The stereotypic image of the Muslim holy warrior with a sword in one hand and the Koran in the other would only be plausible if he was left handed, since no devout Muslim should or would touch a Koran with his left hand which is reserved for dirty chores. All Muslims revere the Koran with a reverence that borders on bibliolatry and superstition. “It is,” as Guillaume remarked, “the holy of holies. It must never rest beneath other books, but always on top of them, one must never drink or smoke when it is being read aloud, and it must be listened to in silence. It is a talisman against disease and disaster.”
In some Westerners it engenders other emotions. For Gibbon it was an “incoherent rhapsody of fable,” for Carlyle an “insupportable stupidity,” while here is what the German scholar Salomon Reinach thought: “From the literary point of view, the Koran has little merit. Declamation, repetition, puerility, a lack of logic and coherence strike the unprepared reader at every turn. It is humiliating to the human intellect to think that this mediocre literature has been the subject of innumerable commentaries, and that millions of men are still wasting time absorbing it.”
For us in studying the Koran it is necessary to distinguish the historical from the theological attitude. Here we are only concerned with those truths that are yielded by a process of rational enquiry, by scientific examination. “Critical investigation of the text of the Qu’ran is a study which is still in its infancy,” wrote the Islamic scholar Arthur Jeffery in 1937. In 1977 John Wansbrough noted that “as a document susceptible of analysis by the instruments and techniques of Biblical criticism [the Koran] is virtually unknown.” By 1990, more than fifty years after Jeffery’s lament, we still have the scandalous situation described by Andrew Rippin: I have often encountered individuals who come to the study of Islam with a background in the historical study of the Hebrew Bible or early Christianity, and who express surprise at the lack of critical thought that appears in introductory textbooks on Islam. The notion that “Islam was born in the clear light of history” still seems to be assumed by a great many writers of such texts. While the need to reconcile varying historical traditions is generally recognized, usually this seems to pose no greater problem to the authors than having to determine “what makes sense” in a given situation. To students acquainted with approaches such as source criticism, oral formulaic compositions, literary analysis and structuralism, all quite commonly employed in the study of Judaism and Christianity, such naive historical study seems to suggest that Islam is being approached with less than academic candor. The questions any critical investigation of the Koran hopes to answer are:
1. How did the Koran come to us.?—That is the compilation and the transmission of the Koran.
2. When was it written, and who wrote it?
3. What are the sources of the Koran? Where were the stories, legends, and principles that abound in the Koran acquired?
4. What is the Koran? Since there never was a textus receptus ne varietur of the Koran, we need to decide its authenticity.
I shall begin with the traditional account that is more or less accepted by most Western scholars, and then move on to the views of a small but very formidable, influential, and growing group of scholars inspired by the work of John Wansbrough.
According to the traditional account the Koran was revealed to Muhammad, usually by an angel, gradually over a period of years until his death in 632 C.E. It is not clear how much of the Koran had been written down by the time of Muhammad’s death, but it seems probable that there was no single manuscript in which the Prophet himself had collected all the revelations. Nonetheless, there are traditions which describe how the Prophet dictated this or that portion of the Koran to his secretaries.
The Collection Under Abu Bakr
Henceforth the traditional account becomes more and more confused; in fact there is no one tradition but several incompatible ones. According to one tradition, during Abu Bakr’s brief caliphate (632-634), ‘Umar, who himself was to succeed to the caliphate in 634, became worried at the fact that so many Muslims who had known the Koran by heart were killed during the Battle of Yamama, in Central Arabia. There was a real danger that parts of the Koran would be irretrievably lost unless a collection of the Koran was made before more of those who knew this or that part of the Koran by heart were killed. Abu Bakr eventually gave his consent to such a project, and asked Zayd ibn Thabit, the former secretary of the Prophet, to undertake this daunting task. So Zayd proceeded to collect the Koran “from pieces of papyrus, flat stones, palm leaves, shoulder blades and ribs of animals, pieces of leather and wooden boards, as well as from the hearts of men.” Zayd then copied out what he had collected on sheets or leaves (Arabic, suhuf). Once complete, the Koran was handed over to Abu Bakr, and on his death passed to ‘Umar, and upon his death passed to ‘Umar’s daughter, Hafsa.
There are however different versions of this tradition; in some it is suggested that it was Abu Bakr who first had the idea to make the collection; in other versions the credit is given to Ali, the fourth caliph and the founder of the Shias; other versions still completely exclude Abu Bakr. Then, it is argued that such a difficult task could not have been accomplished in just two years. Again, it is unlikely that those who died in the Battle of Yamama, being new converts, knew any of the Koran by heart. But what is considered the most telling point against this tradition of the first collection of the Koran under Abu Bakr is that once the collection was made it was not treated as an official codex, but almost as the private property of Hafsa. In other words, we find that no authority is attributed to Abu Bakr’s Koran. It has been suggested that the entire story was invented to take the credit of having made the first official collection of the Koran away from ‘Uthman, the third caliph, who was greatly disliked. Others have suggested that it was invented “to take the collection of the Quran back as near as possible to Muhammad’s death.”
The Collection Under ‘Uthman
According to tradition, the next step was taken under ‘Uthman (644-656). One of ‘Uthman’s generals asked the caliph to make such a collection because serious disputes had broken out among his troops from different provinces in regard to the correct readings of the Koran. ‘Uthman chose Zayd ibn Thabit to prepare the official text. Zayd, with the help of three members of noble Meccan families, carefully revised the Koran comparing his version with the “leaves” in the possession of Hafsa, ‘Umar’s daughter; and as instructed, in case of difficulty as to the reading, Zayd followed the dialect of the Quraysh, the Prophet’s tribe. The copies of the new version, which must have been completed between 650 and ‘Uthman’s death in 656, were sent to Kufa, Basra, Damascus, and perhaps Mecca, and one was, of course, kept in Medina. All other versions were ordered to be destroyed.
