THE ORIGINS OF

MUHAMMADAN

JURISPRUDENCE

By

JOSEPH SCHACHT

OXFORD

AT THE CLARENDON PRESS

Oxford University Press, Walton Street, Oxford OX2 6DP

OXFORD LONDON GLASGOW

First published 1950

Reprinted 1953, 1959, 1967, and 1975

All rights reserved. No part of this publication may be

British Library Cataloguing in Publication Data

Schacht, Joseph

The Origins of Muhammadan jurisprudence.

1. Islamic Law

I. Title

Printed in Great Britain

PREFACE

THIS book is concerned with the origins of Muhammadan jurisprudence. I shall, of course, often have occasion to refer to examples taken from Muhammadan law, which is the material of Muhammadan jurisprudence. But the history of positive law in Islam as such, and the relationship between the ideals of legal doctrine and the practical administration of justice fall outside the scope of the present inquiry.

The sacred law of Islam is an all-embracing body of religious duties rather than a legal system proper; it comprises on an equal footing ordinances regarding cult and ritual, as well as political and (in the narrow sense) legal rules. In choosing the examples I shall concentrate as much as possible on the (properly speaking) legal sphere. This course not only recommends itself for practical reasons; it is also historically legitimate. For the legal subject-matter in early Islam did not primarily derive from the Koran or from other purely Islamic sources; law lay to a great extent outside the sphere of religion, was only incompletely assimilated to the body of religious duties, and retained part of its own distinctive quality. No clear distinction, however, can be made, and whenever I use the term Muhammadan law, it is meant to comprise all those subjects which come within the sacred law of Islam.

I feel myself under a deep obligation to the masters of Islamic studies in the last generation. The name of Snouck Hurgronje appears seldom in this book; yet if we now understand the character of Muhammadan law it is due to him. Goldziher I shall have occasion to quote often; I cannot hope for more than that this book may be considered a not unworthy continuation of the studies he inaugurated. Margoliouth was the first and foremost among my predecessors to make more than perfunctory use of the then recently printed works of Shafi'i; in reviewing the field which is surveyed here in detail he came nearest, both in his general attitude to the sources and in several important details, to my conclusions. Lammens, though his

vi PREFACE

writings rarely touch Muhammadan law and jurisprudence directly, must be mentioned in the preface to a book which is to a great part concerned with the historical appreciation of Islamic 'traditions'; my investigation of legal traditions has brought me to respect and admire his critical insight whenever his ira et studium were not engaged. In the present generation, Bergsträsser, with penetrating insight, formulated the main problems posed by the formative period of Muhammadan law and offered a tentative solution. Although my results are rather different from those which he might have expected, I must pay homage to the memory of my late teacher who guided my first steps in Muhammadan jurisprudence.

All my previous studies in Muhammadan law have led, in a way, to the writing of this book. But, when I came to write it, the refusal of the Egyptian authorities to allow me to return to my work and home in Cairo in 1939 deprived me of the use of my library at the time I needed it most. I particularly regret that I was thereby prevented from consulting the Kitab al-Hujaj by Shaibani, the Kitab al-Sunan by Shafi'i, the Kitab al-Diyāt by Abū 'Āșim Nabīl, the Muntagā min Akhbar al-Aşma'ī, and the materials for my own editions, in varying stages of preparation, of the History of the Judges by Waki', of the Kitab al-Asl by Shaibani, and of the Kitab al-Masa'il by Ibn Hanbal. That I was able, notwithstanding this handicap, to use all essential texts, I owe mainly to the British Museum and to the Griffith Institute in Oxford, and to the unfailing courtesy and helpfulness of their staffs.

I wish to express my deepfelt gratitude to the Governing Body of St. John's College, Oxford, and to Mr. K. Sisam, formerly Secretary to the Delegates of the Clarendon Press, for the active interest they took in my studies in general and in this book in particular, and for the assistance they gave me. Professor F. de Zulueta has accompanied my studies in Muhammadan law and jurisprudence with sympathy and interest since the invitation given by him and by the late H. Kantorowicz to contribute to the projected Oxford History of Legal Science which unfortunately had to be abandoned. Dr. D.

