Author's note: I thank Professor Donald P. Little for his valuable comments on an earlier draft.

¹ For example, Jones, J. M. B., “The Chronology of the Maghāzī—A Textual Survey,” Bulletin of the School of Oriental and African Studies, 19 (1957): 245–80CrossRefGoogle Scholar; idem, “Ibn Isḥāq and al-Wāqidī: The Dream of ʿĀtika and the Raid to Nakhla in Relation to the Charge of Plagiarism,” Bulletin of the School of Oriental and African Studies, 22 (1959): 41–51CrossRefGoogle Scholar; Wansbrough, J., The Sectarian Milieu: Content and Compositionof Islamic Salvation History (Oxford: Oxford University Press, 1978)Google Scholar; Watt, W. M., “The Reliability of Ibn Isḥāq's Sources,” La vie du prophète Mahomet, Colloque de Strasbourg, 10 1980 (Paris: Presses Universitaires de France, 1983), 31–43 (esp. 33 f., 39, and 41, on al-Bukhari)Google Scholar; Landau-Tasseron, E., “Processes of Redaction: The Case of the Tamīmite Delegation to the Prophet Muḥammad,” Bulletin of the School of Oriental and African Studies, 49 (1986): 253–70.CrossRefGoogle Scholar

² Becker, C. H., “Grundsḥtzlichen zur Leben-Muhammadforschung,” in Islamstudien (Leipzig: Verlag Quelle und Meyer, 1924), 1:521 f.Google Scholar; Schacht, J., “A Revaluation of Islamic Traditions,” Journal of the Royal Asiatic Society (1949): 150 f.Google Scholar; Wansbrough, J., Quranic Studies: Sources and Methods of Scriptural Interpretation (Oxford: Oxford University Press, 1977), 183Google Scholar; idem, Sectarian Milieu, 85.

³ Wansbrough, , Sectarian Milieu, 77 f.Google Scholar

⁴ For a critique of the lack or insufficiency of such realization in the scholarship on medieval Islamiclegal theory, for instance, see Hallaq, Wael B., Law and Legal Theory in Classical and Medieval Islam (London: Variorum, 1995).Google Scholar

⁵ al-Ṣanʿānī, ʿAbd al-Razzaq ibn Hammāam, Kitāb al-Muṣannaf, 11 vols., ed. al-Aʿẓami, H. R. (Beirut: al-Majlis al-ʿIlmī, 1972)Google Scholar; Kitāb al-maghāzī: ibid., 5:313–492. On ʿAbd al-Razzaq and his Muṣannaf, see Sezgin, F., Geschichte der arabischen Schriftlums (hereafter GAS), 9 vols. (Leiden: E. J. Brill, 1967), 1:99Google Scholar; Motzki, H., “The Muṣannaf of ʿAbd al-Razzāq al-Ṣanʿānī as a Source of Authentic Aḥādīth of the First Century A.H.,” Journal of Near Eastern Studies, 50 (1991): 1–21CrossRefGoogle Scholar; idem, Die Anfänge der islamischen Jurisprudenz: ihre Entwicklung in Mekka bis zur mine des 2./8. Jahrhunderts (Stuttgart: Franz Steiner, 1991), esp. 50–67.Google Scholar

⁶ Shayba, Ibn Abī, Kitāb al-Muṣannaf, ed. al-Ḥṳt, K. Y. (Beirut: Dār al-Tāj, 1989)Google Scholar; Kitāb al-Maghāzī: ibid., 7:326–445. On Shayba, Ibn Abī, see GAS, 1:108 f.Google Scholar

⁷ Al-Bukhārī, , al-Jāmiʿ al-Ṣaḣiḣ, ed. Krehl, L. (Leiden: E. J. Brill, 1862–1908)Google Scholar; Kitāb al-Maghāzī: ibid., 3:52–193. On al-Bukhārī, , see GAS, 1:115–34.Google Scholar Unless otherwise indicated, all subsequent references to ʿAbd al-Razzaq, Ibn Abi Shayba, and al-Bukhari will be to their respective maghāzī

