more about himsee Ibn ajar al-Asqaln, Tahdhb al-tahdhb, vol. 2, p. 374; Sezgin,
GAS, vol. 1, p. 4, l 13.
maknuhuf l-muaf ql:qf lm
16
wa-kn:kf nnwwfa-jaalahuUbayd
Allh ql:qf alif lmwa-jaala kn kf alif nn wwalif. According to this,
it concerns both the alif of division and the alif of prolongation. However,
it must be said as a critical aside that even if only the alif of prolongation
had been introduced by this Ubayd Allh, the number of two thousand for
the letters which he allegedly added would be far greater. In contrast, this
would correspond approximately to the number of differences of defective
writings between the more cursively written Korans and the early lapidary
Korans. The report might be a reminder of the factand in agreement with
the investigation of the manuscriptsthat Iraq experienced a more seri-
ous transformation of orthography. Sinceas far as we can determine
this also coincided with the change of the ductus, it might give rise to the
assumption that it became the custom there from then on to write Korans
inlapidary script only. This wouldbe one explanationfor calling the lapidary
script Kfic.
This script was introduced also to the ijz, as the considerable num- [iii/256]
ber of ijz lapidary codices attests. In any case, the Korans of the second
groupwrittenincursive script leaning to the right are to be considered more
original and, thus, more closely approximate Uthmnic orthography. The
original ijz script survived even after the introduction of the lapidary
script. The codex Istanbul, Saray, Medina 1 (see plate 10), which shows sev-
eral scripts that deviate considerably from one another, and hardly retains
any inclination to the right, but corresponds down to the least little detail
to the ijz group, is probably one of the last representatives of this type
of script, whereas codex two of the British Museum,
17
as a near complete
specimen is probably the most important one.
(3) Athirdgroupof the older Korans is writteninMaghrib script. The details [iii/257]
of their script and its appearance clearly betray their dependence upon the
Medinan Korans. Their characteristic script, together with the old orthogra-
phy, has remained nearly unchanged until the most recent times. The oldest
example of this type is probably the colossal codex no. 3735, preserved inthe
Istanbul Evkaf Mzesi which, with seven lines to the page, measuring 56 by
63cm, equals the Medinan codex no. 3733 (see above, p. 588).
16
Thus the manuscript but according to what follows wwmust still be added; inaddition
(against the manuscript) the first time and ought to have been written.
17
One page of it is reproduced in table 69 of the Palaeographical Society series, London,
18751883.
The Provision of the Koran with Reading Aids,
Verse Dividers, and Names of the Sras
The fact that all the Korans were originally written in the ambiguous con-
sonantal script, without vowels, headings, or division of verses, posed con-
siderable difficulties in praxis. These problems were gradually solved by the
introduction of reading aids or divisons of verses and, in the case of the indi-
vidual sras, by headings and subtitles respectively. We have a statement
from Yay b. Ab Kathr [al-Yamn Ab Nar,
18
d. 129/746], regarding the
sequences of these innovations:
19
fa-awwal m adath f-hi l-nuqa al l-y
wa-l-t fa-qllbasabihwa-huwanr la-huthummaadathf-hi nuqaan
inda muntaha l-y thumma adath l-fawti wa-l-khawtim. This is in gen-
eral accord with research on manuscripts.
1. The introduction of diacritic points seems to have been accomplished
without opposition. In the earliest manuscripts they are still used very
sparingly. Their origin goes back to pre-Islamic times, and they can already
be found on the earliest coins. In the lapidary script they are nearly always
indicated by strokes. The differentiation of letters by means of points and
strokes respectively happened in the same fashion as is still common today.
There is, however, disagreement over the use of diacritics in the cases of f
and q. The qf is usually found with two strokes above the letter, and f
with one stroke below the letter; one also encounters the differentiation
still used currently in the Maghreb, qf with one stroke above, f with a
stroke below, but also the converse, q with stroke below, and f with stroke
above or without stroke (see plate no. 10, l 4 and 5!). The diacritic points
for t marba apparently make a very late appearance, even in naskh!
In most of the older manuscripts the diacritics are in the same ink as the
vowel marks. It is only later that it becomes a customtoindicate consonantal
variantslike the vowels (see below, p. 596)by different colours.
Verse Dividers
2. The verse dividerswhich originally consisted of several differently ar- [iii/258]
ranged strokesalso seem to have been introduced without opposition.
18
EQ; Juynboll, Encyclopedia, p. 73sqq.; F. Sezgin, Goldziher and hadith, p. xxxv.
19
al-Dn, Kitb al-Bayn (Istanbul Ms, Halis Eff., no. 22) fol. 38
r
, l 14. F. Sezgin, Goldziher
and hadith, p. xxxv.
The sources frequently insert three points to indicate the verse division; for
example, the above-mentioned Yay b. Ab Kathr: m hdhihi l-maif
illhdhihi l-nuqa al-thalthahindarus al-y.
20
Later the verses were sep-
arated by a coloured rosette. Verse dividers, however, are by no means sup-
plied in all manuscripts, nor were they there from the outset. They are sup-
plied or omitted rather indiscriminately, even in one and the same codex.
There are also Korans where not the individual verses are indicated but only
sections of five or ten verses. Tradition ascribes the introduction of these
sections of five or ten verses to Nar b. im al-Layth,
21
who died 89/708 or
90/709 (see above, p. 511). This innovation is opposed by various old author-
ities, for example, IbrhmAL-NAKHA, al-asan al-Bar, Muammad IBN
SRN al-Anr, but particularlywith little credibilityIbn Masd. The
following saying is derived from the latter: jarrid l-Qurn wa-l tukhal-
lihu bi-shay which, incidentally, is also ascribed to asan al-Bar and
Ibrhm AL-NAKHA, and, apart from tashr and takhms, is also to reject
the headings of the sra (see below).
22
The five-verse sections are frequently
indicated by a (according to abjad) or by red, large alifs in a black circle
or larger rosettes, and the sections of ten verses, by painted rosettes,
23
fre-
quently alsoincontrast tothe five-verse sections, by paintedsquares, withor
without a written or the corresponding abjad numeral (see plate no. 5,
l 10).
Names of the Sras
[3.] Yet another innovation applies to the names of the sras that origi- [iii/259]
nally appeared probably as titles written at the end of the sras that later
becoming headings, withor without addedkhtimat srat kadhandftiat
srat kadh respectively, and frequently accompanied by the number of
the verses: wa-hiya ya. The sras carried no names in the oldest
Korans. Aversion to them seems to have prevailed for a long time (see
above). However, it is already reported fromMlik b. Anas (d. 179/795)
24
that
20
al-Dn, K. al-Bayn (Istanbul Ms, Halis Eff., no. 22) fol. 38
r
, l 19. similarly Ibn Ab Dwd
al-Sijistn, K. al-Maif: yuqirrna instead yuarrifna.
21
I.e. Ubayd b. Umayr al-Layth: EQ; Juynboll, Encyclopedia, p. 11 sqq.
22
Ibn Ab Dwd al-Sijistn, K. al-Maif, bb kitbat al-awshir f l-maif. Some of
this also in al-Itqn of al-Suy, naw 76, fal f db kitbatihi (Sprenger ed., p. 868; Cairo
edition, 1318), vol. 2, p. 170.
23
Green, primarily in ijz manuscripts!
24
al-Dn, Kitb al-Bayn, Istanbul Ms, Halis Eff., no. 22, fol. 38
r
, l 2.
he showed a muaf belonging to his grandfather which had been written
during the time of Uthmn, and included titles written at the end of the
sras that later became headings, written in ink on a band ornament over
the whole length of the line: fa-raayn khawtimahu min ibr al amal
al-silsila f l al-sar wa-raaytuhu majm al-y (i.e. with verse divisions
marked). Occasionally only a rest of a line is left vacant between the indi-
vidual sras (see plate 6). But in most cases a blank line is inserted. This
space is frequently used for ornaments or consists of sra headings embel-
lished by leaf ornaments and arabesques (see plate 7). This was the only
place for the decoration of the Koran,
25
its legitimacy being at first contro-
versial.
26
Later on, the first and last leaves of the Korans also became embel-
lished by arabesques, framed in gold, and divided into small fields (circles,
squares) on which the letters of the first and the last sras were spread out.
We find Korans that are divided into seven parts (see below), where the end
of each part is highlighted by ornaments. In ravishingly illuminated Korans
the ten-verse sections already indicated in the text are emphasized by indi-
vidual marginal medallions, as is also the case of passages where sajd is to
be made.
27
4. The division of the Koran into ajz pretty certainly seems to go back to [iii/260]
al-ajjj b. Ysuf (d. 95/713). It is reported fromhim:
28
jamaa al-uff wa-
l-qurr fa-qla akhbirni ani l-Qurn kullihi kamhuwa min arf fa-jaala-
n nasub att ajma anna l-Qurn kullahu thalthamiat alf arf wa-ar-
bana alf wa-saba mia wa-nayyif wa-arban arfn qla fa-akhbirni il
ayy arf yantah nif al-Qurn fa-asab wa-ajma annahu yantah f l-kahf
wa-l-yatalafa f l-f (p. 18, l 18) qla fa-akhbirni bi-asbihi al l-urf.
25
Cf. E. Khnel, Islamische Kleinkunst, p. 26sqq.!
26
Ibn Ab Dwd al-Sijistn, K. al-Maif, bb f taliyat al-maif. According to him,
Ibn Masd allegedly pronounced the decorations of the Koran permissible. The same ques-
tion, with reference to mosques, is already discussed in the previous chapter, kitbat al-
maif bi-l-dhahab. (Manuscripts written entirely in gold are not seldom mentioned in
literature; one copy has survived in Istanbul, Nuruosmaniye, without number.) This is fol-
lowed by a short chapter, f tayb al-maif, which states that Ibn Masd opposed adding
musk scent to produce a pleasant smell. The same also in al-Dn, Kitb al-Bayn (Istanbul
Ms. Halis Eff., no. 22), fol. 37
v
.
27
The indication of sajd is still prohibited by al-Bayhaq [EI
2
; EQ,] who died in 458/1065.
(al-Itqn, Sprenger ed., p. 870, Cairo ed., 1317, vol. 2, p. 171, l 20), naw, p. 76!, fal f db
kitbatih.
28
Ibn Ab Dwd al-Sijistn, K. al-Maif, bb tajziat al-maif, from where the fol-
lowing quotations have been adopted. Cf. in particular the above-mentioned work of Ab al
Qsim Umar b. Muammad IBN ABD AL-KF (above, p. 574!). Older works on ajz have
been listed in Ibn al-Nadms Fihrist (Flgels ed., p. 36, l 28).
According to this, the first seventh extends to the d of man adda anhu
in sra 4:55, the second seventh to the t of abiat in sra 7:145, the third
seventh to the last alif of ukulah in sra 18:31, the fourth seventh to the last
(?) alif of li-kulli ummatin jaaln mansakan in sra 22:66, the fifth seventh
to the h of wa-m kna li-muminin wa-l muminah in sra 33:36, the sixth
seventh to the wof anna l-sawi in sra 48:6, and the last seventh to the end
of the Koran. Enumerated too are the the fractions of three and four.
Another tradition of dividing the Koran into ajz (s.v., juz) leads through [iii/261]
imal-Jadar (d. 128/745); but compared with al-ajjj b. Ysufs arrange-
ment it seems to be less accurate because the divisions coincide mostly with
the end of a sra. But also fractions of five, eight, and ten are listed. In the
older Korans I found only divisions into sevenths. In the margin, however,
other divisions are frequently added by a later hand, particularly fractions
of ten and thirty. The latter become the rule in the naskh Korans. In more
recent editions (as also inthe official Cairene edition) it is customary tohave
the fractions of sixty in the margin.
Whereas the arrangement established by al-ajjj b. Ysuf evidently pur-
ported to guarantee both the inviolability of the consonantal text and to
facilitate its control (cf. this with the minor Masorah of the Hebrew Bible),
later arrangements serve liturgical purposes to establish definite prayer cur-
ricula. They are no longer called ajz as formerly, but now azb (plural of
izb, but also wird), a name from the Koran which was later applied to pri-
vate prayer.
29
5. There is no generally accepted tradition regarding the introduction of the
vowel marks. According to one tradition
30
they are said to have been intro-
duced by [limb. Amr] ABAL-ASWADAL-DUAL (d. 69/688)
31
but only
for the vowel signs and tanwn, whereas the signs for hamza, tashdd, rawm,
and ishmm are attributed to al-Khall b. Amad (b. 100/718).
32
According
to others, it is supposed to have been Nar b. im al-Layth (d. 89/707
29
Cf. article izb in the Encyclopedia of Islam.
30
Most of the traditions here discussedare tobe foundinal-Dn, Kitbal-Naq (editedby
Otto Pretzl, Orthographie undPunktierungdes Koran: zwei Schriftenvonad-Dn, pp. 132133),
further inIbnAb Dwdal-Sijistn, K. al-Maif (see above, p. 575). Most of them, however,
are to be found in al-Itqn (see above, p. 259 n. 22).
31
Sezgin, GAS, vol. 9, pp. 3132; Juynboll, Encyclopedia, p. 415.
32
According to a very credible opinion which, however, is not documented, al-Khall
b. Amad is ascribed to have introduced the arakt which are still in use until now (cf.
al-Suy, al-Itqn, loc. cit., p. 171, l 12!) Neither the sources nor the manuscripts give any clue
that there is supposed to have been a sign for rawm and ishmm (in the sense of a particular
pausal pronunciation; see above, p. 540sq.). [Sezgin, GAS, vol. 8, pp. 5152.]
or 90/708). It is also ascribed to a pupil of AB AL-ASWAD AL-DUAL,
33
Yay b. Yamar who is said to have been commissioned with the task
by al-ajjj b. Ysuf, and that under the caliph Abd al-Malik b. Marwn
(d. 86/705).
34
The innovation was rejected on the part of (Ab l-Khab)
QATDA b. Dima, Abd Allh b. Umar, Ibrhm AL-NAKHA, al-asan al-
Bar, and Muammad IBN SRN. However, according to others, al-asan
al-Bar is said to have approved of the vocalization,
35
whereas it is reported
from Muammad IBN SRN that he owned a pointed manuscript of Yay
b. Yamar. Mlik b. Anas (d. 179/795) prohibited the marking of all offi-
cial (ummaht) Korans, whereas he approved of them for the small copies
(ighr) used for instruction. We indeed find that all the oldest large copies
of the Koran (Samarqand codex, Evkaf, 3733, and the great Cairene codex,
see above, p. 588) have no vowel marks at all. In very many of the other
copies they are only later additions. The attempt to preserve the script of
the text of the Koran unchanged would explain the habit of using as colour
for the vowel marks a colour that clearly stands out against the colour of
the script.
36
Vowels marks are nearly always written in red, for hamza yellow
and green are customary as well (see below, p. 597). Other coloursblue,
orange, yellow and greenare used to indicate variants, a habit, although
disapproved of by al-Dn,
37
which according to the state of manuscripts has
been used very frequently.
38
Blue is nearly always used when the imla is
intended to be expressed. In such a case there is a red point (for a) above
the consonant, a blue one (for i) below.
The use of reading marks is explained in two works, Kitb al-Maif [iii/263]
of Ibn Ab Dwd al-Sijistn (d. 316/929),
39
and Kitb al-Naq of Uthmn
33
EI
2
; Juynboll, Encyclopedia, p. 415; Sezgin, GAS, vol. 9, pp. 3132.
34
The authorities who are ascribed to as having introduced the signs are all Barans.
35
He is even ascribed to have introduced the vowel marks.
36
al-Dn, Orthographie und Punktierung des Koran: Kitb al-Naq (ed. Otto Pretzl), p. 134,
l 1: l astajz al-naq bi-l-sawd li-m fhi min al-taghyr li-rat al-rasm.
37
Ibid., p. 134, l 3.
38
See plate no. 5, third line from the bottom, where l khawfun alayhim is vowelled in
red, whereas l khawfa alayhum is in green. In the Marrakesh manuscript (see plate 3) the
suffix hum when ending with a vowel is with a point on the u; a corresponding variant is
nearly always the sukn sign (frequently followed by tashdd), which is an indication of the
pronunciation, -hum.
39
See above, p. 575sq., bb kayfa tunqa al-maif, rather short and in poor condition.
It begins as follows: qla Ab tim al-Sijistn wa-naqaahu bi-yadihi hdh kitb yustadall
bi-hi al ilm al-naq wa-mawiih. This is followed by expositions by the renowned Ab
tim al-Sijistn (d. 255/869), whose Kitb f l-naq is indeed mentioned in Ibn al-Nadms
Fihrist (edited by G. Flgel, p. 35). In addition, the Fihrist lists the following authors of
works that are no longer extant: al-Khall b. Amad al-Farhd, ABMUAMMADYay b.
b. Sad AL-DN (d. 444/1052).
40
There is a great difference between the two
works occasioned by the Iraqi pointing on the one hand, and the Medinan-
Maghrebi pointing on the other hand. Ibn Ab Dwd al-Sijistn recog-
nizes only vowel marks, including tanwn (the hamza, too, he usually indi-
cates by a vowel mark in a particular position), whereas al-Dn produces a
complicated system of reading aids, which attempts to do justice to all the
refinements of Koranic reading, and which, according to him, is based on
old Medinan manuscripts as well as explicit prescriptions of older Koranic
authorities such as, for example, s b. Mn QLN. Indeed, we find his
system most strictly adhered to in two manuscripts extant in the Maghreb,
namely Ms. Fs (see plate 1) and Ms. Medersa Ben Yusuf, Marrakesh (see
plate no. 3) as well as in many codices written in Maghrebi script. But even
al-Dns work does not do justice to the far more manifold investigations
into manuscripts. In addition, the two works differ since, according to Ibn
Ab Dwd al-Sijistn only the absolute necessary marks are used, whereas
al-Dns presentation presupposes complete pointing.
(a) The indication of vowels: The three vowels, a, i, and u are indicated by [iii/264]
one point each. In the case of fata the point is placed above the consonant
preceding the vowel, in the case of kasra under the same, and in the case of
amma in the middle of the consonant or to the left behind it.
41
Deviating
from their usual method of pointing, the codex Medina 1b in the Istanbul
Saray (see plate 4) has instead of the amma a small vertical stroke imme-
diately after the consonant.
42
al-Mubrak AL-YAZD, further Muammad b. s (al-Ibahn), who has been mentioned
above, p. 574, as a source for al-Dns Muqni, and Ab Bakr IBN AL-ANBR.
40
Edited together with the authors Muqni by O. Pretzl entitled Orthographie und Punk-
tierungen des Koran (1932). In the introduction (p. 123, last line) the author refers to a
larger work on the subject of naq that he composed but which has not survived. Despite
the greater richness compared with the previous one, the book has considerable lacunae,
particularly regarding the exposition of the hamz orthography, which is too much con-
cerned with the qira of Uthmn b. Sad WARSH (with tashl al-hamz), so common in the
Maghreb.
41
al-Dn: fawqa, tata, f wasi or amma l-arf. In Ibn Ab Dwd al-Sijistn the expres-
sion quddmor bayna yaday, also f jabhat al-arf, is found instead of amma. These expres-
sions are based on the conception that the writing is fromright to left so that what follows is
fronted to the preceding letter. Analogously, a sign preceding the letter is also considered f
qaf, standing in the neck.
42
See line 3 from the bottom: yuabna and penultimate line, last word, al-umuru, etc.
(b) Tanwn is indicated by doubling the respective vowel marks. According
to al-Dn (al-Naq, p. 135) the points are placed one upon another (tarkub)
when the n is fully articulated, but next to one another (tatbu) when
idghm or ikhf of the n occurs with the following letter.
43
In manuscripts
this differentiation is not always observed.
44
According to Kitb al-Maif, in case of the ending -an (masc. sing.) the [iii/265]
second point can also be omitted because it is sufficiently identified by the
alif. Al-Dn (Naq, p. 136, l 8) is familiar with the habit of uneducated
scribes in this case to distribute the points in a way that one of them is
positioned on the preceding consonant, while the other is on the alif.
(c) The (over-) prolongation of vowels is indicated by adding the respective
vowel of prolongation (alif, y, ww) in small form and in red colour. In the
Ms. Istanbul, Saray, 50386, however, a small alif is foundas a signof prolonga-
tion also for
45
and . Not infrequently, the - of the suffixes, -kumu, humu,
etc. is expressed by a point on the line with an overhead hook (similarly
to the tashdd, see below, p. 598). Al-Dn (al-Naq, p. 136, l 15) also knows
of a graphical distinction of the ishb (normal pronunciation of a vowel)
and ikhtils (vanishing), and, in the first case, suggests putting an alif ughr
munaria
46
and y or wwughr respectively, in the latter case the normal
point of a vowel, an unfortunate differentiation that I have not encountered
in manuscripts.
(d) Hamz: In the case of initial or final hamz Ibn Ab Dwd al-Sijistn
does not yet know of a proper mark other than the respective vowel point.
However, for the medial hamz he demands two points, a f qaf al-alif,
before alif to indicate the glottal stop, another one after alif (above!) to
indicate a, corresponding to i and u in the middle and below respectively.
Accordingly, the second point is called muqayyida. He recommends the
use of a green point in cases where hamz is pronounced in two different
ways, with and without tashl. For al-Dn, however, the main rule is that
hamz is indicated by a yellow point, but in his explanations he pays far
43
See above, p. 537sq., and also below, p. 599.
44
Very clearly, for example, in plate 3b, l 1, suqufan min, conversely l 3, sururan alayhi. In
the Ms. Saray Medina 1b (plate 4) in the second case, the points are far apart, for example, l
5, baghtatan fa- where the one point is above the , the other, on the line immediately before
the f, likewise (l 7) bi-rusulin min the one point belowthe l (somewhat removed to the side!),
the other point directly before the m.
45
See plate 7, last line, alladh anqaa.
46
This must correspond to a fat sign of the more recent vocalization.
more attention to those cases where hamz is softened (al-Naq, p. 142sqq.).
According to al-Dn, a softening of hamz is indicated when the vowel
hamz is written without particular sign for hamz. Hamzat al-wal is usually
indicated by a red horizontal stroke which, according to the position of
the vowel, runs above for fata, in the middle for amma, and below for
kasra.
47
The vowel of the hamzat al-wal at the beginning of a word might
be indicated by a green vowel mark.
48
In many manuscripts Warshs peculiar
softening of hamz is likewise indicatedby naql. According toal-Dn, inboth
cases a red circle is placed above the alif to indicate that there is no hamz
(see below, p. 600).
According to Kitb al-Maif, is written with a point before the alif, [iii/266]
while with a point after the alif (positioned a bit higher, wa-tarfah
qallan il ras al-alif), a differentiation which is observed in many manu-
scripts.
49
A hamz orthography differing from the common one is to be found in
the Istanbul codex Saray Medina 1b. Here, hamz is indicated by a red hook
opening to the top.
50
In the manuscript Saray Emanet 12, the hamz sign
is found as three red points, for example, [consonantal script without any
points] : anzalnhu, or arranged in the shape of a triangle [as on shn]:
yuminna shayan[onbothwords points only on]. The vowels
are then no longer written individually.
(e) As an indication of sukn a small horizontal stroke is placed above [iii/267]
the vowelless consonant
51
(jarra bi-l-amr, al-Dns al-Naq, p. 137, l 5).
This sign is apparently found nowhere in Iraqi manuscripts, in others only
rarely. However, al-Dn (p. 150, l 1 sqq.) also knows of the small circle as
a sign of sukn, which is still in use today. According to him, the sign is
used in order to indicate (1) that a letter existing in the script is ignored in
pronunciation (2) that a tashdd is missing from the urf al-mukhaffafa,
and (3) that a vowel is missing from the urf al-musakkana. In line 6,
he traces this habit back to Qln: qla f maif ahl al-Madna m kna
47
See plate 1, l 3 m Allhu, stroke at the upper end of alif, also, l 3, wa-yakhsh l-nsa, in
contrast l 9, amru Allhi, stroke in the middle of the alif. Further plate 2, l 2 and 6. Still clearer,
plate 3, 3 a, l 2, min al-qaryatayni, l 4, f l-ayti l-duny, and l 4, mat al-ayti.
48
For example, this is the case in plate 1, l 9, amru Allhi, however not in l 2 in -llha,
which follows wa-l-taqi.
49
See plate 6a, l 5, anzalnhu, and 6b, l 3, min li! Further plate 5, l 10, man, and l 13,
ajruhum.
50
See plate 4, l 8, yastahzina, yaklaukum (over r and y!) and frequently.
51
See plate 3a, l 2, al-qaryatayni (above r and y) and frequently!
arf mukhaffaf
52
fa-alayhi dratun bi-l-umra wa-in kna arfan musakka-
nan fa-ka-dhlika ayan. As regards the use of sukn see also below, Seite
269 [sic, p. 600].
(f) Tashdd is indicated by a small semi-circle open at the top or bottom,
or an acute angle. According to al-Dn (p. 137, l 7) it is always positioned
above the consonant together with the respective vowel mark, in which
case the vowel still needs to be properly indicated, whereas according to
his information, in Maghrebi and Medinan manuscripts it is positioned
where the respective vowel ought to be placed, under omission of a vowel
point.
53
The sign still used currently is derived fromand apparently to be
found only in later manuscripts.
In Koranic reading the doubling of a consonant can be effected also in [iii/268]
Sandhi by the assimilation and incorporation of a vowelless or (in the read-
ing of Ab Amr b. al-Al) even of a vowelled consonant of the preceding
word.
54
In this case different rules apply for the use of tashdd: In the case
of an idghm a, i.e. complete assimilation and incorporation, the dou-
bled consonant receives a tashdd, the vowelless consonant, however, no
sukn. But if assimilationand insertionare not completethis is the case of
ikhf of the n and m, see above, p. 536!then the vowelless n and mreceive
no sukn, but the following consonant also no tashdd, unless n is followed
by a w or y because in this case tashdd can be positioned, but then also
the tashdd must be written above the n to differ from the idghm tmm.
Naturally, the same applies also to tanwn, which, when it is not fully articu-
lated, is identified by two vowel points placed next to each other (see above,
p. 597). Whenever tanwn is followed by a b, the second point of the tanwn
canbe replacedby a small m(inorder to indicate the qalb, see above, p. 536).
52
It seems to be beyond doubt that here mukhaffaf was originally used in contrast to
muthaqqal (vowelled) not in contrast to mushaddad. This, indeed, is in contradiction to
the respective passage where, according to Qln, it is explicitly added: wa-in kna arfan
musakkanan; but for a purely stylistic reason I like to consider this addition a gloss of
the author, which would explain his inaccurate conception of mukhaffaf in contrast to
mushaddad instead of muthaqqal.The small circle as signof sukn is to be found (mostly in
another [blue!] colour) next to the stroke inthe Marrakeshmanuscript (plate 3), for example,
page a, last line, li-man, and in the preceding line, an yakna, whereas, for example, on page
b, l 4, in wa-an, on the other hand, it is missing. It seems to have been added by a later hand.
53
See plate 3a, l 1, nuzzila; l 3 from the bottom, mimm; p. 3b, l 2 from the bottom,
la-yauddnahum; but cf. l 4 of the same page, kullu with tashdd above the l! In both cases
also the vowel mark for u is used, while it is missing in a and i.
54
See above, p. 536; al-Naq, pp. 139142!
(g) urf nqia and zida: According to al-Dn, there is a special rule
that applies to letters which, though pronounced, are not written in the
Uthmnic consonantal text. The most frequent case is the omission of the
alif of prolongation; rare is the coincidence of two hamzas, the omission of
a bearer of hamza, for example, aandhartahum (spelled ) aidh ( )
aunzila (
). As a rule also, when two ys or ws follow one another only

