¹ Neill, Stephen, The Interpretation of the New Testament, 1861–1961 (London, 1964), pp. 338–40.Google Scholar

² Neill, Stephen and Wright, Tom, The Interpretation of the New Testament, 1861–1986 (New York, 1988), pp. 360–64.Google Scholar

³ I use this latter expression in the sense isolated by Martin Kähler's famous distinction, first made in 1892 (cf. Kähler, Martin, The So-Called Historical Jesus and the Historic Biblical Christ, trans. Braaten, Carl E. [Philadelphia, 1964])Google Scholar, between the “historical Jesus” and the “historic Christ,” the latter being the continuous subject of Christian preaching and the object of both Christian faith and Christian piety. Precisely the same distinction is intended when reference is made here to the “historical Muhammad.” While the Prophet's person is not the object of Muslims' faith, as Jesus' is for Christians, his prophethood is, and thus both the person and the role of “the historic Prophet,” to adapt Kähler's expression to the Islamic situation, have had an enormous and continuous influence on Islamic piety, practice, and beliefs (cf. Schimmel, Annemarie, And Muhammad Is His Messenger: The Veneration of the Prophet in Islamic Piety [Chapel Hill, N.C., 1985])Google Scholar, none of which is in question here.

⁴ Rodinson, Maxime, Mohammed, trans. Carter, Anne from the revised French edition of 1968 (London, 1971), p. xi.Google Scholar This was by way of preliminary to writing a 324-page biography of the Prophet!

⁵ What follows does not pretend to be exhaustive on either Muhammad or the Qurʾan, nor does it generally recover—though it occasionally glances at—the ground surveyed by Rudi Paret and Maxime Rodinson down to the early 1960s (Paret, Rudi, “Recent European Research on the Life and Work of the Prophet Muhammad,” Journal of the Pakistan Historical Society, 6 [1958], pp. 81–96Google Scholar; Rodinson, Maxime, “A Critical Survey of Modern Studies of Muhammad,” first published in Revue historique, 229 [1963], pp. 169–220Google Scholar; and translated from French in Swartz, Merlin, Studies on Islam [New York, 1981], pp. 23–85).Google Scholar The state of Qurʾanic studies through the 1970s is reflected in Welch, Alford T., “Kurʾān” in EI², vol. V (Leiden, E. J. Brill, 1981), pp. 400–432Google Scholar; and, more recently, in Neuwirth, Angelika, “Koran,” in Gätje, Helmut, ed., Grundriss der arabischen Philologie, vol. II; Literaturwissenschaft (Berlin, 1987), pp. 96–135.Google Scholar

⁶ See generally, on what might be called the “irenic approach” to Islam, Rippin, Andrew, “Literary Analysis of Qurʾan, Tafsir, and Sira: The Methodologies of John Wansbrough,” in Martin, Richard, ed., Approaches to Islam in Religious Studies (Tucson, 1985), p. 159.Google Scholar

⁷ Ernest Renan, writing in 1851, and cited by Rodinson, Maxime in “The Life of Muhammad and the Sociological Problem of the Beginnings of Islam,” Diogenes, 20 (1957), p. 46.Google Scholar

⁸ Neill, and Wright, , Interpretation, p. 363.Google Scholar

⁹ Abbott, Nabia, Studies in Arabic Literary Papyri, vol. I, Historical Texts (Chicago, 1957), p. 18Google Scholar; and compare Buhl, Frants, Das Leben Muhammeds, translated from the second Danish edition of 1953 by Schaeder, Hans Heinrich (rpt. Heidelberg, 1961), pp. 21ff.Google Scholar The pre-Islamic poetry makes its inevitable appearance in modern surveys on the “background sources” on Muhammad (see Rodinson, , “Critical Survey,” p. 37Google Scholar), but, except for Henri Lammens's work (see nn. 11, 56 below), it is far less in evidence when it comes to actually describing that background.

¹⁰ These are all likewise dutifully reported in surveys of the “sources for the life of Muhammad” (see Rodinson, , “Critical Survey,” pp. 29–39Google Scholar). It is in the north that we come the closest to the environment of Mecca, since both Jewish and Islamic traditions agree that there were Jewish settlements in the northern Hijaz; and, more important, the assertion is confirmed by epigraphical evidence (see Gil, Moshe, “The Origin of the Jews of Yathrib,” Jerusalem Studies in Arabic and Islam, 4 [1984], pp. 203–24).Google Scholar However, the fact remains that there is between the contemporary Greek, Roman, and Sasanian sources about Syria and Arabia and the later Islamic tradition about the same places a “total lack of continuity” (Crone, Patricia, Slaves on Horses: The Evolution of the Islamic Polity [Cambridge, 1980], p. 11).CrossRefGoogle Scholar