This version of events is also open to criticism. The Arabic found in the Koran is not a dialect. In some versions the number of people working on the commission with Zayd varies, and in some are included the names of persons who were enemies of ‘Uthman, and the name of someone known to have died before these events! This phase two of the story does not mention Zayd’s part in the original collection of the Koran discussed in phase one.
Apart from Wansbrough and his disciples, whose work we shall look at in a moment, most modern scholars seem to accept that the establishment of the text of the Koran took place under ‘Uthman between 650 and 656, despite all the criticisms mentioned above. They accept more or less the traditional account of the ‘Uthmanic collection, it seems to me, without giving a single coherent reason for accepting this second tradition as opposed to the first tradition of the collection under Abu Bakr. There is a massive gap in their arguments, or rather they offer no arguments at all. For instance, Charles Adams after enumerating the difficulties with the ‘Uthmanic story, concludes with breathtaking abruptness and break in logic, “Despite the difficulties with the traditional accounts there can be no question of the importance of the codex prepared under ‘Uthman.” But nowhere has it yet been established that it was indeed under ‘Uthman that the Koran as we know it was prepared. It is simply assumed all along that it was under ‘Uthman that the Koran was established in its final form, and all we have to do is to explain away some of the difficulties. Indeed, we can apply the same arguments to dismiss the ‘Uthmanic story as were used to dismiss the Abu Bakr story. That is, we can argue that the ‘Uthmanic story was invented by the enemies of Abu Bakr and the friends of ‘Uthman; political polemics can equally be said to have played their part in the fabrication of this later story. It also leaves unanswered so many awkward questions. What were these “leaves” in the possession of Hafsa? And if the Abu Bakr version is pure forgery where did Hafsa get hold of them? Then what are those versions that seemed to be floating around in the provinces? When were these alternative texts compiled, and by whom? Can we really pick and choose, at our own will, from amongst the variants, from the contradictory traditions? There are no compelling reasons for accepting the ‘Uthmanic story and not the Abu Bakr one; after all they are all gleaned from the same sources, which are all exceedingly late, tendentious in the extreme, and all later fabrications, as we shall see later.
But I have even more fundamental problems in accepting any of these traditional accounts at their face value. When listening to these accounts, some very common-sensical objections arise which no one seems to have dared to ask. First, all these stories place an enormous burden on the memories of the early Muslims. Indeed, scholars are compelled to exaggerate the putatively prodigious memories of the Arabs. Muhammad could not read or write according to some traditions, and therefore everything depends on him having perfectly memorized what God revealed to him through His Angels. Some of the stories in the Koran are enormously long; for instance, the story of Joseph takes up a whole chapter of 111 verses. Are we really to believe that Muhammad remembered it exactly as it was revealed?
Similarly the Companions of the Prophet are said to have memorized many of his utterances. Could their memories never have failed? Oral traditions have a tendency to change over time, and they cannot be relied upon to construct a reliable, scientific history. Second, we seem to assume that the Companions of the Prophet heard and understood him perfectly.
Variant Versions, Verses Missing, Verses Added
Almost without exceptions Muslims consider that the Quran we now possess goes back in its text and in the number and order of the chapters to the work of the commission that ‘Uthman appointed. Muslim orthodoxy holds further that ‘Uthman’s Quran contains all of the revelation delivered to the community faithfully preserved without change or variation of any kind and that the acceptance of the ‘Uthmanic Quran was all but universal from the day of its distribution.The orthodox position is motivated by dogmatic factors; it cannot be supported by the historical evidence…. Charles Adams
While modern Muslims may be committed to an impossibly conservative position, Muslim scholars of the early years of Islam were far more flexible, realizing that parts of the Koran were lost, perverted, and that there were many thousand variants which made it impossible to talk of the Koran. For example, As-Suyuti (died 1505), one of the most famous and revered of the commentators of the Koran, quotes Ibn ‘Umar al Khattab as saying: “Let no one of you say that he has acquired the entire Quran, for how does he know that it is all? Much of the Quran has been lost, thus let him say, ‘I have acquired of it what is available’” (As-Suyuti, Itqan, part 3, page 72). A’isha, the favorite wife of the Prophet, says, also according to a tradition recounted by as-Suynti, “During the time of the Prophet, the chapter of the Parties used to be two hundred verses when read. When ‘Uthman edited the copies of the Quran, only the current (verses) were recorded” (73).
As-Suyuti also tells this story about Uba ibn Ka’b, one of the great companions of Muhammad:
This famous companion asked one of the Muslims, “How many verses in the chapter of the Parties?” He said, “Seventy-three verses.” He (Uba) told him, “It used to be almost equal to the chapter of the Cow (about 286 verses) and included the verse of the stoning”. The man asked, “What is the verse of the stoning?” He (Uba) said, “If an old man or woman committed adultery, stone them to death.”
As noted earlier, since there was no single document collecting all the revelations, after Muhammad’s death in 632 C.E., many of his followers tried to gather all the known revelations and write them down in codex form. Soon we had the codices of several scholars such as Ibn Masud, Uba ibn Ka’b, ‘Ali, Abu Bakr, al-Aswad, and others (Jeffery, chapter 6, has listed fifteen primary codices, and a large number of secondary ones). As Islam spread, we eventually had what became known as the metropolitan codices in the centers of Mecca, Medina, Damascus, Kufa, and Basra. As we saw earlier, ‘Uthman tried to bring order to this chaotic situation by canonizing the Medinan Codex, copies of which were sent to all the metropolitan centers, with orders to destroy all the other codices.