one is actually written, for example, al-nabiyyn ( ) or Dwd ( ). In
all these cases the missing letter can be added in red colour. Conversely,
there are a number of passages in the Koran where there is one letter too
much, which, however, is not reflected in the pronunciation. This is most
frequently the case with alif al-qa, and often in the case of identically
written words as a means of differentiation, e.g., ulika and ilaika.
These are indicated by a small red circle as a sign of deletion (see above,
p. 598). In many of these cases it is doubtful which letter is to be considered
the bearer of hamza. Consequently the signs for the hamz and the circle of
deletion are also different.
(h) From the earliest time a ligature somewhat resembling the Greek letter [iii/269]
had been in use for the letter combination lm alif. For the punctuation
the question which of the two strokes stood for lm and which one for
alif was of practical importance. According to al-Dn (al-Naq, p. 151) the
stroke standing to the right ought to be the respective bearer of the vowel
and hamza sign due to the alif, a solution which likely corresponded to the
scribes habit of, starting the sign from the left and finishing it to the right.
HRN b. Ms AL-AKHFASH, however, held the opposite view, and this
prevailed in the vocalization of the naskh manuscripts.
(i) So far I have not been able to identify signs for waqf (see above, pp. 570
573) in lapidary manuscripts, despite the fact that their use is documented
from the second century onward. Ab Bakr IBN AL-ANBR reports from
al-Kis (d. 189/804)
55
wa-hum (i.e. the people) yasmana wa-yabuna
anhuatt l-maqi wa-l-mabdi.
56
Still, Amad b. al-usayn AL-BAYHAQ
(d. 458/1065) prohibited the use of a sign for waqf.
57
In later naskh manu-
scripts such signs are common.
55
56
Ibn al-Jazar, abaqt, vol. 1, p. 538, l 13.
57
al-Suy, al-Itqn (Cairo ed. 1318, vol. 2, p. 171, l 3), naw 76, fal f db kitbatih.
The Dating and Provenance of Manuscripts
Incontestably dated manuscripts of the Koran are extremely rare. Not until [iii/270]
the fourthcentury didtheir number become more abundant. Other thanthe
copies referredtoby B. Moritz inEnzyklopdie des Islm, vol. 1, pp. 405406, I
have been able to identify two dated copies, the only remaining colophon of
a Damascene Koran (now at the local Museum) from 298/910, and the sup-
plement to a manuscript allegedly written by Al (Topkap) Saray (Mzesi),
Emanet no. 6, from307/919.
58
In contrast, far more frequent are forged dates
in lapidary codices which mostly name Uthmn, Al or al-asan al-Bar as
scribes.
59
Identifying the provenance of Koranic manuscripts also poses con-
siderable difficulties. Based on works of Koranic subjects containing lists of
orthographic and textual variants of the individual amr,
60
theoretically it
ought to be easy to determine the distribution of the manuscripts among
the main centres of Islamic culture. In practice, however, these criteria turn
out to be unreliable. The investigation of several complete Korans in the
Topkap Saray Mzesi Ktphanesi showed that nearly all codices are mis-
cellaneous texts. So far, I have come across only a single copy of the Koran,
Saray 50386, which according to al-Dns list of variants is to be identified as
Medinan, but even this one has a non-Medinan variant in sra 57:24.
61
This
would mean that it is methodologically wrong to desire to localize Koranic
fragments on the basis of single variants. In order to illustrate the scope of
this textual mix, the results of just two manuscripts of the (Topkap) Saray
(Mzesi) must suffice as an example.
Saray codex 50385 in general supplies the special Medinan variants; in
two cases, sras 3:181 and 6:63, it is a Damascene text, in two passages of
sra 23 verses 89 and 91, it is a Baran text. Even in sra 2:126 the text does
not follow the Medinan-Damascene tradition.
Saray Medina 1a has the Damascene text in sras 2:116, 3:127, 10:23, 40:22,
55:11 and 78, 57:10; and in addition in the following passages, where Dam-
ascus and Medina coincide: sras 3:127, 5:58 and 59, 9:108, 18:34, 26:217,
58
It carries the farghnote: wa-hdhihi l-tatimma(i.e. the last four folios) tarkhuhsanat
sab wa-thalthat mia wa-amm ktibuh min awwalih il srat al-qria bi-kha al-imm Al
alayh al-salm.
59
Cf. already above, p. 394 n. 38. The Ms. Saray, Sultan Ahmet, no. 2, has the forgery:
katabahu Al ibn Ab lib f shahr Raman sanat tis wa-ishrn.
60
Cf. pp. 395 and 575.
61
Here and in what follows the quotation of the variants has been supplied. This can
easily be ascertained with the aid of the list above on p. 396sqq.
42:29, 43:71, 57:24, 91:15. Once, in sra 18:94, it has a Meccan peculiarity; in
sra 6:63, possibly a Kfan variant. Conversely, in six passages, 4:69, 6:32,
7:41, 73 and 137, 39:64, the Damascene variants are missing. The important
variant, sra 6:138, cannot be recognized in it because it is written neither
(Damascene) nor (so the others) but . It can safely be said
that the codex belongs to the inner circle formed by the groups of Medina,
Damascus, and Mecca, but on the basis of the lists it is not possible to be
more specific.
An additional means of determining the provenance of manuscripts is [iii/271]
the verse-numbering included in the Korans, which differ according to
amr, and about which we are informed by Islamic tradition.
62
But here
one is faced with the first difficulty, namely that in many codices we can-
not determine with certainty whether or not the verse-numbering coin-
cides with their writing. In many other codices it was later changed by
erasure. Further, as with the lists of variants, the transmission itself is not
absolutely reliable and complete. There are frequently verse endings in the
manuscripts that are not mentioned in the transmission. The sra headings
frequently contain numbers of verses that do not fit any of the systems of
verse-numbering of the amr. Not infrequently the sumtotal of verses cor-
responds to that of one mir whereas in the case of another sra this would
suggest a different mir. In addition, the control of the sumtotal is made dif-
ficult since very frequently, for pages on end the verse signs in the text are
wanting.
To some extent it is possible to infer the provenance of manuscripts [iii/272]
from the vocalization, provided it exists. But even if it does, one can never
be sure that it was done concurrently with the writing of the consonantal
textand at the same place. A precondition for evaluation is competence
in the uncanonical readings, which must first be crystallized from the rich
literature. But in viewof the wide distribution of the better known readings
beyond their place of origin, which we can follow throughout the third and
fourth centuries (see above, p. 508sqq.) it is precisely at the time when
manuscripts are widely vocalized that such vocalization as an indication of
the place of origin might at best serve as yet another confirmation parallel
to other criteria.
The use of the particular character of a script for dating and establish-
ing the origin of a manuscript reminds one of a citation from Muammad
62
A. Spitaler, Die Verszhlung [verse numbering] des Koran nach islamischer berliefer-
ung; see already p. 573.
IBNISQ
63
inthe Fihrist:
64
fa-awwal al-khu al-Arabiyyaal-kha al-Makk
wa-badah al-Madan thumma l-Bar thumma l-Kf. Until now, no single
style from among the great variety of lapidary Korans can be attributed
to any particular mir. That particular work, which after the continuation
of the afore-cited passage in the Fihrist might best be characterized as
Medinan-Meccan script (see above, p. 588sq.), could according to the con-
clusions above (p. 602) just as well have been written at Damascus which,
in view of the prevailing relations between the ijz and Syria under the
Umayyads, and which can be documented also in Koranic readings, would
be quite plausible. Conversely, it must be stated that in one and the same
mir quite different types of script were in use. For example, the fundamen-
tally different scripts evident from plates 5, 7, and 8 to 10 are likely to be
of Medinan origin, possibly also plates 1 and 3. In view of the domination
of ijz influence on Syria and North Africa, a precise differentiation will
be possible by accident only. The purpose of palaeological investigation will
have to be content with grouping together many small scripts according to
the type of script, originating from either one writer or a school of writing,
with the hope that fromother criteria temporal and local relations might be
established.
NewEditions of the Koran
The systematization of the Koranic reading had already been established [iii/273]
in the older lapidary codices, and completely dominates the naskh manu-
scripts. The former thus lost all importance for the history of the text. Apart
fromconservative Maghrebi manuscripts, the manuscript and lithographed
copies of the Koranhave continuously departedfromthe Uthmnic orthog-
raphy. Only some thirty years ago a reform of the Koranic writing arose at
Cairo, culminating in the official Cairene muaf of al-Mabaa al-Amriyya
in 1344/1925.
65
It has since been repeatedly reprinted, with only a change of
the year of printing, and with tacit minor corrections. Based on the above-
mentioned (p. 513sq.) rajaz poem, Mawrid al-amn by al-Kharrz[], the
edition provides a reconstructed consonantal text through an extremely
complicated system of the positioning of reading signs, and attempts (in
63
Ibn al-Nadm himself? It could be also (Ab l-asan Al b. Ubayd Allh) AL-MUSAY-
YIB also known as Ibn al-Musayyib al-Ktib d. 236/850 [Sezgin, GAS, vol. 2, p. 587, l 3] (Ibn
al-Jazar, abaqt, 2847) or Ab Raba d. 294/906 (abaqt, 2849).
64
Flgels edition, p 6.
65
A detailed desciption by G. Bergstrer, Koranlesung in Kairo, pp. 213.
close connection with the above-mentioned (p. 531) exact pointing of the
lapidary codices) graphically most precisely to express the reading of af
b. Sulaymnan im al-Jadar. The edition, produced with unbelievable
care, is froma scholarly point of viewan astonishing performance of Orien-
tal scholars. Since this edition is destined only for practical use, it does not
do justice to the original multifarious nature of the Koran, which is theoret-
ically recognized in Islam to this very day.
For more than a century Occidental scholars were surprisingly content
with G. Flgels rather inadequate edition of the Koran, published first in
1834, but not improved ever since. Only in most recent times have two inde-
pendent projects been launched to provide scholars with a scientific edi-
tion of the Koran. Arthur Jeffery, professor at the American University in
Cairo, is occupied with establishing an apparatus criticus on the basis of
the text of af b. Sulaymn, intending to present a most complete picture
of the Islamic transmission in so far as this can be derived from commen-
taries of the Koran, qirt works, and other literary sources. On the other
hand, the Koran enterprise of the Bayerische Akademie der Wissenschaften
in Mnchen (see above, p. 586) endeavoured to recover the oldest ver-
sion of a non-reconstructed consonantal text based on investigations into
Koranic manuscripts, together with an apparatus criticus derived from the
manuscripts and the variant readings (supplemented by the critically eval-
uated oldest transmissions) to illustrate the textual history of the Koran
during the first centuries.
66
66
A fairly similar research project, the Corpus Coranicum, was launched early in the
twenty-first century at the Berlin-Brandenburgische Akademie der Wissenschaften, Pots-
dam.
PLATES
606 plates
plates 607
608 plates
plates 609
610 plates
plates 611
612 plates
BIBLIOGRAPHY AND ABBREVIATIONS
The sources as they appear in the three indices of the German edition can-
not always be identified with certainty. Those monographs and articles that
have been identified are listed according to current Anglo-American usage.
The source used to identify German periodical articles and similar collec-
tive works is Fuat Sezgins twenty-one volume Bibliography der deutsch-
sprachigen Arabistik und Islamkunde, published between 1990 and 1995 at
Frankfurt am Main by the Institut fr Geschichte der Arabisch-Islamischen
Wissenschaften an der Johann Wolfgang Goethe-Universitt.
Printed Books and Articles
Abd al-Qdir b. Umar al-Baghdd. Khiznat al-adab wa-lubb lubb lisn al-Arab.
Blq, 1291/1874.
bid, Muammad. Hidyat al-nsik al tawd al-mansik. Cairo, 1302/1884
1885.
Ab Dwd Sulaymn b. al-Ashath al-Sijistn. Sunan b. Dd. Cairo, 1280/1863
1864, etc. [Edition or Ms. not supplied; see Sezgin, Geschichte des arabischen
Schrifttums, vol. 1, p. 150.]
Ab l-Faraj al-Ifahn. K. al-Aghn. Blq, 12841285.
Ab l-Fid, Isml b. Al. Abvlfedae Annales mvslemici arabice et latine; opera et
stvdiis I. Iacobi Reiskii. Hafniae (Kbenhavn), 17891794.
Ab l-Layth al-Samarqand, Nar b. Muammad. Tafsr al-Qurn. Cairo, 1310/1892
1893.
Ab Shma, Abd al-Ramn b. Isml. Ibrz al-man. Cairo, 1349/1930.
Ab Tammmabb b. Aws. Hamasae carmina; cumTebrisii Scholiis integris. Bonn,
18281847.
Ab Zayd al-Anr, Sad b. Aws. Nawdir f l-lugha, Beirut 1894.
Adler, Jacob Georg Ch. Descriptio codicvum qvorvndam cvficorvm partes Corani,
Altonae, 1780.
al-Aghn Ab l-Faraj al-Ifahn
Ahikar. The Story of Ahikar from the Aramaic, Syriac, Arabic, Armenian, Ethiopic,
Old Turkish, Greek and Slavonic versions, by F.C. Conybeare, J. Rendel Harris, and
Agnes Smith Lewis. London, 1898.
Ahlwardt, Wilhelm. The Divans of the six ancient Arabic poets, Ennbiga, Antara,
Tharafa, Zuhair, Alqama and Imruulqais, chiefly according to the Mss. of Paris,
Gotha, and Leyden, edited by Wilhelm Ahlwardt. London, 1870.
Andrae, Tor J.E. Die Person Muhammeds in Lehre und Glauben seiner Gemeinde.
Stockholm, 1917.
Anecdota Syriaca; edited by Jan Pieter N. Land. Leiden, Brill, 18621875.
614 bibliography and abbreviations
al-Anr, Ab Yay Zakariy. K. F l-waqf wa-l-ibtid. Blq, 1281/18641865.
K. al-Maqad. Blq, 1281/12641264.
Aqla al-Shib, Qsim b. Firruh
Arberry, Arthur J. The Koran interpreted. London, 1955. Reprint, 1969.
Ascensia Isaiae aethiopice et latine cum prolegomenis, adnotationibus criticis et exe-
geticis additis versionumlatinarumreliquiis, edita ab Augusto Dillmann. Leipzig,
1877.
Aws b. ajar, Ab Shuray. [Gedichte und Fragmente des Aus Ibn Haar, collected,
edited and translated by Rudolf E. Geyer. In: Sitzungsberichte der k. Akademie
der Wissenschaften in Wien, Phil.-hist. Klasse, 126, Abteilung 3 (Wien, 1892.)
al-Azraq, Muammad b. Abd Allh. Akhbr Makka; Geschichte und Beschreibung
der Stadt Mekka, edited by Ferdinand Wstenfeld. Leipzig, 1858. Repr. 1981.
al-Baghaw, al-usaynb. Masd. Tafsr al-Qurn[called] al-Malimal-tanzl. Bom-
bay, 1269/1852.
al-Bjr, Ibrhm b. Muammad. shiyat al-Bjr l-musammt bi-l-Mawhib al-
laduniyya al l-Shaml al-Muammadiyya. Cairo, 1329/1911.
al-Bakr, Ab Ubayd Abd Allh b. Abd al-Azz. Das geographis-
che Wrterbuch des Abu Obeid Abdallah el-Bekri nach den Handschriften zu
Leiden, Cambridge, London und Mailand herausgegeben von Ferdinand Ws-
tenfeld. Gttingen, 18761877.
al-Baldhur, Amad b. Yay. Liber expugnationis regionum; , ed. Michal
J. de Goeje. Leiden, 1866. Reprint, 1968.
El-Beldors Kitb futh el-buldn, (Buch der Eroberung der Lnder), trans-
lated by Oskar Rescher. Leipzig, 19171923.
The Origins of the Islamic state, being a translation from the Arabic ac-
companied with annotations, geographic and historic notes of the Kitb fut
al-buldn, by Philip Kh. Hitti, and Francis C. Murgotten. New York, 1916 and
1924.
al-Blaw (Plaw), mid b. Abd al-Fatt. Zubdat al-irfn, in: Muammad al-
Amn b. Abd Allh, Umdat al-khulln f zubdat al-irfn. Constantinople,
1270/1854.
Bar Hebraeus. Chronicum Syriacum e codicibus Bodleianis, edited by Paul J. Bruns
and George G. Kirsch. Leipzig, 1789.
Barth, Jakob. Die Pronominalbildung in den semitischen Sprachen. Leipzig, Hinrichs,
1913. Reprint, 1963.
Baumstark, Anton Joseph M.D. Geschichte der syrischen Literatur. Bonn, 1922.
al-Bayw, Abd Allh b. Umar. Beidhawii Commentarius in Coranum, ed. Hein-
rich L. Fleischer. Leipzig, 18461848.
al-Bayaq, Ibrhm b. Muammad. al-Masin wa-l-masw, edited by Friedrich
Schwally. Gieen, 19001902.
Becker, Carl Heinrich. Islam: [bibliographischer Jahresbericht] in Archiv fr Reli-
gions-wissenschaft, 15 (1912), pp. 530602.
Beer, Bernhard. Das Leben Abrahams nach Auffassung der jdischen Sage. Leipzig,
1859.
Belin, Franois A. Lettre M. Reinaud sur un document arabe relatif Mahomet
in Journal asiatique, 5e srie, 4 (1854), pp. 482518.
Bergstrer, Gotthelf. Das Kitb al-Lmt des Amad b. Fris in Islamica, 1 (1925).
bibliography and abbreviations 615
Die Koranlesung des Hasan von Basra In Islamica, 9 (1926), pp. 1157.
Koranlesung in Kairo; mit einem Beitrag von K. Huber in: der Islam, 20
(1932), pp. 142, 21 (1933), 110140.
Nichtkanonische Koranlesearten im Mutasab des Ibn inn. Mnchen, 1933.
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Phil.-hist. Ab-
teilung, 1933, Heft 2.)
Plan eines Apparatus criticus zum Koran. Mnchen, 1930. (Sitzungsberichte
der Bayerischen Akademie der Wissenschaften, Phil.-hist. Abteilung, 1930, Heft
7.)
Die Quellen von Jqts Ird, in: Zeitschrift fr Semitistik und verwandte
Gebiete, 2 (19231924), pp. 184218.
Verneinungs- und Fragepartikeln und Verwandtes im urn; ein Beitrag zur
historischen Grammatik des Arabischen. Leipzig, 1914. (Leipziger semitistische
Studien, 5,4.) Reprint, 1968.
Zur ltesten Geschichte der kufischen Schrift, in: Zeitschrift des Deutschen
Vereins fr Buchwesen und Schrifttum, 2 (1919), pp. 4966.
Bernouilli, Johann Jacob. Die erhaltenen Darstellungen Alexanders des Groen; ein
Nachtrag zur griechischen Ikonographie. Mnchen, 1905.
de Biberstein-Kazimisrki, Albert. Le Koran, traduction nouvelle faite sur le texte
arabe, revue et prcde d une introduction par (J.P.) G. Pauthier. Paris, 1840.
Bible. The Holy Bible, King James version.
Bingham, Joseph. Origines sive, Antiqvitates ecclesiasticae, translated from the En-
glish by J.H. Grischov. Halle/Saale, 17241729.
Birt, Theodor. Kritik und Hermeneutik nebst Abri des antiken Buchwesens, 2nd ed.
Mnchen, 19111913. Reprinted, 1978.
de Boulainvilliers, Henri, comte. Vie de Mohamed. Londres, 1730.
Bousset, Wilhelm Die Himmelsreise der Seele in Archiv fr Religionswissenschaft,
4 (1901.) [Off-print reprinted in 1960.]
Boysen, Friedrich Eberhard. Der Koran, oder das Gesetz fr die Muselmnner durch
Mohammed. Halle, 1773.
Bravmann, Meir Max. Materialien und Untersuchungen zu den phonetischen Lehren
der Araber. Gttingen, 1934.
Brockelmann, Carl. Geschichte der arabischen Litteratur. Weimar & Berlin, 1898
1902. 2 v. Leiden, 19371942.
Das Verhltnis von Ibn-el-Ars Kmil fit-Tari zu abaris Ah bar Errusul Wal
Mulk. Dr.phil. thesis, Straburg, 1890.
Browne, Edward G. The Babis of Persia: (2) their literature and doctrines inJournal
of the Royal Asiatic Society, n.s., 21 (1889), pp. 8811009.
Description of an old Persian commentary of the urn, in Journal of the
Royal Asiatic Society, 1894, pp. 417524.
Brnnow, Rudolf Ernst. Chrestomathie aus arabischen Prosaschriftstellern, im An-
schluanSocins Arabische Grammatik. Berlin, 1895. 2ndcompletely newedition
by August Fischer. Berlin, 1913. Repeatedly re-editeduntil 1984, entitledArabische
Chrestomathie aus Prosaschriftstellern.
Das Kitbu-l-Itbi wa-l-Muzwaati des Ab-l-usain Amed b. Fris in
Orientalische Studien Theodor Nldeke zum70. Geburtstag (Gieen, 1906 = vol. 1),
pp. 225248.
Bttner, Carl Gotthilf. Suaheli-Schriftstcke in arabischer Schrift, with transcription,
translation and commentary. Stuttgart & Berlin, 1892.
Buhl, Frants P.W. Muhammeds Liv; med en Indledning om Forholdene i Arabien fr
Muhammeds Optraeden. Kbenhavn, 1903.
al-Bukhr, Muammad b. Isml. a or al-Jmi al-a. Cairo, 1303.
Les traditions islamiques, avec notes et index par Octave Houdas et William
Marais. Paris, 19031914.
Burckhardt, John Lewis. Travels in Arabia, comprehending an account of those ter-
ritories in Hedjaz which the Mohammadans regard as sacred, edited by William
Ouseley. London, 1829.
Burton, Sir Richard F. Personal narrative of a pilgrimage to El-Medinah and Meccah.
London 18551856.
Caetani, Leone. Annali dellIslam. Milano, 19051926.
ChronographiaIslamica; ossiariassuntocronologico dellastoriadi tutti i popoli
musulmani dallanno 1 allanno 922 della Higrah (6221517 dellra volgare) corre-
dato della bibliografia di tutte le principali fonti stampate e manoscritte. 1912.
Canisius, Henricus. Thesaurus monumentorum ecclesiasticorum et historum, ed Ja-
cobcus Basnage. Antwerp, 1725.
Carlyle, Thomas. On heros, hero-worship and the heroic in history. London, 1840.
Casanova, Paul. Mohammed et la fin du monde; tude critique sur l Islam primitif.
Paris, 19111924.
Casiri, Miguel. Biblioteca Arabico-Hispana Escurialensis sive librorumMss. quos Ara-
bice abauctoribus magnampartemArabo-Hispanis compositos bibliothecaCoeno-
bii Escurialensis complectitur (Madrid, 17501770).
Catalogus codicumOrientaliumBibliothecae Academiae Lugduno-Batavae, editedby
R.A. Dozy, M.J. de Goeje, M.Th. Houtsma, and Th.W. Juynboll. Leiden, 18511877.
Caussin de Perceval, Armand Pierre. Le combat de Bedr; pisode de la vie de
Mahomet in Journal asiatique, 3e srie, t. 7 (1839), pp. 99145.
Essai sur l histoire des Arabes avant l islamisme pendant l poque de Mahomet
et jusqu la rduction de toutes les tribus sous la loi musulmane (Paris, 18471848).
Reprint, Graz, 1967.
Cooke, George Albert. Textbook of North-Semitic inscriptions: Moabite, Hebrew, Phoe-
necian, Aramaic, Nabataean, Palmyrene, Jewish. Oxford, 1903.
Corpus inscriptionumGraecarum, vol. iii, edited by August Bckh. Berlin, 1853.
Corpus scriptorum historiae Byzantinae; ed. Barthold G. Niebuhr, et al. Bonn, 1828
1897.
al-Damr, Muammad b. Ms. ayt al-ayawn. Cairo, 1309/18911892.
al-Dn, Ab Amr Uthmn b. Sad. Orthographie und Punktierung des Korans; zwei
Schriften von ad-Dn, ed. by Otto Pretzl. Istanbul, 1932. (Bibliotheca islamica; 3)
K. al-Taysr f l-qirt al-sab = Das Lehrbuch der sieben Koranlesungen, ed. by
Otto Pretzl. Istanbul, 1930. (Bibliotheca islamica; 2) Reprint, 1966.
Derenbourg, Hartwig. Le culte de la desse al-Ouzz en Arabie au IVe sicle de
notre re in: Recueil de mmoires orientaux (Paris, 1905), pp. 3340.
al-Dhahab, Muammad b. Amad. Tadhkirat al-uff. Hyderabad, 13331334/
(19141915) Reprints 19551958, and 19681970.
Tajrd asm al-aba. Hyderabad, 1315/18971898. New edition by lia
Abd al-akm Sharaf al-Dn, Bombay, 1389/1969.
Dieterich, Albrecht. Eine Mithrasliturgie erlutert. Leipzig, 1903.
al-Dimy al-Bann, Amad b. Muammad. Itf fual al-bashar f qirt al-
arbaat ashar. Cairo, 1317/1899. Editedby ShabnMuammadIsml, Beirut and
Cairo, 1407/1987.
Diyrbakr, usayn b. Muammad. Tarkh al-khams f awl anfas nafs. Misr,
1283/1866. Reprint Beirut, 1970.
Dozy, Reinhart Pieter Anne. Essai sur l histoire de l islamisme, traduit du hollandais
par Victor Chauvin. Leiden, 1879.
Die Israeliten zu Mekka von Davids Zeit bis ins 5. Jahrhundert unserer Zeitrech-
nung; einBeitrag zur alttestamentlichenKritik undzur Erforschung des Ursprungs
des Islams, aus dem Hollndischen bersetzt. Leipzig, 1864.
Dussaud, Ren. Les Arabes en Syrie avant l islam. Paris, 1907.
Encyclopdia Biblica; a critical dictionary of the literary, political and religious his-
tory, the archaeology, geography and natural history of the Bible, edited by T.K.
Cheyne and J. Sutherland Black. London, 18991903.
EI
2
: Encyclopaedia of Islam. New edition. Leiden, Brill, 19602004.
EQ: Encyclopaedia of the Qurn, general editor Jane Dammen McAuliffe. Leiden,
Brill, 20012006.
Eutychius, Patriarch of Alexandria. Annales, ed. Louis Cheikho. Beryti, 1906.
Ewald, Georg HeinrichAugust. Grammaticacriticalinguae arabicae cumbrevi metro-
rumdoctrina. Leipzig, 18311833.
Die Propheten des Alten Bundes, erklrt von Heinrich Ewald. 2nd ed. Gttin-
gen, 18671868.
Fell, Winand. Die Christenverfolgung in Sdarabien und die himjarisch-thiopi-
schen Kriege nach abessinischer berlieferung in ZDMG, 35 (1881), pp 174.
al-Fihrist Ibn al-Nadm, Muammad b. Isq
Fischer, August. Eine Qoran-Interpolation in Orientalische Studien Theodor Nl-
deke zum siebzigsten Geburtstag gewidmet von Freunden und Schlern und in
ihremAuftrag herausgegeben von Carl Bezold, pp. 3355. Gieen, 1906.
. Die Quantitt des Vokals des arabischen Pronominalsuffixes hu (hi) in:
Oriental studies publishedincommemorationof the 40thanniversaryof Paul Haupt
as director of the Oriental Seminary of the John Hopkins University, Baltimore.
Leipzig, 1926, pp. 390402.
Fleischer, Heinrich Leberecht. Catalogus librorum manuscriptorum qui in Biblio-
thecasenatoriacivitatis Lipsiensis asservantur: Codices orientalium, descripserunt
Heinrich Leberecht Fleischer et Franz J. Delitzsch, ed. A.G.R. Neumann. Leipzig,
1838.
Kleinere Schriften. Leipzig, 18851888.
ber einenangeblichenBrief Muammads inZDMG, 17 (1863), pp. 385386;
also in his Kleinere Schriften, 3 (1888), pp. 400401. Reprint, 1968.
Flgel, Gustav Leberecht. Die arabischen, persischen und trkischen Handschriften
der kaiserlich-kniglichen Hofbibliothek zu Wien. Wien, 18651867. Reprint, 1977.