⁸ Motzki, , Die Anfänge, 55.Google Scholar On Maʿmar ibn Rashid, see ibid., index, s.v.; Sezgin, , GAS, 1:290 f.Google Scholar

⁹ Motzki notes that ʿAbd al-Razzaq's Kitāb al-Maghāzī, as well as his Kitāb al-Jāmiʿ, which is likewise indebted to Maʿmar, both have traditions other than Maʿmar's. “Deshalb ist es denkbar, daß schon ʿABDARRĀZZAQ oder seine Schüler sie als Teil seines [namely, ʿAbd al-Razzāq's] Traditionswerkes betrachteten”; Motzki, , Die Anfänge, 55 f.Google Scholar On the subsequent transmission of ʿAbd al-Razzaq's Muṣannaf, see al-Ishbīlī, Ibn Khayr (d. 1179), Fahrasa mā rawāh ʿan shuyūkhihi & Abū Bakr Muḥammad ibn Khayr & al-Ishbīlī, ed. Codera, F. and Tarrago, J. R. (Baghdad: Maktabat al-Muthannā, 1963 [reprint ofthe 1894–95 edition]), 127–31.Google Scholar

¹⁰ Schützinger, H., “Ibn Abi Šaiba und sein Taʾrix: Eine Untersuchung an Hand des MS. Berlin 9409,” Oriens 23–24 (1974): 134–46.CrossRefGoogle Scholar (I have not been able to examine ms. Berlin 9409 myself and thereforerely on Schiitzinger's description of its contents.)

¹¹ Schützinger, who did not consult any complete text of the Muṣannaf, thinks that the manuscript of Ibn Abi Shayba's Taʿrikh, which he has described is not a part of the Muṣannaf. His arguments to that effect, however, are not compelling (see “Ibn Abī Šaiba,“ 145: “Hier erhebt sich & die Frage, ob es denkbar ist, dass ein 109 1/2 Blatt umfassendes, chronologisch geordnetes und in Kapitel eingeteiltes Geschichtswerk Bestandteil eines Muṣannaf-Buches sein kann. Es ist kaum vorstellbar, dass sich einsolches Werk in das den Muṣannaf-Büchern eigene Schema einpasst”), and his description of the contents of ms. Berlin 9409 reveals a very marked resemblance to the text of the Kitāb al-Maghāzī in the Muṣannaf. Note that Sprenger, Aloys, who was probably acquainted with the complete text of the Muṣannaf and who used the Taʾrikh for his Das Leben und die Lehre des Mohammad (Berlin, 1869)Google Scholar, had little doubt that the Tarʾīkh did form “part of the Moşannaf of Ibn Aby Shaybah” (see Schützinger, , “Ibn Abī Šaiba,” 140 f.Google Scholar, quoting Sprenger's handwritten note on the manuscript of the Taʾrikh, and cf. ibid., 145). On the history of the transmission of Ibn Abi Shayba's Muṣannaf, see Khayr, Ibn, Fahrasa, 131–33.Google Scholar

¹² Shayba, Ibn Abī, “Kitāb al-Taʾrīkh,” in Muṣannaf, 6:546–64Google Scholar; cf. Schützinger, , “Ibn Abī Šaiba,” 145.Google Scholar

¹³ Landau-Tasseron, E., “Sayf Ibn ʿUmar in Medieval and Modern Scholarship,” Der Islam, 67 (1990):6 ff.CrossRefGoogle Scholar and esp. 9 warns of the dangers of such a distinction but defends it nevertheless. Nabia Abbott, too, makes a sharp distinction between “hadith” and “khabar”—on the basis of their respective contents andthe methods of their transmission, see Abbott's, Studies in Arabic Literary Papyri, II: Qurʿanic Commentary and Tradition (Chicago: University of Chicago Press, 1967), 29, 77, 91, 118, 145, 233, 244, 256.Google Scholar

¹⁴ For the conventional chronology of the maghāzī, see Jones, , “Chronology,” 245–80.Google Scholar

¹⁵ For the position of this ghazwa in the overall chronology of the maghāzī, see Jones, , “Chronology,” 247, 259.Google Scholar

¹⁶ See Hinds, M., “ ‘Maghāzī’ and ‘Sira’ in Early Islamic Scholarship,” in La vie du prophète Mahomet, 65 f.Google Scholar, who makes the same point with reference to the Maghāzī of ʿAbd al-Razzāq.

