INQUIRY into the content, scope, and character of the traditions of Islam must necessarily begin after the death of Muhammad; for the raison d'être of this vast literature is to provide an authoritative standard of belief and conduct based upon the word and deed of Muhammad which shall be binding upon the whole of the Muhammadan world. It is notorious that though the Quran contains a certain number of laws, e.g. rules in regard to marriage, inheritances, and the care of orphans, it cannot be successfully invoked to settle questions arising in such diverse categories as systematic and moral theology, ritual, and civil and military law. The Jews found the Mosaic law with its wealth of detail insufficient by itself without the assistance of case law and tradition, and the Talmud arose to supply this need. Similarly, the Muhammadan community found itself at the death of Muhammad with a holy book and the living memory of a prophet; from these two sources the ecclesiastical and temporal polity of the Islamic world was for all time built up.

The word hadith is a noun formed from the verb hadatha 'to be new' (cf. the Hebrew hadash with the same meaning and the noun hodesh 'new moon'). Properly hadith means 'news' and then a tale or verbal communication of any kind. It may with propriety be used of an account of a tribal raid, of old sagas, of incidents in the life of the prophet, and even of the Quran itself. The great impetus given to religious thought and speculation by Muhammad and the Quran could not fail to influence the language of Muhammadan writers, and thus the word has acquired its narrowed technical connotation of an oral tradition which can be traced back to a Companion or to the prophet Muhammad. Arabic preserves clearly the consciousness of the special connotation given to the word hadith, for Bukhari records a saying of 'Abd Allah b. Mas'ud that 'the best hadith is the book of God'; ¹ and of the prophet in reply to Abu Huraira's question, 'Who will be the happiest on the day of resurrection thanks to your intercession?' 'I thought you would be the first to inquire of me about this hadith because I have noticed your eagerness in regard to the hadith.' ²

Hadith enshrines sunna or 'beaten track' — the custom and practice of the old Mohammadan community inasmuch as hadith were often invoked to prove that a certain act was performed by the prophet, and was therefore to be imitated by all pious believers, it follows that hadith and sunna are sometimes names for one and the same thing. But there is no necessary connexion between them, and we often find that tradition is in conflict with custom. The great merit

¹ Bab I'tisam, ed. Krehl, iv, p. 420.

² Bab Riqaq 51, Krehl, iv, p. 245.

of Malik b. Anas in the eyes of his contemporaries was that he was an authority both on custom law and on oral tradition. Perhaps the best example of the distinction is in the title of a book cited by the Fihrist, 'the book of the sunnas with confirmatory hadith'.¹

The conservatism of the East has long been proverbial, and the Arab may fairly claim a share in the building up of this reputation. The acceptance of monotheism, it is true, marked a break with the past; but the prophet was careful to depart as little as possible from the path of his forefathers. Indeed, it may be said that in the Medina suras he appears as the restorer of the ancient faith of Arabia — the religion of Ibrahim Abu Isma'il. ² The word sunna up to the time of Muhammad meant the practice of antiquity: after his time it acquired significance, and came to denote the practice of the prophet and his immediate successors. The same hatred of innovation finds expression in the utterances of the savants of Medina as in those of their heathen forefathers. Medina naturally became the 'home of the sunna', because there lived the men who had first to adapt their lives to the teaching of the prophet; thus, so far as corporate life was concerned, Medina was the authority on questions of orthodox custom.

The reverence in which the prophet was held by his contemporaries, and more especially by those who had

¹ p. 230, 3. The word hadith throughout this book will be used both as a singular and a collective noun. Hadiths is scarcely possible in English, and the constant employment of the Arabic broken plural ahadith is hardly to be desired.

² See the illuminating observations of Père H. Lammens in Une adaptation arabe du monothéisme biblique.

never seen him in the flesh, naturally led them to preserve and repeat his sayings on all subjects. The feverish desire to know what he had said - and done, which is well marked in the second generation, increased in intensity until it reached its height in the absurdities of the exercise known as Talabu-l-'Ilm. ¹ The foundation of the enormous mass of traditions which afterwards accumulated was laid by the Companions who were scattered throughout the Muhammadan world; but it would be rash to dogmatize as to how much of existing material can be safely ascribed to them. Our estimate of traditions circulated in their name cannot but be adversely affected by the frequent accusations of forgery levelled against many of the professional traditionists, by the many anachronisms they contain, and by the political and sectarian bias they display. ² When all these factors are allowed for, and account is taken of the inevitable mistakes that must occur when traditions are handed down through a long line of speakers, it is difficult to regard the hadith literature as a whole as an accurate and trustworthy record of the sayings and doings of Muhammad. But however sceptical we are with regard to the ultimate historical value of the traditions, it is hard to overrate their importance in the formation of the life of the Islamic races throughout the centuries. If we cannot accept them at their face value as a mirror of the events which preceded the

¹ infra, 36.