‘Uthman’s codex was supposed to standardize the consonantal text, yet we find that many of the variant traditions of this consonantal text survived well into the fourth Islamic century. The problem was aggravated by the fact that the consonantal text was unpointed, that is to say, the dots that distinguish, for example, a “b” from a “t” or a “th” were missing. Several other letters (f and q; j, h, and kh; s and d; r and z; s and sh; d and dh, t and z) were indistinguishable. In other words, the Koran was written in a scripta defectiva. As a result, a great many variant readings were possible according to the way the text was pointed (had the dots added).
Vowels presented an even worse problem. Originally, the Arabs had no signs for the short vowels: the Arab script is consonantal. Although the short vowels are sometimes omitted, they can be represented by orthographical signs placed above or below the letters—three signs in all, taking the form of a slightly slanting dash or a comma. After having settled the consonants, Muslims still had to decide what vowels to employ: using different vowels, of course, rendered different readings. The scripta plena, which allowed a fully voweled and pointed text, was not perfected until the late ninth century.
The problems posed by the scripta defectiva inevitably led to the growth of different centers with their own variant traditions of how the texts should be pointed or vowelized. Despite ‘Uthman’s order to destroy all texts other than his own, it is evident that the older codices survived. As Charles Adams says, “It must be emphasized that far from there being a single text passed down inviolate from the time of ‘Uthman’s commission, literally thousands of variant readings of particular verses were known in the first three (Muslim) centuries. These variants affected even the ‘Uthmanic codex, making it difficult to know what its true form may have been.”
Some Muslims preferred codices other than the ‘Uthmanic, for example, those of Ibn Mas’ud, Uba ibn Ka’b, and Abu Musa. Eventually, under the influence of the great Koranic scholar Ibn Mujahid (died 935), there was a definite canonization of one system of consonants and a limit placed on the variations of vowels used in the text that resulted in acceptance of seven systems. But other scholars accepted ten readings, and still others accepted fourteen readings. Even Ibn Mujahid’s seven provided fourteen possibilities since each of the seven was traced through two different transmitters, viz,
1. Nafi of Medina according to Warsh and Qalun
2. Ibn Kathir of Mecca according to al-Bazzi and Qunbul
3. Ibn Amir of Damascus according to Hisham and Ibn Dakwan
4. Abu Amr of Basra according to al-Duri and al-Susi
5. Asim of Kufa according to Hafs and Abu Bakr
6. Hamza of Kuga according to Khalaf and Khallad
7. Al-Kisai of Kufa according to al Duri and Abul Harith
In the end three systems prevailed, those of Warsh (d. 812) from Nafi of Medina, Hafs (d. 805) from Asim of Kufa, and al-Duri (d. 860) from Abu Amr of Basra. At present in modern Islam, two versions seem to be in use: that of Asim of Kufa through Hafs, which was given a kind of official seal of approval by being adopted in the Egyptian edition of the Koran in 1924; and that of Nafi through Warsh, which is used in parts of Africa other than Egypt.
As Charles Adams reminds us:
It is of some importance to call attention to a possible source of misunderstanding with regard to the variant readings of the Quran. The seven (versions) refer to actual written and oral text, to distinct versions of Quranic verses, whose differences, though they may not be great, are nonetheless substantial. Since the very existence of variant readings and versions of the Quran goes against the doctrinal position toward the Holy Book held by many modern Muslims, it is not uncommon in an apologetic context to hear the seven (versions) explained as modes of recitation; in fact the manner and technique of recitation are an entirely different matter. Guillaume also refers to the variants as “not always trifling in significance.” For example, the last two verses of sura LXXXV, Al Buraj, read: (21) hawa qur’anun majidun; (22) fi lawhin mahfuzun/in. The last syllable is in doubt. If it is in the genitive -in, it gives the meaning “It is a glorious Koran on a preserved tablet”—a reference to the Muslim doctrine of the Preserved Tablet. If it is the nominative ending -un, we get “It is a glorious Koran preserved on a tablet.” There are other passages with similar difficulties dealing with social legislation.
If we allow that there were omissions, then why not additions? The authenticity of many verses in the Koran has been called into question by Muslims themselves. Many Kharijites, who were followers of ‘Ali in the early history of Islam, found the sura recounting the story of Joseph offensive, an erotic tale that did not belong in the Koran. Hirschfeld questioned the authenticity of verses in which the name Muhammad occurs, there being something rather suspicious in such a name, meaning ‘Praised’, being borne by the Prophet. The name was certainly not very common. However the Prophet’s name does occur in documents that have been accepted as genuine, such as the Constitution of Medina.
Most scholars believe that there are interpolations in the Koran; these interpolations can be seen as interpretative glosses on certain rare words in need of explanation. More serious are the interpolations of a dogmatic or political character, which seem to have been added to justify the elevation of ‘Uthman as caliph to the detriment of ‘Ali. Then there are other verses that have been added in the interest of rhyme, or to join together two short passages that on their own lack any connection.
Bell and Watt carefully go through many of the amendments and revisions and point to the unevenness of the Koranic style as evidence for a great many alterations in the Koran:
There are indeed many roughness of this kind, and these, it is here claimed, are fundamental evidence for revision. Besides the points already noticed—hidden rhymes, and rhyme phrases not woven into the texture of the passage—there are the following abrupt changes of rhyme; repetition of the same rhyme word or rhyme phrase in adjoining verses; the intrusion of an extraneous subject into a passage otherwise homogeneous; a differing treatment of the same subject in neighbouring verses, often with repetition of words and phrases; breaks in grammatical construction which raise difficulties in exegesis; abrupt changes in length of verse; sudden changes of the dramatic situation, with changes of pronoun from singular to plural, from second to third person, and so on; the juxtaposition of apparently contrary statements; the juxtaposition of passages of different date, with intrusion of fare phrases into early verses.