= Corani textus arabicus; ad fidemlibrorummanuscriptorumet impresso-

rumet ad praecipuorum. Leipzig, 1834; ( repr. 1858).
Die grammatischen Schulen der Araber nach den Quellen bearbeitet: (1) Die
Schulen von Basra und Kufa und die gemischte Schule. Leipzig, Brockhaus, 1862.
Reprint, 1966.
Mn, seine Lehre undseine Schriften; einBeitragzur Geschichte des Manichis-
mus, aus demFihrist des AblfaradschMuammad, edited, translated, andcom-
mented. Leipzig, 1862. Reprint, 1969.
ber Muammad bin Ishks Fihrist al-ulm, in ZDMG, 13 (1859), pp. 559
650.
Fraenkel, Siegmund. Die aramischen Fremdwrter im Arabischen; eine von het Pro-
vinciaal Utrechtsch Genootschap van Kunsten en Wetenschappen gekrnte Preiss-
chrift. Leiden, 1886. Reprint Hildesheim, 1962.
De vocabulis in antiquis Arabum carminibus et in Corano peregrinis. Leiden,
Brill, 1880.
[Review] Th. Nldeke, Beitrge zur Geschichte des Alexanderromans (1890) in
ZDMG, 45 (1891), pp. 309330.
Frank-Kamenetzky, Israel G. Untersuchungen ber das Verhltnis der dem Umajja
b. Abi alt zugeschriebenen Gedichte zum Qorn. Ph.D. thesis, Knigsberg, 1911.
58.
Freytag, Georg Wilhelm Friedrich. ; Arabum proverbia vocalibus instruxit,
latine vertit, commentario illustravit et sumtibus suis edidit. Bonnae a. Rh., 1838
1843.
Einleitung in das Studium der arabischen Sprache bis Mohammed und zum
Teil spter zum allgemeinen Gebrauche auch fr die, welche nicht Hebrisch und
Arabisch treiben. Bonn, 1861. Reprint Osnabrck, 1972.
Lexicon arabico-latinum praesertum ex Djeuharii Firuzabadiique et . Halle,
18301837. Reprint Osnabrck, 1968.
Friedlaender, Israel. Abdallh b. Sab, der Begrnder der a in Zeitschrift fr
Assyriologie 23 (1909), pp. 296327, 24 (1910), 146.
Heterodoxies of the Shiites in the presentation of Ibn Hazm, in Journal of
the American Oriental Society, 28 (1907/08), pp. 180; 29 (1909), 1183.
Gagnier, Jean. La vie de Mahomet. Amsterdam, 1732. 2 v. German translation, Leben
Mohammeds des Propheten, nachdemFranzsischenmit AnmerkungenvonC.F.R.
Vetterlein. (Coethen, 1802).
GAL Brockelmann, Carl. Geschichte der arabischen Litteratur. Weimar & Berlin,
18981902.
Garcin de Tassy, Joseph Hliodore. Chapitre inconnu du Coran, publi et traduit
pour la premire fois. Journal asiatique, 3e srie, vol. 13 (1842), tome 1, pp. 431
439.
Gardthausen, Victor. Griechische Palographie. 2nd ed. Leipzig, 19111913. Reprint,
1978.
GAS Sezgin, Fuat. Geschichte des arabischen Schriftums. 19672000.
Geiger, Abraham. Einiges ber Plan und Anordnung der Mischnah in Wissen-
schaftliche Zeitschrift fr jdische Theologie, 2 (1856), pp. 474492.
Was hat Mohammed aus dem Judenthume aufgenommen? Eine von der
PreuischenRheinuniversitt gekrnte Preisschrift. Bonn, 1833. 2ndrev. ed., 1902.
Reprint, 1971.
Gerock, Carl (Carolus, Charles) Frdric (Friedrich, Fridericus.) Al-Coranus prophe-
tici munaris Christi laudatur. Argentorari, 1833.
Versuch einer Darstellung der Christologie des Korans Hamburg & Gotha,
Perthes, 1839.
Geyer, Rudolf. [Review] Vollers, Volkssprache und Schriftsprache im alten Arabien
in Gttingische gelehrte Anzeigen, 171 (1909), Nr. 1, cols. 1056.
Glaser, Eduard. Zwei Inschriften ber den Dammbruch von Mrib; ein Beitrag zur
Geschichte Arabiens im 5. und 6. Jahrhundert nach Chr. in Mitteilungen der
Vorderasiatischen Gesellschaft, 2 vi (1897.)
Glycas, Michael. . Annales, a mundi exordio usque ad obitum Alexii
Comneni Paris, 1660. (Corpus byzantinae historiae; 16)
de Goeje, Michael Jan. Die Berufung Mohammeds in Orientalische Studien Theo-
dor Nldeke zum70. Geburtstag gewidmet (1906), vol. 1, pp. 15.
De Legende der Zevenslapers van Efeze in Verslagen en mededeelingen der
Koninklijke Akademie van Wetenschappen, Afd. Letterkunde, reeks 4, deel 3.
Amsterdam 1900.
Goldziher, Ignaz. Abhandlungen zur arabischen Philologie. Leiden, Brill, 18961899.
Reprint, 1982.
Al B Mubrak: al-H itat at-Taufqya al-gadda in Wiener Zeitschrift fr
die Kunde des Morgenlandes, 4 (1890), pp. 347352.
Aus der Theologie des Fachr al-dn al-Rz in Der Islam, 3 (1912), pp. 213
247.
[Review] C.H. Becker, Papyri Schott-Reinhardt, I in Deutsche Literatur-zei-
tung fr Kritik der internationalen Wissenschaft, vol. 27, no. 41 (13 October 1906),
col. 25472550.
Beitrge zur Literaturgeschichte der a und der sunnitischen Polemik. Wien,
1874. (Sitzungsberichte der Akademie der Wissenschaften in Wien, 78, [1874]. pp.
439524.)
[Review] Carra de Vaux, Le Mahomtisme; le gnie smitique et le gnie aryen
dans l islam in ZDMG, 53 (1899), pp. 380387.
[Review] A. de Vlieger, Kitb al Qadr; matriaux pour servir l tude de la
doctrine de la prdestination dans la thologie musulmane. in ZDMG, 57 (1903),
pp. 392402.
Der Dwn des arwal b. Aus al-Huteja in ZDMG, 46 (1892), 153, 173225,
471527, and 47 (1893), 4385, 163201.
IntroductiontoIslamic theology andlaw, translatedfromVorlesungenber den
Islam (Heidelberg, 1910), by Andras and Ruth Hamori. Princeton, N.J., 1981.
Islam in Jewish encyclopedia, vol. 6 (1904), pp. 651659.
Kmpfe und die Stellung des Had im Islam in ZDMG, 61 (1907), pp. 860
872.
Muslimstudies, translated by C.R. Barber and Samuel M. Stern. Chicago, 1967,
1977.
Neue Materialien zur Literatur des berlieferungswesens bei den Moham-
medanern in ZDMG, 50 (1896), pp. 465506.
Neuplatonische undgnostische Elemente imHadth inZeitschrift fr Assyri-
ologie, 22 (1908), pp. 317344.
Die Sabbathinstitution im Islam in Gedenkbuch zur Erinnerung an David
Kaufmann (Breslau, 1900), pp. 86102.
Sulenmnner im Arabischen in ZDMG, 55 (1910), pp. 503505.
Schools of Koranic commentators, translation of Die Richtungen der islamis-
chen Koranauslegung, translated by Wolfgang Behn. Wiesbaden, 2006.
Turb und Haar in zurechtweisenden Redensarten in ZDMG, 42 (1888),
pp. 587590.
Ueber die Dualtitel in WZKM, 13 (1899), pp. 321329.
ber muhammedanische Polemik gegen Ahl al-kitb in ZDMG, 32 (1878),
pp. 341387.
The hirs, translatedfromthe Germanby W.H. Behn. Leiden, 1971, reprinted
2008. (Brill classics in Islam, vol. 3)
Grimme, Hubert. Mohammed: (1. Teil) Das Leben nach den Quellen: (2. Teil) Ein-
leitung in den Koran; Systemder koranischen Theologie. Mnster i. W., 18921895.
(Darstellungen aus demGebiete der nichtchristlichen Religionen, vol. 7, 11.)
Mohammed; die weltgeschichtliche Bedeutung Arabiens. Mnchen, 1904.
Grohmann, Adolf. Corpus papyrorum Raineri archiducis Austriae. 3. Series arabica,
tom. 1, pars 1: Allgemeine Einfhrung in die arabischen Papyri nebst Grundzgen
der arabischen Diplomatik. Wien, 1924. iv, 108 pp.; 2: Protokolle, Einleitung und
Texte. Wien, 1924.
Grnert, Max Theodor. Die Alliteration im Alt-Arabischen in Verhandlungen, 7th
International Congress of Orientalists, 1886, Semitische Sektion, pp. 183227.
Der Imla; der Umlaut im Arabischen. Wien, 1876. (Sitzungsberichte der Phil.-
historischen Classe der kaiserl. Akademie der Wissenschaften, Band 81).
Gnther, Georg L. Louis. Die Idee der Wiedervergeltunginder Geschichte undPhiloso-
phie des Strafrechts; ein Beitrag zur universal-historischen Entwicklung derselben.
Erlangen, 18891895.
Guidi, Ignazio. Testi orientali inediti sopra i Sette Dormienti di Efeso, pubblicati e
tradotti. Roma, 1885. (Atti della Accademia nazionale dei Lincei. Classe di scienze
morali, storiche e filologiche, Ser. 3, no. 12.)
Haas, Hans. Das Bild Muhammeds im Wandel der Zeiten in Zeitschrift fr Mis-
sionskunde und Religionswissenschaft, 31 (1916), pp. 161171, 194203, 225239,
258269, 289295, 321333, 353365. Also published as off-print.
jj Khalfa Ktib elebi
al-alab, Burhn al-Dn Al b. Ibrhm. Insn al-uyn f srat al-Amn al-Mamn
al-marfa bi-l-sra al-alabiyya. Cairo, 1280 or 1281/18631864 or 18641865.
amsa Ab Tammm abb b. Aws
von Hammer-Purgstall, Joseph, Freiherr. Gemldesaal der Lebensbeschreibungen
groer moslemischer Herrscher der ersten sieben Jahrhunderte der Hidschret.
Darmstadt, 18371839.
Der Islam und Mohammed in Jahrbcher der Literatur, Wien, 68 (1834),
pp. 151; 69 (1835), pp. 190.
Literaturgeschichte der Araber; von ihrem Beginne bis zum Ende des zwlften
Jahrhunderts der Hidschret. Wien, 18501856.
Die Handschriften in Gttingen: (3) Nachlsse von gelehrten orientalischen Hand-
schriften. Berlin, 1894
al-arr, al-Qsimb. Al. Durrat al-ghawws f awhmal-aw, editedby Heinrich
Thorbecke. Leipzig, 1871.
Hartmann, Martin. Der islamische Orient; Berichte und Forschungen: Vol. 1, part 1,
Die angebliche sra des Ibn Ishaq pp. 3234.
assn b. Thbit. Dwn. Tunis, 1281/18641865.
Dwn of assn b. Thbit, edited by Hartwig Hirschfeld. Leiden, Brill, 1910.
tim al-. Der Dwn des arabischen Dichters Htim Tej nebst Fragmenten; her-
ausgegeben, bersetzt und erlutert von Friedrich Schulthe. Leipzig, 1897.
Hauleiter, Hermann. Register zumQorankommentar des abar (Kairo, 1321.) Stra-
burg, 1912.
Heller, Bernhard. Elments parallles et origines de la lgende des Sept Dormants
in Revue des tudes juives, 49 (ca. 1902), pp. 190218.
Hippocates. uvres compltes d Hippocrate, traduction nouvelle avex le texte grec
en regard, collation sur les manuscrits et toutes les ditions; accompagne
d une introduction, de variantes et de notes par mile Littr. Paris, 1839
1861.
Hirschfeld, Hartwig. Beitrge zur Erklrung des Koran. Leipzig, 1886.
Jdische Elemente imKoran (Berlin, 1878.) 71 p. (Dr. phil., Straburg.)
New researches into the composition and exegesis of the Qoran. London, 1902.
(Asiatic monographs; 3.)
The History of Alexander the Great, being the Syriac versions, edited from five Mss. of
the pseudo-Callisthenes, with an English translation, accompanied by an historical
introduction on the origins and various Oriental and European versions by
Ernest Wallis Budge. Cambridge, 1889. [Reprint, 1976.]
Horovitz, Joseph. Alter und Ursprung des Isnd in Der Islam, 8 (1918), pp. 3947.
Aus den Bibliotheken von Kairo, Damaskus und Konstantinopel in Mit-
teilungen des Seminars fr orientalische Sprachen zu Berlin, 2nd Abteilung, p.
14sq.; also in Beitrge zur Erschlieung der arabischen Handschriften in Istanbul
und Anatolien, ed. by Fuat Sezgin (1986), pp. 355422.
Hottinger, JohannHeinrich. Historiaquae ex variis orientaliummonumentis collecta.
Tiguri, 1651. 2nd ed., Tiguri, 1660.
Houtsma, Martijn Th. Iets over den dagelijkschen alat der Mohammedaner in
Theologisch tijdschrift, 24 (1890), pp. 127134.
Huart, Clment. Une nouvelle source du Qorn in Journal asiatique, 10e srie, t. 4
(1904), 125167.
The Hudsailian poems, edited by Johann G.L. Kosegarten. London, 1854.
Hunnius, Carl. Das syrische Alexanderlied. Thesis, Gttingen, 1904., also in ZDMG,
60 (1906), pp. 169209.
Ibn Abd Rabbih, Amad b. Muammad. al-Iqd al-fard. Cairo, 1305/18871888.
Ibn Askir, Al b. al-asan. al-Tarkh al-kabr. Damascus, 1329/1911.
Ibn al-Athr, Izz al-Dn. Ibn-El-Athiri Chronicon quod perfectissimum inscribitur;
al-Kmil f l-tarkh, edidit Carolus Johannes Tornberg. Leiden, 18511876. Reprint,
Beirut, 19651967.
Usd al-ghba f marifat al-aba. Cairo, 12851287/18601871; (by a misprint
vol. 5 is given as 1280).
Ibn al-Athr, Majd al-Dn al-Mubrak b. Muammad. Nihya f gharb al-adth
wa-l-athar. Cairo, 1311-1318-22/19001904. [Neither full title nor edition specified
by Schwally.]
Ibn Badrn, Abd al-Malik b. Abd Allh. Commentaire historique sur le pome d Ibn
Abdoun, par Ibn Badroun, publi pour la premire fois, prcd d une introduc-
tion et accompagn de notes, d un glossaire et d un index des noms propres, par
Reinhart P.A. Dozy. Leiden, 1846.
Ibn Durayd, Muammad b. al-asan. Abu Bekr Muhammed ben el-Hasan b. Doreids
Genealogisch-etymologisches Handbuch [Kitb al-Itiqq], edited by Ferdinand
Wstenfeld. Gttingen, 1854.
Ibn ajar al-Asqaln, Amad b. Al. al-Iba f tamyz al-aba, ed. A. Sprenger
and W.N. Lees. Calcutta, 18561888.
Tahdhb al-tahdhb. Hyderabad, 13251327/19071909.
Ibn anbal, Amad. Musnad. Cairo, 1311/18931894.
Ibn azm, Al b. Amad. The Heterodoxies of the Shiites according to Ibn Hazm;
introduction, translation and commentary by Israel Friedlaender. JAOS, 28
(1907), pp. 180 and 29 (1909), pp. 1183.
IbnHishm, Abdal-Malik. : Das LebenMuhammeds nachMuham-
med b. Ishk, edited by Ferdinand Wstenfeld. Gttingen, 18581860. Reprint,
1961. German translation by Gustav Weil, Das LebenMohammeds. Stuttgart, 1864.
The Life of Muammad, a translation of Srat rasl Allh, with introduction
and notes by Alfred Guillaume. London, 1955.
IbnIsq, Muammad. The Makingof the last prophet; areconstructionof the earliest
biography of Muammad (the K. al-Mubtaba), by Gordon D. Newby. Columbia,
S.C., 1989.
Ibn al-Jazar, Muammad b. Muammad. Ghyat al-nihya f abaqt al-qurr =
Das biographische Lexikon der Koranlehrer, edited by G. Bergstrer and Otto
Pretzl. Cairo, Leipzig, 19331935. (Bibliotheca Islamica; 8)
Munjid al-muqrin wa-murshid al-libn. Cairo, 1350/1931. [not identified.]
Ibn Khlawayh. usayn b. Amad. Sammlung nichtkanonischer Koranlesearten =
Mukhtaar f shawdhdh al-Qurn, ed. G. Bergstrer. Leipzig, 1934. (Bibliotheca
Islamica)
Ibn Khaldn, Abd al-Ramn. al-Ibar wa-dwn al-mubtada wa-l-khabar f ayym
al-Arab wa-l-Ajamwa-l-Barbar. Cairo, 1284/1867.
al-Muqaddima. Beirut, 1886. 556 p. [English translation by Franz Rosenthal,
1958]
Ibn Khallikn, Amad b. Muammad. Wafayt al-ayn. Blq, 1299/1881.
Ibn al-Khab al-Dahsha, 1349 or 501431. Tufa dhaw-l-arab: ber die Namen und
Nisben bei Bor, Muslim, Mlik, edited by Traugott Mann. Leiden, 1905.
IbnMjah, AbAbdAllhMuammadb. Yazdal-Qazwn. Sunan. Delhi, 1282, 1289;
Cairo, 1313. [Edition or Ms. not supplied.]
Ibn Mlik, Muammad b. Abd Allh. Alfijjah, carmen didacticum grammaticum,
autore IbnMlik et inAlfijjamcomentarius quemconscripsit IbnAkil, edidit Fried-
rich H. Dieterici. Leipzig, 1851.
Ibn al-Nadm, Muammad b. Isq. al-Fihrist, edited by Gustav Flgel, J. Roediger,
and A. Mller. Leipzig, 18711872. (Edited and translated by Bayard Dodge, 1970.)
Ibn Qutayba, Abd Allh b. Muslim. Ibn Coteibas Handbuch der Geschichte (Kitb
al-Marif ) aus den Handschriften zu Wien, zu Gotha und zu Leyden, edited by
Ferdinand Wstenfeld. Gttingen, 1850. Reprint, 1977.
Liber posis et potarum (al-Shir wa-l-shuar) quem edidit J.M. de Goeje.
(Leiden, 1904). [Partial English translation entitled Introduction to the Book , by
Arthur Wormhoudt (1973.)]
Ibn Rustah, Amad b. Umar. Kitb al-alk (al-Alq) an-nafsa VII; et El-Kitb al-
boldn, auctore Ahmed b. Wdhih al-Jakb (al-Yaqb.) 2nd ed. Leiden, 1892.
(Bibliotheca geographorumArabicorum; 7.) Reprint, Baghdd, ca. 1970.
Ibn Sad, Muammad. ( ), Biographie Muammads bis zur Flucht,
vol. 1, part 1, ed. by Eduard Sachau and Eugen Mittwoch. Leiden, 1905.
( ) Biographie Muhammeds; Ereignisse seiner medinischen
Zeit, Personal-Beschreibung und Lebensgewohnheiten, vol. 1, part 2, edited by
Eugen Mittwoch. Leiden, 1917.
( ) Biographien der Basrier bis zum Ende der zweiten Klasse,
vol. 7, part 2, ed. Bruno Meissner. Leiden. 1915.
( ) Biographiender Basrier vonder drittenKlasse bis zumEnde
und der Traditionarier in anderen Teilen des Islams, vol. 7, part 1, ed. by Eduard
Sachau. Leiden, 1918.
( ) Biographien der Frauen, vol. 8, ed. by C. Brockelmann.
Leiden, 1904.
( ) Biographien der Genossen, die sich noch vor der Eroberung
Mekkas bekehrten, demletzten Termin, wo ein Islamnoch freiwillig und besonders
verdienstlich war, vol. 4, part 2, edited by Julius Lippert. Leiden, 1908.
. ( ) Biographien der Kufier, vol. 6, edited by Karl Vilhelm
Zettersten. Leiden, 1909.
( ) Biographien der medinischen Kmpfer Muhammeds in der
Schlacht bei Bedr, vol. 3, part 2, edited by Josef Horovitz. Leiden, 1904.
( ) Biographien der mekkanischen Kmpfer Muhammeds in
der Schlacht bei Bedr, vol. 3, part 1, edited by Eduard Sachau. Leiden, 1904.
( ) Biographien der Muhirn und Ansr, die nicht bei Bedr
mitgefochten, sich aber frh bekehrt haben, alle nach Abissinien ausgewandert
sind und dann an der Schlacht bei Ohod teilgenommen haben, vol. 4, part 1, ed.
Julius Lippert. Leiden, 1906.
( ) Biographien der Nachfolger in Medina, sowie der Gefhrten
und der Nachfolger in dem brigen Arabien, vol. 5, ed. K.V. Zettersten. Leiden,
Brill, 1905.
( ) Die Feldzge Muhammeds, vol. 2, part 1, ed. Josef Horovitz.
Leiden, Brill, 1909.
( ) Letzte Krankheit, Tod und Bestattung Muhammeds, vol. 2,
part 2, edited by Friedrich Schwally. Leiden, Brill, 1912.
( ) K. al-abaqt al-kabr; Biographien Muhammeds, seiner
Gefhrten und der spteren Trger des Islams bis zum Jahre 230 der Flucht, im
Verein mit C. Brockelmann, J. Horovitz, J. Lippert, B. Meiner, E. Mittwoch, Fr.
Schwally und K.V. Zettersten, edited by Eduard Sachau. Leiden, 19041928.
Ibn Salma, Ab l-Qsim Hibat Allh. al-Nsikh wa-l-manskh f l-Qurn, in the
margin of al-Wids Asbb al-nuzl. Cairo, 1316/18981899.
Ibn Taghrbird, Ab l-Masin Ysuf. Abul-Mahasin b. Tagri Bardii Annales, quibus
titulus est a codd. mss. nunc primum Arabice, ed.
T.G.J. Juynboll et al. Leiden, Brill, 1855 [i.e. 1852]-1861.
Extracts from Ab l-Mahsin b. Tagr Birds chronicle, entitled Hawdith
ad-duhr f mad l-ayym wash-shuhr, edited by William Popper. Berkeley,
Calif., 19301942.
Itf al-Dimy al-Bann, Amad b. Muammad
al-Itqn al-Suy, Jall al-Dn
Jacob, Benno. Im Namen Gottes; eine sprachliche und religionsgeschichtliche
Untersuchung ber im Alten und Neuen Testament in Vierteljahrsschrift fr
Bibelkunde, talmudische und patristische Studien, vol. 1, part 1 (June 1903), pp.
171195.
Jacob, Georg. Altarabisches Beduinenleben nach Quellen geschildert. Berlin, 1895. ix,
179 p. (Studien in arabischen Dichtern, vol. 3.) 2nd enlarged reprint, Berlin, 1897.
Das Leben der vorislamischen Beduinen. Berlin, 1895. (His: Studien in arabis-
chen Dichtern; 3.)
Der muslimische Fastenmonat Ramadn in Jahresbericht der Geographis-
chen Gesellschaft, Greifswald, 6 (1896), pp. 133.
al-Ji, Amr b. Bar.
= Le Livre des beauts et des

antithses, edited by Gerlof van Vloten. Leiden, 1888.
Jamharat ashr al-Arab. Blq, 1308.
al-Ja, Amad b. Al. Akmal-Qurn. Constantinople, 1335/1916 or 1917.
al-Jawhar, Isml b. ammd. (Tj al-lugha wa) al-a (al-Arabiyya). Blq,
1282/1865 and 1292/1875.
al-Jazar, Ab l-Khayr Muammad b. Muammad Ibn al-Jazar, Muammad b.
Muammad
Jeffery, Arthur, ed. Materials for the history of the text of the Qurn: the old codices.
The Kitb al-maf of Ibn Ab Dwd, together with a collection of the variant
readings fromthe codices of Ibn Masd , etc. Leiden, Brill, 1937. Reprint, 1975.
The Jewish encyclopedia. New York & London, 19011906.
al-Jurjn, Al b. Muammad. Definitiones / Sejjid Scherif Ali ben Mohammed
Dschordschani, ed. by Gustav Flgel. Leipzig, 1845.
Juynboll, Gautier H.A. Encyclopedia of canonical adth. Leiden, Brill, 2006.
Juynboll, Theodoor Willem. Handleiding tot de kennis van de Mohammedaansche
wet, volgens de leer der Sjafiitische school. Leiden, Brill, 1903.
Kanz al-umml al-Muttaq, Al b. Abd al-Malik
vonKarabacek, Joseph. Julius Eutings Sinatische Inschriften, in: Wiener Zeitschrift
fr die Kunde des Morgenlandes, 5 (1891), pp. 311326.
Ein Koranfragment des 9. Jahrhunderts aus dem Besitz des Seldschukensul-
tans Kaikubad in his: Zur orientalischen Altertumsfrage, 6. (Sitzungsberichte der
K. Akademie der Wissenschaftten in Wien, 184, no. iii, 1918.)
al-Kasn, Al al-Dn Ab Bakr b. Masd. Bad al-ani f tartb al-shari. Cairo,
1327/19091910. Reprint, Beirut, 1406/1986.
Kasimirski, A. de Biberstein-Kazimirski, Albert de
Ktib elebi (Mustaf b. Abd Allh jj Khalfa). Kashf al-unn = Lexicon biblio-
graphicumet encyclopdicumad codicumVindbonensium, Parisiensium, et Bero-
linensis fidem primum edidit Latine vertit et commentario indicibusque instruxit
Gustav Flgel. London, Oriental Translation Fund, 18351858. Reprint, NewYork,
1964.
Kazem-Beg, Mirza Alexander. Observations sur le Chapitre inconnu du Coran,
publi et traduit par M. Garcin de Tassy in Journal asiatique, 4e srie, 2 (1843),
pp. 373429.
de Kazimirski-Biberstein, A. de Biberstein-Kazimirski, Albert
Keler, Konrad. Mani, Manicher in Realencyklopdie fr protestantische Theolo-
gie und Kirche, 12 (1903).
al-Khans bt. Amr. Dwn (Beirut, 1888), and its translation by Victor de Coppier, Le
Diwand Al Hansa, prcd d une tude sur les femmes potes de l ancienne Arabic
(Beirut, 1889).
al-Khab al-Tibrz, Muammad b. Abd Allh. Mishct-ul-mabh; or, A collection
of the most authentic traditions, regarding the actions and sayings of Muhammed;
exhibiting the origin of the manners and customs, the civil, religious and military
policy of the Musulemns, translated by Arnold N. Matthews. Calcutta, 18091810.
[Arabic ed., Dehli, 1268/1852.]
al-Khzin al-Baghdd, Al al-Dn Al b. Muammad. Tafsr. Cairo, 1309/18911892.
al-Khazraj, Amad b. Abd Allh. Khulat tahdhb al-kaml f asm al-rijl. Cairo,
1322/1904. Reprint, 1391/19171.
Kinberg, Naphtali. A Lexicon of al-Farrs terminology in his Qurn commentray,
with full definitions, English summaries, and extensive citations =

. Leiden, Brill, 1996.