² 'The number of motives leading to the fabrication of traditions was so great that the historian is in constant danger of employing as veracious records what were deliberate fictions.'—Margoliouth, Mohammed and the Rise of Islam, London, 1905, p. vi.

consolidation of Islam into a system. Many of the political, dynastic, religious, and social differences which agitated Islam in the days of its imperial might are illustrated in traditions promulgated by the conflicting parties in the interest of their pretensions. In them we see how the rival forces of militarism and pacifism, asceticism, materialism, mysticism and literalism, free will and determinism, strove fiercely for the mastery.

While the prophet was alive he was the sole guide in all matters whether spiritual or secular. Hadith or tradition in the technical sense, may be said to have begun at his death, for the extraordinary influence or his personality on his companions and associates created from the beginning a demand that believers should be informed what the prophet had done and taught in various circumstances in order that the life of the community and the individual might be modelled on that of the revered leader. But of the resultant mass of tradition few can be confidently regarded as emanating from the authorities whose names they bear. The veneration of a later generation for the prophet is well illustrated in a hadith quoted by Muir: 'Is it possible, father of 'Abd Allah, that thou hast been with Muhammad?' was the question addressed by a pious Muslim to Hudhaifa, in the mosque of Al Kufa; 'Didst thou really see the prophet, and wert thou on terms of familiar intercourse with him?' 'Son of my uncle! it is indeed as thou sayest.' 'And how wert thou wont to behave towards the prophet?' 'Verily we used to labour hard to please him.' 'Well, by the Lord! 'exclaimed the ardent listener, 'if I had been but alive in his time, I would not have allowed

him to put his blessed foot upon the earth, but would have borne him on my shoulders wheresoever he listed.' ¹

During the reign of the first four caliphs the energies of the Arabs were mainly directed to the expansion of their empire. The amazing rapidity of their conquests left little time, even if the inclination were present to preach and teach the faith. A people which within a century had made itself master of the races and lands lying between the Atlantic and the Oxus could not be extensively preoccupied in religious matters. Nor must it be supposed that there was a fixed and established cultus and theory of an ordered religious life even in the prophet's own town. Possibly in Medina, where, under the personal influence of Muhammad, men devoted themselves to the things of religion, an ecclesiastical usage may have developed quite early in the first century; in the provinces where Arabs represented but a mere fraction of proselytized nations no such usage existed. With the army went Companions and Followers, who must have carried with them some traditional religious customs; but in the earliest days, Medina itself had no fixed system, indeed it was then hardly developing. The natural result was a wide divergence in practice between many of the provinces of the empire, which has continued down to the present day in the Muslim world. Echoes of this state of affairs can sometimes be heard in the hadith literature: Abu Daud constantly calls attention to the purely local character of some hadith (infarada ahl of such and such a place).

¹ Life of Muhammad, revised by T. H. Weir, Edinburgh, 1912, p. xxx. (In this edition the references to the original sources are omitted.)

The hadith literature as we now have it provides us with apostolic precept and example covering the whole duty of man; it is the basis of that developed system of law, theology, and custom which is Islam. Now inasmuch as the bulk of this literature is demonstrably the work of the two hundred and fifty years after the prophet's death, it is necessary, in endeavouring to determine the age and early authority of hadith, to examine the very considerable amount of evidence for the existence of hadith written down during the life of the prophet. This evidence has been collected by Sprenger¹, who also quotes what claims to be early evidence to the contrary. Of the series authorizing the writing of hadith we may cite one on the authority of that prolific father of tradition Abu Huraira, who says that one of the Helpers (Ansar) used to sit and listen with admiration to the utterances of the prophet of God, and, being unable to remember what he heard, lamented his weakness to the prophet. The latter replied, 'Call your right hand to your aid,' i.e. write them down. This hadith exists in many different forms associated with the names of Abu Salib and Anas b. Malik. Again, 'Abd Allah b. 'Umar says: 'We said, “O prophet of God, we hear from you hadith which we cannot remember. May we not write them down?” “By all means write them down,” said he.' This hadith exists in no less than thirty versions, which present small differences. Again, Abu Huraira asserts — not without reason! — that none of the Companions preserved more hadith than he, except 'Abd

¹ On The origin and progress of writing down historical facts among the Musulmans. (Journal of the Asiatic Society of Bengal, 1856, pp. 303—29.)

Allah b. 'Umar. 'But he wrote them down, and I did not write them.' This 'Abd Allah (d. 65) says, 'The book I wrote from the prophet of God is Al Sadiqa,' and Mujahid asserts that he saw this book in the possession of its compiler. Anas b. Mãlik states that Abü Bakr wrote down for him the laws regarding alms. Abundant proof could be adduced that books were read and written by the early Arabs; it will suffice to quote a saying attributed to Dhu'l Rumma (d. 117), in which he expresses his dislike of those who rely on their memories instead of writing down poetry: 'Write down my poetry, for the written word is more pleasing to me than memory... -- A book does not forget, nor does it substitute one word for another.'