In many cases a passage has alternative continuations which follow one another in the present text. The second of the alternatives is marked by a break in sense and by a break in grammatical construction, since the connection is not with what immediately precedes, but with what stands some distance back. The Christian al-Kindi (not to be confused with the Arab, Muslim philosopher) writing around 830 C.E., criticized the Koran in similar terms:
The result of all this (process by which the Quran came into being) is patent to you who have read the scriptures and see how, in your book, histories are jumbled together and intermingled; an evidence that many different hands have been at work therein, and caused discrepancies, adding or cutting out whatever they liked or disliked. Are such, now, the conditions of a revelation sent down from heaven?
Skepticism of the Sources
The traditional accounts of the life of Muhammad and the story of the origin and rise of Islam, including the compilation of the Koran, are based exclusively on Muslim sources, particularly the Muslim biographies of Muhammad, and the Hadith, that is the Muslim traditions.
The Prophet Muhammad died in 632 C.E. The earliest material on his life that we possess was written by Ibn Ishaq in 750 C.E., in other words, a hundred twenty years after Muhammad’s death. The question of authenticity becomes even more critical, because the original form of Ibn Ishaq’s work is lost and is only available in parts in a later recension by Ibn Hisham who died in 834 C.E., two hundred years after the death of the Prophet.
The Hadith are a collection of sayings and doings attributed to the Prophet and traced back to him through a series of putatively trustworthy witnesses (any particular chain of transmitters is called an isnad). These Hadith include the story of the compilation of the Koran, and the sayings of the companions of the Prophet. There are said to be six correct or authentic collections of traditions accepted by Sunni Muslims, namely, the compilations of Bukhari, Muslim, Ibn Maja, Abu Dawud, al-Tirmidhi, and al-Nisai. Again it is worth noting that all these sources are very late indeed. Bukhari died 238 years after the death of the Prophet, while al-Nisai died over 280 years after!
The historical and biographical tradition concerning Muhammad and the early years of Islam was submitted to a thorough examination at the end of the nineteenth century. Up to then careful scholars were well aware of the legendary and theological elements in these traditions, and that there were traditions which originated from party motive and which intended “to give an appearance of historical foundation to the particular interests of certain persons or families; but it was thought that after some sifting there yet remained enough to enable us to form a much clearer sketch of Muhammad’s life than that of any other of the founders of a universal religion.” This illusion was shattered by Wellhausen, Caetani, and Lammens who called “one after another of the data of Muslim tradition into question.”
Wellhausen divided the old historical traditions as found in the ninth- and tenth-century compilations in two: first, an authentic primitive tradition, definitively recorded in the late eighth century, and second a parallel version which was deliberately forged to rebut this. The second version was full of tendentious fiction, and was to be found in the work of historians such as Sayf b. ‘Umar (see above). Prince Caetani and Father Lammens cast doubt even on data hitherto accepted as “objective.” The biographers of Muhammad were too far removed from his time to have true data or notions; far from being objective the data rested on tendentious fiction; furthermore it was not their aim to know these things as they really happened, but to construct an ideal vision of the past, as it ought to have been. “Upon the bare canvas of verses of the Koran that need explanation, the traditionists have embroidered with great boldness scenes suitable to the desires or ideals of their paricular group; or to use a favorite metaphor of Lammens, they fill the empty spaces by a process of stereotyping which permits the critical observer to recognize the origin of each picture.”
As Lewis puts it, “Lammens went so far as to reject the entire biography as no more than a conjectural and tendentious exegesis of a few passages of biographical content in the Quran, devised and elaborated by later generations of believers.”
Even scholars who rejected the extreme skepticism of Caetani and Lammens were forced to recognize that “of Muhammad’s life before his appearance as the messenger of God, we know extremely little; compared to the legendary biography as treasured by the faithful, practically nothing.”
The ideas of the Positivist Caetani and the Jesuit Lammens were never forgotten, and indeed they were taken up by a group of Soviet Islamologists, and pushed to their extreme but logical conclusions. The ideas of the Soviet scholars were in turn taken up in the 1970s, by Cook, Crone, and other disciples of Wansbrough.
What Caetani and Lammens did for historical biography, Ignaz Goldziher did for the study of Hadith. Goldziher has had an enormous influence in the field of Islamic studies, and it is no exaggeration to say that he is, along with Hurgronje and Noldeke, one of the founding fathers of the modern study of Islam. Practically everything he wrote between roughly 1870 and 1920 is still studied assiduously in universities throughout the world. In his classic paper, “On the Development of Hadith,” Goldziher “demonstrated that a vast number of Hadith accepted even in the most rigorously critical Muslim collections were outright forgeries from the late 8th and 9th centuries—and as a consequence, that the meticulous isnads [chains of transmitters] which supported them were utterly fictitious.”
Faced with Goldziher’s impeccably documented arguments, historians began to panic and devise spurious ways of keeping skepticism at bay, such as, for instance, postulating ad hoc distinctions between legal and historical traditions. But as Humphreys says, in their formal structure, the Hadirh and historical traditions were very similar; furthermore many eighth- and ninth-century Muslim scholars had worked on both kinds of texts. “Altogether, if hadith isnads were suspect, so then should be the isnads attached to historical reports.”
As Goldziher puts it himself, “close acquaintance with the vast stock of hadiths induces sceptical caution,” and he considers by far the greater part of the Hadith “the result of the religious, historical and social development of Islam during the first two centuries.” The Hadith is useless as a basis for any scientific history, and can only serve as a “reflection of the tendencies” of the early Muslim community.