al-Kind, Muammadb. Ysuf. The Governors andjudges of Egypt; or, Kitbel umar
(el wulh) wa Kitb el qudh, edited by Rhuvon Guest. Leyden and London, 1912.
(E.J.W. Gibb memorial series, 19.)
Koch, John. Die Siebenschlferlegende, ihr Ursprung und ihre Verbreitung; eine my-
thologisch-literaturgeschichtliche Studie. Leipzig, 1883.
Koran name of translator or editor
Krehl, Ludolf. Das Leben und die Lehre des Muhammed, Teil 1. Leipzig, 1884.
Krenkow, Fritz. Tabrzs Kommentar zur Burda des Kab b. Zuhair in ZDMG, 65
(1911), pp. 241279.
Khnel, Ernst. Islamische Kleinkunst. Berlin, 1925.
Kun, Aleksandr L. in: -
: , ed. Nikolai] A. Maev. S.-Peterburg, 1874. Vol. 3,
pp. 401404.
Kurd Al, Muammad. Khia al-Shm. Damascus, 1334/19251347/1928.
Labd b. Raba. Der Dwn des Lebd, nach einer Handschrift zum ersten Male
herausgegeben von Jsuf Dij-ad-dn al-Chlid. Wien, 1880.
Die Gedichte des Lebd nach der Wiener Ausgabe berstetzt; Diwan aus dem
Nachlasse A. Hubers, herausgegeben von Carl Brockelmann. Leiden, Brill, 1891.
de Lagarde, Paul. Sra in Nachrichten von der Kniglichen Gesellschaft der Wis-
senschaften und der Georg-August-Universitt zu Gttingen, 1889, pp. 296298.
Lammens, Henri. L ge de Mahomet et la chronologie de la sra in Journal asia-
tique, 10e srie, vol. 17 (1911), pp. 209250.
Le Berceau de l Islam, l Arabie occidentale la veille de l Hgire; vol. 1: Le
climat, les Bdouins. Roma, 1914.
tudes sur le rgne du Calife Omayade Mowia 1er. Paris, 1908. FromMlanges
de la Facult orientale de l Universit Saint-Joseph, 1 (1906), pp. 1108; 2 (1907),
pp. 1172; 3 (1908), pp. 145312.
Fima et les filles de Mahomet; notes critiques pour l tude de la Sra. Roma,
Sumptibus Pontificii Instituti Biblici, 1912.
Mahomet fut-il sincre? in Recherches de science religieuse, 2 (1911), pp. 25
53, 140166. English abstract Was Mohammed sincere? in Moslem world, 5
(1915), pp. 262267.
Qoranet tradition, comment fut compose la vie de Mahomet inRecherches
de science religieuse, 1 (1910), pp. 2651.
Lane, Edward William. An Account of the manners and customs of the modern Egyp-
tians, written in Egypt during the years 18331834 and 1835, partly fromnotes, made
in the years 18251828. London, 1836.
Lansdell, Henry. Russian Central Asia, including Kuldja, Bokhara, Khiva and Merv.
London, 1885.
Leaves fromthree ancient Qurns, possibly pre-Othmnic, with a list of their variants;
edited by Alphonse Mingana and Agnes Smith Lewis. Cambridge, Engl., 1914.
LeBeau, Charles. Histoire du Bas-Empire / Charles LeBeau & Jean Antoine Saint-
Martin. Paris, 1824.
Leszynsky, Rudolf. Die Juden in Arabien zur Zeit Mohammeds. Berlin, 1910.
Levy, Jacob. Chaldisches Wrterbuch ber die Targumim und eine grossen Theil des
rabbinischen Schriftthums. Leipzig, 18671868. Reprint, 1959.
Neuhebrisches und chaldisches Wrterbuch ber die Talmudim und Midra-
schim. Leipzig, 18761889.
Levy, Moritz Abraham. Bemerkungenzu denarabischenAnalektendes HerrnProf.
Hitzig in ZDMG, 12 (1858), p. 712.
Lewis, Agnes Smith. Leaves fromthree ancient Qurns, possibly pre-Othmnic, ed. by
A.S. Lewis and A. Mingana. Cambridge, 1914.
Liber jubilaeorum qui idem a Graecis H lept Genesis inscribitur versione Graecae
deperdita nunc nonnisi in Geez lingua conservatus nuper ex Abyssinia in Europam
allatus , ed. August Dillmann. Kiel, 1859.
Lindberg, Jacob Christian. Lettre P.O. Brndsted Sur quelques mdailles cufiques
dans le Cabinet du Roi de Danemark, rcemment trouvs dans l le de Flaster, et sur
quelques manuscrits cufiques, avec xii planches. Copenhague, 1830.
Loth, Otto. Tabaris Korancommentar in ZDMG, 35 (1881), pp. 588628.
Luqmn. Fables de Loqmanle Sage; le texte revue de nouveausur les mss. accompagn
d une version franaise et des notes et prcd d une introduction sur la personne
de Loqman et sur l origine de ce recueil de fables par J[oseph] Derenbourg. Berlin
& Londres, 1850.
Lyall, Charles J. The words Hanif and Muslim in: Journal of the Royal Asiatic
Society, London, 1903, pp. 771784.
Mahmud Efendi, astronomer. Sur le calendrier arabe avant l islamisme, et sur la
naissance et l ge du prophte Mohammad in Journal asiatique, 5e srie, 11
(1858), pp. 109192.
Mlik b. Anas. al-Mudawwana al-kubr. Cairo, 1324/1906. Reprints, Beirut and Bagh-
dad, ca. 1982.
al-Muwa. Blq, 1280/18631864.
Mani Flgel, Gustav L.
al-Maqdis, Muammad b. Amad. Descriptio imperii Moslemici = Asan al-taqsm
f marifat al-aqlm, edited by M.J. de Goeje. Leiden, Brill, 1906. Reprint, Frank-
furt a.M., 1992.
al-Maqdis, al-Muahhar b. hir. [Bad al-khalq, or Bad wa-l-tarkh, GAS, vol. 3,
p. 387] Le Livre de la cration et de l histoire d Abou Zeid , edited and translated
by Marie Clment Imbault Huart. Paris, 18991919. Reprint, Baghdad, 1962.
al-Maqrz, Amad b. Al. Die Kmpfe und Streitigkeiten zwischen den Ban Umajja
unddenBanHim[K. al-Niz wa-l-takhumf-mbaynBan Umayyawa-Ban
Hshim,] Dr. phil. thesis by Geerhard Vos, Straburg, 1888. Leiden, Brill, 1888.
Margoliouth, David Samuel. Mohammed and the rise of Islam. New York; London,
1905.
On the origin and import of the names Muslim and Hanif in Journal of the
Royal Asiatic Society, 1903, pp. 467493.
Marracci, Lodovico. Prodromus ad refutationem Alcorani in quod Muhametanae
sectae falistas ostenditur Christianae religionis veritas comprobatur. Roma, 1691.
Refutatio Alcorani in qua ad Mahumetanicae superstitionis radicem secutis
apponitur. Patavii, 1698.
al-Masd, Al b. usayn. Les prairies d or, ed. and transl. Casimir Barbier de Mey-
nard and Abel J.M. Pavet de Courteille. Paris, 18611877.
Historical encyclopaedia, translated by Aloys Sprenger. London, 1864.
K. al-Tanbh wal-ischrf, edited by M.J. de Goeje. Leiden, 1894. (Bibliotheca
geographorumarabicorum, 8.) Reprint, 1967, and 1992.
Meyer, Eduard. Ursprung und Geschichte der Mormonen, mit Exkursen ber die
Anfnge des Islms und des Christentums. Halle a.d. Saale, 1912.
Mez, Adam. The Renaissance of Islam, translated from the German by S. Khuda
Bakhsh and D.S. Margoliouth, London, 1937, reprint, New York, 1975. Spanish
translation, El renacimiento del Islam (Madrid, 1936); and Russian translation,
(Moscow, 1966).
Mingana, Alphonse. An ancient Syriac translation of the Kuran exhibiting new
verses and variants, in Bulletin of the John Rylands Library, 9 (1925), 188235.
. The transmission of the urn in Journal of the Manchester Egyptian and
Oriental Society, 19151916, pp. 2547.
Mishkt al-Khab al-Tibrz, Muammad b. Abd Allh
Miskawayh, Amad b. Muammad. Tajrib al-umam maa nukhab min tawrkh
al-sha , edited by Henry Frederick Amedroz. Cairo, 19141916.
Mller, Johann Heinrich. Palologische Beitrge aus den herzoglichen Sammlungen
zu Gotha; Heft 1: Orientalische Palographie. Erfurt, 1843, and Eisleben, 1844.
Mordtmann, Johannes Heinrich & David Heinrich Mller. Eine monotheistische
sabische Inschrift in Wiener Zeitschrift fr die Kunde des Morgenlandes, 10
(1896), pp. 285292.
Sabische Denkmler in Denkschriften der K. Akademie der Wissenschaf-
ten, Wien, Phil.-historische Klasse, 30 (1883).
Moritz, Bernhard. Arabic palaeography; a collection of Arabic texts from the first
century of the Hidjra till the year 1000, edited by B. Moritz. Cairo & Leipzig, 1905.
(Publications of the Khedivial Library, no. 16.)
Arabia, section: Arabic writing, in: Encyclopaedia of Islam, vol. 1 (19081913),
pp. 381393.
al-Mubarrad, Muammad b. Yazd. The Kmil of el-Mubarrad, edited by William
Wright. Leipzig, 18641892. [Reprint, Cairo, 1308/18911892, and Hildesheim,
Olms, 1992.]
al-Mufaal b. Muammad al-abb. Die Mufaddaljt, nach den Handschriften zu
Berlin, London und Wien herausgegeben und mit Anmerkungen versehen von
Heinrich Thorbecke. Leipzig, 1885.
Muir, William. The Life of Mohamet and history of Islamto the era of the Hegira, with
introductory chapters onthe original sources for the biography of Mahomet, andon
the pre-Islamite history of Arabia. London, 18581861. 2nd ed., 1892; 3rd ed., 1894.
Mller, August. Der IslamimMorgen- und Abendland. Berlin, 18851887.
Mller, David Heinrich. imjarische Studien in Zeitschrift der Deutschen Morgen-
lndischen Gesellschaft, 30 (1876), pp. 671708.
Die Propheten in ihrer urspnglichen Form; die Grund-gesetze der ursemitis-
chenPoesie, erschlosseninBibel, KeilinschriftenundKoranundinihrenWirkungen
erkannt in den Chren der griechischen Tragdie. Wien, 1895.
al-Muqni al-Dn, Uthmn b. Sad, al-Muqni f marifat rasmmaif al-amr.
al-Murta al-Zabd, Muammad b. Muammad. Tj al-ars min jawhir al-
qms. Cairo, 13061307/18881890. [Nldeke-Schwally cite a 1395 edition]
Muslim b. al-ajjj al-Nsbr. [al-Jmi] al-a in the margin of al-Qasallns
commentary on al-Bukhr. Blq, 13031305.
Muslim b. al-Wald, d. 823 or 4. Diwan potae Abul-Walid Moslim ibnol-Walid al-
Anr cognomine arlo-l-ghawn edidit, glossario instruxit Michael Jan de
Goeje. Leiden, 1875.
al-Mutalammis al-uba, Jarr b. Abd al-Mas. Die Gedichte des Mutalammis,
Arabisch und Deutsch, bearbeitet von Karl Vollers, in: Beitrge zur Assyriologie
und semitischen Sprachwissenschaft, 5 (1903), pp. 149231.
al-Muttaq, Al b. Abd al-Malik. Kanz al-umml f sunan al-aqwl wa-l-afl. Hyder-
abad, 13121314/18951897.
al-Nahrawl, Qub al-Dn Muammad b. Amad. Geschichte der Stadt Mekka und
ihres Tempels (al-Ilm bi-alm bayt Allh al-arm), edited by Ferdinand Ws-
tenfeld. Leipzig, 1857. Reprint, 1964.
al-Nasaf, Ab l-Barakt Abd Allh b. Ahmad. Madrik al-tanzl wa-aqiq al-tawl;
tafsr. In the margin of Al al-Dn al-Khzin al-Baghdd. Misr, 12781306/1861
1888.
al-Nas, Ahmad b. Shuayb. al-Mujtab; a collection of traditions, selected by al-
Nas from his larger collection entitled al-Sunan al-kubr, with the commen-
tary of al-Suyt entitled Zahr al-rub. Cairo, 1312 (1894).
al-Nashr Ibn al-Jazar
al-Nawaw, Yay b. Sharaf. Tahdhb al-asm; The Biographical dictionary of illustri-
ous men, chiefly from the beginning of Islamism, now first edited from the colla-
tion of two Mss. at Gttingen and Leiden by Ferdinand Wstenfeld. Gttingen &
London, 18421847.
Neubauer, Adolf. La gographie du Talmud. Paris, 1868.
al-Nsbr, Nim al-Dn al-asan b. Muammad. Gharib al-Qurn wa-raghib
al-furqn, in the margin of al-Tabars Tafsr. Cairo, 1321/19031904.
Nissen, Heinrich. Orientation; Studien zur Geschichte der Religion. Berlin, 1906
1910.
Nldeke, Theodor. [Review] Annali dellIslam, compilati da Leone Caetani. Vols. 1
and 2 (1905 and 1907), in Wiener Zeitschrift fr die Kunde des Morgenlandes, 21
(1907), pp. 297312.
Aufstze zur persischen Geschichte. Leipzig, 1887.
Beitrge zur Geschichte des Alexanderromans. (Denkschriften der Kaiserlichen
Akademie der Wissenschaften, Wien. Philosophisch-historische Classe, 38 (1890),
Abhandlung 5.
Beitrge zur semitischen Sprachwissenschaft. Straburg, Trbner, 1904. Re-
print, Zrich, 1982.
Fnf Moallaqt, bersetzt und erklrt: (2) Die Moallaqat Antaras und
Labds in Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften zu
Wien, 142 (1900).
Geschichte der Perser und Araber zur Zeit der Sassaniden, aus der arabischen
Chronik des Tabari bersetzt und mit ausfhrlichen Erluterungen und Ergnzun-
gen versehen. Leiden, 1879. Reprint, 1973.
Geschichte des Qorns. Gttingen, 1860.
[Review] H. Grimme, Mohammed: 1. Theil, 1892 inLiterarisches Central-blatt
fr Deutschland, Nr. 26 (25 Juli 1892), col. 905908.
Hatte Muammad christliche Lehrer? in ZDMG, 12 (1858), pp. 699708.
Das klassische Arabisch und die arabischen Dialekte in his: Neue Beitrge
zur semitischen Sprachwissenschaft, pp. 114.
Koran in: Encyclopadia Britannica, 11th ed., vol. 15 (1911), pp. 898906.
[Review] L. Caetani, Annali dell Islm, vols. 1 and 2 in Wiener Zeitschrift fr
die Kunde des Morgenlandes, 21 (1907), pp. 297312.
Das LebenMuhammeds nach denQuellenpopulr dargestellt. Hannover, 1863.
Lehnwrter in und aus dem thiopischen in his: Neue Beitrge zur semitis-
chen Sprachwissenschaft (Straburg, 1910), pp. 3166.
Midian in Encyclopdia Biblica, vol. 3, cols. 30793082.
Neue Beitrge zur semitischen Sprachwissenschaft. Straburg, 1910. Reprint,
Zrich, 1982.
Orientalische Skizzen. Berlin, 1892. Reprint, Hildesheim, 1974.
[Review] Testi orientali inediti sopra i Sette Dormienti di Efeso in Gttingis-
che gelehrte Anzeigen, 1886, pp. 453459.
Die Tradition ber das Leben Muhammeds in Der Islam, 5 (1914), pp. 160
170.
[Review] Reste arabischen Heidenthums, by J. Wellhausen (Berlin, 1887), in
ZDMG, 41 (1887), pp. 707726.
Zur Grammatik des klassischen Arabisch. Wien, 1897. Reprint, 1963.
Zur tendenzisen Gestaltung der Urgeschichte des Islms in ZDMG, 52
(1898), pp. 1633.
Obbink, HermanTheodorus. De heilige oorlog volgens denKoran. Leiden, 1901. (Doc-
toral thesis, Utrecht.)
Ockley, Simon. The History of the Saracens. Edition not supplied; the work was first
published in 1708; there are several reprints until 1873.
Oehler, Franciscus [Franz,] ed. Corporis haereseologici. Berlin, 1859.
Orientalische Studien Theodor Nldeke zum siebzigsten Geburtstag gewidmet von
Freunden und Schlern und in ihrem Auftrag herausgegeben von Carl Bezold.
Gieen, 1906.
al-Plaw, mid b. Abd al-Fatt al-Blaw, mid b. Abd al-Fatt
Palaeographical Society, London. Facsimilies of manuscripts and inscriptions. Lon-
don, 18751883.
Palmer, Edward Henry. The Qurn, translated from the Arabic. Oxford, 1880.
Paris, Gaston. La Posie du moyen ge; leons et lectures. Paris, 1885.
Pauli, Johannes. Schimpf und Ernst; edited by Hermann sterley. Stuttgart, 1860.
Pautz, Otto. Muhammeds Lehre von der Offenbarung quellenmig untersucht. Doc-
toral thesis, Universitt Leipzig, 1898.
Pertsch, Wilhelm. Die arabischen Handschriften der Herzoglichen Bibliothek zu
Gotha. Gotha, 18781893.
Pickthall, Mohammed Marmaduke. The Meaning of the Glorious Koran. London,
1953.
Pisarev, Serge Nikolaevich. Coran coufique de
Samarcand crit d aprs la tradition de la propre main du troisime calife Osman.
St. Petersbourg, 1905.
Pretzl, Otto. Die Fortfhrung des Apparatus criticus zum Koran. Mnchen, 1924.
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Philos.-histori-
sche Abteilung, 1934, Heft 5).
Verzeichnis (der handschriftlich erhaltenen lteren Qirt-Werke.) in his:
Die Wissenschaft der Koranlesung, pp. 1447 (nos. 137) and pp. 230243 (nos.
3856.)
Die Wissenschaft der Koranlesung (ilmal-qirah) ihre literarischenQuellen
und ihre Aussprachegrundlagen (ul) in: Islamica, 6 (1933/34), pp. 147, 230
246, 290331; also in: Beitrge zur Erschlieung der arabischen Handschriften in
Istanbul und Anatolien, ed. by F. Sezgin (Frankfurt, 1986), pp. 133238.
Procksch, Otto. ber die Blutrache bei den vorislamischen Arabern und Mohammeds
Stellung zu ihr. Leipzig, 1899.
al-Qasalln, Amad b. Muammad. Irshd al-sr il shar a al-Bukhr. B-
lq, 13041305/18861888.
al-Mawhib al-laduniyya bi-l-mina al-Muammadiyya. Cairo, 1281/1864
1865.
Qays b. al-Kham. Der Dwn des ais b. al-H am, edited and translated by Tadeusz
Kowalski. Leipzig, 1914.
al-Qazwn, Zakariy b. Muammad. el-Cazwinis Kosmographie, edited by Ferdi-
nand Wstenfeld. Gttingen, 18471848.
Quatremre, tienne Marc. Mlanges d histoire et de philologie orientale. Paris
1854.
Sur le got des livres chez les Orientaux in Journal asiatique, 3e srie, t. 6
(juillet 1838), pp. 3578.
von Ranke, Leopold. Die arabische Weltherrschaft und das Reich Karls des Groen.
Leipzig, 1884. (His: Weltgeschichte, vol. 5, parts 1 and 2).
al-Rz, Fakhr al-Dn Muammad b. Umar. Maft al-ghayb al-mushtahar bi-l-
Tafsr al-kabr. Blq, 12791289.
Real-Encyklopdie fr protestantische Theologie und Kirche, begrndet von Johann
Jakob Herzog, 3. verbesserte und vermehrte Auflage herausgegeben von Albert
Hauck. Leipzig, 18961913.
Reckendorf, Hermann. Die syntaktischen Verhltnisse des Arabischen (Leiden, 1895
1898). 2 v. Reprint, 1967.
Recueil de mmoires orientaux, textes et traditions publis par les professeurs de
l Ecole spciale des langues orientales vivantes l occasion du XIVe Congrs des
orientalistes. Paris, 1905.
Renan, Joseph Ernest. tudes d histoire religieuse, 2nd ed. Paris, 1856.
Mahomet et les origines de l islamisme in Revue des deux mondes, 12 (1851),
10631101.
Roberts, Robert. Das Familien-, Sklaven- und Erbrecht im Qoran. Leipzig, Hinrinchs,
1908. 56 p. (Leipziger semitistische Studien, II, 6.) English translation: The social
laws of the Qorn, considered and compared with those of the Hebrew and other
ancient codes (London, 1971).
Rodwell, John Medows. The Koran, translated from the Arabic. London, 1862.
Rsch, Gustav. Die Namen des arabischen Propheten Muhammed und Ahmed in
ZDMG, 46 (1892), pp. 432440, 580.
Rckert, Friedrich. Der Koran, excerpts translated and edited by August Mller.
Frankfurt am Main, 1888. Reprint, Hildesheim, Olms, 1980.
Ruete, Emily. Memoiren einer arabischen Prinzessin. 4th ed., Berlin, 1886. French
translation (1905), and English (1993) entitled An Arabian princess between two
worlds; memoirs.
Rycaut, Paul. Histoire de l tat prsent de l empire Ottoman, augm. d une 2e partie;
traduite de l anglois par Briot. Amsterdam, 1672.
Sachau, Carl Eduard. Muhammedanisches Recht nach schafiitischer Lehre. Berlin,
1897.
Das Berliner Fragment des Ms Ibn Uba; ein Beitrag zur Kenntnis der
ltesten arabischen Geschichtsliteratur in Sitzungsberichte der Knigl. Preui-
schen Akademie der Wissenschaften, Berlin, 11 (1904), pp. 445470.
al-Sajwand, Sirj al-Dn Muammad b. Muammad. K. al-Waqf wa-l-ibtid. Ms.
Wien, Mxt. 717.
Sale, George. The Koran, commonly called the Alkoranof Mohammed; translated into
English from the original Arabic, with explanatory notes taken from the most
approved commentators, to which is prefixed a Preliminary Discourse. London,
1734. Numerous undatedreprints, andits Germantranslationby Theodor Arnold
entitled Der Koran aus dem arabischen Original ins Englische bersetzt von
G. Sale ins Teutsche verdollmetchet von Theodor Arnold. Lemgo, 1746.
al-Samn, Abd al-Karm b. Muammad. al-Ansb, n.d. Reprint, Hyderabad, India,
19621982.
Sarauw, Christian. Die altarabische Dialektspaltung, in: Zeitschrift fr Assyriologie,
21 (1908), pp. 3149.
Sarks, Ysuf Ilyn. Mujam al-mabt al-Arabiyya wa-l-muarraba = Dictionnaire
encyclopdique de bibliographie arabe. Cairo, 1928.
Schaade, Arthur. Sbawaihis Lautlehre. Leiden, Brill, 1911.
Schacht, Joseph. Aus zwei arabischenFurq-Bchern inIslamica, 2 (1927), pp. 505
537.
Schapiro, Israel. Die haggadischen Elemente imerzhlenden Teil des Korans. Leipzig,
1907. (Straburg, Phil. Diss., 1906.)
Schreiber, Theodor. Studie ber das Bildniss Alexanders des Groen; ein Beitrag
zur alexandrinischen Kulturgeschichte mit einemAnhang ber die Anfnge des
Alexanderkultes in Abhandlungen der Schsischen Gesellschaft der Wissen-
schaften zu Leipzig. Philosophisch-historische Classe, 21, no. 3 (1903).
Schreiner, Martin. Zur Geschichte der Polemik zwischen Juden und Mohamme-
danern in ZDMG, 42 (1888), pp. 591675.
Schulthe, Friedrich. Umajja b. Abi--alt. In Orientalische Studien Theodor Nlde-
ke zum70. Geburtstag von Freunden gewidmet (1906), vol. 1, pp. 7189.
Schwally, Ferdinand. Idioticon des christlich palstinensichen Aramisch. Gieen,
1893.
Das Leben nach dem Tode nach Vorstellungen des alten Israel und des Juden-
tums. Gieen, 1892.
Lexikalische Studien in ZDMG, 52 (1898), pp. 132148, 511512; vol. 53 (1899),
197201.
Semitische Kriegsaltertmer. 1. Heft: Der heilige Krieg im alten Israel. Leipzig,
1901.
Die Vision in his Semitische Kriegsaltertmer, pp. 1820.
Schwarz, Paul. Der Diwan des Umar Ibn Abi Rebia, nach den Handschriften zu
Kairo und Leiden; mit einer Sammlung anderweitig berlieferter Gedichte und
Fragmente herausgegeben. Leipzig, 19011909.
Der sprachgeschichtliche Wert einiger lterer Wortschreibungen im Koran
in Zeitschrift fr Assyriologie 30 (1915/16), pp. 4659.
Wann lebte der Verfasser der arib al-Kurn in Zeitschrift der Deutschen
Morgenlndischen Gesellschaft, 69 (1915), pp. 300304.
Sellheim, Rudolf. Materialien zur arabischen Literaturgeschichte, Teil 1. Wiesbaden,
1976. (Verzeichnis der orientalischen Handschriften in Deutschland; vol. 17, series
A.
Sezgin, Fuat. Geschichte des arabischen Schrifttums. Leiden and Frankfurt amMain,
19672000.
Goldziher and adith in I. Goldziher, Schools of Koranic commentators,
transl. W. Behn, pp. xiiixxxviii. Wiesbaden, 2006.
al-Shahrastn, Muammad b. Abd al-Karm. al-Milal wa-l-nial, ed. William Cure-
ton. London, 1846.
ReligionspartheienundPhilosophen-Schulen, vollstndig aus demArabischen
bersetzt und mit Anmerkungen versehen von Theodor Haarbrcker. Halle,
18501851. Reprint, Hildesheim, 1969.
al-Sharn, Abd al-Wahhb b. Amad. Mzn. Cairo, 13111317/189318991900.
al-Shib, Qsimb. Firruh. Aqlat atrb al-qaid f asn al-maqid f ilmal-rasm,
commonly called al-Riyya. Cairo, 1329/1911.
Shebunin, A.F. . . inZapis-
ki Vostochnago Otdelniia Imperatorskago Russkago Arkheologicheskago Obshch-
estva, vol. 6 (1891), pp. 69133.
al-Shrz, Ab Isq Ibrhm b. Al. Jus Shafiiticum: al-Tanbh f l-fiqh; edited by
Abraham Wilhelm Th. Juynboll. Leiden, Brill, 1879.
al-Shshw, al-usayn b. Al. al-Fawid al-jamla al l-yt al-jalla (unidentified
edition. Edition by Idrs Azzz, Rabat, 1989.)
Sibawayh, Amr b. Uthmn. al-Kitb. Blq, 13161318 [18981900/1901.]
al-Sijistn, Abd Allh b. Sulaymn. K. al-Maif, adth no. 407. Ms., Dr al-Kutub
al-hiriyya, Damascus.
Silvestre de Sacy, Antoine I. Commentaire sur le pome nomm Rayya intitul
Akila, par le Scheich Aboulhasan Ali ben Mohammed Schafei in Notices et
extraits des manuscrits de la Bibliothque impriale, 8 i (1810), pp. 333354.
Dumanuscrit Arabe n. 239de la Bibliothque impriale, contenant untrait
sur l orthographie primitive de l Alcoran, intitul
,
par Othman ben-Said [Brockelmann, Geschichte der arabischen Literatur, vol. 1,
p. 407] in Notices et extraits des Mss. de la Bibliothque nationale, 8 i (1810),
pp. 290306, 330332.
Ettaberi annales, [reviewof J.G.L. Kosegartens Latin translation] in: Journal
des savants (1832), pp. 532544.
Mmoire sur l origine et les anciens monuments de la littrature parmi les
Arabes in: Mmoires de la littrature de l Acadmie royale des Inscriptions,
tome 50 (1808), pp. 327342, 419434. [Not sighted].
Notice du manuscrit arabe n. 239 de la Bibliothque impriale, contenant
un trait sur l orthographie primitive de l Alcoran (al-Muqni f marifat kha
maif al-amr), par Othmanben-Said [GAL, i, 407]. Notices et extraits des Mss
de le Bibliothque impriale 8 i (1810), 290332.
Recueil de diffrents traits relatifs l orthographie et la lecture de l Alco-
ran in Notices et extraits des manuscrits de la Bibliothque nationale, t. 8 i (1810),
pp. 333354, 355359.
Trait de la pronunciation des lettres arabes, extrait du manuscrit arabe n
260 de la Bibliothque impriale. Notices et extraits de la Bibliothque nationale
9 i (1813), 175.
Sive Liber Julilaeorum qui idem a Graecis inscribtur versione
Graeca deperdita nunc nonnisi in Geez lingua conservatus nuper ex Abyssinia in
Europamallatus: Aethiopice ad duorumlibrorummanuscriptorumfidem, primum
ed. August Dillmann. Kiel, 1859.
de Slane, WilliamMacGuckin. Catalogue des manuscrits arabes. Paris, Biblio-thque
nationale, 18831895.
Snouck Hurgronje, Christiaan. De Islam in de Gids; algemeen cultureel maand-
blad, 50 (1886), tweede deel, pp. 239273, 454498; derde deel, pp. 90134; vijfde
deel, pp. 4127.
The Mecca Festival, translated by Wolfgang Behn. Wiesbaden, 2012.
Mekka: die Stadt und ihre Umgebung; aus dem heutigen Leben. Den Haag,
18881889.
Het Mekkaansche feest, Leiden, Brill, 1880. 191 p. English translation entitled
The Mecca Festival. Wiesbaden, 2012.
Mekkanische Sprichwrter und Redensarten gesammelt und erlutert. den
Haag, 1886.
Nieuwe bijdragentot de kennis vandenIslam, inBijdragentot de taal-, land-
en volkenkunde van Nederlandsch-Indi, 4th series, 6 (1882), pp. 357421.
[Review] Une nouvelle biographie de Mohammed (H. Grimme, Moham-
med) in Revue de l histoire des religions, 30 (1894), pp. 4870, 149178.
[Review] Sachau, Muhammedanisches Recht nach schafiitischer Lehre in
ZDMG, 53 (1909), pp. 125167.
Socin, Albert, andEugenPrym. Die Dwne des Zuheir undKab inZDMG, 31 (1911),
pp. 710715.
Sperber, Jakob. Die Schreiben Muammads an die Stmme Arabiens in Mitteilun-
gen des Seminars fr orientalische Sprachen, Berlin, 2nd Abteilung, 19 (1916),
pp. 193 (= Dr.phil. thesis, 1916).
Spitaler, Anton. Die Verszhlung des Koran nach islamischer berlieferung; mit ein-
em Vorwort von Otto Pretzl. Mnchen, 1935. (Sitzungsberichte der Bayerischen
Akademie der Wissenschaften, Mnchen, Phil.-hist. Klasse, 11)
Sprenger, Aloys. Das Leben und die Lehre des Mohammad, nach bisher grtenteils
unbenutzten Quellen bearbeitet. Berlin, 18611865. 2nd ed., Berlin, 18681869.
Reprint, 1986.
Life of Moammad fromoriginal sources. Allahabad, 1851.
Mohammads Zusammenkunft mit dem Einsiedler Bayr in ZDMG, 12
(1858), pp. 238249.
Mohammed und der Qorn; psychologische Studie. Hamburg, 1889.
Notes on Alfred von Kremers edition of Wakidys Campaigns in Journal of
the Asiatic Society of Bengal, 25 (1856), pp. 5378, 199220.
Notice of a copy of the fourth volume of the original text of Tabary, by
A. Sprenger; communicated by Sir Henry Elliot in Journal of the Asiatic Society
of Bengal, 19 (1850), pp. 108135.
Observations on the physiology of the Arabic language in Journal of the
Asiatic Society of Bengal, 20 (1851), pp. 115126.
ber das Traditionswesen bei den Arabern, in ZDMG, 10 (1856), pp. 1
17.
ber den Kalender der Araber vor Mohammad in ZDMG, 13 (1859), pp. 134
175.
Strack, Hermann Leberecht. Einleitung in den Talmud; 4. neubearbeitete Auflage.
Leipzig, Hinrichs, 1908.
al-Suhayl, Abd al-Ramn b. Abd Allh. Die Kommentare des Suhaili und des Ab
arr zu den Uud-Gedichten in der Sra des Ibn Him, nach den Hand-schriften
zu Berlin, Straburg, Paris und Leipzig, edited by Arthur Schaade. Leipzig,
1920.
al-Sukkar, Ab Sad al-asan b. al-usayn. The Poems of the Huzailis; edited in the
Arabic, from an original manuscript in the University of Leyden, and translated,
with annotations, by John Godfrey Lewis Kosegarten. London, Oriental Transla-
tion Fund, 1854. (Repint, 1985).
Lieder der Hudhailiten [Arabic and German,] edited and translated by Julius
Wellhausen. Berlin, 1884. (Skizzen und Vorarbeiten, Heft 1.) Reprint, 1985.
al-Suy, Jall al-Dn Abd al-Ramn. Asbb al-nuzl, in the margin of Jallayn f
tafsr al-Qurn al-am. Cairo, 1301/18831884.
usn al-muara f akhbr Mir wa-l-Qhira. Cairo, 1327/19091910.
al-Itqn f ulm al-Qurn; on the exegetic sciences of the Koran, edited
by Basheerood-Deen, and Noorool-Haqq, with an analysis by Aloys Sprenger.
Calcutta, 18521854.
al-Itqn f ulmal-Qurn. Cairo, 1318/19001901. [Not identified]
Muhir f ulmal-lugha wa-anwih. Blq, 1282/1865.