Probably the hadith literature presents us with more contradictory statements on the question as to whether it was permissible to write down traditions of the prophet in the early days of Islam than on any other question. Many express prohibitions can be quoted. Abu Sa'id al Khudri asserts that he asked the prophet's permission to write down hadith, and it was refused. 'The prophet of God came out to us while we were writing hadith, and said, “What is this that you are writing?” We said, “Hadith which we hear from thee.” Said he, “A book other than the book of God! Do you not know that nothing but the writing of books beside the book of God led astray the peoples that were before you?' 'We said, “Are we to relate hadith of you, O prophet of God?” He replied, “Relate hadith of me: there is no objection. But he who intentionally speaks falsely on my authority will find a place in hell.”' In one version Abu Huraira adds that the

writings were heaped together and burned. Further Abu Nadhra relates: 'We said to Abu Sa'id: “Would that you would write down hadith for us, for we cannot remember them.” He answered: “We will not write them, nor will we collect them in books. The prophet of God related them to us orally and we remembered them, so you must do the same.” The comment of Ibn 'Aun (d. 151) on the situation is not without interest. He says: 'The men of the first century who disapproved of writing held that principle in order that the Muslims might not be kept by other books from the study of the Quran. The ancient scriptures have been forbidden because it is impossible to distinguish what is true in them from what is false and the genuine from the spurious: moreover the Quran renders them superfluous.'

As a matter of fact, the controversy as to whether it was lawful or not to write down traditions really belongs to the age when the critical collections of traditions were made. The hadith last quoted do not invalidate the statements that traditions were written down from the mouth of the prophet; the extra-ordinary importance attached to every utterance of his would naturally lead his followers who were able to write to record his words in order to repeat them to those who clamoured to know what he had said; and there 'is nothing at all in any demonstrably early writing to suggest that such a practice would be distasteful to Muhammad. But it cannot be proved that any single tradition or group of traditions now extant were copied from the memoranda of the Companions. The most that can be said is that the canonical collections may preserve some such traditions.

Written hadith were no doubt, objectionable to old-fashioned and orthodox traditionists, who preserved in their memories an enormous number of traditions and enjoyed no small reputation on that account. Objections, too, were raised by those who saw that in many points hadith were contradictory to the Quran. Those also who repeated traditions which were genealogically unsound and accounted unworthy of a place in the written, and soon to become, canonical collections could not but view the corpus of Bukhari and his imitators with acute displeasure.

The basis of hadith is essentially religious, and during the Umayyad period theologians were under a cloud; so that it was not until the second century was well advanced that hadith of a religious character won their way into literature. Of course, a considerable number of traditions which were subsequently incorporated in the canonical collections of hadith were not committed to writing for the first time by the collectors. A goodly number of works on jurisprudence were already in existence besides the well-known works of Abu Hanifa, Shaibani, Shafi'i, and Abu Yusuf.

The earliest date which Muhammadans give for the collection of hadith is contained in the following tradition, said to rest on the authority of Malik b. Anas (94-179). 'Umar b. 'Abdu-l-Aziz wrote to Abu Bakr b. Muhammad b. 'Amr with the order: 'See what hadith of the prophet of God are extant or ancient customs (sunna madiya) or hadith known to Amra, and write them down; for I stand in dread of the disappearance of knowledge and of the death of them that possess it.' This Abu Bakr b. Muhammad

was one of the Ansar whom 'Umar II appointed judge at Medina, and 'Amra was his aunt. Of the statement Sir William Muir writes, ¹ 'About a hundred years after Muhammad, the Caliph 'Umar II issued circular orders for the formal collection of all extant tradition. The task thus begun continued to be vigorously prosecuted; but we possess no authentic remains of any computation of an earlier date than tile middle or end of the second century of the Hijra.' It would seem that this writer accepts the statement at its face value; but the fact that no authentic remains of this alleged first-century compilation are extant, and that the indefatigable students and compilers of tradition in the third century make no mention of an effort to trace such early documents, suggest very strongly that the tradition is not based on fact. It is difficult, if not impossible, to suggest a cogent reason why such an early collection, if it existed, should never have been mentioned by later scholars whose life-work it was to recover the genuine hadith of the apostolic period. For this reason the hadith must be regarded as an invention designed to connect the pious caliph, whose zeal for the sunna was gratefully recognized by theologians, with the tradition literature of Islam. This seems the more likely, as another tradition connects Ibn Shihab Al Zuhri with 'Umar II in this work. Moreover, Malik's statement is only to be found in Al Shaibani's recension of the Muwatta. It is absent from the other versions.