¹⁷ But see Hawting, G. R., “al-Ḥudaybiyya and the Conquest of Mecca: A Reconsideration of the Tradition about the Muslim Takeover of the Sanctuary,” Jerusalem Studies in Arabic and Islam, 8 (1986):16Google Scholar, who seems to make much of ʿAbd al-Razzaq's chronologically awkward juxtaposition of his materials. That there is an implicit “causal sequence” in the way these materials are presented is, however, a gratuitous assumption.

¹⁸ See Hinds, , “‘Maghāzī’ and ‘Sīra’,” 65 f.Google Scholar

¹⁹ Calder, Norman, Studies in Early Muslim Jurisprudence (Oxford: Clarendon Press, 1993), 179.Google Scholar

²⁰ Ibid., 195.

²¹ The expressions “notebooks,” “finished works,” “real books,” and so forth, are Calder's, : Studies, 179, 180, and 171–81Google Scholar, passim. Cf. Gerhardsson, B., Memory and Manuscript: Oral Tradition and WrittenTransmission in Rabbinic Judaism and Early Christianity (Copenhagen: Ejnar Munksgaard, 1964), 157ffGoogle Scholar., on “written notes” and “notebooks” used (illegitimately) in Rabbinic Judaism to study, learn, and better remember the growing materials of the oral Torah. Although his hypotheses concerning “note-books” and other “literary forms” characteristic of the intellectual milieu in early Islam seem indebted, inter alia, to Gerhardsson's work, Calder claims much more for early Islam than Gerhardsson does for Rabbinic Judaism, and it remains unclear, in any case, how much of early Islamic literary practice canbe extrapolated from Rabbinic evidence.

²² Calder, , Studies, 194.Google Scholar

²³ Al-Bukhārī, , Maghāzī, 53 f.Google Scholar

²⁴ On the dīwān: cf. Encyclopaedia of Islam, new edition (EI²), Leiden: E. J. Brill, 1960Google Scholar–); s.v. “Dīwān” (A. A. al-Duri et al.).

²⁵ Al-Bukhārī, , Maghāzī, 71.Google Scholar

²⁶ See notes 50 ff., below.

²⁷ See Mink, L. O., “Narrative Form as a Cognitive Instrument,” in Historical Understanding, ed. Fay, B. et al. (Ithaca: Cornell University Press, 1987), 197Google Scholar: “A narrative must have a unity of its own; this is what is acknowledged in saying that it must have a beginning, middle, and an end.” Also see A Dictionary of Modern Critical Terms, ed. Fowler, R. (London: Routledge, 1987)Google Scholar, s.v. “Narrative,” “Narrative Structure.”

²⁸ Ankersmit, F. R., Narrative Logic: A Semantic Analysis of the Historian's Language (The Hague:Martinus Nijhoff Publishers, 1983), 204Google Scholar and passim.

²⁹ As a narratio, however peculiar in form and content, it presupposes other narratios: Ibn Abi Shayba's account of Badr has some interesting parallels and contrasts with al-Waqidi's, for instance, which deserve to be explored, although the task cannot be undertaken in the present context.