Tadrb al-rw f shar Taqrb al-Nawaw. Cairo, 1307/18891890. Edition by
Amad Umar Hshim, Beirut, 1985.
Tafsr al-Jallayn. Cairo, 1301/18831884.
Le Synaxaire arabe-jacobite, vol. 3, edited & translated by Ren Basset. Paris, 192?
al-abar, Muammad b. Jarr. Annales quos scripsit Abu Djafar Mohammed Ibn
Djarir At-Tabari; cum aliis edidit M.J. de Goeje. Leiden, 18791901.
Chronique de Abou-Djafar Tabari; traduite sur la version persane d Abou
Ali Mohammed Bel ami par Hermann Zotenberg. Paris, 18671874. Reprint,
1958.
Ikhtilf al-fuqah, ed. Friedrich Kern. Cairo, 1902.
Jmi al-bayn f tafsr al-Qurn. Blq, 13211330. Reprinted Beirut, 1986
1987.
al-Tahnaw, Muammad Al b. Al. : A Dictionary of the tech-
nical terms used in the sciences of the Musulmans. Calcutta, Bibliotheca Indica,
1862.
Takprzade, Ahmed b. Mustafa. Miftal-sadawa-mibal-siydaf mawt
al-ulm. Hyderabad, 13281356/19101937.
al-Thalib, Abd al-Malik b. Muammad. Fiqh al-lugha. Cairo, 1317/18991900.
Laif al-marif. [Edition not indicated.]
al-Tirmidh, Muammad b. s. Sunan al-musammt bi-l-Jmi al-a. Delhi, 1315/
18971898.
al-Shaml al-Nab. Calcutta, 1262, 1272; Fs, 1310; Bulq, 1280, 1290; Cairo, 1317,
etc.
Tisdall, Wm. St. Clair. Newlight on the text of the Qurn in Moslemworld, 5 (1915),
143150.
Torrey, Charles Cutler. The Commercial-theological terms in the Koran. Leiden, 1892.
(Universitt Straburg thesis).
Tufa Ibn al-Khab al-Dahsha
al-s, Muammad b. al-asan. Fihrist kutub al-Sha = Tusys list of Shyahbooks, ed.
Aloys Sprenger, et al. Calcutta, 18531855. Reprint, 1980.
Umayya b. Ab l-alt. Umajja b. Abi alt: die unter seinem Namen berlieferten
Gedichtfragmente, gesammelt und bersetzt von Friedrich Schulthess. Leipzig,
1911.
Verlegung des Alcoran Bruder Richardi, Pregiger Ordens, anno MCCC; verdeutscht
durch D. Mar[tin] Lu[ther]. Wittenberg, Lufft, 1542.
de Vog, Charles J. Melchior. Inscriptions smitiques. Paris, 1868.
Vollers, Karl. Volkssprache und Schriftsprache im alten Arabien. Straburg, 1906.
Reprints, 1972 and 1981.
al-Wid, Al b. Amad. Asbb al-nuzl. Cairo, 1315/18971898.
Wahl, Samuel Friedrich Gnther. Der Koran oder Das Gesetz der Moslemen durch
Mohammed den Sohn Abdallahs, auf Grund der vormaligen Verdeutschung
Friedrich E. Boysens von neuem aus dem Arabischen bersetzt, durchaus mit
erluternden Anmerkungen, mit einer historischen Einleitung und einem voll-
stndigen Register versehen. Halle, 1828.
al-Wqid, Muammad b. Umar. History of Muammads campaigns, edited by
Alfred von Kremer. Calcutta, 18541856. Reprint, 1981.
Muhammed in Medina, das ist Vakidis Kitab al-Maghazi in verkrzter deut-
scher Wiedergabe. Berlin, 1882.
Watt, W. Montgomery. Muammad at Mecca. Oxford, 1953.
Weil, Gustav. Geschichte der Chalifen nach handschriftlichen Quellen. Stuttgart &
Mannheim, 18461851. Reprint, 1967.
Historisch-kritische Einleitung in den Koran. Bielefeld, 1844, 2nd revised ed.,
Bielefeld, 1878.
Das Leben Mohammeds nach Mohammed b. Ishak, bearbeitet von Abd el
Malik Ibn-Hischam aus dem Arabischen bers. Stuttgart, 1864.
Mahomet savait-il lire et crire? in Atti del IVCongresso Internazionale degli
Orientalisti, 1878, v. 1 (1880), pp. 357366.
Mohammed der Prophet, sein Leben und seine Lehre, aus handschriftlichen
Quellen und demKoran geschpft u. dargestellt. Stuttgart, 1843.
Muhammed in Medina; das ist Vakidis Kitab al Maghazi in verkrzter deut-
Nldeke, Sprenger, Muir und Reinaud ber Mohammed und den Koran in
Heidelberger Jahrbcher der Literatur, vol. 55, no. 1 (1862), pp. 116.
. Sur un fait relatif Mahomet in: Journal asiatique, 3e srie, 14 (1842),
pp. 108112.
Wellhausen, Julius. Abrider Geschichte Israels und Judas. Berlin, 1884. (His: Skizzen
und Vorarbeiten, Heft 1.) Reprint, 1985.
Das arabische Reich und sein Sturz. Berlin, 1902. viii, 352 p.; 2nd ed., with
introduction by R. Hartmann, 1960.
Die Ehe bei den Arabern in Nachrichten von der Kgl. Gesellschaft der
Wissenschaften zu Gttingen, 1893. pp. 431481.
Medina vor dem Islam; Muammads Gemeindeordnung von Medina, seine
Schreiben, und die Gesandtschaften an ihn. Berlin, 1889. 64 p. (His: Skizzen und
Vorarbeiten, Heft 4.) Reprint, 1985.
Muammads Gemeindeordnung von Medina. Berlin, 1889. (His: Skizzen und
Vorarbeiten, Heft 4), English translation in Arent J. Wensinck, Muammad and
the Jews of Medina (1975.) Excursus Muammads constitution of Medina, pp.
128138.
Muhammed in Medina; das ist Vakidis Kitab al Maghazi in verkrzter deut-
Prolegomena zur ltesten Geschichte des Islams (Berlin, 1899). viii, 160 p. (His:
Skizzen und Vorarbeiten, vol. 6.)
Prolegomena zur Geschichte Israels, 2d ed. Berlin, 1883.
Reste arabischen Heidenthums (1887); 2nd ed., Berlin, 1897. Reprint, 1961.
Seine [Muammads] Schreiben, und die Gesandtschaften an ihn, in His: Skiz-
zen und Vorarbeiten, Heft 4. Berlin, 1889. Reprint, 1985.
Skizzen und Vorarbeiten. Heft 16. Berlin, 18841889.
Wensinck, Arent Jan. Mitteilungen: (Studium der arabischen Traditionssammlun-
gen.) ZDMG, 70 (1916), p. 570; 72 (1918), p. 347.
Mohammed en de Joden te Medina. Leiden, Brill. 1908.
Muammad and the Jews of Medina, translated fromthe Dutch and edited by
W.H. Behn. Freiburg i. Br., 1975; 2nd ed., Berlin, 1982.
Werry, Elwood Morris. Comprehensive commentary on the Qurn, comprising Sales
translation and preliminary discourse. London, 18821886.
Wiener, Anton. Die Fara bad a-idda-Literatur von Madin bis Tan: ein
Beitrag zur arabischen Literaturgeschichte in Der Islam, 4 (1913), pp. 270298,
387420.
Winternitz, Maurice (Moritz.) A History of Indian literature, translated from the
German edition (19081922), by V. Srinivasa Sarma, et al. Delhi, 19811985.
Wright, William. Facsimilies of manuscripts and inscriptions: 100 autotype plates,
with letterpress, edited by William Wright. London, The Palaeological Society
Oriental Series, 18751883.
A Grammar of the Arabic language; translated from the German of Cas-
pari and edited with numerous additions and corrections: third edition revised
by W. Robertson Smith and M.J. de Goeje. Cambridge, 18961898. Numerous
reprints.
Opuscula Arabica, collected and edited fromMss. in the University Library of
Leyden. Leyden, 1859.
WZKM Wiener Zeitschrift fr die Kunde des Morgenlandes.
al-Yaqb, Amad b. Al Yaqb. Ibn Wadhih qui dicitur al-Jaqubi Historiae, ed.
Martijn Th. Houtsma. Leiden, 1883.
El Kitb al-boldn, ed. M.J. de Goeje, 2nd ed. Leiden, 1892 in: Ibn Rustah,
Amad b. Umar, Kitb al-alk. 1892
Yqt b. Abd Allh al-amaw. The Irshad al-arb il marifat al-adb; or, Dictionary
of the learned men of Yqt, edited by D.S. Margoliouth. Leiden, Brill, 19071926.
7 v. Reprint, Cairo/London, 19231930.
Jacuts geographisches Wrterbuch; , edited by Ferdinand Wsten-
feld. Leipzig, 18661873.
al-Zamakhshar, Mamd b. Umar. Kashshf anaqiq al-tanzl. Cairo, 1308/1885
1886. 2 v. Calcutta edition, 1856.
al-Mufaal; edited by Jens P. Broch. Christianiae, 18571879.
ZDMG Zeitschriftt der Deuschen Morgenlndischen Gesellschaft.
ZS Zeitschrift fr Semitistik und verwandte Gebiete
Zuhayr Ahlwardt, Wilhelm. The Divans of the six ancient poets.
Select List of Manuscripts Arranged by the
Numbers of the Respective Catalogues
BERLIN, Ahlwardt, Wilhelm. Verzeichnis der arabischen Handschriften der Knig-
lichen Bibliothek zu Berlin. Berlin, 18871899. 10 v. Reprint, 19791981.
al-Farr, Ab Bakr Yay. Man l-Qurn. Ms. Nuruosmaiye Ktphanesi, no. 459
and Bagdatl Vehbi Efendi, no. 60 [or 66.] Photocopy, Berlin, Staats-bibliothek,
Mss. simulata orientalia, no. 37.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis der arabischen Handschriften, no.
301 (= cod. 1st Petermann, no. 36.) Sras 1, 15, and 16. 119 leaves.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 303 (= cod. 1st Petermann,
no. 325.) Sras 298. 173 leaves.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 305 (= cod. 2nd Wetzstein,
no. 1913.) Sras. 2:2821:109, 22:1824:33, 24:5027:86, 34:4674:1, 78:3590:18. 210
leaves.
Koran fragment (Kfic.) W. Ahlward, Verzeichnis, no. 306 (= cod. 2nd Wetzstein,
no. 1952.) Sra 2:69109. 14 leaves.
no. 37.) Sras. 2:28210:78, 11:1634:10, 35 beginning38:28, 39:1644:2. 164 leaves.
no. 1920.) Sra 3:93:117. 6 leaves.
no. 1953.) Sra 3:3045. 2 leaves.
no. 1924.) Sra 3:403:54. 1 leaf.
no. 1915.) Sras 4:756:20. 83 leaves.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 314 (= cod. Mss. Or. quarto,
no. 372.) Sras 4:8926:211. 167 leaves.
no. 1915.) Sras 4:756:20. 8 leaves.
no. 1916.) Sras 4:1585:84; 10:4210:49. 50 leaves.
Koran fragment (Kfic.) W. Ahlwardt, no. 317 (= cod. 2nd Wetzstein, no. 1922.) Sras
5:885:93; 6:166:25; 7:517:57; 7:757:85. 4 leaves.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 323 (= cod. Mss. Or. folio,
no. 379.) Sras 8:498:62; 9:59:9. 3 leaves.
Koran fragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 325 (= cod. Mss. Or. quarto,
no. 689.) Sra 10:510:13. 2 leaves.
Koranfragment (Kfic.) W. Ahlwardt, Verzeichnis, no. 327 (=cod. Landberg, no. 834.)
Sras 10:5511:69, 15:5416:114, 17:1317:28, 17:5817:62, 17:6917:109, 18:418:13. 50
leaves.
no. 1923.) Sras 11:1322, and 11:12012:9. 2 folios.
no. 1919.) Sras 12:10115:67, and 16:317:1. 18 leaves.
no. 1948.) Sras 15:8116:14; 17:417:25. 2 leaves.
no. 1960.) Sra 16:3416:60. 1 leaf.
no. 1950.) Sras 16:11917:4. 1 leave.
no. 1929.) Sras 18:3320:14. 8 leaves.
no. 38.) Sras 19 to 114, with lacunae. 179 leaves.
no. 1936.) Sras 20:13021:3; 21:4821:61.
no. 1955.) Sras 25:2225:31, 25:4725:56, 27:1027:16, 27:3227:44, 27:6127:67. 6
leaves.
no. 1935.) Sras 26:18826:210; 27:2827:39. 2 leaves.
no. 1927). Sras 28:7329:39. 10 leaves.
no. 379.) Sras 31:333:37. 24 leaves.
no. 1928.) Sras 32:3033:38. 10 leaves.
no. 1946.) Sra 33:432. 6 leaves.
no. 1942.) Sra 39:2439:71. 7 leaves.
no. 379.) Sras 40:7941:11, 42:4643:12. 2 leaves.
no. 1930). Sras 41:2041:37; 42:5043:37. 3 leaves.
no. 1921.) Sras 41:2441:33; 48:449:13; 70:1971:6. 10 leaves.
no. 1949.) Sra 46:146:21. 1 leaf.
no. 1934.) Sras 48:1848:29; 53:3854:25. 2 leaves.
no. 1938.) Sra 48:2648:29. 2 leaves.
no. 1925.) Sras 67:19; 67:10- heading of surah 102; 104:8114 end. 59 leaves.
al-Dn, Ab Amr Uthmn b. Sad. al-Muqni f marifat rasm maif al-amr.
[Wilh. Ahlwardt, Verzeichnis, no. 419 (= cod. Sprenger, no. 376; a Cairo edition
by Muammad al-Sdiq Qamhw was published in 1978).]
IbnAlwnal-Muqr, Amadb. Raba. Introductionto the explanationof the Koran,
without title, in: W. Ahlwardt, Verzeichnis, 420 (= cod. Sprenger, no. 400.)
Anintroductionto the orthography of the Koran, without title; the author died after
816/1413, in: W. Ahlwardt, Verzeichnis der arabischen Handschriften, no. 429 (=
cod. 1st Wetzstein, no. 154, part iv, fols. 17
v
41.)
al-Suy, Jall al-Dn. Itf al-wafd bi-naba sratay al-khal wa-l-afd, in Wilh.
Ahlwardt, Verzeichnis, no. 438 (= cod Landberg, no. 343.) 4 leaves.
Ab Ubayd al-Qsim b. Sallm. Fail al-Qurn. Wilh. Ahlwardt, Verzeichnis, no.
451 (= cod. Petermann, no. 449.) 58 fols.
Ibn Salma, Ab l-Qsim Hibat Allh al-Baghdd. al-Nsikh wa-l-manskh (f
l-Qurn.) Wilhelm Ahlwardt, Verzeichnis, no. 476 (= cod. Sprenger, no. 397.) 29
leaves.
al-Nsikh wa-l-manskh min al-Qurn, anonymous, without title, fols. 6291. W.
Ahlwardt, Verzeichnis, no. 476 (= cod. Sprenger, no. 397, i)
Ibnhir al-Baghdd, Abdal-Qhir. K. al-Nsikhwa-l-manskh. W. Ahlwardt, Verze-
ichnis, no. 478 (= cod. 1 Petermann, no. 555.)
al-Mi f l-tajwd, a work on correct Koranic pronunciation, by an author living
before 785/1383. W. Ahlwardt, Verzeichnis, no. 499 (= cod. Sprenger, no. 391.) 75
leaves.
Makk b. Ab lib al-Qays, Ab Muammad. al-Kashf an wujh al-qirt al-sab
wa-ilalih wa-ujajih. Wilh. Ahlwardt, Verzeichnis, no. 578 (= cod. 2nd Peter-
mann, no. 17.) 524 leaves.
Isml b. Khalaf al-Muqri, Ab hir. Unwn f l-Qurn al-sam. Wilh. Ahlwardt,
Verzeichnis, no. 591 (= cod. Sprenger, no. 382.) 53 leaves.
Ibn al-afrw, Abd al-Ramn. al-Taqrb wa-l-bayn f marifat shawdhdh al-
Qurn. Wilh. Ahlwardt, Verzeichnis, no. 613 (= cod. 2nd Wetzstein, no. 1290.) 66
leaves.
al-Nawaw, Yay b. Sharaf. al-Tibyn f db amalat al-Qurn. Wilh. Ahlwardt,
Verzeichnis, no. 617 (= cod. Sprenger, no. 403 [by an editing hand])
Ibn al-Jazar, Muammad b. Muammad. al-Nashr f l-qirt al-ashr. W. Ahlwardt,
Verzeichnis, no. 657 (= cod. 1st Petermann, no. 159.) 366 leaves. Edited by M.A.
Duhmn, Beirut, 1345/19261927.
al-Thalab, Amad b. Muammad. al-Kashf wa-l-bayn an tafsr al-Qurn. Wilh.
Ahlwardt, Verzeichnis, 737 (= cod. Sprenger, no. 409.) 240 leaves.
Ibn Aiyya, Abd al-aqq b. Ab Bakr b. Abd al-Malik. Al-Jmi al-muarrar al-a
al-wajz f tafsr al-Qurn al-azz. Wilhelm Ahlwardt, Verzeichnis 800 (= cod.
Sprenger, no. 408.) 203 fols.
al-Qurub, Muammad b. Amad. Tafsr al-Qurn. W, Ahlwardt, Verzeichnis, no.
810 (= cod. Sprenger, no. 436.) 230 leaves. Printed, Cairo, 1354/1935.
Muammad b. Murta. K. al-f (Tafsr al-Qurn.) W. Ahlwardt, Verzeichnis, no.
899 (= cod. 1 Petermann, no. 553.)
K. al-Mabn li-nam al-man, in Wilhelm Ahlwardt, Verzeichnis der arabischen
Handschriften, no. 910 (= cod. 1 Wetzstein, no. 103.)
al-Qumm, Al b. Ibrhm. Tafsr Al b. Ibrhm. W. Ahlwardt, Verzeichnis, no. 929 (=
cod. Sprenger, no. 406.) 343 leaves.
Muammad b. Muammad b. Yaqb al-Bukhr, et al. [Collectin of various hands]
in: R. Sellheim, Materialien zur arabischen Literaturgeschichte, Teil 1. Wiesbaden,
1976, p. 12. (Verzeichnis der orientalischen Handschriften in Deutschland; vol. 17,
series A. Or. quarto 1386)
Muammad b. Mamd al-Aqar. Shar arban adthan. W. Ahlwardt, Verzeich-
nis, no. 1512 (= cod. Sprenger, no. 397, ii), fol. 18b27b.
A number of sayings from the Prophet, without name or heading. W. Ahlwardt,
Verzeichnis, no. 1645 (2) (= cod. Sprenger, no. 397, iii), fol. 27b29b.
Anonymous work of a Naqshband, pupil of Muammad al-Mam al-Frq b.
Amad, fl. 1100/1688, on true religion and heresy. W. Ahlwardt, Verzeichnis, Mss.
Or. octavo, no. 2192 (= cod. 2nd Wetzstein, no. 1792, part 2, fol. 4b6b).
Iyb. Ms al-Sabt al-Yaub. al-Shif bi-tarf uqqal-Muaf. Wilh. Ahlwardt,
Verzeichnis, 25592563 (= cod. 1st Petermann, no. 60, cod. Sprenger, nos. 116 and
117, cod. Mss. Or, quarto, no. 45, cod. 2nd Wetzstein, no. 1771, respectively).
al-Dhahab, abaqt. Berlin, Staatsbibliothek, Ms. Or. fol., no. 3140. Not yet cata-
logued in 2008.
K. al-Mabn li-nam al-man, in Wilhelm Ahlwardt, Verzeichnis der arabischen
Handschriften, no. 8790 (= cod. I Wetzstein, no. 94, part iv.)
Ibn Sayyid al-Ns, Muammad b. Ab Bakr. Uynal-athar f l-maghz wa-l-shamil
wa-l-siyar. Wilh. Ahlwardt, Verzeichnis, nos. 9577 and 9578 (= Sprenger, nos. 122
and 123.)
Muammad b. Sad b. Man. al-abaqt al-kabr. W. Ahlwardt, Verzeichnis, no. 9648
(= cod. Sprenger, no. 103.) 302 fols.
Ibn ubaysh, Abd al-Ramn b. Muammad, al-Ghazawt al-mina al-kfila in
W. Ahlwardt, Verzeichnis, no. 9689 (= cod. Wetzstein, no. 173.) [Edited by Suhayl
Zakkr, Beirut, 1992.]
al-Mizz, Ysuf b. Abd al-Ramn. Tahdhb al-kaml f asm al-rijl, part. 4. Wilh.
Ahlwardt, Verzeichnis, 9930 (= cod. Landberg, no. 40.) 340 leaves.
Muammad b. Amad b. Uthmn b. [sic] al-Dhahab. Marifat al-qurr al-kibr
al l-abaqt wa-l-ar. Ahlwardt, Verzeichnis, no. 9943 (= cod. Mss. Or. quarto,
no. 462.) 128 fols.
ab asm al-aba. W. Ahlwardt, Verzeichnis, no. 10169 (= cod. Sprenger, no. 282.)
72 leaves.
GOTHA, Die arabischen Handschriften der Herzoglichen Bibliothek zu Gotha, ver-
zeichnet von Wilhelm Pertsch. Gotha, Perthes, 18781893. 6 v.
Koran fragment (Kfic.) W. Pertsch, Die arabischen Handschriften, no. 427 (= cod.
Mller, 1b, Nr. 1.) Sras 1:16, and 2:68. 2 folios.
Mller, 1b, Nr. 7.) Sras 3:674:112. 20 leaves.
Mller, no. 1b, Nr. 9.) Sra 5:2022. 4 leaves.
Mller, no. 1b, Nr. 36.) Sra 7:6768. 1 leaf.
Mller 1b, Nr. 16); Stz. Kah. 1399. Sras 12:3012:31. 2 leaves.
Mller 1b, Nr. 13.) Sras 10:9311:12, 12:4412:70. 2 leaves.
Mller, no. 1b, Nr.17.) Sra 12:8712:93. 2 leaves.
Mller, no. 1b, Nr. 14.) Sras 13:3814:5. 7 leaves.
Mller, no. 1b, Nr. 15.) Sras 14:614:11. 7 leaves.
Mller, no. 1b, Nr. 22.) Sra 20:7282 and 20:8697. 8 leaves.
Mller, no. 1b, Nr. 33.) Sra 34:2541. 6 leaves.
Mller, no. 1b, Nr. 27.) Sras 36:5037:2. 1 leaf.
Mller, no. 1b, Nr. 31; Stz. Kah. 1399.) Sras 37:7637:169, and 38:3038:86. 4 leaves.
Mller, no. 1.) Sras 43:5146:14. 12 leaves.
Ibn al-Jazar, Muammad b. Muammad. al-Nashr f l-qirt al-ashr, W. Pertsch,
Die arabische Handschriften, no. 549 (= cod. Mller, no. 65); W. Ahlwardt, Ver-
zeichnis, 657 (= codex 1 Petermann, no. 159); W. Ahlwardt, Verzeichnis, 591 (= cod.
Sprenger, 382); ed. M.A. Duhmn. Beirut, 1345/19261927.
Amad b. Zanbal al-Maall. al-Riy al-zhira f waqi Mir wa-l-Qhira, in: W.
Pertsch, Die arabischen Handschriften der Herzoglichen Bibliothek zu Gotha, no.
1671 (= cod. Mller, no. 386; Stz. Kah. 47.) 101 leaves.
GTTINGEN, Niederschsische Staats- und Universitts-Bibliothek. Verzeichnis der
Handschriften imPreussischen Staate (3) Gttingen
Birgevi (Birgili/Birkaw), Muammad b. Pr Al. , also
known as Risale-i Birgili, and Vesiyyetname-i Birgili. Handschriften in Gttingen,
cod. Asch, 88, 90, 87, 97.
ISTANBUL
Ibn Jinn, Uthmn. Mutasab. Istanbul, Ragp Paa Ktphanesi, no. 13.
al-Jabar, Ab Isq Ibrhm b. Uthmn [or Umar.] Kanz al-man. Ms., Fatih
Camii Ktphanesi, Istanbul, no. 52.
LEIDEN, Catalogus codicum orientalium Bibliothecae Academiae Lugduno Batavae,
edited by Reinhart p. A. Dozy, Pieter de Jong, Michal J. de Goeje, M. Th. Houtsma.
Leiden, 18511877. 6 v.
al-Nuwayr, Amad b. Abd al-Wahhb. Nihyat al-arab f funn al-adab. Codd. 2, 19,
Golius and 273, Warner, no. V (5), vol. 1 (1851), pp. 45.
Ab l-Ghanim Muslim b. Mamd al-Shrz. Jamharat al-Islm dht al-nathr
wa-l-nim. Cod. 287 Warner, no. CCCCXI (411), vol. 1 (1851), pp. 274281.
al-Dhahab, Muammad b. Ahmad. al-Mushtabah f l-rijl asmuhum wa-ansbu-
hum. Cod. 325 Warner, no. DCCCLVII (857), vol. 2 (1851) pp. 191193.
IbnAbd al-Kf, Ab l-QsimUmar b. Muammad, titulus operis me latet. Cod. 674
Warner, no. MDCXXXIV (1634), vol. 4 (1866), pp. 56.
al-Dn, AbAmr Uthmnb. Sad. al-Iqtidf rasmal-muaf. Cod. 527 (5) Warner,
no. MDCXXXV (1635), vol. 4 (1866), p. 6.
Opus de lectione Korani secumdum Acim [im]. Cod. 653 (1) Warner, no.
MDCXXXIX (1639), vol. 4, p. 7.
Abd al-Aad b. Muammad b. Abd al-Aad al-anbal al-arrn.
. Cod. 653 Warner, no. MDCXXXIX (1639) vol. 4 (1864), pp. 78, [Brockel-
mann, Geschichte, suppl. vol. 2, p. 212 (sic.)]
Abd al-Aad b. Muammad al-arrn. Nuzhat al-lim f qirat im. Cod. 655,
Warner, pt. 2, no. MDCXL (1640), vol. 4 (1866), p. 9
amd Allh b. Khayr al-Dn, Umdat al-irfn f waf urf al-Qurn. Cod. 46 Golius,
no. MDCXLVI (1646), vol. 4 (1866), pp. 1213.
Hibat Allh Ibn Salma. Nsikh al-Qurn wa-manskhuh. Cod. 411 (1) Warner no.
MDCLV (1655), vol. 4 (1866), pp. 1819.
al-Bukhr, Muammad b. Isml. al-Jmi al-a, alterum exemplar. Cod. 356 a
et b Warner. Catalogus codicum orientalium Bibliothecae Academiae Lugduno
Batavae, no. MDCCVIII (1728), vol. 4 (1864), 296 fol.
al-Naysbr, Najmal-Dnb. Ab Bakr. K. al-Furqf l-fiqh. Cod. 481 Warner, no. MDC-
CCLXI (1861), vol. 4 (1864), p. 155.
Ibn Sayyid al-Ns, Muammad b. Ab Bakr. Uyn al-athar f funn al-maghz wa-
l-shamil wa-l-siyar. Leiden, Cod. 340 Warner (Catalogus codicum orientalium,
MMDCVIII (2608), vol. 5, p. 196).
PARIS, Catalogue des manuscrits arabes de la Bibliothque Nationale, edited by Wm.
MacGuckin de Slane. Paris, 18831895.
Fragments de cinqexemplaires duCoranencriture duidjz, rapprochant duneskhi,
criture du idjz. 97 fol., no. 328, pp. 8990.
Fragments de douze exemplaires du Coran: 4 (Fol. 5062.) Ecriture de Bara (?);
lettres allonges horizontalement., no. 334, p. 94.
1 Le Mokhtaar d Al-Qodor. Copie date de l an 892 de l hgire (1487 de J.C.)
2 (Fol. 167
v
.) Chapitre sur les conditions remplir pour la validit
de la prire. Copie date de 980/1572, no. 828, pp. 177178.
WIEN, Die arabischen, persischenundtrkischenHandschriftender k.k. Hofbibliothek
zu Wien im Auftrag geordnet und beschrieben von Gustav Flgel. Wien, 18651867. 3
v. Reprint 1990.
Takprzade, Amad b. Muaf. Mift al-sada wa-mib al-siyda. Flgel, no.
16 (vol. 1, p. 25).
al-Sjawand. al-Waqf wa-l-ibtid f l-Qurn. Mxt 717 = Flgel, no. 1625
al-Anr, Zayn al-Dn Abu Yay Zakary b. Muammad. al-Daqiq al-mukama
f shar al-muqaddima al-Jazariyya. no. 1636 vol. 1, pp. 7172.
INDEX
Aaron (Prophet), 27, 296297, 301
Abn b. Sad b. al-, 254
Abn b. Uthmn b. Affn, 320
Abbs b. Abd al-Mualib, 104, 150
Abbsid
caliphs, 334, 358
court, 320, 327
Abd al-Aad b. Muammad al-anbal
l-Harrn, 406
Abd Allh b. Abd al-Mutalib (father of
Muammad), 302
Abd Allh b. Amr b. al-, 253
Abd Allh b. af al-Ghnim, 229230
Abd Allh b. mid al-Ibahn, Ab
Muammad, 354
Abd Allh b. Idrs, 524
Abd Allh b. Jash, 148
Abd Allh b. al-Mubrak, 326
Abd Allh b. Qays al-Kind l-Sakn, 393, 517
Abd Allh b. Rawa al-Anr l-Khazraj,
104, 156
Abd Allh b. Sab, 280
Abd Allh b. Sad b. Ab Sar, 37, 131
Abd Allh b. Salm b. al-rith, 52, 133
Abd Allh b. Ubayy b. Sall, 137, 169, 181, 186,
343
Abd Allh b. Umar b. al-Khab, 166, 252,
346, 595
Abd Allh b. al-Zubayr, 251, 253, 257, 286
Abd Allh b. Zurayr al-Ghfiq, 243
Abd al-Malik b. Marwn (caliph), 460, 595
Abd al-Ramn b. Ab Bakr, 281282
Abd al-Ramn b. Awf, 196
Abd al-Ramn b. al-rith, 251, 254, 257
Abd al-Ramn b. Thbit, 160
Abd al-Razzq b. Hammm, 350351
Abhala b. Kab al-Aswad, 44
Abraham (Prophet), 120, 144, 282, 381, 451
and Kaba, 14
religion of (millat Ibrhm), 119, 155, 165
abrogation
of passages by Muammad, 37, 42, 144,
153
of verses, 42, 121, 161, 196, 197, 200, 210,
248, 465
Ab l-Abbs al-Marrkush, 406
Ab Al l-Fris (d. 377/987), 538, 549, 550,
552, 565, 582
Ab Al l-asan b. Al b. Ibrhm al-Hudhal
l-Mir, 566
Ab Al l-asan al-Mlik, 564
Ab l-liyah al-Riy, 455
Ab mir (anf), 182
Ab Amr b. al-Al, 469, 470, 481, 486, 494,
500, 516, 520, 521, 522, 524, 525, 528, 530,
532, 535, 537, 538, 539, 546, 552, 556, 560,
599
Ab l-Aswad al-Dual, lim b. Amr, 594,
595
Ab Ayyb (collector of revelation), 215
Ab Bakr b. Ashtah al-Ibahn, Muam-
mad b. Abd Allh, 256
Ab Bakr al-Ibahn, Muammad b. Abd
al-Ram b. Shabb, 519, 563
Ab Bakr al-iddq, 3, 10, 130, 219, 226, 231,
281, 290, 359
accusations against, 279, 284, 288, 291
collection of Koran, 213, 221, 224,
227229, 261, 286, 312, 327, 331,
473
Ab l-Dard, Uwaymir al-Khazraj, 214, 395,
496
Ab Dd al-Sijistn, 336, 383
Ab l-Fal al-Nab, Jafar b. Muammad,
524
Ab l-Fal al-Rz, 565, 569
Ab l-Faraj al-Ifahn, 367
Ab l-Fid, 373
Ab af Umar b. Al b. Manr, 574
Ab amza Thbit b. Dinr Ab afya, 358
Ab anfa, 571
Ab l-asan al-Ank, Al b. Muammad b.
Isml, 552
Ab l-asan al-Qayj, Al b. Umar, 553,
560
Ab l-asan al-Rz, Al b. Jafar, 561562
Ab l-asan hir b. Ghalbn, 555556, 561
Ab tim al-Sijistn, Sahl b. Muammad,
418, 482, 483, 484, 492, 525, 526, 547, 548,
549, 571, 572
Ab aywa Shuray b. Yazd al-aram,
517
646 index
Ab ayyn (al-Jayyn) al-Andalus, 357,
559, 561, 566, 578
Ab udhayfa, 220221, 230
Ab Hurayra (al-Daws), 280, 317, 318, 335,
346, 369, 371
Ab Isaq al-Qawws l-Marand, Ibrhm b.
Muammad, 566
Ab Jafar al-Rz, sa b. Abd Allh, 455
Ab l-Jrd Ziyd b. al-Mundhir, 358
Ab Khuzayma b. Aws b. Zayd, 225
Ab Lahab (Abd al-Uzza) b Abd al-
Mualib, 7475, 104, 107, 284285
Ab Layth al-Samarqand, Nar b. Muam-
mad, 24, 354
Ab Lubba b. Abd al-Mundhir, 152, 207
Ab Mashar Abd al-Karm al-abar, 549,
562
Ab Mashar Naj b. Abd al-Ramn
al-Sind, 350
Ab Muammad al-Yazd, 486, 522, 525, 531,
544, 564
Ab Ms l-Mawil, Jafar b. Makk, 567
Ab Nash, Ab Jafar Muammad b.
Hrn, 519
Ab Nuaym al-Ifahn, Amad b. Abd
Allh, 339
Ab Qsim al-Amr, Ubayd Allh b.
Ibrhm, 551552
Ab l-Qsim al-Hudhal, Ysuf b. Al, 505,
549
Ab Qays irmah b. Ab Anas, 57
Ab Salma, Nubay b. Shar, 71
Ab li Bdhm, 49, 349, 351
Ab li Manr b. N, 330
Ab Shma, Abd al-Ramn b. Isml, 40,
363, 472, 503, 558, 567
Ab l-Sud, Muammad al-Imd, 580
Ab lib b. Abd al-Mualib, 5859, 104,
182, 302
Ab Tammm abb b. Aws, 367
Ab l-ayyib b. Ghalbn, 516, 555
Ab Ubayda al-Naw, 68
Ab Ubayd al-Qsim b. Sallm, 392,
393396, 406, 411, 430, 446, 455, 476,
482485, 492, 507, 526, 547548, 575
Ab Umma udayy al-Bbil, 344
Ab Wqid al-Layth, 193
Ab Yaqb al-Azraq, Ysuf b. Amr (Ibn
Yasr), 519520, 563
Ab l-Yaqn, 326
Ab Zayd al-Anr (Sad b. Aws b. Thbit).
See Zayd b. Thbit
Abyssinia, 5, 6, 13, 185n294, 221, 378, 379
Christian Najshi (Negs) of, 106
flight to, 58, 83, 84, 119
d, people of, 15, 296
Adam, 129, 190, 282
children/descendants of, 129, 302
dam b. Ab Iys, 351
Adhrut, 122
Ad b. tim al-, 184
Ad b. Zayd b. al-Riq, 217
adultery (zin), 200, 248
ahl al-bayt, 358
ahl al-aramayn (people of Mecca and
Medina), 483
ahl al-kitb, 127. See also People of the Book
Ahlwardt, Wilh., 412
aruf, sing. arf, 40, 259, 306, 464, 465, 471,
472, 527, 600
al-Ahwz, Ab Al l-asan b. Al, 505, 528,
548, 561, 564, 566, 567, 568, 569
azb, 594
isha, 10, 18, 19, 66, 70, 81, 171, 176, 200, 202,
231, 256, 257, 292, 318, 335, 342, 346
on Koran, 250, 390, 392, 497n183
ajz, 593594
alcohol. See intoxicants
Al b. Ab lib, 9, 49, 50, 173, 179, 184, 215,
219220, 243, 249, 252, 257, 288, 292,
296297, 299, 301, 330, 343, 346, 384, 391,
522523, 534, 567, 601
claim to caliphate, 289, 291
and Koran, 250, 327, 359
waiyy, 301
Al b. al-Madn, 361
Allh, 13, 67, 68, 69, 71, 78, 83, 84, 90, 91n163,
93n166, 99, 104, 120, 132, 136, 148, 181, 182,
186, 208, 209, 211, 246, 283, 292, 302, 312,
347, 390, 536
as King, 242n57
as Light of the World, 171, 300
as Lord of the World, 45n147
al-Madn, Ab Ms Muammad b. Ab
Bakr b. Ab s, 339
alms, 114, 131, 150, 181, 190, 359
l Muammad (family of Muammad),
291
alphabet, 270
North Semitic, 269
symbols from, 272
al-Als l-Baghdd, Mamd, 581
al-Amash, Sulaymn b. Mihrn, 428, 524,
531, 532, 544, 564
index 647
mina b. Abd al-Manf (umm Muam-
mad), 182
mir b. ufayl, 132
mma (majority), 482484, 516
al-Ammn, Ab Muammad al-asan b.
Al b. Sad, 572
Amr b. Jish, 207
Amr b. Madikarib, 217
Amr b. al-Musabbi, 184
Amr b. Ubayd, 350
amr, sing. mir (Medina, Mecca, Kfa,
Bara, Damascus), 491, 507, 509, 527, 532,
544, 574575, 601603
Anas b. Mlik, 189, 191, 192, 193, 194, 197, 253,
317, 318, 335, 346, 463
al-Andalus, 519, 578
manuscripts of, 424
Andrae, Tor, 385
angel(s), 11, 63, 7172, 82, 260, 296
Gabriel, 17, 18, 21
malik, 6667
anr, 130, 137, 217, 225n16, 227, 255, 288, 334,
343
al-Anr, Ab Yay Zakariyy, 572
apostates, 297
Aqrab, Battle of, 223, 229
Arabia, 5, 6, 27, 44, 122, 233, 277, 383
pre-Islamic, 8, 260n61, 281, 319, 347,
373
Arabic (language), 8, 32, 51, 260, 269,
303
alphabet, 270
calligraphy, 271
dialects, 40, 41, 259, 414n169, 420, 537
Koranic, 299, 300
sources, 373, 379, 381, 386
Arabiyya, 525526, 565n154
Arab(s)
accepted Islam, 173. See also conversion
(to Islam)
chroniclers, 316, 318
Jews, 185, 383
and language, 477
al-Araj, umayd b. Qays, 549
Aramaic, 8, 27, 91n163
eryna, 209
orthography, 403
Arbad b. Qays, 132