Here I need to interpose a historical digression, if we are to have a proper understanding of Goldziher’s arguments. After the death of the Prophet, four of his companions succeeded him as leaders of the Muslim community; the last of the four was ‘Ali, the Prophet’s cousin and son-in-law. ‘Ali was unable to impose his authority in Syria where the governor was Mu’awiya who adopted the war cry of “Vengeance for ‘Uthman” against ‘Ali (Mu’awiya and ‘Uthman were related and both belonged to the Meccan clan of Umayya). The forces of the two met in an indecisive battle at Siffin. After ‘Ali’s murder in 661, Mu’awiya became the first caliph of the dynasty we know as the Umayyad, which endured until 750 C.E. The Umayyads were deposed by the ‘Abbasids, who lasted in Iraq and Baghdad until the thirteenth century.
During the early years of the Umayyad dynasty, many Muslims were totally ignorant in regard to ritual and doctrine. The rulers themselves had little enthusiasm for religion, and generally despised the pious and the ascetic. The result was that there arose a group of pious men who shamelessly fabricated traditions for the good of the community, and traced them back to the authority of the Prophet. They opposed the godless Umayyads but dared not say so openly, so they invented further traditions dedicated to praising the Prophet’s family, hence indirectly giving their allegiance to the party of ‘Ali supporters. As Goldziher puts it, “The ruling power itself was not idle. If it wished an opinion to be generally recognized and the opposition of pious circles silenced; it too had to know how to discover a hadith to suit its purpose. They had to do what their opponents did: invent and have invented, hadiths in their turn. And that is in effect what they did.” Goldziher continues: Official influences on the invention, dissemination and suppression of traditions started early. An instruction given to his obedient governor al Mughira by Muawiya is in the spirit of the Umayyads: “Do not tire of abusing and insulting Ali and calling for God’s mercifulness for ‘Uthman, defaming the companions of Ali, removing them and omitting to listen to them (i.e., to what they tell and propagate as hadiths); praising in contrast, the clan of ‘Uthman, drawing them near to you and listening to them.” This is an official encouragement to foster the rise and spread of hadiths directed against Ali and to hold back and suppress hadiths favoring Ali. The Umayyads and their political followers had no scruples in promoting tendentious lies in a sacred religious form, and they were only concerned to find pious authorities who would be prepared to cover such falsifications with their undoubted authority. There was never any lack of these.
Hadiths were liable to be fabricated even for the most trivial ritualistic details. Tendentiousness included the suppression of existing utterances friendly to the rival party or dynasty. Under the ‘Abbasids, the fabrications of hadiths greatly multiplied, with the express purpose of proving the legitimacy of their own clan against the ‘Alids. For example, the Prophet was made to say that Abu Talib, father of ‘Ali, was sitting deep in hell: “Perhaps my intercession will be of use to him on the day of resurrection so that he may be transferred into a pool of fire which reaches only up to the ankles but which is still hot enough to burn the brain.” Naturally enough this was countered by the theologians of the ‘Alias by devising numerous traditions concerning the glorification of Abu Talib, all sayings of the prophet. “In fact,” as Goldziher shows, amongst the opposing factions, “the mischievous use of tendentious traditions was even more common than the official party.”
Eventually storytellers made a good living inventing entertaining Hadiths, which the credulous masses lapped up eagerly. To draw the crowds the storytellers shrank from nothing. “The handling down of hadiths sank to the level of a business very early. Journeys (in search of hadiths) favored the greed of those who succeeded in pretending to be a source of the hadith, and with increasing demand sprang up an even increasing desire to be paid in cash for the hadiths supplied.”
Of course many Muslims were aware that forgeries abounded. But even the so-called six authentic collections of hadiths compiled by Bukhari and others were not as rigorous as might have been hoped. The six had varying criteria for including a Hadith as genuine or not—some were rather liberal in their choice, others rather arbitrary. Then there was the problem of the authenticity of the texts of these compilers. For example, at one point there were a dozen different Bukhari texts; and apart from these variants, there were deliberate interpolations. As Goldziher warns us, “It would be wrong to think that the canonical authority of the two [collections of Bukhari and Muslim] is due to the undisputed correctness of their contents and is the result of scholarly investigations.” Even a tenth century critic pointed out the weaknesses of two hundred traditions incorporated in the works of Muslim and Bukhari.
Goldziher’s arguments were followed up, nearly sixty years later, by another great Islamicist, Joseph Schacht, whose works on Islamic law are considered classics in the field of Islamic studies. Schacht’s conclusions were even more radical and perturbing, and the full implications of these conclusions have not yet sunk in.
Humphreys sums up Schacht’s theses as: (1) that isnads [the chain of transmitters] going all the way back to the Prophet only began to be widely used around the time of the Abbasid Revolution—i.e., the mid-8th century; (2) that ironically, the more elaborate and formally correct an isnad appeared to be, the more likely it was to be spurious. In general, he concluded, “NO existing hadith could be reliably ascribed to the prophet, though some of them might ultimately be rooted in his teaching. And though [Schacht] devoted only a few pages to historical reports about the early Caliphate, he explicitly asserted that the same strictures should apply to them.” Schacht’s arguments were backed up by a formidable list of references, and they could not be dismissed easily. Here is how Schacht himself sums up his own thesis: It is generally conceded that the criticism of traditions as practiced 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”… have failed. Goldziher, in another of his fundamental works, has not only voiced his “sceptical reserve” with regard to the traditions contained even in the classical collections [i.e., the collections of Bukhari, Muslim, et al.], 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…
This book [i.e., Schacht’s own book] will be found to confirm Goldziher’s results, and go beyond them in the following respects: a great many traditions in the classical and other collections were put into circulation only after Shafi‘i’s time [Shafi‘i was the founder of the very important school of law which bears his name; he died in 820 C.E.]; the first considerable body of legal traditions from the Prophet originated towards the middle of the second [Muslim] century [i.e., eighth century C.E.], in opposition to slightly earlier traditions from the Companions and other authorities, 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 isnads 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 back to about the year 100 A.H. [718 C.E.]… Schacht proves that, for example, a tradition did not exist at a particular time by showing that it was not used as a legal argument in a discussion which would have made reference to it imperative, if it had existed. For Schacht every legal tradition from the Prophet must be taken as inauthentic and the fictitious expression of a legal doctrine formulated at a later date: “We shall not meet any legal tradition from the Prophet which can positively be considered authentic.”