PREFACE vii

Daube, of Gonville and Caius College, Cambridge, kindly enlightened me on points of Roman law, and Dr. S. Weinstock of Oxford most obligingly translated for me from the Hungarian a paper by Goldziher. Without the unfailing encouragement and help of Professor H. A. R. Gibb this book would hardly have been completed. Lastly, I wish to thank my wife for her truly invaluable aid in preparing the manuscript; to her I dedicate this book as a δόσις όλίγη τε φίλη τε.

I cannot do better than address the reader in the words of Shafi'i (Risala, 59): 'I lost some of my books but have verified what I remembered from what is known to scholars; I have aimed at conciseness, so as not to make my work too long, and have given only what will be sufficient, without exhausting all that can be known on the subject.'

J. S.

OXFORD

PREFACE TO THE FOURTH IMPRESSION

I HAVE made only a few small changes and additions, incorporating some of my more recent conclusions, but have not attempted to add to the book substantially. It remains a work of research that does not aim at giving a comprehensive account of legal science in the first few centuries of Islam. For a general picture of the development of Muhammadan jurisprudence as a whole, from its beginnings to modern times, I may refer the reader to my Introduction to Islamic Law, second impression, Oxford, 1966.

January 1967

CONTENTS

PART I

THE DEVELOPMENT OF LEGAL THEORY

CHAPTER 1. THE CLASSICAL THEORY OF MUHAMMADAN LAW. THE FUNCTION OF TRADITIONS . . . . 1

CHAPTER 2. THE ANCIENT SCHOOLS OF LAW. SHĀFI‘Ī'S

ATTITUDE TO THEM . . . . 6

CHAPTER 3. SHĀFI‘Ī AND LEGAL TRADITIONS . . . 11

CHAPTER 4. TRADITIONS IN THE ANCIENT SCHOOLS OF LAW . 21

A. The Medinese . . . . . 22

B. The Iraqians . . . . . 27

C. The Syrians . . . . . 34

CHAPTER 5. TECHNICAL CRITICISM OF TRADITIONS BY SHĀFI‘Ī AND HIS PREDECESSORS . . . . . 36

CHAPTER 6. ARGUMENTS FOR AND AGAINST TRADITIONS . 40

A. Adversaries of traditions in general . . . 40

B. Arguments against traditions from the Prophet . . 44

C. Arguments in favour of traditions from the Prophet . . 53

D. Conclusions . . . . . 57

CHAPTER 7. SUNNA, ‘PRACTICE’, AND ‘LIVING TRADITION’ . 58

A. General . . . . . . 58

B. The Medinese . . . . . 61

C. The Syrians . . . . . 70

D. The Iraqians . . . . . 73

E. Shāfi‘Ī . . . . . . 77

F. Conclusions . . . . . 80

CONTENTS

CHAPTER 8. CONSENSUS AND DISAGREEMENT . . 82

A. The old idea of consensus . . . 82

B. The Medinese and consensus . . . 83

C. The Iraqians and consensus . . . 85

D. The Mu'tazila and consensus . . . 88

E. Shafi'i and consensus . . . 88

F. The later doctrine of consensus . . . 94

G. Disagreement . . . 95

CHAPTER 9. ANALOGY, SYSTEMATIC REASONING, AND PERSONAL OPINION . . . 98

A. The Umaiyad period . . . 100

B. The Iraqians . . . 103

C. The Medinese . . . 113

D. The Syrians . . . 119

E. Shafi'i . . . 120

F. The Mu'tazila . . . 128

G. The traditionists . . . 128

H. Traditions against human reasoning in law . . . 129

CHAPTER 10. FINAL REMARKS ON LEGAL THEORY . 133

PART II

THE GROWTH OF LEGAL TRADITIONS

CHAPTER 1. PRELIMINARY REMARKS . . . 138

CHAPTER 2. THE GROWTH OF LEGAL TRADITIONS IN THE LITERARY PERIOD. CONCLUSIONS ON THE PRE-LITERARY PERIOD . . . 140

CHAPTER 3. THE CONFLICT OF DOCTRINES AS REFLECTED IN THE GROWTH OF TRADITIONS . . . 152

CHAPTER 4. THE EVIDENCE OF ISNADS . . . 163

CHAPTER 5. THE ORIGIN OF LEGAL TRADITIONS IN THE FIRST HALF OF THE SECOND CENTURY A.H. . . . 176

CHAPTER 6. LEGAL MAXIMS IN TRADITIONS . . . 180

PART III

THE TRANSMISSION OF LEGAL DOCTRINE

CHAPTER 1. UMAIYAD PRACTICE AS THE STARTING-POINT OF MUHAMMADAN JURISPRUDENCE . . . 190

