The Nature of the Literary Evidence
A Dialogue On Methodology  With Apologies To David Hume
by Ibn Warraq (June 2007)
“… [ A]ll we know is what we have been told. With neither artifact nor archive, the student of Islamic origins could easily become victim of a literary and linguistic conspiracy“. John Wansbrough, Res Ipsa Loquitur: History and Mimesis, A. Einstein Memorial Lecture, Jerusalem, 1986.[Preamble: Renan once expressed the view that “in place of the mystery under which the other religions have covered their origins [Islam] was born in the full light of history; its roots are on the surface. The life of its founder is as well known to us as that of any XVI century reformer. We can follow year by year the fluctuations of his thought, his contradictions, his weaknesses”. However, as P.R.Davies put it, credulity does not become an historian. Our knowledge of Islam is based entirely on uncorroborated Arabic literary evidence redacted some two hundred years after the events they purport to describe. The present dialogue below tries to raise and discuss the problems these facts pose for any genuine historians.]
Ibn Cleanthes: “My dear friend, we are indeed lucky in having so many details of the life of the Prophet; we can practically follow his daily movements, listen to his homilies, speeches, read his letters, and have his opinion on countless subjects of the utmost importance to regulate our humble lives. The Prophet has set us an example that we would all do well to emulate, a source of comfort in this profane and irreligious age. What say you, Philo, come let us hear your sentiments.”
Ibn Philo: “Without wishing to vex you or to disturb your piety, I venture to say that you seem unacquainted with science and profound enquiry, particularly into the evidence for our knowledge of the Prophet’s life.”
Ibn Cleanthes: “Come, Philo, enough of your sceptical pleasanteries. You do not deny that there are over seventy historians that have written about the Rise of Islam, the life of Muhammad, and so on. We have more documenatry evidence about Muhammad than Jesus Christ, surely?”
Ibn Philo: “Yes, we do know the NAMES of some seventy historians for the period up to 1000 C.E., but the writings of the earliest historians have not survived. They are quoted by later historians, writing over a hundred years after the events they purport to describe. Thus there is scarcely a sira text that dates back to the first century of Islam.”
Ibn Cleanthes: “What of our beloved Sira by Ibn Ishaq, a copy of which I have here next to my bed? What of all the details that he gives us?”
Ibn Philo: “Ibn Ishaq wrote over a hundred years after the Hijra. As to the putative details, dear Cleanthes, we do not have many for the early years of the Prophet’s life, we know scarce little of his childhood, his early education and development, and we probably never shall. The details that he does give cannot be trusted, and we do not seem to have any objective way to verify them, no external source against which to compare or judge them. We are dealing with theocratic legend. Can we really call figures like Ibn Hisham, Ibn Ishaq, Abu Mikhnaf “ historians”? For, as you know full well, the Greek word for history, historein, simply means to enquire or investigate, which is what Herodotus set out to do, whereas the writers of the sira, give us the impression that they did not set out with the same purpose, they bore witness rather, an altogether different activity and aim. They are indulging in what Collingwood calls “theocratic history” which “means not history proper, that is scientific history, but a statement of known facts for the information of persons to whom they are not known, but who, as worshippers of the god in question, ought to know the deeds whereby he has made himself manifest.” Of course, Muslims do not worship Muhammad, and he is not a God, but mutatis mutandis, Collingwood’s remarks apply equally to Muslims “historians” and their hagiographic attitude to the Prophet and his supposed deeds. They did not distinguish the ‘duty of reporting from the legitimacy of believing’. 
Ibn Cleanthes: “Yes but Ibn Ishaq must have talked to many people who had known the Companions of the Prophet.”
Ibn Philo: “True but we cannot rely upon a purely oral tradition to scientifically reconstruct the events at the dawn of Islam. The chances are that the material being transmitted will have undergone a considerable amount of change, people’s memories may have failed them, their prejudices, even fears of being accused of impiety, will have affected, distorted or altered the contents of what was being transmitted. In any case, it is not at all certain that people living the events that are later seen to be of significance saw them as such, and highly improbable that anyone (assuming he was literate) thought of recording the details that he was living through. Tendency to exaggerate, or even to fabricate details will have been great. Furthermore, we do not even have Ibn Ishaq’s book.”