archetypal (text), 50, 207, 209, 272
Arnold, Theodor, 386
Asad Efendi, 292
asir al-awwaln, 12
asbb al-nuzl, 361362
al- b. al-Wil, 76
Ascension to Heaven (mirj), 18, 20, 22, 59,
385
al-Ash, Maymn b. Qays, 28, 366
al-Ashar, Ab Ms, 192, 193, 196, 235, 236,
237, 243, 248249, 251, 308
shr, fasting of, 146n37
im b. Ad, 170
im al-Jadar, 391, 393, 418, 470, 483, 486,
492, 493494, 520525, 527, 530, 532, 533,
538539, 594, 604
Asm bt. Ab Bakr, 257
al-Ama, Ab Sad Abd al-Malik, 546
assimilation, 411, 417, 426, 437, 535, 539, 541,
555, 599
of consonants, 535n182, 536, 554
rules of, 543
astrology, 327
al-Aswad b. Abd Yaghth, 236
A b. Ab Rab Aslam al-Qurash, 49, 191,
349
al-Ar al-Hamadhn, al-asan b. Amad,
562, 566, 567, 574
awil literature, 185n294
Aws b. mir al-Rhib, 136
Aws b. Khlid, 217
Aws b. Mlik, 160
Aws b. al-mit, 172
Aws b. Suwayd, 160
Aws b. Thbit al-Anr, 159, 160
yt (symbols, signs, verses), 272
yat al-kurs (Verse of the Throne),
150n51
Ayysh b. Ab Raba, 163
Badr, 284, 292
Badr, Battle of, 37, 47, 87, 100, 102, 104, 109,
118, 124, 126, 141, 145, 151n63, 152, 153, 154,
158, 165, 173, 174, 196, 235, 236, 325, 341,
342, 379, 383, 385
al-Baghaw, Ab Muammad al-usayn b.
Masd al-Farr, 52, 337, 354, 478, 504,
576, 579
Baghdad, 508, 518, 519, 521, 524
al-Baghdd, al-usayn b. Uthmn b.
Thbit, 557
al-Bhil, Abd al-Wid b. Muammad, Ab
Muammad, 557
Bar (or Nestorios), 13, 23
Barayn, 305, 306
Bajla b. Abda, 202
648 index
al-Bakk, Ziyd b. Abd Allh, 321, 327
Bakr (tribe), 5
Balaam (of Bible), 130
al-Baldhur, Amad b. Yay, 327
al-Balam, Ab Al Muammad, 330
al-Blaw, mid b. Abd al-Fatt, 557,
559
Ban mir b. aaa, 132
Ban Amr b. Umayr, 345
Ban Asad b. Khuzayma, 179
Ban Aws, 135, 137, 139, 155
Ban Ghaafn, 163, 164, 167
Ban Ghifr, 119
Ban Hshim, 74
Ban Hudhayl, 390, 437
Ban Khuza, 169
Ban Kilb, 366
Ban Makhzm, 163, 257
Ban Mualiq, 119, 162, 168, 169, 171, 173, 178,
292
Ban Nar, 156, 158, 161, 163, 167, 342
Ban Qaynuq, 153, 155, 185, 186
Ban Quraya, 152, 163, 167, 168, 169, 186,
377
Ban Slim, 182, 207
Ban Salima, 107
Ban Tamm, 178
Ban ayyi, 217
Ban Thaqf, 112
Ban Umayya, 138, 284, 286
Ban afar, 164
Ban Zubayr, 262
al-Bqir, Muammad b. Al b. usayn b. Al
b. Ab lib, 358
bqn-system, 484, 544, 551
Bar alb, Dionysios, 460
basmala, 91n163, 9495, 243, 246, 249,
277278
of the Jews, 221
Bara, 249, 305, 306, 308, 334, 346, 395, 400,
401, 446, 468, 473, 478, 508, 521, 522, 532,
533, 583
manuscript of Koran, 399
Battle of the Trench, 47, 151, 158, 167168,
169, 171
al-Bay, 162
al-Bayw, Abd Allh b. Umar, 112,
123, 354, 356, 358, 368, 491, 577, 578,
580
al-Bayhaq, Amad b. al-usayn, 206, 600
al-Bazzr, Ab hir Abd al-Wid b. Ab
Hshim, 484, 524, 550
al-Bazz, Ab l-asan Amad b. Muam-
mad, 518, 540
Bedouins, 175, 179, 181, 217
believers, 87, 145, 155, 168, 169, 173, 176, 188,
296, 301, 343
marks of true, 110, 137138
on mortality of Muammad, 280
Bergstrer, G., 544, 566, 585
Bethge, Fr., 281
Bible, 68, 71, 94, 98, 208, 209, 233. See also
New Testament; Old Testament
Hebrew, 278, 573, 594
and Muammad, 12
bida, 476, 571
al-Bilbays, Shihb al-Dn Amad b. Al b.
Abd al-Ramn, 560
biographies, 367
collections of, 505, 507
of Muammad, 4, 83, 368
al-Birkaw (al-Birgl), Muammad b. Pr Al,
240, 570
Bir Mana, 197, 198
Bishr b. Raba, 217
Brockelmann, Carl, 315, 331
Buth, Battle of, 135
Buhl, Frants, 379
al-Butur, 367
al-Bukhr, Muammad b. Isml, 20, 70, 191,
192, 193, 197, 198, 205, 321, 332, 334, 335,
336, 337, 348, 351, 353, 361, 365, 370, 377,
383
al-Bulqn, Alam al-Dn, 363
al-Bulqn, Jall al-Dn, 363
Bura, 122
Byzantines, 20, 58, 122, 181
Caetani, Leone, 83, 84, 135, 229, 230, 281, 315,
371, 379380, 383
calendar, Islamic, 180
caliphate, 130, 289
caliphs, 300, 327
first/early, 200, 211, 213, 226, 231, 288, 291,
317, 329, 330, 331, 359
calligraphers, 261, 495
calligraphy, 271, 412, 587
campaigns, 126, 133, 148, 162, 163, 164, 165,
169, 171, 172, 180, 181, 237, 306, 320, 323,
324, 325, 347, 369, 378
Caussin de Perceval, Armand P., 373, 379
Celestial Tablets, 312
Christianity, 5, 6, 195, 248, 311, 312, 366, 384
influences on Islam, 45, 382
index 649
Christian(s), 2, 13, 59, 84, 119, 120, 125, 139,
143, 144, 146n37, 155, 165, 187, 277, 381
apologist, 460
Arab tribes, 136
communities, 1, 6
embassy, from Najrn, 100, 144n28,
154n86
influence of, 6, 82, 92, 242, 341, 378, 380,
381
orders to fight, 181
prayers/liturgies, 94, 246
scholarship of the Occident, 268
writings/scriptures, 12, 111, 311
codices, 586, 589, 596
in Istanbul, 590, 598
lapidary, 601, 603, 604
Medinan, 574
model, 256, 258
Samarqand, 586n4, 587, 588, 589n11, 595
Uthmnic. See Koran
commentaries/commentators (of Koran),
341, 342, 346, 349, 350, 353, 355, 356, 359,
361, 364, 386, 407, 430, 471, 497, 549, 564,
575, 582, 604
Shite, 581
Sufi, 360
Companions (of Muammad), 196, 198, 249,
254, 268, 278, 317, 334, 338, 340, 345, 507,
508
biographies of, 326, 369
earliest/old, 251, 262, 281, 286, 307, 346,
347
on Koranic text, 430
memorization of Koran, 36
readings of, 39
conjurations, 63, 78, 8890, 246
consonantal script, 474, 476, 491, 526, 542,
543, 545, 589, 591, 594, 600, 602, 603,
604
consonants, 409, 469, 490, 536, 568, 569, 570,
595
Constitution (of Medina), 6n16, 8, 95, 136,
322, 375, 382, 383
conversion (to Islam), 104, 151n63, 155, 325
Muammads goal of, 59
new converts of Medina, 137n11
outward, 179
Coptic Patriarch (of Egypt), 154
Cordoba, 552
correspondence/letters, 326. See also
epistles
of Muammad/Prophet, 154, 324, 382
creation, story of, 142
creed, 8, 88, 358
al-abb, Al b. Muammad, 559
al-ak b. Muzim, 110, 145, 349, 352, 588
Daya b. Khalfa al-Kalb, 18, 19, 23, 151
al-Djn (qri), 547
Damascus, 237, 251, 305, 306, 400, 401, 473,
492, 496, 497, 508, 509, 516, 521, 532, 533
manuscripts of Koran, 399
al-Dmir, 383
al-Dn, Uthmn b. Sad, 253, 396, 403, 404,
405, 411, 418, 420, 425, 479, 484, 493, 506,
527, 528, 550, 551, 553, 555, 558, 561, 565,
567, 568, 571, 574, 595596, 597, 598, 599,
600, 601
al-Draqun, Al b. Umar, 527, 551, 555
David (Prophet), 107, 256
de Sacy, Silvestre, 279, 405, 406, 568
de Tassy, Garcin, 293, 299
al-Dhahab, Shams al-Dn Ab Abd Allh b.
Al, 285, 339, 506
Dht al-Jaysh, 162
Dht al-Riq, 162
Dh l-Nn, 269. See also Jonah (Prophet)
Dh l-Qarnayn (Alexander the Great),
115116
diacritical marks, 65, 108, 476, 490, 542, 591
dialects, 259, 260, 420, 425, 437, 458, 478
differences in readings, 489, 490
dietary prescriptions, 131, 145
al-Dimy l-Bann, Amad b. Muammad,
407, 564, 574
al-dn (religion), 178
irr b. Amr, Ab Amr, 465
disbelief, 101, 165, 285
disbelievers, 6, 9, 23, 81, 87, 122n35, 139n16,
145, 151, 173
divorce, 166, 168, 169
dwn (collections), 365366
al-Diyrbakr, usayn, 49, 197, 374
doctrines/dogma, 5, 98, 139, 353
false, of Christians, 165
Dozy, R.P.A., 332, 370, 376
dreams, 18, 19, 22, 66, 67, 69, 82, 110, 111, 282
du, 240, 243
al-Dr, Ab Umar, 532, 539, 547
Egypt, 305, 306, 516, 517, 518, 521, 562
elegies, 55n1, 365
elisions, 414
emigrants, 118. See also muhjirs
650 index
emigration, 6, 56, 57, 59, 103, 121, 123, 131, 135,
136, 140, 141, 154, 158, 173, 176, 186. See also
hijra
to Abyssinia, 119, 379
to Jufa, 125
to Medina/Yathrib, 47, 53, 86, 102, 127,
288
epilepsy/fits, 20, 21, 25, 72
epistles (to pagan tribes), 95, 277. See also
correspondence/letters
eschatology, 77, 111
eternal life, 45, 59
Ethiopic, 8, 319, 411
etymology, 351, 415
exegesis, 347, 348, 352, 355, 356, 361, 362, 366,
372, 475, 478, 575
exegetes, 64, 75, 76, 77, 80, 86, 93, 94, 106, 107,
121, 124, 130, 145, 151, 153, 154, 174, 343, 344,
348, 354, 364, 446
Shite, 302, 359
exegetic
fancy/conjecture, 165, 172, 182, 187, 290
scholarship, 171, 253
traditions, 47, 68, 160, 179, 202, 268,
272n93, 281, 344, 352, 446
Ezekiel, 282
al-Fal b. Ibrhm al-Naw l-Kf, Ab
l-Abbs, 486
faith, 71, 106, 139, 143, 157, 289, 381, 491. See
also Islam
dn, 65, 178
enemies of, 76, 77, 79, 106, 154, 173, 180
Fn, Musin, 293
far, 491
Farewell Pilgrimage, 148, 150, 165166, 183,
342
al-Fris, Ab l-asan Al b. Muammad,
559, 562
al-Fris l-Shrz, Ab l-usayn Nar b. Abd
al-Azz, 562
al-Farr, Yay b. Ziyd, 471, 485, 543, 582,
583
farsh (private variant readings), 521, 551
farsh al-urf, 539, 540, 544, 555, 564
faa (linguistic purity), 534
al-Fs, Ab Abd Allh Muammad b.
asan b. Muammad, 558
fasting, 45, 129, 146, 146n37, 205
fat (victory), 158, 178
Ftia, 25, 90n158, 90n159, 93n166, 94, 242,
243, 246, 267, 357358
Fima (daughter of Muammad), 101, 169,
384
fight(ing), 142, 153
command to, 149, 173
Final Judgement, 6, 35, 59, 71, 81, 84, 100, 101,
300, 301, 385
fiqh, 475, 486, 487, 501, 518, 526, 568
canonical schools of, 491
Fischer, August, 81
Fleischer, Heinrich L., 356
Flgel, Gustav, 50, 264, 604
Followers, 259, 288, 317, 507, 508n16
food (prohibitions), 145, 146n37, 188
forgeries (of Koran), 279, 282, 284, 287, 290,
291, 299, 303
fornication, 170171
fuqah (jurisconsults), 287n40, 466, 468,
476. See also fiqh
furqn (word and meaning), 27
Gabriel, 1720, 23, 41, 67, 260, 269, 465. See
also Angel
Gagnier, Jean, 373, 379
Geiger, Abraham, 93, 380, 387
Gerock, Karl Friedrich, 380
gharnq, 82, 84, 113
gharb (curiosities; strange), 221, 337
al-Ghz b. Qays, Ab Muammad, 404, 519,
574
gnostic sects, 312
God. See also Allh
essence of, 88
of the Kaba, 75
Koran as Word of, 472
rest after Creation, 102
as speaker in Koran, 65
unity of, 59
voice of, 3
goddesses, 59n29, 75n60, 8283, 84, 101n22,
174, 342
Gog and Magog, 115
Goldziher, Ignaz, 84, 315, 332, 370, 371,
379380, 381, 439, 440, 441, 443, 444,
545
Gospels, 5, 6, 42, 129. See also New Testa-
ment
grammar, 303, 351, 352, 353, 355, 356, 358,
364, 430, 447, 469, 478, 479, 481, 482, 487,
504, 526, 549, 568, 575, 583
feminine/masculine endings, 405, 415,
418, 435, 477, 570
-lexical commentary, 497, 565
index 651
grammarians, 408n134, 416n178, 418419,
420, 455, 474, 476, 478480, 505, 516, 563,
583
Greek, 8
Grimme, Hubert, 61, 377, 378, 379, 381
Grohmann, Adolf, 586
adth, 68, 84, 194, 197, 203, 214, 253, 316, 319,
322, 338, 343, 347, 361, 370, 413, 526, 545.
See also traditions
arrangement of, 335336
biographical, 346
canonical, 331333, 335, 346, 369
collection of, 208, 253, 547
credibility of, 346
af (weak), 337
exegetic, 344, 346, 348, 447
falsified, 333, 371
field of, 356
form of, 332
as holy/Gods words, 193, 208
literature, 334, 353, 374, 375, 376, 383
Muslim criticism of, 351, 371
vs. revelation/Koran, 192, 198, 206
Shite, 359
adr, 533, 534, 535, 537, 568
afa bt. Umar, 175176, 225, 229, 230, 231,
235, 247, 249, 252, 342, 498n183
codex of, 256, 258, 260, 261, 266, 273, 277,
286, 306, 312
af b. Sulaymn, Ab Umar, 470, 486, 492,
521, 523, 525, 527, 532, 533, 604
ajj, 14, 147148, 179
minor, 145
ajjj b. Muammad al-Awar, 446
al-ajjj b. Ysuf, 462, 464, 593, 594, 595
jj Khalfa, 356, 560, 567
al-akam II al-Mustanir Billh, 552
al-alab, Al b. Ibrhm, 374
al-alab, Burhn al-Dn, 521
Hammer-Purgstall, Joseph von, 240
amr al-Asad, 157
hamz, 538, 554, 555, 574, 597, 598, 600
pronunciation of, 537
hamza, 409, 412, 419, 420, 421, 423, 424, 428,
429, 535, 544, 594, 596, 600
amza b. Abd al-Mualib, 109, 148n43
amza b. abb al-Taym, 118, 405, 470, 486,
495, 523, 524, 525, 529, 531, 533, 535, 537,
538, 539, 546, 554, 555, 571
anafite (school), 195, 466, 521, 524
Hanfiyya, 248
anfs, 6, 15n53, 182
Hni al-Yazd, 453
arf (letter, reading). See aruf
al-rith b. Yazd al-aram, 163
Hrn al-Akhfash, 455, 516, 551, 600
Hrn b. Ms, Abd Allh, 446
al-asan b. Ab l-asan, 49
al-asan b. Al b. Ab lib, 101, 359
al-asan al-Bar, 87, 112, 114, 350, 466467,
474, 490, 491, 498, 520521, 531, 544, 592,
595, 601
assn b. Thbit, 28, 36, 57, 104, 159, 165, 366
ib b. Ab Baltaa, 177
aywa b. Shuray al-im, 517, 561
heaven, 81, 82, 108, 273, 282
ascension to, 111. See also mirj
Hebrew, 8, 27, 68
Bible, 278, 573, 594
literature, 91n163
mishn, 94
hell, 59, 73, 74, 81, 296, 297, 378
Heraclius (Emperor), 154
hereafter, 101, 367
heresy, 43, 465
Hibat Allh b. Salma al-Baghdd, 42, 76,
79, 85, 86, 128, 189, 192, 248
ijz, 346, 401, 415, 418, 521, 603
dialect of, 420, 425
hijra, and dating of sras, 51, 53, 60, 112, 118,
127, 141142, 172174. See also emigration
Hill b. Umayya, 170171
im (Homs), 237, 251, 516, 517
ir, Mount, 66, 71
al-ra, 217
Hirschfeld, Hartwig, 61, 68, 108, 114, 274, 275,
280, 380, 382
Hishm b. Ammr al-Sulam, 537, 547, 551,
554
Hishm b. al-akm b. izm, 38, 39
Hishm b. al-Kalb, 344345
historiography, 319, 333, 366
history, 324, 330, 342, 373, 385
Arabic books of, 365
early Islamic, 324, 326
maghz, 322
pre-Islamic, 347, 348, 373
Shite work of, 326
world-, 326, 328, 329, 330
Horovitz, Joseph, 319
Hottinger, Johann H., 373
Houtsma, M. Th., 326
Huart, Clment, 14
652 index
Hd (Prophet), 12, 123
al-udaybiyya
campaign of, 147, 158, 172, 175, 179, 185, 187
Pact of, 6n16, 89, 11, 47, 95, 132, 138, 174,
177, 180, 342
Hudhalites, 367
udhayfa b. al-Yamn, 251, 258, 262
uff, 505
ujaj, 582
ukm, 487, 488
al-ulwn, 547
umrn b. Ayan al-Shaybn, Ab amza,
523
unayn, Day of, 181, 284
urf. See aruf, sing. arf
usayn b. Ab usayn, 361
al-usayn b. Al b. Ab lib, 101, 359
al-usayn b. Wqid, 49
al-ur, Ab l-asan Al, 493
Hypocrites, 73, 76, 87, 123, 137, 154, 158, 161,
163, 171, 172, 174, 176, 181, 188, 343. See also
munfiqn
Ibn Abbs, Abd Allh, 10, 49, 51, 191, 192, 193,
243, 253, 268, 317, 345, 346351, 353, 361,
368, 369, 371, 499
Ibn Abd al-Barr, Ab Umar Ysuf, 338, 339
Ibn Abd al-Kaf, Ab l-Qsim Umar b.
Muammad, 48, 114, 574
Ibn Ab Dwd al-Sijistn, 575, 589, 595,
596, 597
Ibn Ab Layl l-Anr, Muammad b. Abd
al-Ramn, 494, 523
Ibn Ab Shihb, 509
ibn dam (man), 194, 248
Ibn mir al-Yaub, Abd Allh, 308, 395,
470, 494, 509, 516, 517, 524, 528, 530, 532,
538, 551, 554
Ibn al-Anbr, Ab Bakr Muammad b.
al-Qsim, 291, 392, 405, 406, 571, 572, 574,
600
Ibn al-Arab, Muy l-Dn, 360
Ibn al-Athr, Izz al-Dn Ab l-asan Al, 165,
321, 330, 331, 339, 340
Ibn Aiyya al-Murib, Abd al-aqq, 41,
255, 362
Ibn Aws, Amad b. Muammad, 571
Ibn al-Bdhash, Ab Jafar Amad b. Al, 561
Ibn al-Briz, Hibat Allh b. Abd al-Ram,
557
Ibn Bunn, Ab Muammad Umar b. Abd
al-amad, 518
Ibn Dakhwn, 539, 546
Ibn al-Fam, Ab l-Qsim Abd al-
Ramn al-iqill, 560, 561
Ibn al-Fa al-Hamadhn, Fakhr al-Dn
Amad b. Al, 559
Ibn ajar al-Asqaln, 159, 160, 340
Ibn anbal, Amad, 333, 334, 336, 520
Ibn ayyuwayh, Muammad b. al-Abbs,
323
Ibn azm, Ab Abd Allh, 189
Ibn ibbn al-Bust, Ab tim Muam-
mad, 40
Ibn Hishm, Abd al-Mlik, 14, 67, 75, 106,
118, 181, 191, 197, 316, 321, 322, 324, 325, 327,
328, 330, 331, 352, 353, 365, 368, 374, 376,
377
Ibn al-Imm al-Mir, 362
Ibn Isq, Muammad, 58, 162, 317, 320,
321, 322, 323, 324, 325, 326, 327, 328,
329, 330, 331, 339, 352, 353, 365, 368,
602603
Ibn al-Jawz, Ab l-Faraj Abd al-Ramn b.
Al, 363
Ibn al-Jazar, Ab l-Khayr Muammad b.
Muammad, 306, 464, 471, 472, 481, 493,
494, 502, 503, 505, 506, 521, 528, 534, 545,
547, 550, 553, 554, 556, 557, 560, 562, 563,
564, 569
Ibn Jinn, Uthmn, 478, 481, 500, 549, 565,
566
Ibn Jubayr al-Ank, Ab Jafar, 547
Ibn Jurayj, Abd al-Malik b. Abd al-Azz, 349,
446
Ibn Kathr, Isml b. Umar, 576, 579
Ibn Kathr al-Dr, Abd Allh, 516, 517, 518,
528, 530, 532, 533, 535, 537, 538, 539, 540,
560
Ibn Khalaf al-iqill, Ab hir, 552, 560
Ibn Khlawayh, al-usayn b. Amad, 565,
566
Ibn Khaldn, 356, 407
Ibn Mjah, Ab Abd Allh Muammad b.
Yazd, 336
Ibn Mandah, Ab Abd Allh, 339
Ibn Masd, Abd Allh, 214215, 235236,
237, 243, 244, 245246, 247248, 251252,
258, 267, 287, 308, 346, 429, 430, 437,
443445, 473, 477, 496, 499, 523, 534, 582,
592
text of, 229, 266, 273, 431432, 447455,
456, 457, 459, 465466
Ibn Mihrn, Ab Bakr Amad, 493, 562
index 653
Ibn Miqsam, Muammad b. al-asan, 475,
476, 477, 480, 481, 552, 567
Ibn Muayin, Muammad b. Abd
al-Ramn, 518, 531, 544, 564
Ibn Mujhid, Ab Bakr Amad b. Ms, 352,
468, 472, 476, 478, 480, 484, 488, 490, 492,
493, 494, 495, 496, 497, 500, 502, 509, 522,
524, 526, 527, 528, 547, 548, 549, 550, 551,
552, 555, 556, 565, 567
Ibn al-Mund, Amad b. Jafar, 505
Ibn Muqlah, al-asan b. Al, 468
Ibn al-Nadm, Muammad b. Isq, 361, 505,
588
Ibn al-Qaq al-Makhzm, Ab Jafar Yazd,
489, 509, 518, 519, 531, 544, 547, 564
Ibn al-Qi, Ab l-Qsim Al b. Uthmn,
559
Ibn Qutayba, Abd Allh b. Muslim, 326, 393,
484
Ibn Sad, Muammad, 11, 84, 184, 197, 206,
213, 232, 255, 321, 324, 325, 326, 328, 330,
337, 338, 365, 375, 382
Ibn al-Sarrj, Ab Bakr, 551
Ibn Sayyid al-Ns, Muammad b. Ab Bakr,
323
Ibn Shannabdh, Muammad b. Amad,
467, 475, 548, 550, 567
Ibn Shurayh al-Ruayn, Ab Abd Allh, 560
Ibn Srn, Muammad, 263, 592, 595
Ibn Siwr al-Baghdd, Ab hir Amad b.
Al, 563
Ibn al-Sumayfa, 518, 549
Ibn Umar, Abd Allh (b. Umar al-Khab).
See Abd Allh b. Umar b. al-Khab
Ibn Umm Maktm, 79, 342, 343
Ibn Wi al-Yaqb, 305
Ibn al-Zibar, Abd Allh, 365
Ibn Zubayr, 190, 192
ibtid, 572
idghm, 537, 556, 599
idolaters, 73, 122n35, 142, 145, 151, 182, 185
Muslims ordered to attack, 180
idolatry, 2, 13, 101, 165, 301
irm (state of ritual consecration), 148
ijm (majority), 403, 465, 482, 483, 485, 486,
488
ijtihd, 487, 500n193
ikhf, 536
ikhtiyr (free choice), 465, 482, 484, 486, 487,
492, 494495, 501, 517, 523, 525526, 548, 549
Ikrima b. Abd Allh al-Barbar, 49, 190, 193,
263, 271, 349, 350, 351, 361
illiteracy, 260n61, 273
imla, 415, 416, 417n187, 418, 425, 454, 525,
538, 539, 544, 554, 564, 568, 573, 574, 595
Immite, 302
Immiyya, 289
imm(s), 289291, 296, 300, 301302, 327,
359
of Koranic variants, 495, 526, 528, 537,
546, 563, 565
Imrn b. Uthmn al-Zubayd, Ab Ibrhm,
517
incantations, 89
indecency, 160
infidels, 119, 301
inheritance, 159, 160, 165, 166, 383
intercession (from God), 82, 84
intermission (fatra) in revelation, 71
interpolations, 52, 73, 81, 104, 107, 128, 143,
210, 226n25, 279, 280, 281, 283, 289n51
intoxicants, 148n43, 161
invocation(s), 63, 89, 98
particle of, 412
sras of, 246
irb, 582
Iraq, 258, 306, 307, 308, 334, 401, 516, 517, 519,
524, 590
irtijl, 476
Isaac (Prophet), 124
s b. Maryam (Jesus), 119, 188. See also Jesus
s b. Umar al-Thaqaf, Ab Umar, 474
Ifahn, 524
Ishmael (Prophet), 124, 432
ishmm, 540, 555, 568, 594
Islam, 2, 158, 162. See also faith
beginnings of, 217, 310, 312
and Christianity, 381, 384
conflict within, 301
converts to, 112, 138, 139, 179, 366
development of, 38, 382, 383, 488
enemies of, 284
establishment of, in future, 196, 345
history of, 47, 110
influence of, 374
and Mormonism, 379
polemics against, 460
as religion of Abraham, 119, 381
as social movement, 377378
spread of, 59, 103, 317, 375. See also
propaganda
success at Yathrib, 136
tenets of, 45, 136, 146n37
as a term, 14n53
654 index
Isml b. Isq al-Mlik l-Azd, Ab Isq,
466, 548
Isml b. Jafar al-Anr, Ab Isq, 519
isnds (chain of authorities)
criterion of, 316319
to establish variant readings, 480, 495,
508509, 527, 543, 580
and versions of Koran, 230
al-isr. See Night Journey
Israel, Children of, 148, 255
Israelites, 1, 311
prophets of, 3
istana, 242
Istanbul, 586, 590, 598
istibb, 487
ihr, 536
al-Jabar, Ab Isq Ibrhm b. Uthmn
(or Umar), 407, 481, 559, 570
Jbir b. Abd Allh b. Amr al-Khazraj, 71,
165
Jbir b. Zayd al-Azd, Ab l-Shatha, 50
al-Jbiya, 516
Jacob (Prophet), 124
Jafar sl-diq, Ab Abd Allh, 291, 327
Jhiliyya, 14, 185n294, 366
al-Ji, 206
jiz, 573
Jall al-Dn al-Maall, 358, 580
Jall al-Dn Rm, 360
jamaa l-Qurn, 213, 219n2
al-Jannb, 373
jar, 392
al-Jawhar, 240
Jeffery, Arthur, 575, 604
Jerusalem, 108, 110111, 122, 142, 282, 283
Night Journey to, 82
Jesus, 1, 6, 119, 186n299, 188, 311, 381
verses on, 106107
Jewish
-Aramaic mathnth, 94
Day of Atonement, 146n37
laws, 148
poet Kab b. al-Ashraf, 130
prayers/liturgies, 94, 246
Quraya, 377
Sabbath, 119
scriptures, 5, 12, 264
sources, 95, 341
Jew(s), 2, 73, 119120, 126, 277, 284, 343
in Arabia, 5, 383
customs of, 145, 149
as godless, 142143
influence of, 5, 13, 82, 92, 173, 242, 347,
378, 381, 383
of Khaybar, 175, 185, 384
of Medina/Yathrib, 77, 113, 128, 129, 136,
138139, 165
Muammads relations with, 59, 88, 114,
116, 155, 162, 165, 383
verses directed to, 102103, 123124,
130131, 141, 143, 151, 156, 186187
winning over to Islam, 125, 146n37
Jibl, 521
jihd, 196, 382. See also war, religious
jinns, 14, 59, 64, 105, 108109, 130
al-Jirna, 207
John (of New Testament) /v, 107
Jonah (Prophet), 206, 269
Joseph (Prophet), 124, 380
story of, 288
Judaism, 1, 144, 195, 248, 271, 311, 312
influence on Islam/Koran, 380, 382, 383,
387
Judeo-Christian (tradition), 67
Judgement, Day of, 101, 296. See also Final
Judgement
al-Jumzr, Sulaymn b. usayn, 570
juz, pl. ajz (sections or reading parts), 220.
See also ajz
Kaba, 1, 14, 124, 136, 142n25, 144, 147, 173, 180,
282, 283
circumambulation of, 129
incorporation into Islam, 83
Kab b. al-Abr, 347
Kabbalistic (figures), 271
Kab b. al-Ashraf, 76, 130, 162
Kab b. Mlik, 104, 156
Ka b b. Zuhayr, 366
al-Kfiyaj, Ab Abd Allh Muy l-Dn, 363
khin, pl. kuhhn (soothsayers), 1n1, 30, 63,
105
Kalb (tribe), 5
al-Kalb, Muammad b. Sib, 24, 49, 350,
351, 352
Karabaek, Joseph von, 588
al-Ksh, Muammad b. Murta, 302, 307,
359
Kathr b. Afla, 253
al-Kawwsh, Amad b. Ysuf, 572
Kazem-Beg, 41, 293, 299, 302
Khadja (wife of Muammad), 59, 67, 72
al-Khafj, Shihb al-Dn, 581
index 655
Khalaf b. Hishm al-Bazzr, 476, 486, 495,
525, 531, 532, 544, 562, 564
al-Khall b. Amad, 594
al-Khqn, Ab Muzim b. Khqn, 534,
568, 569
Khrijite (sect), 288, 351
al-Kharrz(), Muammad b. Muammad,
407, 575, 603
khab, 217
al-Khab al-Tibrz, Muammad b. Abd
Allh, 192, 337
khatma (recitation of entire Koran), 493
Khawla (of Khuwayla), 172
al-Khawln, Ab Idrs, 202
Khaybar, 175, 185, 236, 384
siege of, 161
al-Khzin al-Baghdd, Al al-Dn Al b.
Muammad, 355, 362, 576, 578
Khazraj, 135, 136, 137, 139, 155
Khuzayma b. Thbit al-Anr, 261
al-Kis, Ab l-asan Al b. amza, 395,
486, 490, 494, 495, 522, 524, 525, 529,
531, 532, 533, 538, 539, 543, 554, 582,
600
kitb, pl. kutub (books), 209, 285
Koran, 2, 4, 34, 6869, 70, 98, 196, 198, 204,
312, 367. See also abrogation; revelation;
sras; verses
Alid, 220, 358
canonical edition, 263264, 308, 463, 465,
469
chronology of sras, 48, 51, 61, 263
circumstances of revelation, 22
codices of, 252, 290, 305306, 309
collection of, 183, 208, 211, 213, 214, 219,
220, 221, 226, 227, 228, 230, 231, 234,
235, 249, 256, 331, 362, 431, 473
commentaries on, 291, 302, 307, 335,
336, 341, 346, 352, 353, 355, 364, 386,
407, 430, 471, 497, 549, 564, 575, 582,
604
commission, 256, 258, 259, 262, 286, 391
composition of, 33, 43, 270
controversy on genuine, 288, 464
copies (Medinan, Meccan, Kfan,
Baran, Damascene), 237, 392, 393,
395, 396, 399, 402, 427, 453, 456, 586,
590, 601
copyists of, 403, 419, 463
diction of, 32, 51, 63, 117, 195
divisions of, 593, 594
errors in, 389, 391, 392
establishing, 194, 199, 255, 262, 574
exegesis of, 343, 347
foreign words in, 195, 417
forgeries of, 14, 279, 282, 290, 299, 303
adth vs. revelations, 205206
histories of prophets, 5
of Ibn Masd, 499, 522
Iraqi manuscripts, 420, 422
Jewish/Christian elements of, 5, 242, 380
language of, 117, 353, 383, 408
lapidary, 603
legal passages in, 166
local texts of, 400
manuscripts of, 234, 399, 403, 424, 430,
447, 457, 495, 520, 585, 586, 588, 590,
591, 598, 599, 601, 604
materials written on, 224, 232233
memorization of, 214, 217, 225, 228, 258
Musaylimas imitation of, 45
names in, 284, 285, 290
naskh, 594, 600
non-Uthmnic readings, 430431, 444,
446, 459, 462463, 464, 465, 467, 475,
488, 585
numbering of verses, 573574
official (ummaht), 595
opening formula of, 90
organization of sras, 24, 25, 75
origin of, 44, 52, 99
and ornaments, 587, 593
pauses in, 33, 51
pre-Uthmnic redactions, 263, 286, 287,
308, 457, 497
readings of, 4142, 306, 468, 471, 474, 487,
508, 509, 516, 520, 540, 543, 554, 599,
603. See also readings
recitation of, 39, 40, 108, 306, 416, 466,
471, 477, 494, 533, 554, 568, 582
reciters of, 217, 223, 229, 230, 235236,
287, 309, 351
sciences of, 362364, 392, 465, 520, 526,
550, 583, 585
script of, 588, 595
as source for biography of Muammad,
375
standardization of the text, 306, 488490
style of, 28, 52, 125, 300, 301
teaching of, 478, 494, 495, 496, 497, 524,
533, 543, 569, 573, 574
textual variants of, 394, 395, 443
translations of, 386387
of Ubayy b. Kab, 238, 239, 240, 244, 247
656 index
Koran (cont.)
Uthmnic, 267, 273, 389, 392, 402,
406, 427, 430, 431432, 434, 436,
437, 438, 441, 443445, 448451,
453455, 458459, 462, 464, 466, 468,
600
written text of, 404, 425, 426, 472
Krehl, Ludolf, 377
Kremer, Alfred von, 323
Kfa/Kfan, 237, 251252, 257, 258, 305308,
334, 400, 401, 446, 456, 473, 478, 496, 497,
508, 523, 532
Alid sympathies, 522
manuscripts of Koran, 234, 399, 417, 418,
427, 508
readers at, 455
schools of, 583
Kfic
fragments of Koran, 403, 408, 411, 489
ornamental style of, 587
parchment codices, 233
Labd (Medinan Jew), 89
Labd b. Raba, 132, 366
lan, 569
Lammens, Henri, 209, 344, 371372, 384
Laq b. Mlik, 44
al-Lt, 82, 174
law (well-guarded tablet), 67
law maf (Heavenly Tablets), 385
law(s), 112, 140, 146, 165, 185n294, 186, 188,
338, 383
canonical, 347, 348, 351, 353
criminal, 200
family and property, 169
al-Layth b. Khlid al-Baghdd, Ab l-rith,
539
Lent, 146n37
Leszynsky, Rudolf, 383
letters. See also logograms
alif, 423, 425, 427, 428, 570, 587, 590, 597,
600
shapes of, 587
as signs, 269
transposition of, 271
Lewis, A.S., 426, 457, 459
lexicography, 368, 430, 497, 549, 575
ligature, 426n258, 600
linguistic(s), 390, 455, 475, 481, 484, 489, 490,
500, 543, 565, 568
accuracy, 477478
exegesis, 551
expression, 91, 242, 444, 454
inaccuracies, 283, 301, 391, 479, 501,
569n184
purity (faa), 534
usage, 52, 67, 91n163, 140, 146n37, 150n51,
303, 326n50, 383, 445, 458, 502
literature, 365
Arabic national, 319, 366, 368
dalil al-nubuwwa, 325
liturgical, 6
littrateurs, 505
liturgies, 8, 68, 93n166, 123, 211, 277n1, 359
Jewish and Christian, 91n163, 242, 246
logograms, 267269, 268269, 270, 271, 272,
273, 274, 277278, 353, 554, 558, 559,
572
Loth, Otto, 271, 272, 273
lugha, 568, 582
Luqmn al-akm, 128
al-Mafir, Abd Allh Muammad b.
Amad b. Muammad, 559
man, 582
madd (prolongation), 477n41, 536, 537,
538n192, 554, 570
madhhab, 485n97, 524
al-Madn, Isml b. Jafar, 395
madrasa, 558
Madyan, 123n45
maghz
books, 320, 334, 335, 365
history, 322
Maghreb, 417, 519, 562, 566, 575, 591, 596
magic, 89, 150n51
al-Mahd l-qim, 289
majnn (possessed person), 64
majority, dogma of, 486487, 488, 490. See
also ijm
Makk b. Ab lib, 472, 480, 481, 482, 484,
500, 501, 502, 520, 521, 527, 552, 553, 554,
555, 556, 560, 563, 567, 568
Mlik b. Anas, 338, 393, 403, 466, 499, 519,
592, 595
Mamar b. Rashd, 50
Mant, 82, 174
Manichaeans, 146n37
manskh (abrogated), 189n1, 199, 465. See
also abrogation
Manr b. Amad al-Irq, Ab Nar, 562
manuscripts. See also Koran
Iraqi, 598