³⁰ See the familiar example of the omission, in Ibn Hisham's recension of Ibn Ishaq's account, of the tradition about the capture of al-ʿAbbas at Badr and of the Prophet's demand that he ransom himself; al-Tabari does have a report to this effect, via Ibn Ishaq, which suggests that the report in question was a part of Ibn Ishaq's original text of the Sīra. See Guillaume, A., The Life of Muḥammad: A Translation of [Ibn] Isīāq's Sīrat Rasūl Allāh (London: Oxford University Press, 1955), 312 f.Google Scholar Note too that al-ʿAbbas does not figure in the list of prisoners at Badr, although 43 men are supposed to have been captured and only 42 are named: Guillaume, , Life of Muhammad, 338 f.Google Scholar and n. 1 on 338; see also Sellheim, R., “Prophet, Chalif und Geschichte: Die Muḥammed-Biographie des Ibn Isḥāq,” Oriens, 18–19 (1965–1966): 49.Google Scholar For another example of the suppression, or rather the editing, of a tradition unfavorable to al-ʿAbbas, see Kister, M. J. and Plessner, M., “Notes on Caskel's Ǧamharat an-Nasab,” Oriens, 25–26 (1976): 64 f.Google Scholar What Funkenstein says of ancient and medieval—Greco-Roman, Jewish, and Christian—historians would seem to hold for Muslim writers, as well: “it had to occur to some ancient and medieval authors—as indeed it did—that the historian, rather than being a mere spectator, possesses a ius vitae nocendi of sorts over that which he should record. He or she can make and unmake history, can obliterate names, events, identities by not recording them, for evil or good purposes. Activity and memory belong together: without memory, the political activity cannot affect future generations”: Funkenstein, A., Perceptions of Jewish History (Berkeley: University of California Press, 1993), 30.Google Scholar

³¹ Al-Bukhārī, , Maghāzī, 68 f.Google Scholar

³² al-Razzāq, ʿAbd, Maghāzī, 353 (no. 9729).Google Scholar

³³ Shayba, Ibn Abī, Maghāzī, 357 (no. 36679).Google Scholar

³⁴ Ibid., 363 (no. 36717).

³⁵ Ibid., 361 (no. 36700).

³⁶ Ibid. (no. 36702).

³⁷ See al-Baghdādī, al-Khaṭīb, Taʾrikh Baghdād, 14 vols. (Cairo, 1931), 10:66–71 (no. 5185), esp. 67 f.Google Scholar

³⁸ L. O. Mink's view of historical narrative as a “configurational” mode of comprehension, where “the actions and events of a story comprehended as a whole are connected by a network of overlapping descriptions” can probably be applied to a work such as Ibn Abi Shayba's, as well (see Mink, , “History and Fiction as Modes of Comprehension,” in Historical Understanding, 42–60Google Scholar; the quotation is from p. 58). The “overlap of descriptions,” Mink notes, “may not be part of the story itself (as one thing after another) but only of the comprehension of it as a whole” (ibid.). In Ibn Abi Shayba's Maghāzī, but also in chronicles such as al-Tabari's, the overlap—of traditions and descriptions—is a part of the story, however; it is precisely such overlapping traditions that construct and comprise the “image” of the pastthat emerges from their configuration. (On the centrality of constructing such an “image” [or “narrative substance”] to the function of a narratio, see Ankersmit, Narrative Logic, chap. 5 and passim.)

³⁹ It was, in any case, on the wrong side that al-ʿAbbas had taken part in the battle.

⁴⁰ For al-Bukhari's, traditions on Hudaybiyya, see his Maghāzī, 110–19.Google Scholar The outline given here takes note of the major themes but does not encompass all the traditions that make up al-Bukhari's material on Hudaybiyya.

⁴¹ Shayba, Ibn Abī, Maghāzī, 381–90.Google Scholar

⁴² al-Razzāq, ʿAbd, Maghāzī, 330–43.Google Scholar

⁴³ On “common links,” see Schacht, J., The Origins of Muhammad Jurisprudence (Oxford: Clarendon Press, 1950), 171 f.Google Scholar

⁴⁴ To the coherent narrative account of Hudaybiyya in ʿAbd al-Razzaq's recension are appended two further traditions, both of which attest that it was ʿAli ibn Abi Talib who wrote the document (kitāb) of the agreement at Hudaybiyya. al-Razzāq, ʿAbd, Maghāzī, 342 f.Google Scholar (nos. 9721 f.).

⁴⁵ Wansbrough, , Sectarian Milieu, 87Google Scholar (emphasis added); also cited in Hinds, , “ ‘Maghāzī’ and ‘Sīra’,” 63.Google Scholar

⁴⁶ See al-Bukhārī, Maghāzī, 187 f., 190.Google Scholar

⁴⁷ Ibid., 185.

⁴⁸ Ibid., 191.