Ibn Cleanthes: “Ah, Philo you indulge in raillery and artificial malice, otherwise what can one say of your paradoxes. What, do you expect to be able to explain away the existence of this volume that I have perused nearly every day for the last twenty years? Although I do not hold the original book in my hand, it must, nonetheless, lie somewhere in some museum or library, with the paper slightly yellowed undoubtedly, but unmistakably bearing the handwriting of Ibn Ishaq himself. In short, a volume that has survived inviolate from the 8th century.”
Ibn Philo: “No I trifle not, Cleanthes, first, it seems unlikely that Ibn Ishaq had access to paper which was only just being manufactured in Baghdad in the late 8th century; the oldest paper with Arabic writing on it found in Egypt dates from between 796 – 815. Second the original work of Ibn Ishaq has been lost, we only possess the recension of Ibn Hisham, someone who died over two hundred years after the Hijra. Ibn Hisham got Ibn Ishaq’s work from the latter’s pupil al-Bakkai. So the work that you optimistically call Ibn Ishaq’s work is really Ibn Hisham’s version, much abbreviated, edited and even altered by Ibn Hisham, of Al Bakkai’s copy. Thus we depend entirely on Ibn Ishaq’s transmitters, of which there are at least fifteen; in other words, his biography of the Prophet, that you keep by your side, dear Cleanthes, has not been preserved as a single work. We do not possess an autograph manuscript of his work, we are entirely dependent for our knowledge of what he wrote on manuscripts at a certain number of removes from the original. And even the various versions of his text that we do possesss are full of discrepancies and contradictions, concerning both dates and the contents, reminding us that texts copied by hand are very quickly liable to corruption; it really is not so easy as you think to make an accurate copy. In any case, there seems to be no non-arbitrary way of deciding between the various versions.”
Ibn Cleanthes: “What of al -Tabari’s use of Ibn Ishaq, does he not frequently quote him? Can we not reconstruct Ibn Ishaq’s original from al -Tabari, and perhaps from some of Ibn Ishaq’s more reliable pupils?”
Ibn Philo: “Al Tabari was as selective as Ibn Hisham, and as Conrad has argued, “here too personal criteria of selection and al-Tabari’s explicitly stated intention of giving versions of events which enjoyed the general assent of the Community would lead us to expect from him not a verbatim transcription of the text, but a redaction which Muslims in the late third century A.H. would have found appropriate and useful”. In any way it is probable that al Tabari was often quoting from memory, in which case again it is unlikely that we would have verbatim transcriptions.
As for Ibn Ishaq’s pupils, I perhaps gave the impression that it was all only a matter of correcting the errors of the scribes and copyists. But as Conrad has observed, “Ibn Ishaq’s numerous students and their successors took what they received from the master and redacted and transmitted it in different ways. Witness, for example, the differences between Ibn Hisham, the quotations in al-Tabari, the recension of Yunus ibn Bukayr, and that of Muhammad ibn Salama al–Harrani. As different lines of transmission represent potentially different redactions, efforts to reconstruct the original form of a text cannot simply combine quotations from different lines of transmission, as if Ibn Ishaq’s students and their successors were making no revisions or changes of their own….Transmitters did not limit themselves to passing on what they had received from their teachers, but rather laid claim to the role of adapting and revising their materials as they saw fit, not just by the well-known means of the collective isnad, but also by rearranging, abbreviating, expanding, and recasting”.
Ibn Cleanthes: “But, dear Philo, you do tire me. I must own, I never know if you are in earnest or in jest. But arguments in the hands of a man of ingenuity like you do acquire an air of probability. Did we not find in recent years papyri bearing the works of Wahb ibn Munabbih who was born in 654, hence, by the time he was twenty, must have met many people who were eye-witnesses to the events described in the sira?”
Ibn Philo: “Yes indeed. But one must distiguish between when the actual physical papyrus that we possess was written and when the text that it contains was first composed. In this case the papyrus dates back to only the 9th century, though a part of the text seems to go back to Wahb. So once again we cannot be sure that we have an unadulterated text, too great a time gap exists between the composition of the text, and the copying of the papyrus that we possess. But, my dear Cleanthes, you show yourself to be far too uncritical, nay even credulous, when assuming that access to an eye-witness solves all our problems. Do not forget that a reliable witness should have been able to see, and understand what he saw, that he was not so engrossed that his preconceptions would have shaped, altered or distorted what he saw. Warriors in the midst of a battle see only a fraction of what is happening, and often through screens of sand and dust. Eyewitness accounts are as much filtered through personal experience and emotions as written ones, and are not infallible. In any case, many rogues will pretend for various reasons, to gain prestige for example, to have taken part in events when they had done no such thing.”