Kfic, 234, 399, 417, 418, 427, 508
index 657
lapidary, 589, 600
Maghrebi, 599, 603
Medinan, 599
naskh, 600, 603
non-Uthmnic, 585
al-Maqdis (al-Muqaddas), 484, 516, 517, 519,
521, 522
Maqn (Jews of), 384
al-maq wa-l-mawl (separation or
combination), 402403, 405
Margoliouth, David S., 379
Marracci, Ludovico, 373, 386
marriage, 161, 168, 201n109, 203, 342
Marwn b. akam (Umayyad caliph), 307
Mary, 107
Mary(am) (Coptic slave), 175176, 342
mashhra (canonical), 472, 487, 503, 565
Maslama (prophet), 223, 329. See also
Musaylima
Maslama b. Mukhallad al-Anr, 195, 196
Masoretic (texts), 48
al-Masd, Al b. al-usayn, 327, 330
mathn, 9394
matn (content; text), 316, 317, 318, 333
Mauricius, 122
maw (freely forged [variant readings]),
489
mawl, pl. mawl, 284, 477
Maymniyya, 288
al-Mzin, Ab Uthmn Bakr b. Muam-
mad, 522
Mecca, 2, 5, 55, 56, 57, 77, 80, 84, 110, 125, 142,
175, 284, 305, 306, 334, 400, 517, 520, 532,
533, 602
conquest of, 86, 113, 121, 138, 147, 155, 158,
173, 177180, 185, 188
manuscripts of Koran of, 396n47, 399,
418
period of, 69, 318
sanctity of, 120, 124, 127, 147, 187188
Medina, 25, 7677, 112113, 162, 251, 258, 307,
334, 400, 401, 489, 508, 519, 532, 533
codices of, 574, 590
Constitution of, 6n16, 8, 95, 136, 322, 375,
382, 383
converts of, 136n4, 142n25
emigration to, 47, 53, 86, 102, 127, 288
manuscripts of Koran of, 244n69, 392n27,
393, 394n42, 399, 404, 424, 588, 590,
596, 599, 601
mosque of, 107
people of, who did not fight, 181, 182
period of, 61, 73, 95, 99, 100, 128, 139,
146n37, 150, 178, 184n283, 196, 200,
271, 301, 328, 365, 372, 381, 385386
pre-Islamic, 135
reading of, 509, 518
script of, 412, 588n10, 596, 603
theocracy of, 132, 318, 382
variants of, 601
memory/memorization, 316, 348, 473, 554,
559
to preserve the Koran, 211
recitation from, 309
Mesopotamia, 305
Messenger (of Allh/God), 6364, 89,
148, 180, 183, 210, 213, 232, 248, 280n7,
296, 301, 302n115, 316, 340, 366. See also
Muammad; Prophet
messenger(s), 72n48, 98, 123, 282, 296297,
301
named Amad, 6
Messianic kingdom, 289
Meyer, Eduard, 384
mir, 26
milla (religion), 119n13
millat Ibrhm, 155, 165
Mingana, Alphonse, 458, 459, 460
Miqdd b. Amr (or al-Aswad), 235, 236, 237,
251, 308
miracles, 110, 282
mirj (Ascension to Heaven), 18, 22, 385
mir. See amr
monograms, 270, 274
monotheism, 84, 88n151, 122n35
of South Arabia, 378
moon, 269
Al as, 359
Muammad splitting, 344
oaths sworn by, 63
Moritz, Bernhard, 412, 417, 418, 588, 601
Mormonism, 379, 384
Moses (Prophet), 27, 114, 115n137, 123, 129,
136, 280n7, 296, 297, 301, 381
history of, 98
as Ms, Ibn Imrn, 119
al-Muaddil, Ab Isml Ms, 549, 556, 574
Mudh b. Jabal, 214, 517
muallaqa, 366
Muwiya b. Ab Sufyn, 237, 257, 289, 346
muawwidhatni, 8990
al-Mubarrad, Ab l-Abbs Muammad b.
Yazd, 475, 551, 583
Muar, clan of, 243
658 index
mufradt (individual variant readings), 522,
566567
al-Mughra b. Ab Shihb Abd al-Ramn
Abd Allh al-Makhzm, 270
muhjirs, 137, 230, 288, 343. See also
emigration
Muammad, 2, 3, 14, 19, 33, 44, 82, 133, 183,
202, 243, 269, 273, 280, 282, 290, 299, 301,
324, 344, 346, 359, 368, 430, 567. See also
campaigns; Messenger
ability to read/write, 912, 273
abrogating verses, 121
-Al (twin-constellation), 302
amanuenses of, 227, 235, 254, 256, 274,
286, 327
authorship of Koran, 194, 273
biographies of, 2, 318, 323, 324, 327, 331,
333, 365, 368, 369, 371, 373374, 375,
378, 381, 385, 386, 387
call to prophethood, 5758, 66, 7072, 91,
97, 103
characteristics of (ifa), 324326, 337, 385
childhood of, 324, 344, 385
Christian biographies on, 379380
daughters of, 168169, 384
death of, 178, 196, 211, 219, 220, 232, 235,
248, 279280, 282, 301, 312, 324, 341,
456
diction of, 51, 8687
domestic affairs of, 140, 175, 185, 332, 342
enemies, 65, 70, 284, 288, 365
history of, 141, 367
Ibrhm (Ibn Muammad), 76, 176
influence of Judaism and Christianity, 8,
13, 68, 387
and Jews, 88, 377
last years of, 169, 183184
letters and epistles of, 277
life of, 5556, 58, 168, 316, 320321, 327,
330331, 341, 372, 374
at Mecca, 377, 382
Medinan period, 210, 365, 381
name of, 6n16, 280281
personality of (shamil), 372
as political leader, 136, 166, 175, 318, 386
public life of, 65, 332
receiving revelation, 20, 21, 25, 207, 260,
372, 384, 465
recording of revelation, 36, 38, 230, 250,
270, 312
sayings of, as word of God, 201, 205, 242
as Seal of the prophets, 14
as social reformer, 378
wives of, 168, 175176, 292, 318
Muammad b. s l-Ibahn, 574
Muammad b. Isq b. Ibrhm al-Marwaz,
525
Muammad b. Ms l-Khwrizm, 327
Muammad b. Numn b. Bashr, 50
Muammad b. Sadn, Ab Jafar al-arr,
476, 547
Muarram, 146n37
Muir, William, 6061, 6466, 82, 85, 86, 87,
88, 89, 91, 102, 151, 370, 376
Mujhid b. Jabr al-Makk, 121, 349, 350, 351,
353
Mujammi b. Jriya, 215
mujtahid, 487
Mller, August, 79, 377, 385, 387
Mller, David H., 34, 79, 105
Mumin b. Al b. Muammad l-Rm
l-Qalkbdh, 407
munfiqn (Hypocrites, waverers), 137, 138,
139, 140, 141, 154, 174n217, 176, 284, 290,
291, 359
Muqtil b. Sulayman, 352
muqrin, 471n1, 516
murder, 112, 164, 393
musabbit (sras), 151, 197, 249
Muab b. Sad, 255
Ms b. Uqba, 320
Musfir b. al-ayyib al-Bar, 522
muannaf books, 334
Musaylima, 4445, 91n163, 131, 148n43. See
also Maslama (prophet)
muaf, pl. maif, 221, 232n52, 234, 256,
393, 431, 473, 477, 479, 481, 482, 488, 495,
496, 500, 504, 565n154, 589, 593
book, 233
Medinan, 424
official Cairene, 603
Muslim (Ibn al-ajjj) al-Naysbr, Ab
l-usayn, 191, 192, 193, 196, 197, 198, 205,
335, 336, 337, 348, 383
Muslims, 73, 137, 139, 143
community of, 104, 135
duties of, 112
signs of true, 130
victory over pagans, 82
musnad literature, 333334
muala, 503
muta (temporary marriage), 161
Muta, 181
battle of, 187
index 659
mutawtir, 503
al-Muawwi, 428
Mutazilites, 288, 355, 356, 430n273
al-Muttaq l-Hind, Al al-Dn Al b.
usm, 190, 200, 201, 202, 249, 255
al-Nbigha al-Dhubyn, 28
Nfi b. Abd al-Ramn al-Layth, 404, 470,
483, 486, 489, 490, 494, 509, 518, 519, 520,
528, 529, 532, 533, 537, 538, 539, 546, 560,
567
al-Nas, Ab Jafar Amad b. Muam-
mad, 571, 583
al-Nakha, Ibrhm b. Yazd, 391, 446, 588,
592, 595
naql, 475n28, 485, 598
al-Naqqsh, Ab Bakr Muammad, 506, 516,
548, 574
al-Nasaf, Ab l-Barakt, 200, 578
al-Nas, Amad b. Al b. Shuayb, 336, 383
nasakha, 210n8
al-Nashshr, Sirj al-Dn Ab af Uthmn
b. Qsim al-Anr, 557
naskh, 465
naskh Korans, 594
Nar b. im al-Layth, 592, 594
na (textual evidence), 403, 502, 545
al-Nawaw, Yay b. Sharaf, 335
al-Naysbr, al-Qumm, 579
New Testament, 5, 107, 311
stories in the Koran, 6
Night Journey (al-isr), 22, 82, 108, 110111,
282, 385
Night of Power (laylat al-qadr), 69, 78,
146n37
al-Nsbr, al-asan b. Muammad, 200
Noah (Prophet N), 101, 123, 129, 491
Nldeke, Theodor, 270, 271, 272, 273, 274, 280,
359, 371, 376, 382
non-Muslims, 155
numismatic (inscriptions), 587
nunation, 409, 416, 420, 421
nr (light), 299, 301
al-Nr l-afqus, Al b. Muammad, 481,
557
Nuayr al-Rz, Ab l-Mundhir, 575
nuskha (records; notebook), 545546, 567
oaths, 6364
Obbink, Herman Theodorus, 382
odes, 366
Old Testament, 56, 311
ordinances, 4, 147
marital, 149
Oriental (sources), 373, 375, 578, 604
Orientalists, 293, 373, 387
orphans, 160
orthography/orthographic, 89, 293, 391,
394, 401n87, 402403, 405, 407, 409, 413,
414419, 422423, 426, 428429, 443445,
458459, 466467, 556, 573, 587, 588, 590,
596n40, 598
Arabic vs. Western Semitic, 411
defective, 555
inaccuracies/errors, 390, 542
Koranic, 395n43, 406, 408, 424, 425n247,
428, 431n281, 438n374, 540, 543,
574575
licence, 468469
omissions (in written Koran), 411413,
420421
rules, 408419
transmission of, 404
of Ubayy b. Kab, 454
of Uthmnic Koran, 427, 585, 590, 603
variants, 397n56, 400, 402, 430, 431, 601
Ottomans, 521, 524
paganism, 6, 90, 195n71, 246, 342, 379
pagans, 88n151, 99, 119, 129, 139, 162, 185, 187,
281, 343
as adversaries of Muammad, 59
averruncation formulae of, 89, 172
conversion of, 84
epistles to, 95, 277
goddesses of, 84, 342
institutions of, 146n37, 157, 166
Muslim victory over, 82
superstitions of, 188
palaeography, 586, 603
Palestine, 113, 122, 340
palimsests, 426427, 428, 429, 457, 459
Paradise, 86, 198, 297, 378
Pautz, Otto, 382
People of the Book, 120, 127, 129, 131, 136, 151,
155, 188, 312, 381, 435, 473
persecution (of Muslims), 65, 83, 106, 119, 121,
127, 157, 301
Persian, reciting Koran in, 464n609
Persians, 58, 121, 122
Pertsch, Wilh., 412413
Pharaoh, 296
philology, 348, 352, 356, 375, 387, 497, 549,
568
660 index
pilgrimage, 147, 155, 158, 173, 175, 186, 187, 342,
359
ceremonies of, 145
Meccan, 129, 381
poetry, 28, 29, 155, 319, 322, 347, 348, 352, 355,
365367, 368, 423, 424, 473, 569
forms of (metre, rhyme, stanza), 32, 117
pre-Islamic (pagan), 260, 366
poet(s)
Muammad as, 97
pre-Islamic, 5, 28, 34, 37
referred to (in Koran), 104105
politics, 150
polytheism, 84, 101
prayers, 123, 161, 190, 211, 240, 242243, 267,
359, 378, 442, 466, 467, 488, 521
appointed times of (three), 45
of danger (alt al-khawf), 149, 163, 362
direction of, 142143, 142n25, 144, 145
five daily, 80
qunt, 243, 244, 247
alt, 142n25
Pretzl, Otto, 39, 529
priests/monks, 187
prisoners, 153, 178
pronunciation, 26n47, 29n70, 33n89,
41n130, 392, 401n81, 414416, 417419,
422, 425426, 428, 471n1, 490, 494, 497,
523524, 526, 533, 537, 540, 543, 551, 557,
568, 574, 600
of consonants, 536, 545
rules of, 521, 535, 544, 550, 554
of vowels, 538539
propaganda, 106, 136, 142n25, 150
prophecy, 1, 3, 345
Prophet (Muammad). See Muammad
prophethood, 1, 2, 22, 69, 71
call to, 60
early period of Muammads, 73
signs of, 324, 325
prophets, 34, 35, 44, 207, 280, 302, 491
ancient, 98, 120, 129
assembled on Night Journey, 108
false, 131
histories of, in Koran, 5, 33, 100
Jewish/of Israel, 15, 63, 101
prose, 29n70, 3031, 33, 45, 51, 319, 365n303,
367, 387n81, 559
prostitution, 170
Protosemitic, 27
Psalms, 107
punishment(s), 170, 200201, 468
Qadarites, 100, 344, 345
al-Q l-Fil, 558
qs, 236, 256, 468
al-Qdisiyya, Battle of, 217
qfiya (rhyme), 29
al-Qalnis, Ab l-Izz Muammad b.
Bundr, 562, 566
Qln, Ab Ms s b. Wirdn, 519, 529,
596, 598
al-Qannj, Muammad Bashr al-Dn
Uthmn, 579
qaraa, 385. See also qira (reading)
Qrn (Biblical ora), 125
qadas, 31n77, 366
al-Qasalln, Amad b. Muammad, 18, 255,
564, 574
Qatda b. Dima, Ab l-Khab, 350, 353,
446, 595
al-Qazwn, Ab Ysuf Abd al-Ramn b.
Muammad, 357n251
qibla, 142n25, 144, 146n37
qira, pl. qirt (reading; recitation), 174,
309, 425, 447, 465, 471n1, 474, 486, 493,
496, 497, 500, 508, 518520, 546547, 552,
571, 574, 578, 579, 582. See also readings;
recitations
science of, 537, 553, 563, 567, 575, 585
works, 522, 528, 562, 604
qiys, 487, 501502, 526
al-Qubqib, Muammad b. Khall, 564
al-Qumm, Al b. Ibrhm, 291, 302, 358
Qunbul, Ab Amr Muammad b. Abd
al-Ramn b. Muammad, 518, 538
qunt prayers, 244, 247
qurns, 207208, 210, 219
word and meaning, 2527, 209
Quraysh, 75, 83, 107, 113114, 124, 138, 150,
162163, 167, 173, 175, 177, 179, 180, 183, 188,
227, 254, 255, 261262, 290291
dialect of, 413, 420
envoys to Abyssinia, 106
poets of, 105
al-Qura, Muammad b. Kab, 350
qurr, 223n4, 468, 476, 478479
of Kfa, 473, 496
al-Qurub, Muammad b. Amad, 51, 199,
255, 357, 362
Qurub, Muammad b. al-Mustansr, 549
al-Rab b. Anas al-Bakr, 455
al-Ramn (divine name), 45, 91n163, 99,
132, 271
index 661
Raman, 41, 69
Ranke, Leopold von, 377
rasm, 405, 407n130, 570
rw (transmitter), 319, 500. See also riwya
rawm, 29n70, 540, 554, 568, 594
ray, 475, 486
Rayna (slave girl), 169
al-Rz, Fakhr al-Dn Muammad, 68, 357,
577, 580
readings
Baran, 511, 513, 515, 520, 528
canonical, 202, 391, 500, 502503, 526,
529, 547, 549, 565566
collection of, 547
complete, 498, 500, 526, 527, 547
Damascene, 533
Kfan, 511, 513, 515, 522, 523, 528, 533
oral, 403
standarization of, 543
uncanonical, 526, 566, 575577, 578, 602
ul of, 537, 540
variant, 42, 259, 261, 270, 306307, 352,
354, 355, 358, 395, 399, 400, 402407,
408n138, 414n169, 420, 429, 430, 431,
432, 444445, 458, 468469, 471,
474476, 478, 480, 481, 484485,
488489, 492, 495, 502, 507, 518, 520,
526, 541542, 551, 552553, 562563,
604
Eight, 529, 561562
Four, 497
Fourteen, 497, 521, 554, 564
individual, 494, 498, 500, 532, 548,
555, 557, 566, 567
local, 525, 526
Meccan/Medinan, 510, 512, 514, 518
science of, 478, 479, 482, 486487,
493, 497, 500, 503, 505, 509, 525,
534, 550, 552, 554, 569, 570
Seven, 353, 362, 464, 479, 489490,
493, 495, 497, 501503, 508, 509,
516518, 520, 522529, 529531,
532, 533, 535, 540, 542, 544,
547549, 551, 553555, 557558,
560, 561562, 564, 565, 567,
576580, 582
teaching of, 557
Ten, 497, 501502, 518, 521, 522,
524, 526, 529531, 544, 549,
563564, 578, 580
transmission of, 498, 499, 532, 565
writings on, 556557, 560
recensions (of Koran), 189, 194, 196197,
203204, 213, 226, 230, 234, 247248, 286,
307, 327, 429, 431, 454, 456
canonical, 259, 278, 309
of Ibn Masd, 258, 496, 499
model, 252
pre-Uthmnic, 277, 497
Uthmnic, 220, 225n16, 230, 234, 235,
246, 247, 252, 254255, 258259,
261262, 266, 279, 286287, 289290,
292, 305308, 310, 312, 390, 392, 394,
402403, 411, 427, 429, 462463,
473474, 496, 585
recitations (of Koran), 416, 430, 466
Kfic, 582
method of, 554, 570
oral (qirt), 395n43
seven aruf, 399n72, 465
system of, 534
types of, 535, 569
redactions, 259, 278, 308309, 364. See also
recensions
pre-Uthmnic, 234, 263
of Ubayy b. Kab, 266
of Uthmn, 290, 292, 306307, 310, 312
of Zayd b. Thbit, 273, 286
refrain (in sras), 33, 35, 85, 88, 105, 210
Reland (Reeland), 373
religion
of Abraham/Ibrhm, 142n25, 144, 155,
165, 174
of the Book, 313
no compulsion in, 150
repetition (in Koran), 97, 117, 299
resurrection, 59, 84, 101
Day of, 197, 249, 297
oaths sworn by, 64
revelation, 2, 4, 56, 11, 17, 22, 26, 27, 48, 53,
57, 67, 68, 140, 189, 194, 199, 211, 232, 259,
381
to Abraham and Moses, 12
collection of, 247
and contemporary events, 341
diction of, 98
on Final Judgement, 81
historical occasions of, 47, 74, 207, 343,
344, 345, 352, 358, 361
intermission (fatra) in, 7172
length of, 24
oaths sworn by, 64
as qurns, 207
revealed Word of God, 272
662 index
revelation (cont.)
subjects of, 343
types of, 18, 19
during vision/dream, 6667, 69
written, 67, 209, 284
rhyme, 3033, 51, 74, 80, 94, 100101, 106, 107,
114, 116, 117, 118, 124, 129, 140, 193, 195, 197,
200, 210, 248249, 282283, 292, 413415,
417, 426, 441, 442, 443, 539
rhythm, 69, 88, 101, 210
Rifa b. Amr/Rifa b. Waqash, 159
rijl (authorities), 554
riwya (canonical transmissions), 354, 355,
493, 494, 501, 518, 523, 564, 576, 580, 582
Roberts, Robert, 383
al-Rus, Ab Jafar Muammad b. Ab Sra,
486
Rckert, Friedrich, 387
ruy (daytime vision), 111
Sabians, 6, 143, 378
al-Sab, Ab Isq Amr b. Abd Allh
al-Hamdn, 446
Sachau, Eduard, 315
adaqt, 181, 378. See also alms
Sad b. Ab Waqqs, 126, 217
Sad b. al-Rab, 159
Sad b. Ubayd, 215
af and Marwa, 145, 433
al-Saff (Abbsid caliph), 334
al-afrw, Abd al-Ramn, 554, 566
afwn b. al-Muaal, 342
Sad b. al-, 227, 236, 251, 253, 254, 255, 257,
258, 262, 308
Sad b. Jubayr, 349, 350
Sad b. al-Musayyab, 49
saj, 29, 33, 51, 63
Saj (prophetess), 44
al-Sajwand, Muammad b. ayfr, 572
al-Sakhw, Alam al-Dn Al b. Muammad,
363, 406, 558, 569
sakt, 537, 556
alt al-khawf (prayer of danger), 163, 362.
See also prayers
Sale, George, 386, 387, 388
li (Prophet), 123, 129
Slim b. Maqil, 215, 220, 221, 230
Salmn (the Persian), 121
Smnids, 330
Sandhi, 408n137, 536, 537, 599
Satan, 45, 64n4, 82, 85, 90, 148, 376
say (ceremony), 433
sayl al-Arim (Flood of the Dyke), 14
Schapiro, Israel, 380
script (of Koran), 408, 426, 445, 470, 595, 603
ijz, 588, 590
Kfic, 542, 586, 590
lapidary, 587591
Maghrebi, 590, 596
style of, 587588
Scripture(s), 10, 125, 209
of Jews and Christians, 5, 12, 73, 360
sermons, 14, 32n88, 81n101, 84, 104n48, 109,
127, 284, 327, 377
Seven Sleepers, legend of, 116
shdhdh, pl. shawdhdh (uncanonical), 430,
487, 488, 497, 502, 526, 544, 563, 565, 566,
575, 582, 583, 585
shahda (formula), 437
shahd/shhid, 133
al-Shahrastn, Muammad b. Abd
al-Karm, 302303
al-Shahrazr l-Baghdd, Ab l-Karam
al-Mubrak, 563
Shanfar (poet), 28
shar (commentaries), 368, 552
shara (Islamic canon law), 491
Shark b. al-Sam, 170
Shas b. Qays, 155, 156
al-Shib, al-Qsim b. Firruh, 406, 493, 558,
559, 575
shawhid (evidence), 355, 368
al-Shawkn l-Yaman, Muammad b. Al,
581
Shayba b. Ni, 518
Shaydhalah, Ab l-Mal Azz b. Abd
al-Malik, 363
Sha, 288, 301, 327, 359
Shite(s), 220, 288, 289, 291292, 293, 300,
307, 327
on aruf, 40
commentaries, 219, 358359
deification of Al, 289
exegetes, 302, 359
fabrications, 36, 250, 302
history, 326
literature, 291, 292
on Muammads ability read/write, 9
al-Shirbn, Ysuf b. Muammad al-Khab,
580
shuar (shir), 105. See also poet(s)
Shuayb (Prophet), 123, 129
Shuba b. Ayysh, 523, 527, 533, 539
Shuba b. al-ajjj, 350
index 663
Shula, Muammad b. Amad, 559
Shurabl b. Sad (Sad), 320
al-Shshw, al-usayn b. Al, 51
Sbawayh, Amr b. Uthmn, 401, 447, 481,
489, 492
Sib al-Khayy, Ab Muammad Abd
Allh, 561, 563, 564
sidrh (Heb.), 26
iffn, Battle of, 473, 496
signs, 269, 270, 271, 296
of division of sras, 278
of division of verses, 600
of God, 98, 282, 302
of prophethood, 324325
reading, 495, 545, 603
of true Muslim, 130
vowel, 594, 600
sra, 320, 321n23, 324326, 332, 371, 383
slaves/slavery, 3, 70, 121, 169, 175, 185n294,
326, 343, 349, 351, 378, 383
Smith, Joseph, 384
Snouck Hurgronje, Christiaan, 124, 147148,
381
Solomon (Prophet), 95, 114, 256, 277
soothsayers, 97. See also khin, pl. kuhhn
South Arabia, 378, 518
spelling, 408, 420426, 428, 437, 443, 451
(hij), 404
spoils of war, 152153, 156157
Sprenger, Aloys, 12, 13, 14, 18, 23, 57, 68, 71, 82,
83, 271, 281, 341, 349, 369370, 372, 374,
375, 376, 377, 379, 383
succession (after Muammad), 10, 38, 289
al-Sudd, Isml b. Abd al-Ramn, 446
al-Sudd, Muammad b. Marwn, 351, 353
Sufis, 359360
Sufyn b. Uyayna, 350, 351
al-Suhayl, Abd al-Ramn b. Abd Allh b.
Amad, 191, 193, 197
uuf (leaves), 23, 232233
sukn, 568, 598, 599
al-Sulam, Ab Abd al-Ramn (d. 73 or
74/692 or 693), 496, 523
al-Sulam, Ab Abd al-Ramn (d. 412/1021),
360
sunna, pl. sunan
books, 332, 336, 337, 359, 369
Sunnites, 130, 290, 291
on Al, 288
historical works of, 219
on Muammads inability to read/write,
9
superstitions, 14, 89, 90, 146n37, 188, 266,
267, 307
sra (word and meaning), 2526
Sra of the Two Lights, 293, 294296, 299,
301, 302
sras, 3, 24, 32, 47, 53, 78, 352, 353
arrangement of, 25, 37, 6061, 75, 210, 213,
220, 245, 247, 263, 274, 312
beginning of, 277
chronology of, 48, 99, 118, 147, 234, 383
divisions of, 594
headings of, 592, 593, 602
of invocation, 246
length of, 61, 64, 117, 140
Meccan, 31, 32, 47, 4852, 55, 58, 59,
6061, 75, 99, 127, 129, 139, 301, 327
Medinan, 17, 31, 32, 37, 47, 4852, 80, 301,
327
names of, 592
style of, 30, 60, 382
Srat al-afd (or al-Khal or al-Qunt),
240241
al-Suy, Jall al-Dn, 18, 41, 42, 48, 49, 50,
77, 78, 107, 109, 124, 189, 192, 193, 194, 197,
200, 213, 221, 227, 237, 239, 240, 244, 245,
246, 255, 256, 345, 346, 357, 358, 359, 360,
361, 363, 364, 406, 407, 472, 580
synonyms, 352, 444, 451, 452, 454455, 458,
466, 474
Syria, 13, 306, 307, 379, 521, 524, 562, 603
Syriac, 27, 91n163, 319
story of Ahikar/Ahiqar, 191n40, 192
translation of Koran, 460
abaqt (book of classes), 324, 326, 338, 339,
506, 554
al-abar, Muammad b. Jarr, 52, 67, 76, 122,
192, 197, 198, 200, 255, 261, 271, 317, 321,
328330, 331, 333, 338, 353, 354, 362, 365,
368, 374, 375, 376, 392, 430, 446447, 455,
465, 466, 471, 475, 485, 548, 552, 576, 579
al-Tabars, Ab Al l-Fal b. al-asan, 581
tbin (Followers), 508
Tabk, 109, 171
battle of, 187
tafsr, 351, 356357, 465, 499, 565, 580, 581
Taghlib (tribe), 5
taqq, 524, 534, 535, 537, 568
tazn (sorrowful recitation), 535
al-if, 108, 109, 123, 150, 346, 367
journey to, 5859
tajwd, 533, 535, 536, 537, 561, 563, 568570
664 index
ala b. Muarrif al-Hamdn, 523, 549
ala b. Ubayd Allh, 270
tall, 497, 552, 553, 563, 582
talqn, 493n156, 524
Tamm (tribe), 5
embassy of, 342
Tamm al-Dr, 214
tam, 535
tamyz, 193
Tankh (tribe), 5
tanwn, 29n70, 536, 568, 594, 596597, 599
tard, 535
arq, pl. uruq, 519n71, 520n77, 529532, 549,
561, 563, 564
tarqq (thinning out), 535, 570
tartl, 533534, 568
tashdd, 594, 598599
taf, 476
tashl, 574, 597
Takprlzade, 240, 507
tasmiya, 277. See also basmala
tarb, 535
tawtur, 485, 502504
tawl, 347, 359
taxes, 178, 378, 381
communities liable for, 108, 127, 142
Tayyi (tribe), 5
Temple at Jerusalem, 256
Thbit b. Qays b. Shamms, 160
al-Thalab, Muammad b. Ibrhm, 350,
354, 355, 362
Thamd, 15
Thamdites, 78, 296
Thaqf (tribe), 85, 390
theft, 164165, 185186, 343
theocracy, 136, 382
theology, 84n118, 356, 378, 497
al-Tirmidh, Muammad b. s, 189, 191, 193,
194, 198, 261, 325, 336, 337, 353
Torah, 129. See also Bible
Torrey, Charles, 383
traditionalism, 488, 492, 500
traditionists, 121, 253, 327, 336, 455, 481, 495,
505, 518, 549
traditions, 322, 329, 331, 332, 334, 370, 384,
526. See also adth
arrangement of, 333, 337
collections of, 430
consensus of, 74
criticism of, 381
deviating/divergent, 253, 256
dogma of, 480, 481, 482
fabricated, 9, 84, 281, 291n62, 317, 318, 332,
343, 359, 372
form of, 230, 336
Muslim criticisms of, 333, 336, 374, 375,
380
rankings of, 336n109, 337
reliability of, 194, 208, 336
science of, 321, 336, 447, 532
weak (af), 255, 336n109
translations, 373, 460
of Koran, 386387, 388
transmissions, 370, 391392, 394, 456, 567.
See also riwya
oral, 316, 444, 445, 454455, 459, 470, 473,
474, 477, 500, 545, 582
reliability of, 342
transmitters, 551, 554, 556. See also rw
Transoxiana, 524
treaties, 8, 316
tribes. See Ban
ulaya, 44
uma b. Ubayriq (or Ab uma), 186, 343
al-uraythth, Al b. al-usayn, 566
Ubda b. al-mit, 215
Ubayd Allh b. Ziyd, 589, 590
Ubayd b. Muwiya b. Zayd b. Thbit b.
al-ak, 215
Ubayd b. Umayr b. Qatda, 67
Ubayy b. Kab, 39, 189190, 192194, 200, 201,
214215, 229, 235, 237, 238, 239, 240, 242,
243, 244, 247249, 254, 255, 266, 267,
289n51, 308, 430, 431, 436, 447, 453456,
459, 465, 499, 582
Uud, Battle of, 109, 121, 126, 145, 149, 151153,
156160, 162163, 165, 171, 186187, 198,
235, 325, 342, 359
Uk, 108
al-Ukbar, Ab l-Baq, 566, 577
Ult al-Jaysh, 162
Ullmann, Ludwig, 387
Umar b. Ab Raba, 423
Umar b. al-Khab, 3, 36, 3839, 119, 148,
196, 199, 201, 217, 223, 225, 226228, 229,
231, 236, 243, 256, 279280, 312, 317, 359,
430, 473, 569
conversion of, 58, 102103, 153
and recensions of Koran, 230, 327
Umar b. Qsim al-Anr, 560
Umar b. afar, 566
Umayya (family), 262, 359
Umayya b. Ab l-alt, 14, 28, 130, 367
index 665
Umayyads, 217, 220, 257, 284, 327, 347, 359,
460, 516, 603
in Spain, 552
umm, 11n28
Umm Kua, 159
Umm Kulthm, 168169
umra (visits, to Kaba), 147148
unbelievers, 155, 169, 176, 187, 435
Muslims to fight, 182
Uqba b. al-Muay, 76
Urwa b. al-Zubayr, 66, 320, 322
Usma b. Zayd, 164
Usayd, 207
al-Ushmn, Amad b. Muammad b. Abd
al-Karm, 572
ul, 487, 525, 526, 539, 550, 551, 554556,
563564, 571
al-fiqh, 503
of variant readings, 537, 544
usury, 150
al-Utaq, Abd al-amad b. Abd al-Ramn,
546
Utba b. Raba, 118
Uthmn b. Affn, 214, 236, 237, 251252,
253254, 256, 257, 259260, 262, 277, 280,
286, 291, 306, 320, 389, 391, 392, 447, 464,
527, 593, 601
accused of forgery, 284, 285, 287, 288
rebellion against, 258
text of, 260, 261, 263, 327, 331, 428, 430,
473, 474, 477, 479, 488, 501, 574
Uwaymir b. rith, 170171
al-Uzz, 82, 174
variants/deviations, 431, 446447, 453456,
459, 461462, 464468, 484, 495, 595,
601602
orthographic, 396399, 400
textual, 574, 575, 601
Verse of Stoning, 199, 200, 201, 226, 248
Verse of Summer, 166n166
Verse of the Throne (yat al-kurs), 150n51,
272
verses, 36, 210, 353
abrogated, 196
arrangement of, 31
authenticity of, 279283, 446
context of, 52
division of, 293, 591, 592, 593
length of, 51, 61, 6364, 80, 98, 117, 264
lost, 248, 249
numbering of, 602
number of Koranic, 573574
visions. See dreams
vocalizations, 68n23, 293, 417, 432, 436, 448,
454, 462n604, 476, 489491, 497, 540, 595,
600, 602
vowels, 260, 403, 409, 411, 413, 414, 415, 416,
418, 419, 420, 421422, 423, 427, 535, 536,
537, 538, 539, 540, 541, 554555, 568, 569,
596, 598
marks of, 591, 594, 596, 597, 599
al-Wid, Ab l-asan Al b. Amad, 119,
127, 346, 361362, 493
Wahl, Samuel F.G., 387
Wash b. arb al-abash, 109110, 125
way (revelation), 17, 209
Wak b. al-Jarr, 350
al-Wald b. al-Mughra, 79
Wald b. Uqba b. Ab Muay, 178, 257
waqf (pause), 29, 551, 570573, 574, 583, 600
al-Wqid, Muammad b. Umar, 20, 50, 119,
162, 197, 203, 254, 323, 324, 325, 326, 327,
328, 330, 331, 341, 352, 353, 358, 365, 375,
376, 381
war, religious, 147, 156, 158, 174, 185, 196
Warsh, Uthmn b. Sad, 470, 486, 518, 519,
529, 532, 537, 538, 539, 563, 598
al-Wsi, Amad b. Ibrhm, 374
wal, 409, 413, 414, 416, 551, 573
Weil, Gustav, 20, 22, 23, 36, 38, 51, 61, 68, 81,
89, 97, 100, 108, 112, 113, 120, 122, 124, 153,
157, 186, 279, 280, 281, 282, 283, 284, 285,
286, 373374, 375, 377, 382, 387
Wellhausen, Julius, 135, 323, 325, 376, 381,
383, 385
Wensinck, Arent Jan, 383
wine. See intoxicants
wird, 594
women, 168, 334, 339
Wright, Wm., 417
Wstenfeld, Ferdinand, 316, 317, 324, 328
Yay b. Ab Kathr, Ab Nar, 591, 592
Yay b. al-rith al-Dhimr, 404, 509, 516
Yay b. Man (Mun), 525
Yay b. Yamar, 595
Yamma, Battle of, 217, 226, 228, 229, 232
al-Yamma, 223
Yaqb al-aram, 522, 529, 531, 532, 544,
547, 561, 562, 567
al-Yaqb, Amad b. Ab Yaqb, 49, 220, 227,
263, 326
666 index
Yarmk, Battle of, 255
Yathrib, 5, 124, 125, 135, 137, 139, 153, 284, 383.
See also Medina
Biblical ideas at, 136
siege of, 167
Yazd b. Hrn al-Wsi, 350
Yazd b. Quayb al-Sakn l-im l-Sham,
517
Yazd al-Fris, 589
Yemen, 5, 305, 306, 378, 521
Ynus b. Bukayr (Ibn Wil al-Shaybn),
339
Ysuf Efendi-Zade, Ab Muammad Abd
Allh b. Muammad, 501
al-hir, Dwd b. Khalaf, 475
al-Zajjj, Ab Isq Ibrhm b. Muammad
b. al-Sar, 583
zakt, 378, 381. See also alms
al-Zamakhshar, Mamd b. Umar, 112, 200,
240, 249, 355, 356, 368, 394, 402, 409, 411,
430, 447, 499, 501, 504, 577
al-Zarkash, Badr al-Dn Muammad b. Abd
Allh, 364
Zayd b. Akhzam, 326
Zayd b. Amr b. Nufayl, 13
Zayd b. ritha, 168, 284, 285, 342
Zayd b. Thbit (b. al-ak), 38, 213, 214,
215, 219, 223225, 227, 228, 231, 232, 236,
243, 247, 251, 254, 255, 256, 258, 259, 260,
261, 263, 266, 270, 271, 273, 274, 286, 287,
306, 473
Zayd al-Khayl, 184
Zaynab bt. Jash, 168, 285
Zaynab bt. Kab b. Ujra, 342
al-Zaynab, Muammad b. Ms Ab Bakr,
518
Zechariah, 107
zin (adultery), 200
Zindq, 345
Zirr b. ubaysh, 523
Zotenberg, H., 330
al-Zubayr b. al-Awwm, 270, 320, 345
al-Zuhr, Muammad b. Muslim, 50, 253, 322
We hope you learned what you expected to learn from
this eBook. Find more such useful books on
Learn more and make your parents proud :)
Regards
Books?
Want More
Thank You