A. Preliminary remarks . . . 190

B. Umaiyad popular practice . . . 192

C. Umaiyad administrative practice . . . 198

D. The attitude of the ancient schools of law to Umaiyad practice . . . 213

CHAPTER 2. COMMON ANCIENT DOCTRINE AND CROSS-INFLUENCES . . . 214

A. The common ancient doctrine . . . 214

B. Early cross-references and cross-influences . . . 218

C. Later polemics and influences . . . 222

D. Conclusions . . . 222

CHAPTER 3. THE KORANIC ELEMENT IN EARLY MUHAMMADAN LAW . . . 224

CHAPTER 4. THE IRAQIANS

A. Shuraih . . . 228

B. Hasan Başrī . . . 229

C. Sha'bī . . . 230

D. Ibn Mas'ud and his Companions . . . 231

E. Ibrahim Nakha'ī . . . 233

F. Hammad . . . 237

G. The Iraqian opposition . . . 240

H. Sufyan Thauri . . . 242

CHAPTER 5. THE MEDINESE AND MECCANS

A. The 'seven lawyers of Medina' . . . 243

B. Zuhri . . . 246

C. Rabi'a . . . 247

xii CONTENTS

d. Yahyā b. Sa'īd . . . . . 248

e. The Medinese opposition . . . . . 248

f. The Meccans . . . . . 249

CHAPTER 6. THE TRADITIONISTS . . . . . 253

CHAPTER 7. THE MU'TAZILA . . . . . 258

CHAPTER 8. KHĀRIJI LAW . . . . . 260

CHAPTER 9. SHI'A LAW . . . . . 262

PART IV

THE DEVELOPMENT OF TECHNICAL LEGAL THOUGHT

CHAPTER 1. THE DEVELOPMENT OF LEGAL REASONING IN GENERAL . . . . . 269

CHAPTER 2. SYSTEMATIZING AND ISLAMICIZING . . . . . 283

CHAPTER 3. AUZA'I'S REASONING . . . . . 288

CHAPTER 4. THE REASONING OF INDIVIDUAL IRAQIANS . . . . . 290

A. Ibn Abī Lailā . . . . . 290

B. Abū Hanīfa . . . . . 294

C. Abū Yūsuf . . . . . 301

D. Shaibānī . . . . . 306

CHAPTER 5. MĀLIK'S REASONING . . . . . 311

CHAPTER 6. SHĀFI'Ī'S REASONING . . . . . 315

EPILOGUE . . . . . 329

APPENDIX I. CHRONOLOGY OF SHĀFI'Ī'S WRITINGS . . . . . 330

APPENDIX II. LIST OF PARAGRAPHS IN SHĀFI'Ī'S TREATISES . . . . . 331

BIBLIOGRAPHY AND LIST OF ABBREVIATIONS . . . . . 336

Arabic . . . . . 336

European . . . . . 339

INDEX OF LEGAL PROBLEMS . . . . . 341

GENERAL INDEX . . . . . 344

ADDENDA . . . . . 349

PART 1

THE DEVELOPMENT OF LEGAL THEORY

CHAPTER 1

THE classical theory of Muhammadan law, as developed by the Muhammadan jurisprudents, traces the whole of the legal system to four principles or sources: the Koran, the sunna of the Prophet, that is, his model behaviour, the consensus of the orthodox community, and the method of analogy.¹ The essentials of this theory were created by Shāfi'ī, and the first part of this book, which is concerned with the development of legal theory, centres in a study of Shāfi'ī's achievement.² Closely connected with and not second to his material contribution to Muhammadan jurisprudence, is the part Shāfi'ī played in the formation of technical legal thought: he carried it to a degree of competence and mastery which had not been achieved before and was hardly equalled and never surpassed after him. The fourth part of this book, therefore, is devoted to a study of technical legal thought in Shāfi'ī and his predecessors. The second part starts from the conclusions which can be drawn from Shāfi'ī's attitude to the second of the principles of law, the sunna of the Prophet as laid down in traditions, and aims at working out a method by which these legal traditions may be used for following the development of legal doctrine step by step through the still largely uncharted period before Shāfi'ī. The results so gained will enable us to realize that the starting-point of Muhammadan jurisprudence lies in the practice of the late Umaiyad period, and the third part of this book accordingly tries to trace the transmission of legal doctrine from its start down to the beginnings of the literary period.