¹¹ Compare Lammens, Henri, La Mecque à la Veille de l'Hégire (Beirut, 1924)Google Scholar, where the Arab literary evidence is collected (and perhaps distorted), with Crone, Patricia, Meccan Trade and the Rise of Islam (Princeton, N.J., 1987), passimGoogle Scholar; and Peters, F. E., “The Commerce of Mecca before Islam,” in Kazemi, Farhad and McChesney, R. D., eds., A Way Prepared. Essays … Richard Bayly Winder (New York, 1988), pp. 3–26.Google Scholar A more sober approach than that of Lammens to the same pre-Islamic milieu has been taken over the last quarter-century by M. J. Kister of the Hebrew University (see Kister, M. J., Studies in Jahiliyya and Early Islam [London, 1980]Google Scholar, and n. 83 below). In the face of the complete dearth of Hijaz evidence, Yehuda Nevo and Judith Koren have recently attempted to extrapolate the pre-Islamic Meccan milieu from what appears to have been a collection of pagan shrines still flourishing in the mid-8th century at Sde Boker in the Negev (Nevo, Yehuda D. and Koren, Judith, “The Origins of the Muslim Descriptions of the Jahili Meccan Sanctuary,” Journal of Near Eastern Studies, 49 [1990], pp. 23–44).CrossRefGoogle Scholar The argument is seductive, but whether the buildings in question were indeed shrines does not appear to be at all clear.

¹² For Muhammad, see Buhl, , Das Leben, p. 366.Google Scholar While there is some material evidence for the Galilee and Jerusalem of Jesus' day, the latter conveniently summarized in Wilkinson, John, Jerusalem as Jesus Knew It: Archaeology as Evidence (London, 1978)Google Scholar, there has been no archaeological exploration in either Mecca or Medina, nor are the prospects good that there will be (Peters, F. E., Jerusalem and Mecca: The Typology of the Holy City in the Near East [New York, 1986], pp. 72–74).Google Scholar The almost total absence of archaeological evidence for early Islam is particularly striking when contrasted with the role that the excavation of sanctuaries and the discovery of legal and liturgical inscriptions have played in controlling the purely literary material that constitutes the “Hebrew Epic.”

¹³ In all that follows I have left aside the question of “revelation” and “inspiration” and taken as my starting point the historian's normal assumption that the religious documents in question, the New Testament and the Qurʾan, are entirely and uniquely the products of human agents, whoever those latter may turn out to be.

¹⁴ These latter reports are the hadith or Prophetic traditions allegedly reproducing the actual words of Muhammad on a variety of subjects. Their authenticity, which is of crucial importance to the historian, will be taken up in due course; here it need only be noted that while they do not share the cachet of divine inspiration attached by Christians to the entire New Testament, they have for Muslims a high degree of authority. Though that authority may have originated in their promotion, like that of the Mishna and Talmud, to magisterial authority in legal questions, the hadith soon began to enjoy the same status as purely historical documents.

¹⁵ If anything, the gap between the events of Jesus' life and their final redaction in the preserved Gospels appears to be growing narrower as time passes (see Robinson, John A. T., Redating the New Testament [Philadelphia, 1976]Google Scholar; and Neill, and Wright, , Interpretation, p. 361).Google Scholar

¹⁶ Conceivably even fewer, or perhaps many, many more. Though the later Muslim tradition came to agree that the “collection” of the Qurʾan took place in the caliphate of Uthman (644–656), some early Muslim authorities dated it to the Caliph Abu Bakr (632–634) and others to Umar (634–644). This early uncertainty about what would appear to be a critical event in Islamic history is by no means atypical, and two modern scholars have rejected the traditional “Uthmanic” consensus out of hand. One (Burton, John, The Collection of the Qurʾan [Cambridge, 1977]Google Scholar) would make the “collection of the Qurʾan” the work of the Prophet himself, while the other (Wansbrough, John, Qurʾanic Studies: Sources and Methods of Scriptural Interpretation [Cambridge, 1977]Google Scholar) would postpone it to the 9th century. It is still early in the career of each hypothesis, but neither seems to have been widely embraced.

¹⁷ Jeffery, Arthur, Materials for the History of the Text of the Qur'an (Leiden, 1937)Google Scholar; Paret, Rudi, “Der Koran als Geschichtsquelle,” Der Islam, 37 (1961), pp. 24–42CrossRefGoogle Scholar, cited from its reprint in Paret, Rudi, ed., Der Koran (Darmstadt, 1975), pp. 141–42.Google Scholar

¹⁸ Eliash, J., “The Shiʿite Qur'an: A Reconsideration of Goldziher's Interpretation,” Arabica, 16 (1969), pp. 15–24CrossRefGoogle Scholar; Kohlberg, E., “Some Notes on the Imamite Attitudes toward the Qur'an,” in Islamic Philosophy and the Classical Tradition: Essays … Richard Walter (Columbia, S.C., 1973), pp. 209–24.Google Scholar