The History of The Quran by Theodor Noldeke

Uploaded by

Copyright:

Available Formats

The History of The Quran by Theodor Noldeke

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The History of The Quran by Theodor Noldeke

Uploaded by

Copyright:

Available Formats

E

, the promised messenger, . Cf. Ibn Sad (abaqt) Biographie

. In Talmud and Midrash the

, like i.e. = (cf. Diyrbakr, Tarkh

would not be an insurmount-

( ); K. al-Shur, 15; Muslim, a vol. 2, p. 170sq.; (al-

, words that nearly all commentators take to mean the

who have a poor understanding of

4); al-Khab al-Tibrz, Mishct-ul-

, he replied with the words

, and frequently in traditions,

, a book on dogmas and legends.

. It makes even less sense when Sprenger also declares

, to write (sras 17:60, 52:2,

(compare the Hebrew designation of a civil servant,

sra 17:60 and are considered

(sra 25:5), and

(sra 16:103). The Commenta-

(sra 48:29.) Although

(four times). D.S. Margoliouths idea (On the origin

( .) The identical form occurs outside the rhyme in a poem

, 1; Muslim in al-Qasalln, vol. 9, p. 333; al-abar, Ab

or . The sources, p. 16 n. 17, above.

, 2 end (vol. 3, p. 145) = , 10 (vol. 1, p. 202); al-Khab

) which, as will be demonstrated below, was a vision. That

, sra 17:81), and that he fasted a great deal. Such

), with the excep-

; cf. the more frequent

, al-abar in the introduction to the Tafsr, Cairo ed.,

or, with a weak hamza,

( ) and (Ibn Hishm, p. 713, l 1 = Dwn, p. 45, l 9)

not fromto read, rather

, with the meaning, unite, collect

. This was the opinion of

) and rhymed prose ( ).

which is also frequently found in rhymed verse.

(Dwn, ed. by Brockelmann, no. xxi, 4) for

, Labd, vol. 2, no. xxxix, l 1); in al-Ash:

, sra 101:7. This, too, is possible in rhymed

to avoid the un-Arabic form

). These forms caused Muslims a great deal of

) we find that the last word takes the place

made minute observations (listed in al-Itqn, p. 699sqq.) although at times going

(cf. al-Qasalln, vol. 9, p. 6sq.), in a slightly different arrangement,

, fal 2, 8; [AB AL-QSIM] Umar b. Muammad

; and sra 71:5,

; in Labd (Ibn Hishm, p. 941, l 10:)

(in al-Buturs [EI

); sra 2:181 ( and ; cf. sras 65:7 and 92:7 and

(sra 104:1, cf. AbZayd, al-Nawdir, p. 76,

) as soon as it was written down and frequently

(Sprenger, Leben und die Lehre,

. In this case the verses had already been communicated to others.

); al-Khab al-Tibrz, Mishct, fail al-Qurn,

), Persian, Syriac, Nabataean, Ethiopic!

. Al-Qurub, vol. 1, Jmi akm al-Qurn,

(sra 44:44) to say

, 37; al-Khab al-Tibrz, Mishct, ibid., bb 3, fal 1, 2;

, al-abar, vol. 1, p. 1738, l 17; Ibn