Though Shāfi'ī laid down the essentials of the classical theory

2 THE CLASSICAL THEORY OF MUHAMMADAN LAW

of Muhammadan law, he did not say the last word with regard either to consensus or to analogy. Analogy was the last of the four principles to gain explicit recognition, and even after Shāfi‘ī’s time had to overcome much negative resistance and positive disapproval; the history of this process has been studied by Goldziher in one of his fundamental works which also contains an analysis of Shāfi‘ī’s contribution to legal theory.¹ As regards consensus, Snouck Hurgronje has made clear its all-important function as the ultimate mainstay of legal theory and of positive law in their final form:² the consensus guarantees the authenticity and correct interpretation of the Koran, the faithful transmission of the sunna of the Prophet, the legitimate use of analogy and its results; it covers, in short, every detail of the law, including the recognized differences of the several schools. Whatever is sanctioned by consensus is right and cannot be invalidated by reference to the other principles. Thus the classical doctrine, but we shall find that for Shāfi‘ī consensus played a much more modest part. It is easy to see that the element of retrospective guarantee embodied in the classical doctrine of consensus is hardly compatible with the free movement and violent conflict of opinions, such as we witness in the creative period of Muhammadan law to which Shāfi‘ī belongs.

We are therefore left, as far as Shāfi‘ī and his predecessors and contemporaries are concerned, with two recognized material sources, the Koran and the sunna. We may take the importance of the Koranic element in Muhammadan law for granted, though we shall have to qualify this for the earliest period;³ but for Shāfi‘ī the sunna takes a place comparable to that filled by the consensus in the later system. It is one of the main results of the first part of this book, that Shāfi‘ī was the first lawyer to define sunna as the model behaviour of the Prophet, in contrast with his predecessors for whom it was not necessarily connected with the Prophet, but represented the traditional, albeit ideal, usage of the community, forming their ‘living tradition’ on an equal footing with customary or generally agreed practice. For Shāfi‘ī, therefore, only the actions of the Prophet carry authority, and he admits on principle only traditions from the Prophet

THE FUNCTION OF TRADITIONS 3

himself, although he still shows traces of the earlier doctrine by admitting traditions from the Companions of the Prophet, and opinions of their Successors and even later authorities as subsidiary arguments.

His predecessors and contemporaries, on the other hand, while certainly already adducing traditions from the Prophet, use them on the same level as they use traditions from the Companions and Successors, interpret them in the light of their own 'living tradition' and allow them to be superseded by it. Two generations before Shafi'i reference to traditions from Companions and Successors was the rule, to traditions from the Prophet himself the exception, and it was left to Shafi'i to make the exception his principle. We shall have to conclude that, generally and broadly speaking, traditions from Companions and Successors are earlier than those from the Prophet.

In the preceding paragraphs I have referred repeatedly to traditions from the Prophet and others. They are not identical with the sunna but provide its documentation, whether we take sunna with Shafi'i and the later theory as the model behaviour of the Prophet, or in its older meaning as the traditional usage of the community which is to be verified by reference to ancient authorities. All alleged information from the Prophet and others is couched in the form of single statements generally short, each preceded by a chain of transmitters (isnād) which is intended to guarantee its authenticity.¹ To serve this purpose the isnad must be uninterrupted and must lead to an original eye- or ear-witness, and all transmitters must be absolutely trustworthy. The criticism of traditions as practised by Muhammadan scholars was almost invariably restricted to a purely formal criticism of isnads on these lines.

The traditions, mainly from the Prophet, that passed the more or less severe tests of this kind applied to them, were collected in the third century A.H. in a number of works, six of which were later invested with particular authority and form together the classical corpus of orthodox Muhammadan tradition. They are the works of Bukhari, Muslim, Abu Dawud,

4. THE CLASSICAL THEORY OF MUHAMMADAN LAW

Tirmidhi, Ibn Māja, and Nasā'ī. Other well-known collections of traditions, to which we shall have occasion to refer, are by Ibn Hanbal, Dārimī, Dāraquṭnī, and Baihaqī. This concentration of interest on traditions from the Prophet, and the almost complete neglect of traditions from Companions, not to mention Successors and later authorities, reflects the success of Shāfi'ī's systematic insistence that only traditions going back to the Prophet carry authority.