Ibn Cleanthes : “But dear friend you seem to wish to belittle our venerable ancestors. What of their prodigious memories? We know they were capable of the most astonishing feats of recollection. Come do not deny that the Arabs of the first years of Islam were men of remarkable gifts.”
Ibn Philo: “Yes, indeed they were, but I do not denigrate them in particular, I only point to some common failings of all humans. Modern clinical psychologists have shown how “autobiographical memories are not accurate historical accounts of events as they happened at the time of encoding, but rather a reconstruction based on a number of affective and motivational factors. Memories are contaminated with information from similar events and so change over the years as we encounter new experiences. What we remember about an event depends on when and for what purpose we are remembering, reflecting our beliefs about ourselves and the world at present. Thus memory is continually reprocessed and reinterpreted with changing contexts and perceptions.””.
Ibn Cleanthes: “But how on earth can the Companions of the Prophet have failed to have remembered the momentous events they were living through? Surely their united testimony must be of some authority. They do furnish us with so many vivid details it seems natural to assume the Companions were indeed overwhelmingly struck by their experiences.”
Ibn Philo: “As I have already remarked, Cleanthes, it is not certain they were aware at the time that they were living through momentous events; but what is more important here is that, as David B. Pillemer shows, “despite the vividness of our memory for such momentous events, our recall of the actual facts of the event and the circumstances we were in at the time are often incorrect. Neither the vividness of a memory nor the strength of our certainty in a memory can ensure its veracity… As Jerome Bruner once said, “a life is not ‘how it was’ but how it is interpreted and reinterpreted, told and retold””. 
Ibn Cleanthes: “Ah, Philo, you do rank me in that class of fools you call “the credulous”. So be it, but you do also shake me to the very foundations of my being. For you the Sira, the biography of the Prophet, is but a legendary history, for me however, as John Milton put it, “ Yet those old and inborn names of successive kings, never any to have been real persons, or done in their lives at least some part of what so long hath been remembered, cannot be thought without too strict an incredulity”.”
Ibn Philo: “That is indeed a feeble argument, Cleanthes, for all it amounts to is the claim that the tales in the Sira are credible just because they have been “so long remembered”, in other words, so often repeated. The rise of historical criticism should have destroyed faith in that sort of argument, but I see that it has not. You are a conservative critic, Cleanthes, you are swayed by not reason but by your passions, and as Housman put it, the faintest of all passions is the love of truth. And in this matter at least, if you will forgive me Cleanthes, you are indeed the average man who “believes that the text of ancient authors is generally sound, not because he has acquainted himself with the elements of the problem, but because he would feel uncomfortable if he did not believe it”.”
Ibn Cleanthes: “It is indeed true, I do feel decidedly uncomfortable. Coming back to something more substantial and solid. There are, you must avow, well-attested traditions, with an impeccable chain of transmitters that take us back to the very days, nay even into the very presence of the Prophet himself. We have the work of our intrepid hadith collectors who sifted through thousands of traditions to arrive at an authentic core. You cannot dismiss the work of Bukhari, Muslim, Tirmidhi, Maja, Dawud, Nisai, and Ibn Hanbal so casually.”
Ibn Philo: “But these traditions were collected many years after the death of Muhammad; Bukhari died over two hundred years after 632 C.E., while al Nisai died 280 years after. The traditions they gathered are tendentious in the extreme, a reflection of the society in which they were collected rather than any past historical reality. You will find that any change in social organization or practice is immediately accompanied by a corresponding proliferation of new traditions, which are often justifications for existing conditions or political situations. Though 7th century Arabia was not uniquely an oral society, from what we can make out from our meager sources, it seems that written accounts were rare. So we have a predominantly oral society, and all oral societies as Vansina reminds us, “tend to have a simpler notion of historical casuality, one which negates gradual change altogether. They tend to view institutions and techniques as unitary phenomena that came into existence fully fledged as they are in the present.”  But as Patricia Crone has argued, “new religions do not spring fully-fledged from the heads of prophets, old civilisations are not conjured away”.