Two other second-century writers have been cited as authors of compilations of hadith, namely, 'Abd al Malik b. Juraij and Sa'id b. Abi 'Aruba. Their works

¹ Op. cit., pp. 33 f.

are not extant; but from the description of them given by later writers there is little doubt but that they were books of jurisprudence (Fiqh), drawn up with a view to stabilizing the sunna. Therefore they only incidentally contained traditions; their primary purpose was to serve as handbooks for lawyers. The need for such works increased when the free development of the public religious life of the community was no longer hindered by the worldly régime of the Umayyads.

Of a similar nature, though of far greater importance, is the Muwatta of Malik b. Anas. This work, which has always been highly-prized by Muhammadans, is not a collection of traditions. The author's interest was in jurisprudence, and his aim to establish a system of law based on the sunna of Medina. Thus he appeals to legal precedents as often as to hadith, which it was only incidentally his purpose to record for the sake of their legal significance. His object was not, like that of the later collectors, to ascertain what traditions of the prophet were current throughout the Muhammadan world and to test their authenticity by a series of artificial canons; but he had the severely practical and limited aim of establishing a system of law according to the agreement or consent (ijma) of the people of Medina. Thus he appeals, in matters in which his paragraphs coincide roughly with the hadith literature, not to hadith carried back by a chain of guarantors to the prophet, but to the sunna, which, with the legal decisions of recognized authorities and the consensus of opinion of the Medinotes, in his view constituted a system of law binding on the whole community. His method was to collect under distinct heads the sunna of Medina in regard to legal and religious

Matters: where tradition failed he appealed to ijma. Thus he had none of the theoretical interest in hadith which characterizes the traditionists of the next century. When necessary he did not hesitate to express his own opinion (ray) on difficult points where the evidence seemed to be at all dubious or self-contradictory. The following extract from the Mudawwanat will best illustrate Malik's method: 'Malik was asked concerning certain persons who went raiding and disembarked in Cyprus, where they proceeded to buy sheep, honey, and butter, and paid for these articles with dinars and dirhems; Malik disapproved. He further said to us of his own initiative: “I strongly object to coins which contain the mention of God and His book being taken and given to one that is unclean. I disapprove most strongly of such a practice.” I asked him whether we might make purchases with dirhems and dinars of traders who disembarked on our coast, or of members of the tolerated cults. He replied that he disapproved. He was asked whether money might be changed by changers in Muslim markets who belonged to these cults. He replied that he disapproved.' ¹

That Malik was no collector of traditions in the later sense is clear from his independent handling of his material. He does not always take care to trace back his isnad² or chain of guarantors to the prophet, nor are all the links in the chain set out. Thus, although his Muwatta saw the light a century before the canonical collections, it contains many hadith which have no place in the later works, because they

¹ x, 102, as quoted by Margoliouth, Early Development of Muhammadanism, p. 119.

² Infra, p. 23.

are not supported by a list of names reaching in uninterrupted succession from Muhammad to Malik.

It is unfortunate for the study of Muhammadan origins that the extant versions of the text of the Muwatta differ so radically one from another. The explanation of these different versions is probably to be sought in the practice of Ijaza and Munawala, ¹ to which Malik frequently resorted when pressed by a number of pupils. ² The textus receptus of Malik is the version of his Spanish pupil Yahya b. Yahya al Masmudi; and it is this version which is commonly quoted as the Muwatta. But there are no less than fifteen other versions all differing from the Muwatta Yahya and from each other. Of these by far the most important is the work of Muhammad b. Al Hasan Ali Shaibani, who was the pupil of Malik and of Abu Hanifa. This is generally cited as the Muwatta Muhammad. It contains certain matter which is not to be found in the received text and has been worked over by Al Shaibani and brought into accord with the tenets of his master Abu Hanifa. In such cases he prefixes his own views and comments with the words 'Muhammad says'.³

The growing importance of tradition as an authoritative force in the establishment of the legal and ritual life of the community created a demand for hadith on every conceivable subject, a demand which, as will be seen in the following chapters, produced an unfailing

¹ See Additional Note to Chap. I.

² Cf. Sprenger in ZDMG, x, pp. 9 ff.

³ A translation of a short section of the Muwatta illustrating the differences between the recensions of Yahya and Shaibani will be found in MS, ii, pp. 224—6.

supply; and naturally the vast accessions to current tradition thus generated necessitated some sort of systematic arrangement of material. The earliest collections are at one with the later in this, that attention was focused not on the matn or subject matter of the tradition but on the isnad or chain of guarantors going back to a Companion of the prophet. The characteristic of the Musnad, the earliest type of collection, was that hadith, quite irrespective of their contents and subject matter, were arranged under the name of the Companion on whose authority they were supported (musnad). The person who could repeat a respectable number of such musnad traditions received the honorific Musnid, or Musnida in the case of a woman. The isnad must always be in direct speech, thus: 'A told me, saying that B said C had informed him, saying, D mentioned that he heard E relate, “I heard F ask the Apostle of God so and so.”' The name musnad, which properly belonged to the individual tradition, passed over to the whole collection. A large number of such collections was current in the third century, though comparatively little of the literature survives to-day. The most important would seem to have been the Musnad of Ahmad b. Hanbal. On account of its great bulk¹ this work was seldom to be found in its entirety even in antiquity. It contains about thirty thousand hadith grouped under the names of some seven hundred Companions. Though the author follows a plan of his own, dividing his work into books of traditions emanating from Muhammad's family, the Helpers, Women, and so on, he makes no

¹ The Cairene edition (1890) is in six volumes, containing in all 2,885 pages.

attempt to group his gigantic store of tradition with any regard to the encyclopaedic range of the several subjects dealt with.