Traditions were formulated polemically in order to rebut a contrary doctrine or practice; Schacht calls these traditions “counter traditions.” Doctrines, in this polemical atmosphere, were frequently projected back to higher authorities: “traditions from Successors [to the Prophet] become traditions from Companions [of the Prophet], and traditions from Companions become traditions from the Prophet.” Details from the life of the Prophet were invented to support legal doctrines.
Schacht then criticizes isnads which “were often put together very carelessly. Any typical representative of the group whose doctrine was to be projected back on to an ancient authority, could be chosen at random and put into the isnad. We find therefore a number of alternative names in otherwise identical isnads.”
Shacht “showed that the beginnings of Islamic law cannot be traced further back than to about a century after the Prophet’s death.” Islamic law did not directly derive from the Koran but developed out of popular and administrative practice under the Ummayads, and this “practice often diverged from the intentions and even the explicit wording of the Koran.” Norms derived from the Koran were introduced into Islamic law at a secondary stage.
A group of scholars was convinced of the essential soundness of Schacht’s analysis, and proceeded to work out in full detail the implications of Schacht’s arguments. The first of these scholars was John Wansbrough, who in two important though formidably difficult books, Quaranic Studies: Sources and Methods of Scriptural Interpretation (1977) and The Sectarian Milieu: Content and Composition of Islamic Salvation History (1978), showed that the Koran and Hadith grew out of sectarian controversies over a long period, perhaps as long as two centuries, and then was projected back onto an invented Arabian point of origin. He further argued that Islam emerged only w hen it came into contact with and under the influence of Rabbinic Judaism—”that Islamic doctrine generally, and even the figure of Muhammad, were molded on Rabbinic Jewish prototypes.” “Proceeding from these conclusions, The Sectarian Milieu analyses early Islamic historiography—or rather the interpretive myths underlying this historiography—as a late manifestation of Old Testament ‘salvation history.’”
Wansbrough shows that far from being fixed in the seventh century, the definitive text of the Koran had still not been achieved as late as the ninth century. An Arabian origin for Islam is highly unlikely: the Arabs gradually formulated their creed as they came into contact with Rabbinic Judaism outside the Hijaz (Central Arabia, containing the cities of Mecca and Medina). “Quranic allusion presupposes familiarity with the narrative material of Judaeo-Christian scripture, which was not so much reformulated as merely referred to…. Taken together, the quantity of reference, the mechanically repetitious employment of rhetorical convention, and the stridently polemical style, all suggest a strongly sectarian atmosphere in which a corpus of familiar scripture was being pressed into the service of as yet unfamiliar doctrine.” Elsewhere Wansbrough says, “[The] challenge to produce an identical or superior scripture (or portion thereof), expressed five times in the Quranic text can be explained only within a context of Jewish polemic.”
Earlier scholars such as Torrey, recognizing the genuine borrowings in the Koran from Rabbinic literature, had jumped to conclusions about the Jewish population in the Hijaz (i.e., Central Arabia). But as Wansbrough puts it, “References in Rabbinic literature to Arabia are of remarkably little worth for purposes of historical reconstruction, and especially for the Hijaz in the sixth and seventh centuries.
Much influenced by the Rabbinic accounts, the early Muslim community took Moses as an exemplum, and then a portrait of Muhammad emerged, but only gradually and in response to the needs of a religious community. This community was anxious to establish Muhammad’s credentials as a prophet on the Mosaic model; this evidently meant there had to be a Holy Scripture, which would be seen as testimony to his prophethood. Another gradual development was the emergence of the idea of the Arabian origins of Islam. To this end, there was elaborated the concept of a sacred language, Arabic. The Koran was said to be handed down by God in pure Arabic. It is significant that the ninth century also saw the first collections of the ancient poetry of the Arabs: “The manner in which this material was manipulated by its collectors to support almost any argument appears never to have been very successfully concealed.” Thus Muslim philologists were able to give, for instance, an early date to a poem ascribed to Nabigha Jadi, a pre-Islamic poet, in order to “provide a pre-Islamic proof text for a common Quranic construction.” The aim in appealing to the authority of pre-Islamic poetry was twofold: first to give ancient authority to their own Holy Scripture, to push back this sacred text into an earlier period, and thus give their text greater authenticity, a text which in reality had been fabricated in the later ninth century, along with all the supporting traditions. Second, it gave a specifically Arabian flavor, an Arabian setting to their religion, something distinct from Judaism and Christianity. Exegetical traditions were equally fictitious and had but one aim, to demonstrate the Hijazi origins of Islam. Wansbrough gives some negative evidence to show that the Koran had not achieved any definitive form before the ninth century: Schacht’s studies of the early development of legal doctrine within the community demonstrate that with very few exceptions, Muslim jurisprudence was not derived from the contents of the Quran. It may be added that those few exceptions are themselves hardly evidence for the existence of the canon, and further observed that even where doctrine was alleged to draw upon scripture, such is nor necessarily proof of the earlier existence of the scriptural source. Derivation of law from scripture… was a phenomenon of the ninth century….A similar kind of negative evidence is absence of any reference to the Quran in the Fiqh Akbar I…. The latter is a document, dated to the middle of the eighth century, which was a kind of statement of the Muslim creed in face of sects. Thus the Fiqh Akbar I represents the views of the orthodoxy on the then prominent dogmatic questions. It seems unthinkable had the Koran existed that no reference would have been made to it.
Wansbrough submits the Koran to a highly technical analysis with the aim of showing that it cannot have been deliberately edited by a few men, but “rather the product of an organic development from originally independent traditions during a long period of transmission.”