It is generally conceded that the criticism of traditions as practised by the Muhammadan scholars is inadequate and that, however many forgeries may have been eliminated by it, even the classical corpus contains a great many traditions which cannot possibly be authentic. All efforts to extract from this often self-contradictory mass an authentic core by 'historic intuition', as it has been called, have failed. Goldziher, in another of his fundamental works,¹ has not only voiced his 'sceptical reserve' with regard to the traditions contained even in the classical collections,² but shown positively that the great majority of traditions from the Prophet are documents not of the time to which they claim to belong, but of the successive stages of development of doctrines during the first centuries of Islam. This brilliant discovery became the corner-stone of all serious investigation of early Muhammadan law and jurisprudence,³ even if some later authors, while accepting Goldziher's method in principle, in their natural desire for positive results were inclined to minimize it in practice.

The importance of a critical study of legal traditions for our research into the origins of Muhammadan jurisprudence is therefore obvious. This book will be found to confirm Goldziher's results, and to go beyond them in the following respects: a great many traditions in the classical and other collections were put into circulation only after Shāfi'ī's time; the first considerable body of legal traditions from the Prophet originated towards the middle of the second century, in opposition to slightly earlier traditions from Companions and other autho-

THE FUNCTION OF TRADITIONS 5

rities, and to the 'living tradition' of the ancient schools of law; traditions from Companions and other authorities underwent the same process of growth, and are to be considered in the same light, as traditions from the Prophet; the study of isnāds often enables us to date traditions; the isnāds show a tendency to grow backwards and to claim higher and higher authority until they arrive at the Prophet; the evidence of legal traditions carries us back to about the year 100 A.H. only; at that time Islamic legal thought started from late Umaiyad administrative and popular practice, which is still reflected in a number of traditions.

CHAPTER 2

THE ANCIENT SCHOOLS OF LAW.

SHĀFI‘Ī’S ATTITUDE TO THEM

SHĀFI‘Ī is known as the founder of one of the four surviving orthodox schools of law. It was not his intention to found such a school, and Muzani, the author of the earliest handbook of the Shāfi‘ite school, declares at the beginning of his work:¹ ‘I made this book an extract from the doctrine of Shāfi‘ī and from the implications of his opinions, for the benefit of those who may desire it, although Shāfi‘ī forbade anyone to follow him or anyone else.’ Shāfi‘ī devotes a considerable part of his writings to discussions with and polemics against his opponents, but always with a view to making them acknowledge and follow the sunna of the Prophet, and he speaks repeatedly against the unquestioning acceptance of the opinion of men.²

The older schools of law to which Shāfi‘ī is opposed, know a certain degree of personal allegiance to a master and his doctrine.³ Amongst the Iraqians, we find Abū Yūsuf refer to Abū Ḥanīfa as ‘the prominent lawyer’, and Shaibānī to ‘the companions of Abū Ḥanīfa’; Shāfi‘ī refers to those ‘who follow the doctrine of Abū Ḥanīfa’, or to his ‘companions’, and calls him ‘their master’; but also Abū Yūsuf has followers of his own. The most outspoken passage is one in which an Iraqian opponent, presumably Shaibānī, acknowledges Shāfi‘ī’s doctrine as good, but Shāfi‘ī retorts that, as far as he knew, neither the opponent had adopted it nor another of his ilk who lorded it over them, presumably Abū Ḥanīfa.⁴

Some of the Medinese rely on Mālik for their knowledge of traditions, and consider Mālik’s Muwaṭṭa’ as their authoritative

THE ANCIENT SCHOOLS OF LAW 7

book 'which they prefer to all others and which they are accustomed to follow'; they are the 'followers' of Mālik and he is their 'master'; they regard his opinion as if it were the consensus, and there is no consensus for them besides Mālik in Medina. But they are only a fraction of the Medinese, just as the followers of Abū Hanīfa are only part of the Iraqians.

The real distinguishing feature between the ancient schools of law is neither the personal allegiance to a master nor, as we shall see later, any essential difference of doctrine, but simply their geographical distribution. Shāfi'ī is explicit about it: 'Every capital of the Muslims is a seat of learning whose people follow the opinion of one of their countrymen in most of his teachings.'¹ Shāfi'ī goes on to mention the local authorities of the people of Mecca, Basra, Kufa, Syria; elsewhere, he refers to the Iraqians and Medinese, the Basrians and Kufians, the scholars of each place where knowledge of traditions is to be found, the people of the different countries, and he gives detailed lists of these local authorities.