⁴⁹ See Hinds, , “ ‘Maghāzī’ and ‘Sira’,” 65 f.Google Scholar ʿAli figures very prominently in many of ʿAbd al-Razzaq's traditions, but these do not necessarily have a pro-ʿAli, much less a Shiʿi, character. ʿAbd al-Razzaq isnevertheless supposed to have been a Shiʿi.

⁵⁰ al-Akbar, Pseudo-al-Nāshiʾ, “Masāʾil al-Imāma,” in Frühe muʿtazilitische Häresiographie, ed. Ess, J. van (Wiesbaden: Franz Steiner, 1971), 66Google Scholar (ad the aṣḥāb al-ḥadīth of Baghdad). More generally, see Madelung, W., Der Imām al-Qāsim ibn Ibrāhim und die Glaubenslehre der Zaiditen (Berlin, 1965), 225 ff.CrossRefGoogle Scholar

⁵¹ Van Ess, , Häresiographie, 65 f.Google Scholar

⁵² See, for instance, al-Khallāl, al-Musnad min Masāʾil AbīʿAbdallāh Aḥmad b. Muḥammad b. Ḥanbal, British Library Ms. Or. 2675, fol. 63a, and fols. 56a ff.; more generally, see Madelung, , Der Imām al-Qāsim, 225 ff.Google Scholar, and Nagel, T., “Das Problem der Orthodoxie im frühen Islam,” in Studien zum Minderheitenproblem im Islam (Bonn, 1973), 1:7–44.Google Scholar

⁵³ Shayba, Ibn Abī, Maghāzī, 440Google Scholar (no. 37075).

⁵⁴ Ibid., 440 f. (no. 37078), 442 (no. 37090).

⁵⁵ Ibid., 441 (no. 37079).

⁵⁶ Ibid., 441 (no. 37079), 442 (nos. 37087 f.).

⁵⁷ On the murder of ʿUthman, see al-Ṭabari, , Taʾrīkh al-Rusul waʿl-Mulūk, 16 vols., ed. Goejeet, M. J. de et al. (Leiden: E. J. Brill, 1879–1901), 1:2980–3025Google Scholar; and Humphreys, R. S., “Qurʾanic Myth and Narrative Structure in Early Islamic Historiography,” in Tradition and Innovation in Late Antiquity, ed. Clover, F. M. and Humphreys, R. S. (Madison: University of Wisconsin Press, 1989), 279 ff.Google Scholar, for a perceptive analysis of al-Tabari's accounts.

⁵⁸ Shayba, Ibn Abī, Muṣannaf 7:532–64.Google Scholar Note that Ibn Abi Shayba is said to have compiled, interalia, both a Kitāb al-Jamal and a Kitāb Ṣiffīn: al-Nadīm, Ibn, Kitāb al-Fihrist, ed. Flügel, G. (Leipzig: F. C. W. Vogel, 1870–1871), 229.Google Scholar If the contents of these two works are identical with those of the aforementioned sections of the Muṣannaf, the subsumption—in the latter—of traditions relating to Siffin as a subsection of the Kitāb al-Jamal would seem to be the work of some later redactor or editor.

⁵⁹ Cf. Kitāb al-Jamal, 535 (no. 37768), 542 (no. 37807), and 547 (no. 37841 ff.). Needless to say, the point these traditions make is also important for the juristic precedent it establishes (namely, how totreat opponents or rebels when they are Muslims).

⁶⁰ ʿAʾisha: ibid., 536 (no. 37771 f.), 542 (no. 37811); ʿAli: 536 (no. 37774), 539 (no. 37795 f.), 541(no. 37802), 543 (no. 37812), 545 (no. 37832), 548 (no. 37852); Talha: 545 (no. 37827); al-Zubayr,545 (no. 37828).

⁶¹ Kitāb al-Jamal, passim, and esp. 540 f. (no. 37798), 544 (no. 37817), 551 (no. 37871). On the “betrayal” motif in early Islamic historiography, see Humphreys, , “Qurʿanic Myth and Narrative Structure,” 278 ff.Google Scholar

⁶² Hinds, , “ ‘Maghāzī’ and ‘Sira’,” 63.Google Scholar

⁶³ In his article on “Maghāzī” in EI2, Hinds only summarizes his earlier findings as set out in“ ‘Maghāzī’ and ‘Sira’.”