Ahmad was an indefatigable Talibu-l-'Ilm, and to his journeys we owe the Musnadu-l-Shamiyyin and other geographical groups of traditions. His vast collection was edited and published by his son 'Abd Allah Abu 'Abdu-l-Rahman (d. 290). The Musnad preserves a great many traditions which are not to be found elsewhere. Like all other collectors, Ahmad practically always confined his criticism to the isnad; especially was he strict in his scrutiny of traditions from any one suspected of Qadarite leanings. Post eventum prophecies are to be found in the canonical collections; but not in the profusion and with the detail and exactitude of Ahmad's collection. Hadith dealing with conquests, the geographical advantages of certain cities, and royal personages are the clearest examples of this development.

The Musnad is marked by a fearless indifference to the susceptibilities of the 'Abbasids. Whereas the two great works of Bukhari and Muslim may be searched in vain for any generous recognition of the merits of the Umayyads, Ahmad, who forsooth had little to thank their successors for, preserves many of the numerous traditions extolling the glories of the Banu Umayya which must at one time have been current in Syria. A similar liberal attitude is adopted towards hadith which support the claims of the Shi'as. The great importance of this gigantic collection of tradition lies in its wealth of detail. Its value as a witness to events in the prophet's life, real and fictitious, is best illustrated in Margoliouth's Muhammad.

Later writers in the opposing schools often edited the collections of their leaders, arranging the authorities in alphabetical order, so that we read of the Musnadu-l-Shafi'i or the Musnadu-Malik. The reference is to the works mentioned on page 20 arranged as musnads. The word musnad is often misapplied by Muhammadan scholars, who speak of tradition works in general as musnads, e.g. of the Musnadu-l-Bukhari, where 'Jami' would be correct.

By the middle of the third century hadith had attained such importance as a means of determining the practice and beliefs of the community that a more practical collection and arrangement than the musnads became imperative. It was felt generally that hadith must be (a) brought into closer relation with jurisprudence, and (b) put on an unassailable footing. The controversy among the doctors of Islam which determined whether the community might develop its customs and re-interpret them according to the needs of each age, or whether it must rigidly conform to the practices and precedents of the apostolic generation, was probably responsible for the great impetus given to the collection and codifying of hadith. At this time traditions were written down with the definite aim of establishing an unerring authority for law and custom; thus, though the collectors devoted scrupulous attention to the isnad, so far as the arrangement of traditions was concerned the isnad was subordinate to the matn. Traditions were recorded according to their subject and the subject-matter was arranged under the headings of law books. Such collections were Musannafat. The object of the musannaf was to provide the lawyer with a handbook of tradition in

which he could readily look up the ipsissima verba of the prophet and thus silence an objector. The musnads were obviously unsuited to such a purpose: unless one knew the name of the original guarantor of a particular hadith one might have to read through the whole corpus of tradition to find it. While no ordered arrangement of oral and written tradition existed it was impossible to ask that young men who were being trained for the office of judge in the various provinces should be made to study hadith.

The task of the compilers was to demonstrate the practical value of hadith for the practical lawyer. The first and most important of the Musannafs is the Salih, 'the Genuine', of Abu 'Abd Allah Muhammad b. Isma'il Al Bukhari (194—256). This is a corpus of tradition pure and simple, compiled with the object of providing an orthodox criterion in all matters of jurisprudence. ¹ The author applied himself to the task by adopting as the framework of his book headings which covered the whole range of fiqh. His work is divided into ninety-seven books, which again are divided into 3,450 babs or chapters. The traditions themselves are preceded by a tarjama or rubric designed to lead the reader to a decision where authorities in the various Madhahib or schools differ. As Al Qastallani has said, 'Bukhari's fiqh is contained in the headings of his chapters.' The tarjama consists of a text from the Quran, or quite as often of a fragmentary hadith for which no isnad is

¹ As Chapter V will show: jurisprudence is by no means the only interest of hadith. Inasmuch as the Sahih is a Jami' it contains inter alia historical, biographical, and eschatological matter. It is precisely its many-sided character which lends charm to the study of the hadith literature.

forthcoming. It is generally supposed that the author regarded these last as genuine, but was unable to find a genuine chain of transmittors. There is great acumen behind the selection of these texts and fragmentary traditions with which most chapters are furnished, for they suggest a connexion in thought and a consequent interpretation of a hadith which is otherwise neutral or susceptible of a contrary meaning. Moreover, in cases where the Quran was invoked the author appeared to have the support of an unimpeachable authority. An interesting feature of these paragraphs is that they sometimes occur without any hadith following. It has been inferred from this that although Bukhari planned to cover the whole field of fiqh he sometimes failed to find confirmatory hadith and left the heading in the air, as it were, in the hope that the hiatus would some day be filled. This theory of the gaps is in keeping with the general purpose of this and other musannafat, and accords with the scrupulous honesty and exactitude of the author himself. M. Houdas, however, makes the interesting suggestion¹ that they are in reality a polemic against the Murjiites 'El-Bokhari . . . usa d'un moyen détourné pour atteindre plus sûrement l'hérésie qui menaçait l'existence même de l'islamisme; et, tandis que ses confrères se bornaient á classer et étiqueter leurs hadits, il songea á en faire une arme offensive contre les Mordjiites et contre tous ceux qui attentaient á la pureté primitive de la religion musulmane.' Thus he sees a peculiar significance and purpose in the first hadith recorded by Bukhari, 'Works are only to be judged by there inten-

¹ El Bokhari. Les Traditions islamiques . . . par O. Houdas et W. Marcais, ii, Paris, 1906. pp. x

tion, and in the chapter on Faith. In his view, the explanation given above does not suffice. It is to be regretted that M. Houdas has not worked out his interesting theory in detail. It is not difficult to find examples to which his explanation could hardly apply. It might be held, for instance, that the first hadith, which continues, 'As for him who migrates to obtain worldly possessions or to marry, his migration will be rewarded according to its object,' is rather a protest against the usurpation of the term Muhajir than an attack on the Murjiites. Nevertheless, as M. Houdas well says: 'Quoi qu'il en soit de cette question bien secondaire en somme, El Bokhari a, d'une part, rendu un signalé service à l'islamisme en conjurant le peril Mordjiite, et, d'autre part, il a, pour ainsi dire, fixé d'une manière définitive la constitution pratique de la religion du Prophète.'

Tradition reports that this remarkable man took cognizance of 600,000 hadith¹, and himself memorized more than 200,000. Of these he has preserved to us 7,397, or, according to other authorities, 7,295. If one adds to these the fragmentary traditions embodied in the tarjama the total is 9,082. On the other hand, the same tradition is often repeated more than once under different chapters (Abwab), so that if repetitions are disregarded the number of distinct hadith is reduced to 2,762, which are to be found in the 3,450 abwab into which his book is divided. ² When one reflects

¹ Ibn Khallikan, Wafayatu-l-A'yan, ed. Wüstenfeld, p. 580. These figures must be taken with a grain of salt. It is hardly likely that a man of Bukhari's ability would commit to memory 200,000 hadith in order to utilize less than a twentieth of them.

² I have taken these figures from Houdas, op. cit.

from these figures furnished by a Muslim historian that hardly more than one per cent of the hadith said to be openly circulating with the authority of the prophet behind them were accounted genuine by the pious Bukhari, one's confidence in the authenticity of the residue is sorely tried. Where such an enormous preponderance of material is judged false, nothing but the successful application of modern canons of evidence can restore faith in the credibility of the remainder. This is not, of course, to assert that the hadith literature is destitute of any historical foundation: such a conclusion would be unwarranted. But the undoubted historical facts do demand that each individual hadith should be judged on its merits.

So far as one is able to judge, Bukhari published the result of his researches into the content of what he believed to be genuine tradition with all the painstaking accuracy of a modern editor. Thus he records even trifling variants in the hadith, and wherever he feels that an explanatory gloss is necessary either in isnad or matn it is clearly marked as his own. When a variant has been given he sometimes adds his comment at the end, Qala Abu 'Abd Allah . . . ashbah, in my opinion the words so and so are more probably correct. In the Appendix will be found a translation of the Kitabu-l-Qadar from this most important of all hadith collections. ¹

Though the text of the Sahih does not present the unusual divergences in type to be found in the Muwatta it has survived in several recensions. It is only to be expected that a collection which, according to Ibn

¹ Reproduced, without critical and historical notes, from my article in JRAS, January 1924 by permission of the Society.

Khallikan, ¹ was read before ninety thousand hearers, should now be extant in several different forms. Of these the best known is that founded on a critical edition made by Muhammad Al Yunini (d. 658). This was printed at Bulaq in 1314, is carefully vocalized, and contains marginal notes of variant readings. The commentary of Ahmad b. Ahmad Al Qastallani (d. 923), Irshadu-l-Sari,² is of such value as to be wellnigh indispensable. The version of Abu Dharr, represented by Krehl's text, is also of considerable value: it is frequently at variance with the received text.

Nothing is more eloquent of the exalted position of hadith in the islamic community from the third century onwards, and of the prominent position of Al Bukhari among the Ashabu-l-hadith, than the extravagant homage which was paid to him and his work. A man who laboured sixteen years on the compilation of his corpus, who sought the aid of prayer before committing a tradition to writing, ³ and who interrogated over one thousand shaikhs living in places so distant as Balkh, Merv, Nisapur, the principal towns of Mesopotamia, the Hijaz, Egypt, and Syria, deserved well of his co-religionists. If Muslims since his death have

¹ Loc. cit.

² Bulaq, 1305, ten vols. For the literature see Brockelmann, Geschichte d. arabischen Literatur, i, pp. 158 ff.

³ Bukhari's work from first to last was an act of consummate piety. He was inspired to undertake the task, he says, by a dream in which he seemed to be driving away flies from the prophet's person. An interpreter of dreams told him that the flies were lies which had gathered round apostolic tradition. He never afterwards inserted a hadith in his collection until he had made an ablution and offered up a prayer of two rak'as.

canonized him, made pilgrimages to his tomb, and invoked his saintly aid in the difficulties of this life, they have but shown their devotion to the man who holds the position next their prophet. The latter, indeed, is reported to have been seen in a dream awaiting the arrival of Bukhari at the gates of Paradise. ¹

Another Musannaf on which Islam has also conferred the title Al Sahih is that of a younger contemporary of Bukhari, Muslim b. Al Hajjaj (d.261). Its contents are practically identical with Bukhari's collection except in the isnads, and the difference in treatment is really not very great. The principal difference is the absence of the paragraph headings characteristic of Bukhari. Muslim's work is arranged according to Fiqh, but he does not follow his plan so scrupulously: thus, while Bukhari often arranges the same tradition with a different isnad under different paragraphs when it is suitable to support more than one point of law and custom, Muslim places the parallel versions together. He does not plunge straightway into his task, but prefaces his book with a statement of the conditions a tradition must fulfill before it can be regarded as genuine and authentic. ² Every hadith which could serve as a support for fiqh must itself rest upon the authority of men whose trustworthiness was above suspicion (thiqat). Further, the authorities must stand in unbroken succession (ittisal), it must be demonstrable that they were contemporaries, and were actually in personal intercourse. Such a hadith must contain the words haddathani, sami'tu, or some other

¹ Houdas, II, xxiii.

² Some extracts from his introduction will be found in Chap. IV.

word implying personal intercourse. Another category of hadith which was not so highly esteemed was called mu'an'an; in these in place of a verb of hearing or telling it is only asserted that A narrates from ('an) B. Muslim was willing to accept such hadith if it could be established that A and B were contemporaries, but Bukhari demanded a further proof, that they should have been in personal contact: it was not sufficient that A should report—though in good faith—that B had related a certain hadith unless it could be proved that he himself had met him and therefore could be presumed to have heard it from his mouth, not through a third person. Muslim, however, excluded many traditions, not because he questioned their genuineness, but because they were not supported by ijma'. His assertion of this prepared the way for a more thorough Fami'.¹

The man Bukhari has always been immeasurably greater in the popular estimation than Muslim, and the tendency has been for the work of the former to take precedence of the latter. The one is prized for its range over the whole field of fiqh and the strictness of the shurut or rules for determining the trustworthiness of rawis, while the other is preferred for its more concise treatment of the material. Together they form an almost unassailable authority, subject indeed to criticism in details, yet deriving an indestructible influence from the ijma' or general consent of the community in custom and belief, which it is their function to authenticate.

¹ It has not been considered necessary to give a translation of any section of Muslim's Sahih, because the great majority of the hadith given in Ch. V are muttafaq, i. e. to be found in the Sahihun.

Besides these 'Two Genuine Books' there are others which Islam has elevated to canonical rank, the whole being known as 'The Six Books' (Al-kulubu-l-sitta). These are the

Sunan of Abu Daud (d. 275).

Fami' of Abu 'Isa Muhammad al Tirmidhi (d. 279).

Sunan of Abu 'Abd Allah Muhammad b. Maja (d. 283).

Sunan of Abu'Abd Al-Rahman al Nasa'i (d. 303). ¹

The same motive which inspired Bukhari moved the authors of the Sunan to their labours. Their aim was a narrower one, the compilation of a collection which would provide hadith dealing with all that was permissible and unpermissible to a Muslim. If so much genuine material had existed it is a priori inconceivable that it would have been passed over by Bukhari and Muslim; consequently much greater latitude had to be given to all who narrated hadith that were desiderated. Apostolic authority was obtained for all the enactments in jurisprudence, but at the cost of a principle. Hadith which were only fairly sound hasan were included. As the author of Masabihu-l-Sunna tells us, support for most of the ahkam comes from fair hadith.

Abu Daud, a contemporary of Bukhari, was a pupil of Ahmad b. Hanbal, and the master of Al Nasa'i. These two reversed the principle of the Sahihan that only hadith which rested on the authority of men universally esteemed trustworthy could be accepted, and they rejected only those which were universally

¹ These are loosely called 'the four sunnas', but inasmuch as Tirmidhi does not confine himself to matters of jurisprudence but deals with the whole field of hadith, his collection is properly a Jami'.

deemed unworthy of credence. They noted that some theologians were stricter than others in their scrutiny of the guarantors (rijal), and where a favourable verdict had been accorded by a lenient scholar they accepted the hadith despite the weight of adverse criticism. At the same time they did not attempt to exalt the mass of additional matter to the same degree of respect as that accorded to a hadith admittedly sahih. They expressed their opinion of the guarantors in no unmeasured terms. Abu Daud says that he wrote down half a million hadith, from which he selected 4,800; he calls these authentic, those which seem to be authentic, and those which are nearly so. ¹

The Sunan naturally fill up the gaps left by Bukhari in his scheme of fiqh, and supply hadith in support of the most pettifogging details in the ritual and legal life of the community, a pedantry which threatened to bring the whole system of hadith into ridicule. Al Nasa'i in his Sunan² takes notice of this ridicule.

Al Tirmidhi studied under Bukhari, Abmad b. Hanbal, and Abu Daud. He was the first to classify the various hadith under the three headings sahih, hasani; and hasan sahih.

The great value of these sunan is in their witness to the extent to which the rival schools of Islam had established their ritualistic and legal systems in the third century. Al Tirmidhi in particular with his wealth of inconsistent hadith, shows clearly how the

¹ Muir, op. cit., xlii.

² Al Nasri, Shahdra 1282, i, p. 6, qala-l-mushriknna inna nara sahibakum yu'alllmukum-l-khara'ata. In truth, what was really being said by men who objected to all the insignificant activities and even decencies of life being governed by apostolic tradition is here put into the mouth of the idolatrous contemporaries of Muhammad.

divergences in the orthodox schools of to-day were as clearly marked in his time.

Besides the 'six books' there were several other collections in circulation which failed to establish themselves in catholic consent. Of these the most important is that of Abu Muhammad 'Abd Allah al Darimi (d. 255). The chief characteristic of his work is its eclectic and subjective method. He often records his own opinion as to whether a hadith is binding on the community or not; and he does not hesitate to give decisions contradicting earlier authorities. It is a little difficult to say whether his book is with propriety to be called Sunan or Fami'. It is in the main a manual of hadith necessary in jurisprudence, but it does not confine itself to purely legal matters. The author is indifferent to the minute questions of law and ritual so fully dealt with in the Madhhab collections of Abu Daud and Al Nasa'i, so that even with the inclusion of matter of a general character his work is barely a third of the size of the other sunan. The tendency was for the pressure of the School (Madhhab) to increase, and a collection which failed to apply that pressure everywhere could not compete with sunan of a comprehensive character: thus Al Darimi's work never won its way to canonical rank.

It would unduly prolong this work to enumerate and discuss other collections which enjoyed a certain amount of popularity, e. g. Baqi b. Makhlad Al-Qurtubi (d. 276), Al 'Assal (d. 349), which are sometimes quoted by writers with the encyclopaedic knowledge of Al Suyuti and Ibn Khaldun. An account of them will be found in Brockelmann's Geschichte, and their hadith in some of the later compendia.

It will have been seen that the Sahih of Bukhari owed its existence to the arduous journeys undertaken by its author in search of hadith in circulation throughout the Muslim world. So great was the prestige of one who could narrate, as the last link in a chain, a tradition from the mouth of the prophet that those who were prevented by the responsibilities of life from undertaking these journeys in quest of 'ilm had to find some other means to secure their admission to this apostolic succession. The means lay ready to hand. Malik b. Anas had been known to give his pupils a written text with his authority to repeat its contents with the formula haddathana, as though the conveyance had been by word of mouth. This process was called Munawala, 'personal transmission' or 'handing over'. Ijaza, 'permission', was more lax. A teacher or rawi was asked to allow a person to promulgate a collection of hadith in his name. Malik is said to have allowed a pupil to do this without examining his text. The extraordinary differences between extant texts of the Muwatta are probably the sequel.

Bukhari evidently felt some hesitation about traditions by way of munawala: but in the following centuries, when the zeal of the 'Searchers after Knowledge' was at its height, ijaza could be given by letter or by proxy by an authority living in one end of the Muhammadan world to an applicant in the other. It would seem that when the content of tradition had been committed to writing, and partially established by ijma', the continued pursuit of 'knowledge' merely represented the activities of the credulous, who believed that there was still a residue of genuine apostolic tradition to be recovered, and who hoped to add their finds to the collections which had been accepted or were then winning their way to recognition in the second rank.

The Traditions of Islam [Table of Contents]
Answering Islam Home Page