One of these lists shows the variety of doctrines within the great geographical divisions: 'In Mecca there were some who hardly differed from 'Atā', and others who preferred a different opinion to his; then came Zanjī b. Khālid and gave legal opinions, and some preferred his doctrine, whereas others inclined towards the doctrine of Sa'īd b. Sālim, and the adherents of both exaggerated. In Medina people preferred Sa'īd b. Musaiyib, then they abandoned some of his opinions, then in our own time Mālik came forward and many preferred him, whereas others attacked his opinions extravagantly. I saw Ibn Abīl-Zinād exaggerate his opposition to him, and Mughīra, Ibn Hāzim and Darāwardī follow some of his opinions, whereas others attacked them [for it]. In Kufa I saw people incline towards Ibn Abī Lailā and attack the doctrines of Abū Yūsuf, whereas others followed Abū Yūsuf and disagreed with Ibn Abī Lailā and with his divergences from Abū Yūsuf, and others again inclined towards the doctrine of Sufyān Thaurī and that of Hasan b. Sālih. I have also heard of other instances of this kind, similar to those which I have observed and described. Some Meccans even think of 'Atā' more highly than of the Successors, and some of their opponents place Ibrāhīm Nakha'ī.

8. THE ANCIENT SCHOOLS OF LAW

[of Kufa] at the top. Perhaps all these adherents of different masters exaggerate.'¹

Shafi'i insists on the fact that the reputation of all these authorities varies much, and that they hardly agree on a single point of law or a general principle. If Shafi'i denies here the existence of reasoned agreement even between the several prominent scholars in each centre, he does not, on the other hand, imply the existence of any clear-cut, fundamental differences in legal theory between the local schools; it was exactly their common reliance on 'living tradition' and their free exercise of personal opinion, in other words, their lack of strict rules such as were elaborated only by Shafi'i, that led to wide divergences in doctrine.

There was as yet no trace of the particular reputation of Medina as the 'true home of the sunna',² a reputation incompatible with Shafi'i's terse statement: 'We follow this [tradition from the Prophet], and so do all scholars in all countries except Medina, and so do the great authorities',³ and with his sustained polemics against the Medinese.

The three great geographical divisions that appear in the ancient texts are Iraq, Hijaz, and Syria. Within Iraq, there is a further division into Kufians and Basrians. Although occasional references to the Basrians are not lacking,⁴ little is known about their doctrine in detail,⁵ and our knowledge of the ancient Iraqians is mainly confined to the Kufians. In Hijaz there are also two centres, Medina and Mecca,⁶ and again our infor-

SHAFI'I'S ATTITUDE TO THEM 9

Information on Medina is incomparably more detailed. The Syrian school is mentioned rarely,¹ but we have some authentic documentation on its main representative Auzā'ī.

Egypt did not develop a school of law of its own, but fell under the influence of the other schools. There were followers of the Iraqian doctrine in Egypt, but most of the scholars there belonged to the Medinese school of which they formed a branch. Shafi'i refers to them in the writings of his later, Egyptian, period as 'Egyptians' or as 'some of the people of our country'.²

Shafi'i considers himself a member of the Medinese school, and references to the Medinese or Hijazis as 'our companions', and to Malik as 'our master' or 'our and your master' occur over the whole range of his writings, from his early to his late period. Also his Iraqian opponents regard him as one of the Medinese, or a follower of Malik, or one of the Hijazis in general. But Shafi'i does not identify himself with the particular adherents of Malik within the school of Medina, although he is eager to defend Malik against an undeserved attack. In other contexts, Shafi'i keeps his distance from the Medinese in general and denies responsibility for those of their opinions which he does not share.

No compromise was possible between Shafi'i and the Medinese, nor indeed any other ancient school of law, on their essential point of difference in legal theory, concerning the overriding authority of traditions from the Prophet, as opposed to the 'living tradition' of the school. When he comes to this subject Shafi'i attacks the Medinese with the strongest possible words. The whole of Tr. III is a sustained attack on the Medinese for their failure to follow the traditions from the Prophet which they relate themselves (and, failing that, their own traditions from Companions and Successors), and an effort to convert them to his own point of view. In this connexion Shafi'i even uses arguments which do less than justice to the Medinese.³