In any case these traditions were transmitted by storytellers, orally, that is. But not only did they transmit them, and coloured them with their prejudices, but actually created them. As Crone says, “ …the sound historical tradition to which they are supposed to have added their fables simply did not exist”. Theirs was very much a performance tailored to the moods, needs, and expectations of their audiences. Their function was very similar to the role that poets played in preliterate Ancient Greece. As Finley reminds us, “The audiences …had no more genuine knowledge of the “once upon a time” world of the heroes than had the poets, but they too knew that it was in esssentials unlike their own. Heroes fought in chariots and feasted in great palaces, and so on, and the wildly inaccurate ways in which the poets described these things seemed right. How could they possibly know that they were in fact wrong? In a world wholly without writing, and therefore without records, disagreements could be resolved only by subjective judgements, by reference to the way it had been said before, by the superior auctoritas of one poet as against another, by anything other than reference to a document. Whatever dropped out of the poetic repertoire soon disappeared from ‘memory’ for ever; whatever innovation proved ‘successful’ soon became accepted as part of the tradition ever since once upon a time”.
Ibn Cleanthes: “Dear Philo, while I am full of admiration for your logical acumen, I am unable to approve of your doctrines, but do not know how to confound your arguments, they seem but the cavils of atheists, libertines and freethinkers. You seem not sensible how perturbing your conclusions are for a man of faith, you disturb a general harmony. You fail to place yourself in the mind of a believer. You start abstruse doubts, and objections, and ask me countless questions. I know not, I care not. That concerns not me. I have found a deity, and his apostle and here I stop my inquiry. Let those go farther, who are wiser or more original.
Ibn Philo: I pretend to be neither. But it is true that I do not concern myself with actual Muslim practice and faith, I am not studying sociology, however worthy that discipline is. I am studying history; I am not straining for originality but striving for truth; objective truth; trying to establish what really happened. “Am I therefore become your enemy because I tell you the truth”.  It does seem to me a mistake, however, to base one’s faith on putatively historical facts, which do not stand up to critical examination. For surely, if there were a deity, He would approve of the search for truth.With apologies to David Hume, Dialogues Concerning Natural Religion, Harmondsworth: Penguin Classics, 1990 (Ist Published 1779)  “ …in the full (clear) light of history” is now a famous phrase but is often used without any acknowledgement to its rightful coiner, e.g. R.S.Humphreys, Islamic History – A Framework For Inquiry, Princeton, 1991, p.69; and Andrew Rippin, Muslims, Their Religious Beliefs and Practices, Vol.1, London: Routledge, 1990, p.ix. M.I.Finley, Aspects of Antiquity, Harmondsworth: Penguin Books, 1978, p177. R.G.Collingwood, The Idea of History, Oxford, Ist Edition, 1946, Reprinted 1951, pp.14-15;
quoted by M.I.Finley, op.cit ., p.177Robin Lane Fox, The Unauthorized Version: Truth and Fiction in the Bible, Harmondsworth:Allen Lane, 1991 , p167; it is not clear who Mr Fox is quoting in this instance , but probably Collingwood, though I have been unable to locate the phrase in The Idea of History . Lawrence I.Conrad, Recovering Lost Texts: Some Methodological Issues, JAOS, Vol.113, No 2 April-June 1993, pp 258-263 Lawrence I.Conrad, Recovering Lost Texts: Some Methodological Issues, JAOS , Vol.113, No 2, April -June 1993, pp 258 – 263. Jan Vansina, Oral Tradition as History, London & Nairobi, 1985, p.4 . Janet Feigenbaum, How we tell it and how it was “, The Times Literary Supplement, Oct .30 1998, pp14-15  David B.Pillemer, Momentous Event. Vivid Memories, Harvard, 1998 Janet Feigenbaum, TLS , Oct., 30 , 1998 M.I.Finley, Aspects of Antiquity, Harmondsworth:Penguin Books, 1978, p.35. A.E.Housman, Selected Prose, Cambridge, 1961, p.43.  Jan Vansina, op.cit., pp130 -131. P.Crone, Slaves on Horses: the Evolution of the Islamic Polity, London, 1980 p12  P.Crone, Meccan Trade and the Rise of Islam, Cambridge, 1987. M.I.Finley, The World of Odysseus, Harmondsworth, 1979, p149.(My emphasis , I.W. ) Galatians IV .16