¹⁹ As noted, one who has failed to be convinced is John Wansbrough who, in two major studies (Wansbrough, , Qurʾanic Studies, and Wansbrough, The Sectarian Milieu: Content and Composition in Islamic Salvation History [Oxford, 1978]Google Scholar) has attempted to demonstrate that (1) the Qurʾan was not finally fixed (“collected”) until the early 9th century, and (2) it was shaped out of biblical and other materials by redactors influenced by contemporary Judeo-Christian polemic. For a sympathetic appreciation of Wansbrough's work, see Rippin, “Literary Analysis”; and for a Muslim's criticism of both Wansbrough and Rippin, see Rahman, Fazlur, “Approaches to Islam in Religious Studies: Review Essay,” in Martin, Richard, ed., Approaches to Islam in Religious Studies (Tucson, 1985), pp. 198–202.Google Scholar

²⁰ Graham, William, “Qurʾan as Spoken Word: An Islamic Contribution to the Understanding of Scripture,” in Martin, Richard, ed., Approaches to Islam in Religious Studies (Tucson, 1985), p. 31Google Scholar: “Fundamentally, the Qurʾan was what its name proclaimed it to be: the recitation given by God for human beings to repeat (cf. Sura 96:1).”

²¹ This is believed according to the universal Muslim tradition (Watt, W. Montgomery, Bell's Introduction to the Qurʾan [Edinburgh, 1970], pp. 37–38).Google Scholar

²² Bultmann, Rudolf, History of the Synoptic Tradition (New York, 1963), p. 205.Google Scholar

²³ See Davies's, W. D. judicious remarks (Paul and Rabbinic Judaism [Philadelphia, 1980], p. 3)Google Scholar: “While it is clear that Rabbinic sources do preserve traditions of an earlier date than the second century … [i]t must never be overlooked that Judaism had made much history during that period. It follows that we cannot, without extreme caution, use the Rabbinic sources as evidence for first century Judaism.” Study of the life of Muhammad suffers, as we shall see (see n. 80), from the selfsame problem.

²⁴ Watt, , Bell's Introduction, pp. 77–79Google Scholar; compare Serjeant, R. B., “Early Arabic Prose,” in Beeston, A. F. L. et al., eds., Arabic Literature to the End of the Umayyad Period (Cambridge, 1983), pp. 126–27.Google Scholar

²⁵ Harvey, A. E., Jesus and the Constraints of History (Philadelphia, 1982), pp. 6ff.Google Scholar

²⁶ Rippin, , “Literary Analysis,” p. 159Google Scholar, commenting on Wansbrough's delineation of this style (Wansbrough, , Qurʾanic Studies, pp. 40–43, 47–48, 51–52ff.Google Scholar; Wansbrough, , Sectarian Milieu, pp. 24–25)Google Scholar: “The audience of the Qurʾan is presumed able to fill in the missing details of the narrative, much as is true of work such as the Talmud, where knowledge of the appropriate biblical citations is assumed or supplied by only a few words.” Far more than this is assumed by the Mishna and Talmud, of course. There, the reader is expected to understand the lines of both the issues and the current state of the debate on those issues when the text opens.

²⁷ Watt, , Bell's Introduction, pp. 77–82, 127–35.Google Scholar All these would fall within what the New Testament Form critics would call “Paränesis” or “Sayings and Parables” (cf. Stein, Robert H., The Synoptic Problem: An Introduction [Grand Rapids, Mich., 1987], pp. 168–72)Google Scholar, though with far greater variety than the Gospel examples show.

²⁸ Neill, and Wright, , Interpretation, p. 264.Google Scholar

²⁹ See, most recently, Boullata, Issa, “The Rhetorical Interpretation of the Qurʾan: Iʿjāi and Related Topics,” in Rippin, Andrew, ed., Approaches to the History of the Interpretation of the Qurʾān (Oxford, 1988), pp. 140–41.Google Scholar

³⁰ For Richard Bell's ingenious but unconvincing hypothesis, see Watt, , Bell's Introduction, pp. 101–7.Google Scholar

³¹ Namely, that the present suras were the original units of revelation, and that the hadith, and the historical works incorporating them, provide a valid basis for dating the suras (cf. Neuwirth, , “Koran,” p. 100Google Scholar). These premises, which roughly correspond to standard rabbinic theory about the books of the Bible, would, of course, rule out even the possibility of a “documentary hypothesis” for either the Bible or the Qurʾan.

³² The standard statement of what has become the Western position is found in the first volume of Nöldeke's, TheodorGeschichte des Qorans (Göttingen, 1860), revised by Schwally, Friedrich in 1909.Google Scholar Others have slightly revised the Nöldeke-Schwally sequence, but it remains the basic sura order used in the West (Neuwirth, , “Koran,” pp. 117–19).Google Scholar

³³ Watt, , Bell's Introduction, p. 114Google Scholar: “Like all those who have dated the Qurʾan, Bell accepted the general chronological framework [and much else besides] of Muhammad's life as this is found in the Sira … and other works.” The value judgment is that expressed in Welch, , “Kurʾān,” p. 417.Google Scholar

³⁴ This was somewhat disingenuously conceded by Watt, W. Montgomery (Muhammad at Mecca [Oxford, 1953], p. xiii)Google Scholar, and, more helpfully, by Sellheim, Rudolf (“Prophet, Calif und Geschichte: Die Muhammad Biographie des Ibn Ishaq,” Oriens, 18–19 [1965–1966], pp. 33–91)CrossRefGoogle Scholar; and Wansbrough, (Qurʾanic Studies and Sectarian Milieu), among others.Google Scholar

³⁵ The Life of Muhammad: A Translation of Ishaq's Sirat Rasul Allah, with Introduction and Notes by Alfred Guillaume (Oxford, 1955), p. 691.Google Scholar

³⁶ Cook, Michael, Muhammad (New York, 1983), p. 68.Google Scholar

³⁷ See Qurʾan 10:38–39, where, as usual, God is speaking: This Qurʾan is not such as could ever be invented in despite of God; but it is a confirmation of that which was before it and an exposition of that which is decreed for men—there is no doubt of that—from the Lord of the Worlds. Or do they say he [that is, Muhammad] has invented it? Then say: If so, do you bring a sūra like it, and call for help on all you can besides God, if you have any doubts.

For the Gospels, see John 21:24: “It is this same disciple who attests what has here been written. It is in fact he who wrote it, and we know that his testimony is true”; and cf. Luke 1:1–4.

³⁸ Summarily described, from a Muslim point of view, in Rauf, Muhammad Abdul, “Hbdot;adīth Literature—I: The Development of the Science of Ḥadīth,” in Beeston, A. F. L. et al., eds., Arabic Literature to the End of the Umayyad Period (Cambridge, 1983), pp. 271–88.CrossRefGoogle Scholar However, they may have included, even by their own criteria, far more chaff than has been suspected; compare Juynboll, G. H. A., “On the Origins of Arabic Prose: Reflections on Authenticity,” in Juynboll, G. H. A., ed., Studies on the First Century of Islamic History (Carbondale, Ill., 1982), pp. 171–72Google Scholar: “Classical Muslim isnād criticism has not been as foolproof as orthodox circles, and in their wake many scholars in the West, have always thought.”

³⁹ Consider, for example, what might be taken, were it genuine, as a prime example of early Islamic kerygma, Muhammad's own “farewell discourse” on the occasion of his last pilgrimage before his death. It is reported in substantially similar versions by three major historians, Ibn Ishaq, Waqidi, and Tabari, but, remarked R. B. Serjeant, a generally conservative critic, “patently signs of political ideas of a later age, coupled with internal and external contradictions, largely discredit the attribution of much of the extant versions to the Prophet” (Serjeant, , “Early Arabic Prose,” p. 123).Google Scholar For another example, see n. 62.

⁴⁰ This is not to say that, as Wright put it (Neill, and Wright, , Interpretation, p. 362)Google Scholar: It is still universally agreed that our picture of the earliest Church must begin with the study of Paul, and in particular of the letters generally agreed to be authentic.… These writings, which almost certainly antedate the earliest written Gospel, remain central for both the theology and history of the period.

Islam lacks a Paul, that is, an authoritative contemporary interpretation of the founder's message. The Islamic sources for early Islam are, like those on the life of Muhammad himself, later by a century and a half. Paul may have done theological mischief in the Christian context by providing an interpretation before the message, but all in all, it is better to have Paul than Tabari, as either a historian or an exegete.

⁴¹ Neill, and Wright, , Interpretation, p. 294.Google Scholar

⁴² Buhl, , Das Leben, p. 366.Google Scholar Michael Cook succinctly summed up the contemporary historical data provided by the Qurʾan: Taken on its own, the Qurʾan tells us very little about the events of Muhammad's career. It does not narrate these events, but merely refers to them; and in doing so, it has a tendency not to name names. Some do occur in contemporary contexts: four religious communities are named (Jews, Christians, Magians, and the mysterious Sabians), as are three Arabian deities (all female), three humans (of whom Muhammad is one), two ethnic groups (Quraysh and the Romans), and nine places. Of the places, four are mentioned in military connections (Badr, Mecca, Hunayn, Yathrib), and four are connected with the sanctuary (Safa, Marwa, Arafat, while the fourth is “Bakka,” said to be an alternative name to Mecca). The final place is Mount Sinai, which seems to be associated with the growing of olives. Leaving aside the ubiquitous Christians and Jews, none of these names occurs very often: Muhammad is named four or five times (once as “Ahmad”), the Sabians twice. Mount Sinai twice, and the rest once each.” (Cook, , Muhammad, pp. 69–70Google Scholar)

⁴³ Watt, W. Montgomery, Muhammad's Mecca: History in the Qurʾān (Edinburgh, 1988), especially pp. 26–38.Google ScholarWelch, Alford T. (“Muhammad's Understanding of Himself: The Koranic Data,” in Hovannisian, Richard G. and Vryonis, Speros, eds., Islam's Understanding of Itself [Malibu, Calif, 1983], pp. 15–52)Google Scholar has likewise attempted a biographical sketch of Muhammad's “self-understanding” as revealed by the Qurʾan.

⁴⁴ Welch, Alford T., “Allah and Other Supernatural Beings: The Emergence of the Qurʾanic Doctrine of Tawhid,” in Welch, Alford T., ed., Studies in Qurʾan and Tafsir, JAAR Thematic Issue 47, 1979, 1980), pp. 733–58.Google Scholar

⁴⁵ Welch, , “Muhammad's Understanding,” p. 16Google Scholar; and compare the significant omission of the personal pronoun in “A thorough analysis of the Qurʾanic contexts involving Allah, other deities, and the ‘lower’ members of the spirit world shows a clear and unmistakable development of ideas or teachings” (Welch, , “Allah,” p. 734).Google Scholar

⁴⁶ Ibid., pp. 751–53.

⁴⁷ On the genre, see Kister, M. J., “The Sīra Literature,” in Beeston, A. F. L. et al., eds., Arabic Literature to the End of the Umayyad Period (Cambridge, 1983), pp. 352–67.CrossRefGoogle Scholar

⁴⁸ The consensus opinion—and reservations—are rendered in Welch, “Kurʾān,” p. 414. Similar, and stronger, reservations are expressed by Wansbrough (Qurʾanic Studies, p. 141); Cook (Muhammad, p. 70); and Rippin (“Literary Analysis”), who wrote: Their [the “occasions of revelation” narratives] actual significance in individual cases of trying to interpret the Qurʾan is limited: the anecdotes are adduced, and thus recorded and transmitted, in order to provide a narrative situation in which the interpretation of the Qurʾan can be embodied. The material has been recorded within exegesis not for its historical value but for its exegetical value. Yet such basic literary facts about the material are frequently ignored within the study of Islam in the desire to find positive historical results, (p. 153)

⁴⁹ On Qurʾanic exegesis posing as biography, see Watt, W. Montgomery, “The Materials Used by Ibn Ishaq,” in Lewis, Bernard and Holt, P. M., ed., Historians of the Middle East (London, 1962), pp. 23–34Google Scholar; and on the “raids of the Prophet,” which Watt regards as the “essential foundation for the biography of the Prophet and the history of his times,” see ibid., pp. 27–28, and also Jones, J. M. B., “The Maghāzi Literature,” in Beeston, A. F. L. et al., eds., Arabic Literature to the End of the Umayyad Period (Cambridge, 1983), pp. 344–51.CrossRefGoogle Scholar

⁵⁰ Alternatively, as Patricia Crone dramatically stated it (Slaves on Horses, p. 203n. 10): “Consider the prospect of reconstructing the origins of Christianity on the basis of the writings of Clement or Justin in a recension by Origen.”

⁵¹ Goldziher, Ignaz, “On the Development of the Hadith,” in Stern, S. M., ed., Muslim Studies (London, 1971), vol. II, pp. 17–254Google Scholar; originally published in 1890.

⁵² Schacht, Joseph, The Origins of Muhammadan Jurisprudence (Oxford, 1950).Google Scholar

⁵³ Compare Stein's recent assessment of the materials attributed to Jesus in the Gospels: “The lack of such material [dealing with the most pressing problems facing the earliest Christian communities] in the Gospels witnesses against the idea that the church created large amounts of the gospel materials and in favor of the view that the church tended to transmit the Jesus traditions faithfully.” Moreover, citing G. B. Caird, “There is not a shred of evidence that the early church ever concocted sayings of Jesus in order to solve any of its problems” (Stein, , Synoptic Problem, p. 189Google Scholar).

⁵⁴ Thus argues Watt, W. Montgomery in The History of al-Tabari, vol. VI, Muhammad at Mecca, trans, and annotated Watt, W. Montgomery and McDonald, M. V. (Albany, N.Y., 1988), p. xviii.Google Scholar

⁵⁵ On these latter see the trenchant Form criticism analysis by Noth, Albrecht, Quellenkrilische Studien zu Themen, Formen und Tendenzen frühislamischer Geschichtsüberlieferung, vol. I, Themen und Formen (Bonn, 1973).Google Scholar

⁵⁶ On Lammens's, Henri approach, see “Qoran et tradition: Comment fut composée la vie de Mohamet?” Recherches de Science Religieuse, 1 (1910), pp. 25–61Google Scholar, and Fatima et les filles de Mahomet (Rome, 1912)Google Scholar; and compare Becker, C. H., “Grundsätzlichen zur Leben-Muhammadforschung,” in Islamstudien, 2 vols. (Leipzig, 1924; rpt. Hildesheim, 1967), vol. I, pp. 520–27Google Scholar, and Salibi, K. S., “Islam and Syria in the Writings of Henri Lammens,” in Lewis, Bernard and Holt, P. M., Historians of the Middle East (London, 1962), pp. 330–42Google Scholar; for Rodinson's characterization, see Rodinson, , “Critical Survey,” p. 26Google Scholar, and compare Buhl, , Das Leben, p. 367Google Scholar: “H. Lammens … dessen Belesenheit und Scharfsinn man bewundern muss, der aber doch oft die Objectivität des unparteischen Historikers vermissen lässt.”

⁵⁷ Watt, , Muhammad, p. xiiiGoogle Scholar, and compare Watt, , “Materials,” p. 24Google Scholar. Kister's cautiously worded opinion seems similar: The development of Sirah literature is closely linked with the transmission of the Ḥadith and should be viewed in connection with it.… Although some accounts about the recording of the utterances, deeds and orders dictated by the Prophet to his companions are dubious and debatable and should be examined with caution (and ultimately rejected), some of them seem to deserve trust.” (Kister, , “Sira Literature,” p. 352Google Scholar)

⁵⁸ Compare Rodinson, , “Critical Survey,” p. 42Google Scholar: “Orientalists are tempted to do as the Orientals have tended to do without any great sense of shame, that is, to accept as authentic those traditions that suit their own interpretation of an event and to reject others.” Rodinson, who, as we shall see shortly, had even less faith than Watt in the source material, may have himself done precisely that in his own biography of the Prophet.

⁵⁹ Crone, , Slaves on Horses, pp. 14–15Google Scholar:

Among historians the response to Schacht has varied from defensiveness to deafness, and there is no denying that the implications of his theories are, like those of Noth, both negative and hard to contest.… That the bulk of the Sīra … consists of second century ḥadīths has not been disputed by any historian, and this point may be taken as conceded. But if the surface of the tradition consists of debris from the controversies of the late Umayyad and early Abbasid period, the presumption must be that the layer underneath consists of similar debris from the controversies that preceded them, as Lammens and Becker inferred from Goldziher's theories.

According to Crone, Watt “disposes of Schacht by casuistry,” but Shaban, Paret, Guillaume, and Sellheim have likewise been unwilling to deal squarely with the critical issue he has raised (ibid., p. 211, n. 88). Watt's, brief rebuttal is in his “The Reliability of Ibn Ishaq's Sources” in La vie du prophète Mahomet (Colloque de Strasbourg 1980) (Paris, 1983), pp. 31–43Google Scholar; and Watt, and McDonald, , Muhammad, pp. xvii–xix.Google Scholar

⁶⁰ Becker, , “Grundsätzlichen,” p. 521Google Scholar, cited in Watt, , “Materials,” p. 23.Google Scholar

⁶¹ Cited in n. 4 above; compare his similar remarks in n. 58 above and, earlier, Buhl, , Das Leben, pp. 372–77.Google Scholar

⁶² The earliest example of such a summary, in both the serial and the absolute chronology, appears in Ibn Ishaq's Life (1:336) on the occasion of some Muslims emigrating to Abyssinia in 615, when the ruler there was given a summary presentation of Islamic “good news.” This apparently early Muslim “kerygma” has been analyzed in Wansbrough, , Qurʾanic Studies, pp. 38–43Google Scholar, and Wansbrough, , Sectarian Milieu, pp. 100–101.Google Scholar That author concludes (Qurʾanic Studies, p. 41) that “the structure of the report suggests a careful rhetorical formulation of Qurʾanic material generally supposed to have been revealed after the date of that event,” and, even more sweepingly (Sectarian Milieu, p. 100), “Save for the Meccan pilgrimage, no item in these lists falls outside the standard monotheist vocabulary, and is thus of little use in the description of origins.”

⁶³ From Mark onward—“Here begins the Gospel of Jesus Christ the Son of God”—all the Gospels make a similar declaration at their outset.

⁶⁴ In Ibn Ishaq's original “world history” version, before Ibn Hisham removed the “extraneous material,” the story began with Creation, and Muhammad's prophetic career was preceded by accounts of all the prophets who had gone before. The life of the man was the “seal” of their line (see Abbott, , Studies, pp. 87–89Google Scholar). This earlier, “discarded” section of Ishaq's work can be to some extent retrieved (Newby, Gordon Darnell, The Making of the Last Prophet: A Reconstruction of the Earliest Biography of Muhammad [Columbia, S.C., 1989])Google Scholar, and while its remains are revealing of Ibn Ishaq's purpose and the milieu in which the work was finally composed (Abbott, , Studies, p. 89Google Scholar), they add nothing of substance to the portrait of the historical Muhammad.

⁶⁵ Ibn Ishaq 3 in Guillaume, , Life of Muhammad, p. 3.Google Scholar

⁶⁶ Ibid., pp. 102–7 in Guillaume, , Life of Muhammad, pp. 69–73Google Scholar; and compare what Ibn Ishaq calls “Reports of Arab Soothsayers, Jewish Rabbis and Christian Monks” about the birth of the Prophet (ibid., pp. 130ff. in Guillaume, , Life of Muhammad, pp. 90ff.Google Scholar).

⁶⁷ Sellheim, , “Prophet,” pp. 38–39, 59–67Google Scholar; Kister, , “Sira Literature,” pp. 356–57Google Scholar; and, for a more general consideration of “polemic as a history-builder,” see Wansbrough, , Sectarian Milieu, pp. 40–45 and n. 77 below.Google Scholar

⁶⁸ Kister, , “Sira Literature,” pp. 356–57Google Scholar, on the early Sīra of Wahb ibn Munabbih (d. 728 or 732) and the “popular and entertaining character of the early Sīra stories, a blend of miraculous narratives, edifying anecdotes and records of battles in which sometimes ideological and political tendencies can be discerned.” (Compare Cook, , Muhammad, p. 66.Google Scholar)

⁶⁹ The New Testament critic Joseph Fitzmyer defined a “theologoumenon” as “a theological assertion that does not directly express a matter of faith or an official teaching of the Church, and hence in itself is not normative, but that expresses in language that may prescind from facticity a notion which supports, enhances or is related to a matter of faith” (Fitzmyer, Joseph A., “The Virginal Conception of Jesus in the New Testament,” originally published in 1973, rpt. in Fitzmyer, Joseph A., To Advance the Gospel [New York, 1981], p. 45).Google Scholar

⁷⁰ Sellheim, , “Prophet,” pp. 49–53Google Scholar; Kister, , “Sīra Literature,” pp. 362–63.Google Scholar

⁷¹ Wansbrough, , Sectarian Milieu, p. 35Google Scholar; and compare Noth, , Quellenkritische Studien, pp. 40–45, 155–58.Google Scholar The reason for the vague “distributional chronology,” as Wansbrough called the pre-Hijra system, was certainly not, as Watt has suggested (in Watt, and McDonald, , Muhammad, p. xxiGoogle Scholar), that “there were fewer outstanding events.” The call of the Prophet, the earliest revelation of the Qurʾan, and the making of the first converts would all appear to be supremely important, though the Muslim tradition had little certainty, chronological or otherwise, about them (ibid., pp. xxii, xxv-xli), likely because there was either no way or no reason to remember the date.

⁷² Crone, , Slaves on Horses, p. 13Google Scholar: The inertia of the source material comes across very strongly in modern scholarship on the first two centuries of Islam. The bulk of it has an alarming tendency to degenerate into mere arrangements of the same old canon—Muslim chronicles in modern languages and graced with modern titles. Most of the rest consists of reinterprelation in which the order derives less from the sources than from our own ideas of what life ought to be about—modern preoccupations graced with Muslim facts and footnotes.

⁷³ One attempt to substitute “genuine” eyewitness testimony (if not to Muhammad himself, then to the first appearance of the Islamic movement on the early 7th-century Near East) has been Patricia Crone and Cook's, MichaelHagarism: The Making of the Islamic World (Cambridge, 1977)Google Scholar, and while a brave and provocative book, it has tempted few others to follow its suggestion: “The historicity of the Islamic tradition is … to some degree problematic: while there are no cogent internal grounds for rejecting it, there are equally no cogent external grounds for accepting it. … The only way out of the dilemma is thus to step outside the Islamic tradition altogether and start again” (p. 3). What the external testimony to early Islam amounts to (and it is not a great deal) is summarized in Cook, , Muhammad, pp. 73–76Google Scholar; and the limitations of this approach are underscored in Wansbrough, , Sectarian Milieu, pp. 115–16.Google Scholar

⁷⁴ Watt, “Materials”; Sellheim, “Prophet”; Wansbrough, “Qurʾanic Studies”; Wansbrough, Sectarian Milieu.

⁷⁵ Wansbrough, , “Qurʾanic Studies,” p. 66.Google Scholar

⁷⁶ See n. 80 below. Michael Cook (Muhammad) reflects the far more modest aims of contemporary searchers after “influences”: For the most part we are reduced to the crude procedure of comparing Islam with the mainstream traditions of Judaism and Christianity, and trying to determine which elements came from which. The answers are often convincing, but they fail to tell us in what form those elements came to Muhammad, or he to them. (p. 77)

⁷⁷ This was done as early as Geiger's, AbrahamJudaism and Islam (originally published in Latin in 1832; rpt. from the translation published in 1898 [New York, 1970])Google Scholar; and then later, Torrey, Charles Cutler, The Jewish Foundations of Islam ([New York, 1933; rpt. New York, 1967]).Google Scholar There have been a number of suggestive portraits of the “Jewish Muhammad,” followed by the arguments of Bell, Richard, The Origin of Islam in Its Christian Environment (London, 1926; rpt. London, 1968)Google Scholar, Ahrens, Karl, “Christliches im Qoran” (Zeitschrift der Deutschen Morganlandischen Gesellschaft, 84 [1930], pp. 15–68, 148–90)Google Scholar; and Andrae, Tor, Les origines de l'Islam et le Christianisme (Paris, 1955), for a “Christian Muhammad.”Google Scholar

⁷⁸ The political hypothesis, first argued by Eisler and Brandon, took this more recent form in Maccoby, Hyam, The Mythmaker: Paul and the Invention of Christianity (New York, 1987)Google Scholar: Though all these [just cited Jewish] writers have their individual approaches, it is characteristic of the school as a whole to use the Talmud to show that Jesus' life and teaching are entirely understandable in terms of the Judaism of his time, particularly rabbinical or Pharisaic Judaism. The corollary is that, since Jesus did not conflict with Judaism, his death took place for political reasons, later camouflaged as religious by the Christian Church in its anxiety to cover up the fact that Jesus was a rebel against Rome. (pp. 208–9)

Cf. Bammel, Ernst, “The Revolutionary Theory from Reimarus to Brandon,” in Bammel, Ernst and Moule, C. D. F., eds., Jesus and the Politics of His Day (New York, 1984), pp. 11–68.CrossRefGoogle Scholar

⁷⁹ Harvey, , Jesus, p. 6Google Scholar, was cited in n. 25 above on the “constraints of history.” However, he went on to add:

This is not to say, of course, that he [Jesus] must have been totally subject to these constraints. Like any truly creative person, he could doubtless bend them to his purpose.… But had he not worked within them, he would have seemed a mere freak, a person too unrelated to the normal rhythm of society to have anything meaningful to say.

⁸⁰ Cook, , Muhammad, p. 77.Google Scholar Moreover, it is here that the Islamicist, like the New Testament scholar (see n. 23), runs into the problem of the usefulness of the “rabbinic sources”: to what extent can the Mishna, the Talmud, and the Midrashim (many of these latter sources being, in fact, post-Islamic and so possibly influenced by, rather than influencing, early Islam) be used to illuminate the pre-Islamic milieu of Mecca? Geiger, Torrey (Jewish Foundations, p. 34), and, notoriously, Katsh, Abraham, Judaism in Islam: Biblical and Talmudic Backgrounds of the Koran and Its Commentaries. Surahs II and III (New York, 1954)Google Scholar, invoked them almost as if Muhammad had a personal yeshiva library at his disposal, or, as Torrey thought, even a rabbinic teacher (Jewish Foundations, pp. 40–42).

⁸¹ Wensinck, Arent Jan, Muhammad and the Jews of Medina, with the excursus, Muhammad's Constitution of Medina by Wellhausen, Julius, trans, and ed. Behn, Wolfgang (Freiburg, 1975), p. 73.Google Scholar

⁸² Paret, , Der KoranGoogle Scholar; Watt, , “Materials,” p. 28Google Scholar; Watt, and McDonald, , Muhammad, pp. xxi–xxvGoogle Scholar; Sellheim, , “Prophet,” pp. 73–77Google Scholar; Kister, , “Sīra Literature,” pp. 352–53.Google Scholar

⁸³ Kister and his students have painstakingly compared variants in early, and largely unpublished, Muslim traditions on various topics—thus, for example, his analysis of a rather mysterious pre-revelation religious practice of Muhammad, called taḥannuth (“Al-Tahannuth: An Inquiry into the Meaning of a Term,” Bulletin of the School of Oriental and African Studies, 31 (1968), pp. 223–36)Google Scholar—and attempted to construct the original understanding behind them, on the assumption that the “original” tradition derived, to some degree, from a historical “fact.” They did not, however, directly address the critical question of the authenticity of any of the hadith materials with which they are so scrupulously dealing, though Kister for one, as we have seen (n. 68 above), was well aware of the historiographical problems posed by the inauthenticity of the hadith.

⁸⁴ Wellhausen, Julius, Reste arabischen Heidentums, 2nd ed. (Berlin, 1897).Google Scholar

⁸⁵ Hawting, G. R., “The Origins of the Islamic Sanctuary at Mecca,” in Juynboll, G. H. A., ed., Studies on the First Century of Islamic History (Carbondale, Ill., 1982), pp. 25–47.Google Scholar

⁸⁶ It is instructive of the two methods to compare Hawting, “Origins,” with Rubin, Uri, “The Kaʿba, Aspects of Its Ritual, Functions, and Position in Pre-Islamic and Early Islamic Times,” Jerusalem Studies in Arabic and Islam, 8 (1986), pp. 97–131, both of which deal with the pre-Islamic sanctuary at Mecca.Google Scholar