Wansbrough was to throw cold water on the idea that the Koran was the only hope for genuine historical information regarding the Prophet; an idea summed up by Jeffery, “The dominant note in this advanced criticism is ‘back to the Koran.’ As a basis for critical biography the Traditions are practically worthless; in the Koran alone can we be said to have firm ground under our feet.” But as Wansbrough was to show: “The role of the Quran in the delineation of an Arabian prophet was peripheral: evidence of a divine communication but not a report of its circumstances…. The very notion of biographical data in the Quran depends on exegetical principles derived from material external to the canon.”
A group of scholars influenced by Wansbrough took an even more radical approach; they rejected wholesale the entire Islamic version of early Islamic history. Michael Cook, Patricia Crone, and Martin Hinds writing between 1977 and 1987 regard the whole established version of Islamic history down at least to the time of Abd al-Malik (685-705) as a later fabrication, and reconstruct the Arab Conquests and the formation of the Caliphate as a movement of peninsular Arabs who had been inspired by Jewish messianism to try to reclaim the Promised Land. In this interpretation, Islam emerged as an autonomous religion and culture only within the process of a long struggle for identity among the disparate peoples yoked together by the Conquests: Jacobite Syrians, Nestorian Aramaeans in Iraq, Copts, Jews, and (finally) peninsular Arabs.
The traditional account of the life of Muhammad and the rise of Islam is no longer accepted by Cook, Crone, and Hinds. In the shore but pithy monograph on Muhammad in the Oxford Past Masters series, Cook gives his reasons for rejecting the biographical traditions:
False ascription was rife among the eighth-century scholars, and…in any case Ibn Ishaq and his contemporaries were drawing on oral tradition. Neither of these propositions is as arbitrary as it sounds. We have reason to believe that numerous traditions on questions of dogma and law were provided with spuriousus chains of authorities by those who put them into circulation; and at the same time we have much evidence of controversy in the eighth century as to whether it was permissible to reduce oral tradition to writing. The implications of this view for the reliability of our sources are clearly rather negative. If we cannot trust the chains of authorities, we can no longer claim to know that we have before us the separately transmitted accounts of independent witnesses; and if knowledge of the life of Muhammad was transmitted orally for a century before it was reduced to writing, then the chances are that the material will have undergone considerable alteration in the process.
Cook then looks at the non-Muslim sources: Greek, Syriac, and Armenian. Here a totally unexpected picture emerges. Though there is no doubt that someone called Muhammad existed, that he was a merchant, that something significant happened in 622, that Abraham was central to his teaching, there is no indication that Muhammad’s career unfolded in inner Arabia, there is no mention of Mecca, and the Koran makes no appearance until the last years of the seventh century. Further, it emerges from this evidence that the Muslims prayed in a direction much further north than Mecca, hence their sanctuary cannot have been in Mecca. “Equally, when the first Koranic quotations appear on coins and inscriptions towards the end of the seventh century, they show divergences from the canonical text. These are trivial from the point of view of content, but the fact that they appear in such formal contexts as these goes badly with the notion that the text had already been frozen.”
The earliest Greek source speaks of Muhammad being alive in 634, two years after his death according to Muslim tradition. Where the Muslim accounts talk of Muhammad’s break with the Jews, the Armenian version differs strikingly:
The Armenian chronicler of the 660s describes Muhammad as establishing a community which comprised both Ishmaelites (i.e., Arabs) and Jews, with Abrahamic descent as their common platform; these allies then set off to conquer Palestine. The oldest Greek source makes the sensational statement that the prophet who had appeared among the Saracens (i.e., Arabs) was proclaiming the coming of the (Jewish) messiah, and speaks of the Jews who mix with the Saracens, and of the danger to life and limb of falling into the hands of these Jews and Saracens. We cannot easily dismiss the evidence as the product of Christian prejudice, since it finds confirmation in the Hebrew apocalypse [an eighth-century document, in which is embedded an earlier apocalypse that seems to be contemporary with the conquests]. The break with the Jews is then placed by the Armenian chronicler immediately after the Arab conquest of Jerusalem.
Although Palestine does play some sort of role in Muslim traditions, it is already demoted in favor of Mecca in the second year of the Hegira, when Muhammad changed the direction of prayer for Muslims from Jerusalem to Mecca. Thereafter it is Mecca which holds center stage for his activities. But in the non-Muslim sources, it is Palestine which is the focus of his movement, and provides the religious motive for its conquest. The Armenian chronicler further gives a rationale for this attachment: Muhammad told the Arabs that, as descendants of Abraham through Ishmael, they too had a claim to the land which God had promised to Abraham and his seed. The religion of Abraham is in fact as central in the Armenian account of Muhammad’s preaching as it is in the Muslim sources; but it is given a quite different geographical twist.
If the external sources are in any significant degree right on such points, it would follow that tradition is seriously misleading on important aspects of the life of Muhammad, and that even the integrity of the Koran as his message is in some doubt. In view of what was said above about the nature of the Muslim sources, such a conclusion would seem to me legitimate; but is fair to add that it is not usually drawn.
Cook points out the similarity of certain Muslim beliefs and practices to those of the Samaritans (discussed below). He also points out that the fundamental idea developed by Muhammad of the religion of Abraham was already present in the Jewish apocryphal work called the Book of Jubilees (dated to c. 140-100 B.C;), and which may well have influenced the formation of Islamic ideas. We also have the evidence of Sozomenus, a Christian writer of the fifth century who “reconstructs a primitive Ishmaelite monotheism identical with that possessed by the Hebrews up to the time of Moses; and he goes on to argue from present conditions that Ishmael’s laws must have been corrupted by the passage of time and the influence of pagan neighbors.”