⁶⁴ See Landau-Tasseron, , “Sayf Ibn ʿUmar,” 9Google Scholar: “the fact is that certain scholars may be classifiedeither as muḥaddithūun or as akhbāriyyūn, whereas others may not, which means that the differentiation is not baseless. That the classification of some of the early historians is not clear-cut does not turn the two fields into one. Nor does it alter the impression that, in general, the attitude of the early unmistakable muḥaddithūn towards the historians was one of suspicion and distrust.” Also see Leder, S., “The Literary Use of the Khabar: A Basic Form of Historical Writing,” in The Byzantine and Early Islamic Near East, ed. Cameron, A. and Conrad, L. I. (Princeton: Darwin Press, 1992), 313 ff.Google Scholar

⁶⁵ See Lecker, M., “The Ḥudaybiyya-Treaty and the Expedition against Khaybar,” Jerusalem Studies in Arabic and Islam, 5 (1984): 6 ff.Google Scholar, ad Ibn Abī Shayba, Taʾrikh, Ms. Berlin 9409, fol. 57a. For a parallel passage (with minor variants, which, in all probability, are copyists' mistakes) in Shayba's, Ibn AbiMaghāzī, see Maghāzī, 382 (no. 36839).Google Scholar

⁶⁶ Another specimen of “traditionist historiography” may perhaps be seen in the so-called ṣaḥīfa of the Egyptian traditionist and judge Ibn Lahīʿa. He has traditions here about, inter alia, the murder of ʿUthman and the revolt of ʿAbdallah ibn Zubayr. On the ṣaḥīfa and its author, see Khoury, R. G., ʿAbd Allāh ibn Lahīʿa (97–174/715–790): Juge et grand maître de l'école égyptienne (Wiesbaden: Otto Harrassowitz, 1986).Google Scholar Khoury argues (p. 181) that Ibn Lahiʿa was at once a muḥaddith and a historian, whichmay have been the case. His traditions however are hadith in an apocalyptic idiom more than they areanything else, and they are of interest primarily for showing how aspects of the first and the second fitna may have been remembered by, and recounted in, certain traditionist circles in Egypt.

⁶⁷ See Hinds, , “ ‘Maghāzī’ and ‘Sira’,” 60 f.Google Scholar for a list of scholars—all of whom died in the second half of the second century A.H.—who are said to have written on the maghāzī.

⁶⁸ Al-Sakhāwī's, (d. 1497) al-Iʿlān bi'l-tawbīkh li-man dhamma ahl al-tawrīkh (Damascus, 1349 A.H.;trans, by Rosenthal, F.Google Scholar in idem, A History of Muslim Historiography, 2nd ed. [Leiden: E. J. Brill, 1968], 269–529Google Scholar; the following references are to this translation) is not only an elaborate defense of the legitimacy and usefulness of historical studies, it also gives a broad sampling of the kinds of criticism traditionist scholars had for history (see, especially, Rosenthal, , Historiography, 338 ff.Google Scholar). Despite the late dateof this treatise, many of the traditionist criticisms it quotes, and seeks to refute, purport to go back to the first centuries of Islam. It is noteworthy that, besides criticisms directed at historical studies per se, many a traditionist seems to have been critical even of the jarḥ wa taʿdīl genre, a traditionist stronghold, for it was thought to involve slandering the reputation of scholars. (Whether jarḥ wa taʿdīl was “history” at all is a question neither al-Sakhawi nor the critics he is engaged with seem to ask.) Note, too, that al-Sakhawi himself, in outlining a “legal classification” of history, recognizes that certain aspectsof it do fall into the category of the “forbidden”: “This applies, especially, to stories told in connection with the biographies (siyar) of the prophets. Then, there is the information about disputes among themen around Muhammad (which is also forbidden), because the historical informants (akhbārī) who report it as a rule exaggerate and mix things up” (Rosenthal, , Historiography, 335Google Scholar).