Sozomenus goes on to describe certain Arab tribes who, on learning of their Ishmaelite origins from Jews, adopted Jewish observances. Again there may have been some influence on the Muslim community from this source. Cook also points out the similarity of the story of Moses (exodus, etc.) and the Muslim hijra. In Jewish messianism, “the career of the messiah was seen as a re-enactment of that of Moses; a key event in the drama was an exodus, or flight, from oppression into the desert, whence the messiah was to lead a holy war to reconquer Palestine. Given the early evidence connecting Muhammad with Jews and Jewish messianism at the time when the conquest of Palestine was initiated, it is natural to see in Jewish apocalyptic thought a point of departure for his political ideas.”
Cook and Patricia Crone had developed these ideas in their intellectually exhilarating work Hagarism: The Making of the Islamic World (1977). Unfortunately, they adopted the rather difficult style of their “master” Wansbrough, which may well put off all but the most dedicated readers; as Humphreys says, “their argument is conveyed through a dizzying and unrelenting array of allusions, metaphors, and analogies.” The summary already given above of Cook’s conclusions in Muhammad will help non-specialists to have a better grasp of Cook and Crone’s (henceforth CC) arguments in Hagarism.
It would be appropriate to begin with an explanation of CC’s frequent use of the terms “Hagar,” “Hagarism,” and “Hagarene.” Since a part of their thesis is that Islam only emerged later than hitherto thought, after the first contacts with the older civilizations in Palestine, the Near East, and the Middle East, it would have been inappropriate to use the traditional terms “Muslim,” “Islamic,” and “Islam” for the early Arabs and their creed. It seems probable that the early Arab community, while it was developing its own religious identity, did not call itself “Muslim.” On the other hand, Greek and Syriac documents refer to this community as Magaritai, and Mahgre (or Mahgraye) respectively. The Mahgraye are the descendants of Abraham by Hagar, hence the term “Hagarism.” But there is another dimension to this term; for the corresponding Arabic term is muhajirun; the muhajirun are those who take part in a hijra, an exodus. “The ‘Mahgraye’ may thus be seen as Hagarene participants in a hijra to the Promised Land; in this pun lies the earliest identity of the faith which was in the fullness of time to become Islam.”
Relying on hitherto neglected non-Muslim sources, CC give a new account of the rise of Islam: an account, on their admission, unacceptable to any Muslim. The Muslim sources are too late, and unreliable, and there are no cogent external grounds for accepting the Islamic tradition. CC begin with a Greek text (dated ca. 634-636), in which the core of the Prophet’s message appears as Judaic messianism. There is evidence that the Jews themselves, far from being the enemies of Muslims, as traditionally recounted, welcomed and interpreted the Arab conquest in messianic terms. The evidence “of Judeo-Arab intimacy is complemented by indications of a marked hostility towards Christianity.” An Armenian chronicle written in the 660s also contradicts the traditional Muslim insistence that Mecca was the religious metropolis of the Arabs at the time of the conquest; in contrast, it points out the Palestinian orientation of the movement. The same chronicle helps us understand how the Prophet “provided a rationale for Arab involvement in the enactment of Judaic messianism. This rationale consists in a dual invocation of the Abrahamic descent of the Arabs as Ishmaelites: on the one hand to endow them with a birthright to the Holy Land, and on the other to provide them with a monotheist genealogy.” Similarly, we can see the Muslim hijra not as an exodus from Mecca to Medina (for no early source attests to the historicity of this event), but as an emigration of the Ishmaelites (Arabs) from Arabia to the Promised Land.
The Arabs soon quarreled with the Jews, and their attitude to Christians softened; the Christians posed less of a political threat. There still remained a need to develop a positive religious identity, which they proceeded to do by elaborating a full-scale religion of Abraham, incorporating many pagan practices but under a new Abrahamic aegis. But they still lacked the basic religious structures to be able to stand on their two feet, as an independent religious community. Here they were enormously influenced by the Samaritans.
The origins of the Samaritans are rather obscure. They are Israelites of central Palestine, generally considered the descendants of those who were planted in Samaria by the Assyrian kings, in about 722 B.C.E. The faith of the Samaritans was Jewish monotheism, but they had shaken off the influence of Judaism by developing their own religious identity, rather in the way the Arabs were to do later on. The Samaritan canon included only the Pentateuch, which was considered the sole source and standard for faith and conduct.
The formula “There is no God but the One” is an ever-recurring refrain in Samaritan liturgies. A constant theme in their literature is the unity of God and His absolute holiness and righteousness. We can immediately notice the similarity of the Muslim proclamation of faith: “There is no God but Allah.” And, of course, the unity of God is a fundamental principle in Islam. The Muslim formula “In the name of God” (bismillah) is found in Samaritan scripture as beshem. The opening chapter of the Koran is known as the Fatiha, opening or gate, often considered as a succinct confession of faith. A Samaritan prayer, which can also be considered a confession of faith, begins with the words: Amadti kamekha al fatah rahmeka, “I stand before Thee at the gate of Thy mercy.” Fatah is the Fatiha, opening or gate.
The sacred book of the Samaritans was the Pentateuch, which embodied the supreme revelation of the divine will, and was accordingly highly venerated. Muhammad also seems to know the Pentateuch and Psalms only, and shows no knowledge of the prophetic or historical writings.
The Samaritans held Moses in high regard, Moses being the prophet through whom the Law was revealed. For the Samaritans, Mt. Gerizim was the rightful center for the worship of Yahweh; and it was further associated with Adam, Seth, and Noah, and Abraham’s sacrifice of Isaac. The expectation of a coming Messiah was also an article of faith; the name given to their Messiah was the Restorer. Here we can also notice the similarity of the Muslim notion of the Mahdi.
We can tabulate the close parallels between the doctrines of the Samaritans and the Muslims in this way: