We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 20 January 2011
ʿAbdallāh ibn Sabaʾ is a figure generally regarded as Islam's first heretic by Sunnī scholars and also vilified by Shīʿī scholars. In this article an anonymous, esoteric work known as Umm al-Kitāb is examined as it contains an exceptional narrative that adopts a strikingly sympathetic approach to Ibn Sabaʾ. It is also argued that the work's unique take on the Ibn Sabaʾ legend sheds considerable light on the date and elusive provenance of this early Shīʿī text.
I am grateful to Professors Wadād al-Qāḍī, Fred Donner, and Wilferd Madelung who read this paper in its earliest form and offered many valuable comments and suggestions. Thanks are also due to Mushegh Asatryan of Yale who kindly read my draft and shared valuable insights from which the final version profited.
2 al-Ṭabarī, Abū Jaʿfar Muḥammad b. Jarīr, Taʾrīkh al-rusul wa-l-mulūk, (ed.) de Goeje, M. J. et al. . (Leiden, 1879–1901), i, p. 2942Google Scholar; al-Baghdādī, ʿAbd al-Qāhir, K. al-farq bayna l-firaq wa-bayān al-firqa al-nājiya minhum (Cairo, 1910), p. 225Google Scholar; al-Ṣafadī, Khalīl b. Aybak, al-Wāfī bi-l-wafayāt, Vol. xvii, (ed.) Krawulsky, D., Bibliotheca Islamica 6q (Wiesbaden, 1982), p. 190Google Scholar (quoting Ibn Abī l-Dam, d. 1244). The trope also appears in Shīʿī works as well. See: al-Nawbakhtī, Abū Muḥammad Ḥasan b. Mūsā, Firaq al-shīʿa, ed. Ritter, H., Bibliotheca Islamica 4 (Istanbul, 1931), p. 20Google Scholar; al-Qummī, Saʿd b. ʿAbdallāh al-Ashʿarī, K. al-maqālāt wa-l-firaq, (ed.) M. Mashkur, J. (Tehran, 1963), p. 20Google Scholar; al-Ṭūsī, Ikhtiyār maʿrifat al-rijāl (= abrg. of Muḥammad b. ʿUmar al-Kishshī's al-Rijāl), (ed.) Ḥ. Mustafavī (Mashhad, 1970), pp. 108–109; Abū Ḥātim al-Rāzī, K. al-zīna, in: al-Samarrāʾī, ʿA. S., al-Ghulūw wa-l-firaq al-ghāliya fī l-ḥaḍāra al-islāmīya (Baghdād, 1972), p. 305Google Scholar
3 Khaythama, Ibn Abī, al-Taʾrīkh al-kabīr, (ed.) Halal, Ṣ. F. (Cairo, 2004), iii, p. 177Google Scholar; al-Qāḍī, ʿAbd al-Jabbār, Tathbīt dalāʾil al-nubūwa, (ed.) ʿUthmān, ʿA.-K. (Beirut, 1966), i, pp. 546–547Google Scholar; idem, Faḍl al-iʿtizāl wa-ṭabaqāt al-muʿtazila, (ed.) F. al-Sayyid (Tunis, 1974), p. 143; al-Iṣfahānī, Abu Nuʿaym, Ḥilyat al-awliyāʾ (Cairo, 1910), viii, p. 253Google Scholar.
4 ʿĀṣim, Ibn Abī, K. al-sunna, (ed.) al-Albānī, Muḥammad Nāṣir al-Dīn (Damascus, 1980), ii, p. 476Google Scholar; al-Mawṣilī, Abū Yaʿlā, al-Musnad, (ed.) Asad, Ḥusayn Salīm (Damascus, (1984–1994), i, pp. 349–350Google Scholar.
5 Ps.-Nāshiʾ al-Akbar, K. uṣūl al-dīn, pp. 22–23, in: van Ess, J., Frühe muʿtazilitische Häresiographie: Zwei Werke des Nāšiʾ al-Akbar (gest. 293 H.), Beiruter Texte und Studien 11 (Beirut, 2003 2)Google Scholar; l-Dunyā, Ibn Abī, K. maqtal amīr al-muʾminīn ʿAlī b. Abī Ṭālib, (ed.) Ṣāliḥ, I. (Damascus, 2001), pp. 83–84Google Scholar; al-Baghdādī, al-Khaṭīb, Taʾrīkh Madīnat al-Salām, (ed.) Maʿrūf, B. ʿA. (Beirut, 2001), ix, pp. 516–517Google Scholar.
6 Kishshī, Rijāl, p. 106–107; Shahrāshūb, Ibn, Manāqib Āl Abī Ṭālib, (ed.) al-Biqāʿī, Y. (Qum, 2000), i, p. 325Google Scholar. For two early, non-Shīʿī versions, see: al-Dīnawārī, Ibn Qutayba, al-Maʿārif, (ed.) ʿUkāsha, Th. (Cairo, 1969), p. 266Google Scholar and Rusta, Ibn, K. al-aʿlāq al-nafīsa, (ed.) de Goeje, M. J., Bibliotheca Geographorum Arabicorum 7 (Leiden, 1967), p. 218Google Scholar.
7 Composed by the ʿUthmānī akhbārī al-Tamīmī, Sayf b. ʿUmar (see his K. al-ridda wa-l-futūḥ wa-K. al-jamal wa-masīr ʿĀʾisha wa-ʿAlī, (ed.) al-Samarrāʾī, Q., Leiden, 1995, pp. 135–137)Google Scholar and the Imāmī muʿtazilī Hishām b. al-Ḥakam, respectively (see: Nawbakhtī, Firaq, pp. 19–20 and Saʿd b. ʿAbdallāh, Maqālāt, pp. 19–21). That these two later works of Nawbakhtī and Saʿd b. ʿAbdallāh preserve an earlier treatise of Hishām has been argued in Madelung, W., “Bemurkungen zur imamischen Firaq-Literatur,” Der Islam, XLIII (1967), pp. 37–52Google Scholar; however, see now: Modarressi, H., Tradition and Survival: A Bibliographical Survey of Early Shīʿite Literature (Oxford, 2003), i, pp. 265–266Google Scholar.
8 Ivanow, , “Notes sur l'Ummu'l-kitāb des Ismaéliens de l'Asie Centrale”, REI, VI (1932), pp. 426–427Google Scholar.
9 Ibid., p. 425.
10 Idem, The Alleged Founder of Ismailism (Bombay, 1946), pp. 99–101; idem, Studies in Early Persian Ismailism (Leiden, 1948), p. 108.
11 Idem, Ismaili Literature: A Bibliographical Survey (Tehran, 1963), pp. 193–195.
12 “Ummu'l-kitāb,” Der Islam, XXIII (1936), pp. 1–132.
13 Ummu'l-Kitāb (Naples, 1966).
14 Tijdens, E. F., “Der mythologisch-gnostische Hintergrund des (Umm al-kitāb),” Acta Iranica, VII (1977), pp. 241–526Google Scholar.
15 See: Halm, , “‘Das Buch der Schatten’: Die Mufaḍḍal-Tradition der Ġulāt und die Ursprünge des Nuṣairitiertums (II)”, Der Islam, LVIII (1981), pp. 36 ff.Google Scholar; idem, Gnosis, pp. 113 ff.
16 History of Islamic Philosophy, (trans.) L. Sherrard (London, 1993), pp. 75–76.
17 Religion in Iran: From Zoroaster to Bahaʾullah, (trans.) J. M. Marchesi (New York, 2000), pp. 150–162.
18 Halm, Gnosis, p.120; idem, “The cosmology of the pre-Fāṭimid Ismāʿīliyya,” in: Farhad Daftary, (ed.), Mediaeval Ismāʿīlī History and Thought (Cambridge, 1996), pp. 82–83.
19 According to a ḥadīth recognised by both Sunnīs and Shīʿa alike, one morning when Muḥammad was wearing a black coat, his daughter Fāṭima, her husband ʿAlī, and their two sons, al-Ḥasan and al-Ḥusayn, arrived one after another, and taking all four under his cloak the Prophet blessed them reciting the qurʾānic verse, “God desires to remove from you, the people of the house [ahl al-bayt], impurity and to purify you completely” (Q. 33:33). See: Wensinck, A. J. et al. ., Concordances et indicies de la tradition musulmane (Leiden, 1936–1988), viii, p. 398.52–4Google Scholar and EI 3, s.v. “Ahl al-Kisāʾ” (F. Daftary).
20 See: Wolff, P., “Auszüge aus dem Katechismus der Nossairier,” ZDMG, III (1849), p. 307Google Scholar (nr. 66); Strothmann, R., Esoterische Sonderthemen bei den Nusairi: Geschichten und Traditionen von den heiligen Meistern aus dem Prophetenhaus (Berlin, 1958), p. 4 (§5)Google Scholar.
21 See: Kohlberg, Etan, “Authoritative Scriptures in Early Imami Shiʿism,” in Les retours aux écritures fondamentalismes présents et passes, (eds.) Patlagean, E. and Le Boulluec, A. (Paris, 1993), pp. 297 ffGoogle Scholar.
22 Modarressi, Tradition, i, pp. 4–12; cf. Kohlberg, E. and Amir-Moezzi, M. A., Revelation and Falsfication: The Kitāb al-qirāʾāt of Aḥmad b. Muḥammad al-Sayyārī (Leiden, 2009), pp. 24–30CrossRefGoogle Scholar.
23 Modarressi, Tradition, i, pp. 17–20.
24 Ibid., i, p. 7 and n. 43 thereto.
25 Most famous among these incidents is that of Ibn Tūmart (d. 534/1130), whose Mahdist pretensions were allegedly derived from his obtaining the K. al-jafr from which “he had gained knowledge from ahl al-bayt” and realised his destiny as the Qāʾim and the identity of his successor, ʿAbd al-Muʾmin. See: Khallikān, Ibn, Wafayāt al-aʿyān wa-abnāʾ al-zamān, ed. ʿAbbās, Iḥsān (Beirut, 1968–1972), v, pp. 47–8Google Scholar and iii, pp. 238, 240 f. and Goldziher, I., “Materialien zur Kenntnis der Almohadenbewegung,” ZDMG, XLIV (1890), pp. 123 ffGoogle Scholar.
26 Kohlberg, “Authoritative Scriptures,” pp. 301 ff.; Modarressi, Tradition, i p. 12.
27 Toufic Fahd, “al-Djafr,” EI 2, ii, pp. 375b-377; cf. Amir-Moezzi, M. A., The Divine Guide in Early Shīʿism: The Sources of Esotericism in Islam, (trans.) Streight, D. (Albany, 1994), pp. 69 ffGoogle Scholar.
28 Fillipani-Ronconi argued for both Manichaean and Buddhist origins in the introduction to his translation of the text as well as in his article, “Note sulla soteriologia e sul simbolismo cosmico dell’ Ummu'l-kitāb,” AIOUN, XIV (1964), pp. 111 ff. Against this view, see the reviews of his translation by J. van Ess (Der Islam, XLVI, 1970, pp. 95–100) and W. Madelung (Oriens XXV/XXVI, 1976, pp. 352–358) as well as Halm, Gnosis, pp. 116–117. Tijdens sought to detect within the text two layers, one authored by a Judaeo-Christian sect with Mu'tazilite sympathies and the other by a redactor under the influence of Avincennian cosmology. Halm, in my view rightly, rejects this hypothesis for reasons discussed in Halm, “Das Buch der Schatten,” pp. 37 ff. For a recent effort to situate the text within a Mesopotamian context, see: Hämeen-Antilla, Jaako, “Ascent and Descent in Islamic Myth”, in Mythology and Mythologies: Methodological approaches to intercultural influences, (ed.) Whiting, R.M. (Helsinki, 2001), pp. 47–67Google Scholar.
29 Halm, “Buch der Schatten,” pp. 35–36; idem, Gnosis, p. 120. Modarressi has compiled a substantial corpus of citations putatively derived from Jābir al-Juʿfī's tafsīr that strongly suggests otherwise; see his Tradition, i, pp. 94–97.
30 For the conventional heresiographical account, see: Nawbakhtī, Firaq, pp. 37 ff., 58 ff. See also Halm, Gnosis, pp. 199–206.
31 For which reason Madelung postulated a date for UaK no earlier than the sixth/twelfth century; see his review of Fillipani-Ronconi's translation in Oriens, XXV-XXVI (1976), p. 355
32 Halm, “Buch der Schatten,” p. 39; idem, Gnosis, p. 120.
33 See note 6 above.
34 Later authors resolved this contradiction by creating a harmonised account in which Ibn Sabaʾ escapes execution through exile, but how and why this occurs is beyond the scope of this essay. For a fuller account see: S. Anthony, The Caliph and the Heretic: Ibn Sabaʾ and the Origins of Shīʿism (Leiden, forthcoming), Chapters 4–5.
35 Das Kitāb al-Nakṯ des Naẓẓam und seine Rezeption im Kitāb al-Futyā des Ğāḥiẓ: Eine Sammlung der Fragment mit Übersetzung und Kommentar (Göttingen, 1972), pp. 54 ff.
36 The earliest version appears in a musnad attributed to Zayd b. ʿAlī (d.122/740), but it is likely it was compiled in the middle of the second/eighth century. See: Musnad Zayd b. ʿAlī (Beirut, 1966), p. 340 and Madelung, W., Der Imām al-Qāsim ibn Ibrāhīm und die Glaubenslehre der Zaiditen (Berlin, 1965), pp. 53–56CrossRefGoogle Scholar. For others early versions of the tradition, see: al-Ṣanʿānī, ʿAbd al-Razzāq, al-Muṣannaf, (ed.) al-Aʿẓamī, Ḥ. ʿA.-R. (Beirut, 1970–72), x, p. 213Google Scholar (no.9413); al-Shāfiʿī, Muḥammad b. Idrīs, al-Umm, (ed.) Muḥammad, ʿA. and Aḥmad, ʿA. (Beirut, 2001), x, p.560Google Scholar. Not all Baṣran versions go back to ʿIkrima; some are attributed to Anas b. Malik via Qatāda b. Diʿāma. This version, however, is textually corrupt. Cf. the analyses of Suliman Bashear, Arabs and Others in Early Islam, SLAEI 8 (Princeton, 1997), p. 78 and Juynboll, G. H. A., Encyclopedia of Canonical Ḥadīth (Leiden, 2008), p. 146aGoogle Scholar.
37 On whom, see: Günther, S., “al-Nawfalī's Lost History: The Issue of a 9th Century Shīʿite Source Used by al-Ṭabarī and Abū l-Faraj al-Iṣfahānī,” in Al-Ṭabarī: A Medieval Muslim Historian and His Work, (ed.) Kennedy, H., SLAEI 15 (Princeton, 2008), pp. 157–174Google Scholar.
38 l-Ḥadīd, Ibn Abī, Sharḥ Nahj al-balāgha, (ed.) Ibrāhīm, M. A. (repr., Beirut, 2001), viii, pp. 94–95Google Scholar. Cf. Ṭūsī, Ikhtiyār, p. 109; al-Kulaynī, , al-Kāfī, (ed.) al-Ghaffārī, ʿA. A. (Tehran, 1971), vii, pp. 291–292Google Scholar.
39 See his redaction of Nawfalī's aforementioned account in Ibn Abī l-Ḥadīd, Sharḥ, v, pp. 5–6. On his importance as a redactor of Nawfalī's materials, see: Günther, S., Quellenuntersuchungen zu den Maqātil al-Ṭālibiyyīn des Abū ’l-Farağ al-Iṣfahānī (Hildesheim, 1991), pp. 133 ffGoogle Scholar.
40 Ṭūsī, Ikhtiyār, pp. 106–107. Kishshī's material may be the earliest; if his isnād is somewhat reliable, his account of ʿAlī's execution of ibn Sabaʾ may derive from the Kitāb al-radd ʿalā l-ghulāt of Hishām b. al-Ḥakam's student Yūnus b. ʿAbd al-Raḥmān (d. 208/823–4). I am skeptical that it does, however, and believe the tradition requires further corroboration for an early dating.
41 al-Maqdisī, Muṭahhar b. Ṭāhir, K. al-badʾ wa-l-taʾrīkh, (ed.) Huart, Cl. (Paris, 1916), v, p. 125Google Scholar; cf. Ibn Abī l-Ḥadīd, Sharḥ, v, p. 5. This is a slightly modified version of Ibn ʿAbbas’ dictum, which appears throughout the original Baṣran narrative of ʿAlī's immolation of the apostates (see: note 36 above), intended to strengthen the case of the ghulāt.
42 See: Reynolds, G. S., A Muslim Theologian in the Sectarian Milieu: ʿAbd al-Jabbār and The Critique of Christian Origins (Leiden, 2004), pp. 61–62Google Scholar.
43 Tathbīt, ii, pp. 549–550. A similar story to that of UaK, although far shorter, appears also in the Haft Bāb-i Bābā Sayyednā, in: Ivanov, W., Two Early Ismaili Treatises (Bombay, 1933), p. 15 (Prs.)Google Scholar; Eng. trans. in: Hodgson, M.G.S., The Order of the Assassins: The Struggle of the Nîzârî Ismâʿîlîs against the Islamic World (The Hague, 1955), p. 294Google Scholar. S. J. Badakhchani has recently identified the previously unknown author of the Haft Bāb as Ṣalāḥ al-Dīn Ḥasan-i Maḥmūd, a Nizārī dāʿī and poet with close associations with Naṣīr al-Dīn al-Ṭūsī, thanks to a recently discovered manuscript indentifying him as the individual who authored the work in 602/1205. See: Badakchani, , (ed.) and (trans.), The Paradise of Submission: A New Persian Edition and English Translation of Naṣīr al-Dīn Ṭūsī's Rawḍa-yi taslīm (London, 2005), pp. xv–xviGoogle Scholar and p. 244 and n. 15 thereto.
44 Donohue, J. J., The Buwayhid Dynasty in Iraq 334h./945–403h./1012 (Leiden, 2003), 51 ff.Google Scholar, 149 ff. On Khaṣībī's relationship with ʿIzz al-Dawla in particular, see: Friedman, Y., The Nuṣayrī-ʿAlawīs: An Introduction to the Religion, History, and Identity of the Leading Minority in Syria (Leiden, 2010), pp. 29–30CrossRefGoogle Scholar.
45 Ibid., 33 f., 253 f.
46 Khaṣībī, al-Risāla al-rāstbāshīya, in: Rasāʾil al-ḥikma al-ʿalawīya (Beirut, 2006), 34 f.
47 Idem, Fiqh al-risāla al-rāstbāshīya, in: Rasāʾil al-ḥikma, 108; cf. Friedman, Nuṣayrī-ʿAlawīs, 126 ff.
48 M. J. Kister, “Djābir b. ʿAbd Allāh al-Anṣārī,” EI 2, suppl., p. 231.
49 See: Kohlberg, E., “An Unusual Shīʿī Isnād,” IOS, V (1975), pp. 142–149Google Scholar.
50 Kohlberg, “Authoritative Scriptures,” p. 304.
51 Kister, “Djābir,” p. 231a.
52 K. al-ridda, p. 136.
53 Tūsī, Ikhtiyār, p. 45; al-Qummī, Abū l-Ḥasan ʿAlī b. Ibrāhīm, Tafsīr, (ed.) al-Jazāʾirī, Ṭayyib al-Mūsawī (Najaf, 1967), ii, p. 147Google Scholar; al-Ḥillī, Ḥasan b. Sulaymān, Mukhtaṣar Baṣāʾir al-darajāt, (ed.) al-Muẓaffar, Mushtāq (Qumm, 2000), pp. 151, 155Google Scholar.
54 Saʿd b. ʿAbdallāh, Maqālāt, p. 56.
55 Ibid. Saʿd b. ʿAbdallāh lists another pentadist sect which posits ʿAlī, rather than Muḥammad, as the maʿnā founded by Bashshār al-Shaʿīrī, a devotee of Jaʿfar al-Ṣādiq, known as the ʿAlyāʾiyya (ibid., pp. 59–60); cf. Halm, Gnosis, p. 218 ff.
56 Saʿd b. ʿAbdallāh, Maqālāt, pp. 56–57.; cf. Halm, Kosmologie und Heilslehre, pp. 157 ff. and idem, Gnosis, pp. 218 ff.
57 Abū Ḥātim al-Rāzī, K. al-zīna, p. 307: “The Mukhammisa are those who claim that Muḥammad, ʿAlī, Fāṭima, al-Ḥasan and al-Ḥusayn, all five of them, are one thing [shayʾun wāḥidun]. The Spirit dwells in them equally, preferring neither one over the other. They claim that Fāṭima was not a woman and are loathe to speak of Fāṭima in feminine terms and thus call her Fāṭim”.
58 Zahr al-maʿānī in: W. Ivanow, Rise of the Fatimids (Calcutta, 1942), p. 64; cf. idem, Ismaili Literature, pp. 77 ff.
59 Ivanow, Rise, p. 256, n. 2; cf. Poonawala, I. K., Biobibliography of Ismāʿīlī Literature (Malibu, Calif., 1977), pp. 133 ffGoogle Scholar.
60 Thus, in a Nuṣayrī work authored by Maḥmūd Baʿamrā, Muḥammad al-Bāqir appears to his disciple Dhū l-Dawr in the prayer niche (miḥrāb) of the Prophet's mosque in Medina seated before in five bodily manifestations: Muḥammad, Fāṭir (i.e., Fāṭima), al-Ḥasan, al-Ḥusayn and Muḥsin. See: R. Strothmann, Esoterische Sonderthemen, pp. 19–20 (§ 71) and Halm, Gnosis, p. 387, n. 689.
61 UaK may have also appropriated the Mazdakite notion of the kūdak-i dānā—i.e., the ‘omniscient child’. See: Yarshater, E., “Mazdakism,” in Cambridge History of Iran, (ed.) Yarshater, E. (Cambridge, 1983), iii/2, p. 1014Google Scholar and Madelung, W., Religious Trends in Early Islamic Iran (Albany, 1988) pp. 8–9Google Scholar.
62 The infancy gospel has long been notorious for its extremely complex and broad textual history: versions of Inf. Ps.-Th. appear in at least 13 different languages. The pioneering study is that of Gero, S., “The Infancy Gospel of Thomas: A Study of the Textual and Literary Problems”, Novum Testamentum, XIII (1971), pp. 46–80Google Scholar. This essay, however, should now be read in tandem with the more updated studies of Voicu, S., “Verso il testo primitive dei . . . ‘Racconti dell’ infanzia del Signore Gesù,” Apocrypha, IX (1998), pp. 7–95CrossRefGoogle Scholar and Chartrand-Burke, T., “The Greek Manuscript Tradition of the Infancy Gospel of Thomas,” Apocrypha, XIV (2003), pp. 129–151CrossRefGoogle Scholar.
63 On which, see: Gijsel, J., Libri de nativitate mariae I, Psuedo-Matthaei Evangelium: textus et commentaries, Corpus Christianorum, Series Apocryphum 9 (Turnhout, 1997)Google Scholar.
64 Rousseau, A. and Doutrieleau, L. (eds.), Contre les heresies, Livre I, Tome II, Sources chrétiennes 264 (Paris, 1979), pp. 288–289Google Scholar. On the Marcosians and their founder, see: Förster, Niclas, Marcus Magus: Kult, Lehre und Gemeindeleben einer valentinianischen Gnostikergruppe (Tübingen, 1999)Google Scholar.
65 E.g., see: al-Ṭabarī, , Jāmiʿ al-bayān ʿan taʾwīl āy al-Qurʾān, (eds.) Shākir, A. M. and Shākir, M. M. (Cairo, 1954), i, pp. 121–122Google Scholar (no. 140) and n. 2 thereto; al-Thaʿlabī, , ʿArāʾis al-majālis, (trans.) Brinner, W. M. (Leiden, 2002), pp. 647–8Google Scholar; al-Kisā’ī, , The Tales of the Prophets, trans. Thackston, Wh. (Boston, 1978), pp. 332–333Google Scholar; al-Ṭarafī, Ibn Muṭarrif, The Stories of the Prophets, (ed.) Tottoli, Roberto, Islamkundliche Untersuchungen 258 (Berlin, 2003), pp. 170.ult-171 (Ar.)Google Scholar, and see also the references cited by Tottoli in his annotation of al-Ṭarafī's text in ibid., p. 99 (§445). Cf. also UaK 60 ff., where Muḥammad al-Bāqir explains the basmala to Jābir al-Juʿfī.
66 Bābawayh, Ibn, Maʿānī al-akhbār, (ed.) al-Ghaffārī, ʿA. A. (Tehran, 1959), pp. 45–46Google Scholar; idem, al-Tawḥīd, (ed.) Hāshim al-Ḥusaynī al-Ṭihrānī (Tehran, 1967), p. 236; al-Majlisī, , Biḥār al-anwār (Tehran, 1954), xiv, p. 286Google Scholar. Cf. Ayoub, M., “Towards an Islamic Christology: An Image of Jesus in Early Shīʿī Literature,” MW, LXVI (1976), pp. 163–188Google Scholar.
67 The namesake of the Zaydite sect known as the Jārūdīya, Abū l-Jārūd al-Hamdānī (d. ca. 150/767), was, in addition to a fervent supporter of Zayd b. ʿAlī's revolt, a close disciple of Muḥammad al-Bāqir, from whom he reputedly also transmitted a tafsīr (See: W. Madelung, “Abū Jārūd,” EIr, i, p. 327; Modarressi, Tradition, i, p. 122). Kathīr b. ʿAyyāsh, who precedes him in the isnād, is the main transmitter of both his tafsīr and his notebook of ḥadīth (known as al-Aṣl); see: Modarressi, Tradition, i, pp. 122–123.
68 i.e., the next three letters of the alphabet.
69 As above, the following three letters of the alphabet: ḥāʾ, ṭāʾ and yāʾ.
70 G. Weil and G. S. Colin, “Abdjad,” EI 2, i, p. 96.
71 Sike, H., Evangelium Infantiae vel liber apocryphus de Infantia Salvatoris ex manuscripto edidit ac Latina versione et notis illustravit (Ultrecht, 1697)Google Scholar. All references to the Ar. Inf. below refer to the edition published in Thilo, Codex Apocryphus Novi Testamenti (Leipzig, 1832), i, pp. 63–158 (with the emendations of E. Rödiger)
72 See the edition and study of Provera, Mario, Il vangelo arabo dell’ infanzia (Jerusalem, 1973)Google Scholar. G. Graf, however, lists several more MSS, in both Arabic and Garshūnī, of which the latter, to my knowledge, remains unstudied; see: Geschichte der chistlichen arabischen Literatur (Vatican City, 1944–1953), i, pp. 226–227. Several other studies of the Arab. Inf. omit the school anecdote as well. Cf. Burmeister, K. H. S., “Fragments of an Arabic Version of Two Infancy Gospels”, Studia Orientalia Christiana Collectanea, VII (1962), pp. 103–114CrossRefGoogle Scholar; Noja, Sergio, “L’Évangile arabe apocryphe de Thomas, de la ‘Biblioteca Ambrosiana’ de Milan (G 11 sup),” in Biblische und Judistische Studien: Festschrift für Paolo Sacchi, (ed.) Vivian, A. (Paris, 1990), pp. 681–690Google Scholar; idem, “À propos du text arabe d'un évangile apocryphe de Thomas de la Ambrosiana de Milan”, in YAD-NAMA: Im memoria di Alessandro Bausani, (eds.) B. S. Amoretti and L. Rostagno (Rome, 1991), i, pp. 335–341.
73 Wright, William, Contributions to the Apocryphal Literature of the New Testament (London, 1865), pp. 11–16Google Scholar (Syr.), pp. 6–11 (Eng.). Wright based his edition on an important sixth-century MS (British Library, Add. 14484, foll. 12v-16r); however, another important, early MS (Göttingen, Universitätsbibliothek, Syr. 10), dated to the fifth or sixth century a.d., remained neglected for some time until the study of W. Baars and J. Helderman collated the manuscript with Wright's edition in their study, “Neue Materialien zum Text und zur Interpretation des Kindheitsevangelium des Psuedo-Thomas,” OrChr, LXVII (1993), pp. 191–226 and LXVIII (1994), pp. 1–32.
74 In this story, as in UaK 12–13, the Christ-child is sent by his family to school at age five to receive instruction in the alphabet, whereupon the lesson is interrupted by the teacher's failed attempt to teach Jesus the letters alpā and bītā; see: Budge, E. A. Wallis, The History of the Blessed Virgin Mary and the History of the Likeness of Christ (London, 1899), i, p. 71Google Scholar (Eng.) and ii, pp. 66–67 (Syr.). Cf. Mimouni, S. C., “Les Vies de la Vièrge; État de la question”, Apocrypha, V (1994), pp. 239–246Google Scholar. A Jacobite Life of Mary also exists preserved in various, hitherto unedited manuscripts. See: Mignana, A., “The Vision of Theophilus, Or the Book of the Flight of the Holy Family in Egypt,” Bulletin of the John Rylands Library, XIII (1929), pp. 383–474Google Scholar and S. C. Mimouni, op. cit., p. 239 and n. 128 thereto.
75 Halm, H., “‘Das Buch der Schatten’: Die Mufaḍḍal-Tradition der Ġulāt und die Ursprünge des Nuṣairitiertums (I)”, Der Islam, LV (1978), pp. 219 ff.Google Scholar; Capezzone, L., “Il Kitāb al-Ṣirāṭ attributto a Mufaḍḍal Ibn ʿUmar al-Ğuʿfī: Edizione del ms. unico (Paris, Bibliothèque Nationale Ar. 1449/3) e studio introduttivo”, RSO LXIX (1995), pp. 295–416Google Scholar; Modarressi, Tradition, i, pp. 333–337; Friedman, Nuṣayrī-ʿAlawīs, pp. 241–247.
76 Oriens, XXV/XXVI (1976), p. 555.
77 To my own observations about the significance of the Ibn Sabaʾ materials for dating UaK, one may now add the fascinating observations of Bernd Radtke. Radtke has demonstrated that there exists a profound overlap between the vocabulary utilised in the so-called Jābir-Apocalypse identified within UaK by Halm, especially the sections containing discourses on the macro- and micro-cosmos, and between the vocabulary utilised in the theosophical writings of the mystic al-Ḥakīm al-Tirmidhī (d. ca. 300/910). Ḥakīm's case is somewhat exceptional and, in Radtke's estimation at least, the possibility of cross-pollination between his writings and those of the ghulāt is slim. Although his “theosophy,” as Radtke puts it, “had no immediate sequel” (Ṣūfīs of the generations immediately following him almost entirely neglected his thought), al-Ḥakīm's writing offers important evidence that many of the ideas present in Halm's earliest dated unit of UaK, the Jābir-Apocalypse, could have plausibly been originated within al-Ḥakīm's lifetime or shortly thereafter, i.e. the late third/ninth century or slightly later. See: Radtke, B., “Iranian and Gnostic Elements in Early Taṣawwuf: Observations concerning the Umm al-Kitāb,” in Proceedings of the First European Conference of Iranian Studies, part 2: Middle and New Iranian Studies, (eds.) Gnoli, G. and Panaino, A. (Rome, 1990), pp. 519–530Google Scholar.
78 ʿAbdallāh-i Ṣabbāḥ is clearly an attempt to render ʿAbdallāh-i Sabaʾ into a form resembling the name of the first Nizārī dāʿī of Alamūt, Ḥasan-i Ṣabbāḥ. As Ivanow notes, “the form Ṣabbāḥ is merely one of the ‘corrections’ of the copyists” (“Ummu'l-kitāb,” p. 7; cf. idem, “Notes,” p. 428 and n. 2 thereto; van Ess in Der Islam, XLVI, 1970, p. 97; Tijdens, p. 279).
79 In the text: wāsiʿ al-maqālāt; that wāsiʿ is a corruption of sabʿ as made clear by the following clause; cf. Halm, Gnosis, p. 369 n. 235.
80 See: Q. 40:15.
81 Although corrupt, this passage does not in fact seem to have been an interpolation. UaK in actuality only mentions seven, not ten, dīvāns, but it does speak of 10 discourses late into the ‘Jābir-Apocalypse’ (UaK 247); cf. Halm, Gnosis, p. 369 and n. 237 thereto.
82 In the text: dar maḥalla-ye quraysh va ibn-i hāshim dar khāna-ye ʿAbd al-Manāf.
83 Ṭūsī lists a certain Jaʿfar b. Ibrāhīm al-Juʿfī as a companion of Muḥammad al-Bāqir (Rijāl, p. 129.-4), with whom van Ess and Halm attempt to identify with the person named here in the text (cf. Halm, Gnosis, p. 369 n. 239 and van Ess in Der Islam, XLVI, 1970, p. 96). However, Ṭūsī's text might have also been corrupted; one should perhaps read the nisba as al-Jaʿfarī rather than al-Juʿfī. Jaʿfar b. Ibrāhīm al-Jaʿfarī—called ‘al-Jaʿfarī’ because he was a descendent of ʿAlī's revered brother, Jaʿfar b. Abī Ṭālib—was the companion of three imāms: ʿAlī Zayn al-ʿĀbidīn, Muḥammad al-Bāqir (if the emendation to Ṭūsī's text is accepted), and Jaʿfar al-Ṣādiq (see: Ṭūsī, Rijāl, 111.7, 175.4). It seems, though, that it is only from ʿAlī Zayn al-ʿĀbidīn that one finds a significant body of reports transmitted on his authority; cf. Ibn Ḥajar al-ʿAsqalānī, Lisān al-mīzān (Beirut, 1987), ii, p. 135 and Majlisī, Biḥār, lxxii, p. 40 and lxxiii, p. 287. Hence, the text of UaK might have read ‘Jaʿfar al-Jaʿfarī’, but with the nisba corrupted to read ‘al-Juʿfī’ due to a later copyist's attempt to ‘correct’ the text so as to resemble the nisba of the most prominent companion of al-Bāqir in UaK, Jābir al-Juʿfī. On the other hand, it is also possible that the text jumps chronologically here to refer to the actions of Abū al-Muṭṭalib Jaʿfar b. Muḥammad b. al- Mufaḍḍal al-Juʿfī, a grandson of Mufaḍḍal b. ʿUmar al-Juʿfī (fl. late second/eighth century) and a scholar to whom is attributed a book titled Kitāb ādāb al-dīn. For example, see: al-Ḥasan b. Shuʿba al-Ḥarrānī, Ḥaqāʾiq asrār al-dīn, in Majmūʿat al-Ḥarrāniyyīn, vol. 1, Silsilat al-Turāth al-ʿAlawī 4 (Beirut, 2006), pp. 117.9, 136.1, 142.1. The chronology is somewhat problematic but not impossible. This latter identification would place the provenance UaK squarely in proto-Nuṣayrī circles. I owe this latter observation to Mushegh Asatryan of Yale University.
84 Unknown. Ivanow speculates that he may be the son of the Ḥasanid ʿAbd al-ʿAẓīm b. ʿAbdallāh b. ʿAlī, a revered Shīʿī ascetic buried in Rayy for whom the Būyid vizier of the city, al-Ṣāḥib Ibn ʿAbbād, wrote an epistle in praise of his virtues. However, as a companion of the imāms Muḥammad al-Jawād (203–20/818–35) and ʿAlī al-Hādī (220–54/835–68), no son of his could have possibly been a contemporary of Hārūn al-Rashīd. Cf. W. Madelung, “ʿAbd al-ʿAẓīm al-Ḥasanī,” EIr, i, pp. 96–97. Halm's attempt to identify this ʿAlī b. ʿAbd al-‘Aẓīm with the third/ninth-century compiler of UaK is a bit more inspired (Gnosis, p. 123), but is even more speculative and just as unconvincing.
85 Reading with Ivanow's text: na-hama-ye muʾminān ʿilm be-dīn-i rowshānī bar-tābad and rejecting Halm's emendation of na-hama to be-hama (see: Gnosis, p. 369 n. 243).
86 This appears to allude to the well-known Imāmī tradition (which has many iterations), “Our (i.e., the imāms’) teaching [ḥadīthunā] is difficult [ṣaʿb], even arduous [mustaṣʿab]. None can bear it except a prophet sent by God [nabī mursal], an archangel [malak muqarrab], or a servant whose heart God has tested for faith [ʿabdun imtaḥana llāhu qalbahu li-l-īmān]”. See: al-Ṣaffār, Muḥammad b. al-Ḥasan, Baṣāʾir al-darajāt, (ed.) Kūchabābaghī, M. (Tehran, 1983), pp. 20 ffGoogle Scholar. and Kulaynī, Kāfī, ii, pp. 253 ff. On its role in the Imāmī tradition, see: Amir-Moezzi, A., “Seul l'homme de Dieu est humain: Théologie et anthropologie mystique à travers l'exégèse imamite ancienne (aspects de l'imamologie duodécimaine iv),” Arabica, XLV (1998), p. 250Google Scholar.
87 Prs. mālik-i taʿālā; the most common designation for God throughout UaK.
88 Āmina is, of course, the name of the Prophet Muḥammad's mother who, according to the Islamic tradition, died soon after his birth. Muḥammad al-Bāqir's real mother was named not Āmina, but rather Fāṭima Umm ʿAbdallāh, and was a daughter of al-Ḥasan b. ʿAlī (making al-Bāqir the direct descendent of both al-Ḥasan and al-Ḥusayn b. ʿAlī).
89 In the text: farā īzadī. There is also an unlikely variant reading: divine statutes (farāʾiż īzadī); see: Ivanow, “Ummu'l-kitāb,” p. 180.
90 Prs., ḥikmat-o ʿilm-i taʾyīdī; an allusion to the divine knowledge granted to the qurʾānic Jesus (see: Q. 2:87 and 257).
91 Cf. Tijdens, p. 279. The Arabic alphabet contains 28 letters in total, 29 if one counts the ligature lām-alif.
92 Ibid. A similar tablet often appears in the literature of other Shīʿīs as well. According Imāmī-Shīʿī accounts, Jābir al-Anṣārī allegedly saw in Fāṭima's house a tablet including the names of the Twelve Imāms (Kohlberg, “An Unusual Shīʿī Isnād”, p. 144 n. 10).
93 In the text: alif khodāvand ast va-lām bālā-ye ān muḥammad ast. The pericope is without an obvious interpretation since what exactly is meant by the “lām above the alif” is not entirely clear. Halm (see: Gnosis, 379 n. 253) suggests tentatively that the hamza 
is intended; cf. Tijdens, p. 280.
94 In the Arabic script, the name of the letter 
is written using three letters as 
; the diacritic dot (Ar. nuqṭa) referred to here appears above the final letter fāʾ 
.
95 The nūn comes from the initial letter in word for diacritic, nuqṭa.
96 Perhaps this passage represents a fusion of pentadist belief with the identification of God with letters of the alphabet ascribed to the Mughīrīya See: Wasserstrom, S., “The Moving Finger Writes: Mughīra b. Saʿīd's Islamic Gnosis and the Myth of Its Rejection”, History of Religions, XXV (1985), pp. 15 ffGoogle Scholar.
97 Reading with Halm (Gnosis, p. 129) “peykar-i vey”, rather than “be-yak rūy,” as in the text.
98 Tijdens (pp. 279–280) regards this passage as a later interpolation of a copyist, pointing to the tautological nature of its re-explanation of the significance of alif.
99 Again, an expansion on the meaning of alif that may be a later interpolation; see: Tijdens, pp. 285 ff.
100 This passage seems to provide evidence that the ostensible differences between the Mukhammisa and the ʿAlyāʾiyya—i.e., according priority to either Muḥammad or ʿAlī—may in fact amount to a misleading distinction between what must have certainly amounted to two mutually intelligible discourses.
101 Viz., the diacritic of the letter alif 
only becomes apparent when it appears over the fāʾ 
when its name is written out fully as alif 
.
102 Ahrīman, the traditional name of God's adversary in the Zoroastrian religion (cf. J. Duchesne-Guillemin, “Ahriman,” EIr, i, pp. 670 ff.), should be understood here as referring to the devil insofar as UaK speaks of the devil as Ahrīman a total of six times (see: Tijdens, p. 292). Bausani suggest reading kandū as Kondav and, thus, as a variant of Konī-Dēv, the sinister general of Ahrīman in Manichaen myth (see: Religion in Iran, pp. 151–152).
103 A somewhat obscure passage that has been translated differently: “. . . il logos dei Credenti, poiché mediante il logos si profferisce [discorso]” (Filippani-Ronconi, p. 9); “. . . das Wort der Frommen, das durch das Wort hervorgerufen wurde” (Tijdens, p. 292); “. . .das Reden der Gläubigen, denn durch das Reden hat er verkündigt” (Halm, Gnosis, p. 130).
104 In the text, the Persian reads: “alif bozorgtar ast yā oshtor?”—lit. “Is alif bigger or a camel?” Here, oshtor, or camel, clearly arose from a misguided attempt at the translation of jamal, which in Arabic means ‘camel’, but which in other Semitic languages (Hebrew, Aramaic, Syriac, etc.) is also the name for the equivalents of the Arabic letter jīm 
. In Syriac, the corresponding letter is gamal 
. Aside from the structural similarities discussed above, this passage provides the strongest evidence of a Syriac Vorlage for the Bāqir story. This passage, as noted by van Ess (Der Islam, XLVI, 1970, p. 96), also provides the strongest textual evidence that UaK was originally an Arabic composition later translated into Persian. Pace Ivanow, Guide, p. 193.
105 Prs., oshtor.
106 Prs., oshtor.
107 The third/ninth-century Arabic translation of the so-called Theology of Aristotle also employs the term al-nafs al-nāṭiqa as a translation of λοιζομένην φυχήν of Plotinus, Enneads, iv 7, 85, 17; cf. Radtke, “Iranian and Gnostic Elements”, p. 525.
108 In the text, “ke agar alif rā chandān ke be-kashī be-shayād kashīdan”; the passage may be corrupt. Cf. Filippani-Ronconi, p. 10 and Tijdens, p. 103.
109 In the text: sajda; the text should probably read instead, “ashhadu” (see: Tijdens, pp. 300f).
110 Al-Salsal is to be identified with Salmān al-Fārisī. Tijdens's translation also adds Abū l-Khaṭṭāb to the litany of names invoked (op. cit., p. 303); however, it does not appear in Ivanow's text or his list of textual variants (cf. Ivanow, “Ummu'l-kitāb,” p. 108).
111 Prs., oshtor.
112 Viz., the highest heaven of the divine dīvāns; cf. UaK 96–119 and Halm, Gnosis, pp. 149 ff.
113 Prs., Bā Dharr; cf. UaK 131–3.
114 Properly speaking, the letter ṭāʾ actually consists of three, not two, letters: ṭāʾ 
, alif 
, and hamza 
.
115 Halm's translation reads “Diese sieben [sic.] Punkte . . .” (Gnosis, p. 131).
116 As noted by Halm (Gnosis, p. 370 n. 260), the ṭāʾ, which has a value of 9, has been dropped and replaced with the tāʾ—by which the tāʾ marbūṭa 
(lit., tied tāʾ) occurring at the end of the word nuqṭa 
is meant—in order to make the equation work.
117 This passage, as noted first by Ivanow (“Notes”, p. 423 n. 3), seemingly must originate after the advent of the minor occultation in 260/874. It is odd, however, insofar as it speaks of twelve “from the loins of ʿAlī” rather than Muḥammad, since it should read eleven if ʿAlī is counted as the first imām. Should this be interpreted esoterically (i.e., ʿAlī is Muḥammad, Muḥammad is ʿAlī, etc.), be attributed to an authorial error, or something else? The Jābir-Apocalypse also speaks of “the twelve lights of the house of the Prophet [davāzdah nūr-i ahl-i bayt]” (UaK 71), a passage which both Tijdens (op. cit., p. 313) and Halm (Gnosis, p. 142) dismiss, without much discussion, as a later interpolation.
118 Viz., the tāʾ marbūṭa; see note above.
119 Or, “the Holy Spirit”: see note above.
120 i.e., the clothing worn by the inhabitants of heaven.
121 Cf. Halm, Kosmologie und Heilslehre, p. 151.
122 The text reads “sajda”, which is incomprehensible; read instead, with Tijdens (op. cit., p. 347; cf. Halm, Gnosis, p. 370 n. 261), “ashhadu”.
123 Reading with Tijdens (op. cit., p. 347) and Halm (Gnosis, p. 371 n. 262) “wa-walīhi” rather than “wa-ālīhi” as in the text.
124 Abū l-Khaṭṭāb's name here appears likely as a result of a later Khaṭṭābī redactor; see: Halm, Gnosis, p. 371 n. 264.
125 This passage is entirely in Arabic: anta l-awwal wa-anta l-ākhir, anta al-ẓāhir wa-anta al-bāṭin, wa-anta bi-kulli shay’in ʿalīmun.
126 Or, “this man is possessed [dīvāna]!”
127 Jābir b. Yazīd al-Juʿfī (d. 128/746 or 132/750).
128 See note above.
129 One of the prominent qurrāʾ forming the opposition to ‘Uthmān and, later, a celebrated partisan of ʿAlī who died during the caliphate of Mu ʿāwiya b. Abī Sufyān. His appearance here poses an intractable chronological difficulty, for Muʿāwiya's reign ended a mere three years after Muḥammad al-Bāqir's birth in 57/677 whereas the events here putatively occur while the imām was five years old.
130 Lit., “the standing” or “the riser”; cf. Madelung, “Ḳāʾim Āl Muḥammad,” EI 2, iv, p. 456. A common name of a messianic figure often identified with the imām or mahdī among the Shīʿa, the full significance of the term as employed is not entirely clear, but it should not be conflated with its later, Ismāʿīlī expansion: “al-qāʾim bi-amr allāh”—i.e., “he who undertakes the command of God”. Antecedents of the term can also be found in Samaritan and Gnostic texts; see: Halm, Gnosis, pp. 362–363 at n. 77.
131 Cf. the Nuṣayrī work of al-Ṭabarānī, Maymūn b. al-Qāsim (d. 426/1034–5), Majmūʿ al-aʿyād, (ed.) Strothman, R. in Der Islam, XXVII (1944), p. 381.7Google Scholar in which the death of Ibn Sabaʾ is described in terms of a ‘trial’ (Ar. miḥna).
132 Abū l-Khaṭṭāb's execution and activities transpired, not during ʿAlī's caliphate, but during the imāmate the sixth imām, Jaʿfar al-Ṣādiq. This passage could have been the handiwork of a later Khaṭṭābī redactor (Halm, Gnosis, p. 371 n. 371).
133 Having been born in 57 a.h., Bāqir as a five year old child would be speaking here in the year 62 a.h. 62 years according to the lunar hijrī-calendar equals 60 solar years, thereby implying that the coming of the qāʾim would correspond to a thousand years after the hijra? See: Halm, Gnosis, p. 371 n. 371.
134 The mention of a son of Ibn Sabaʾ named Ṭālib is solely attested to by UaK in this passage; however, Ibn Babawayh in his Risāla fī l-iʿtiqādāt includes a statement made by the Imāmī mutakallim Zurāra b. Aʿyān (d. ca. 149–9/765–6) to Jaʿfar al-Ṣādiq that a descendent of ʿAbdallāh b. Sabaʾ adhered to the doctrine that God had delegated his powers to the imāms, i.e., tafwīḍ. See: Babawayh, Ibn, Risāla fī l-iʿtiqādāt, in: Muṣannafāt al-Shaykh al-Mufīd (Qumm, 1993), v, p. 100Google Scholar
135 Here the text identifies ‘madhhab-i fidāʾī’ of Ṭālib the Nizārī-Ismāʿīlī ‘Assassins’ of Syria, which would date the passage to at least the second half of the sixth/twelfth century. Those who argue for the antiquity of UaK regard it as a late addition to the older strata of UaK. See: Tijdens, pp. 361 ff. and Halm, Gnosis, p. 371 n. 273. However, there is nothing integral to the structure of the text that warrants this conclusion.
136 The identification of the followers of Abū l-Khaṭṭāb with followers of the descendents of Ismāʿīl b. Jaʿfar al-Ṣādiq goes back to at least the second/eighth century; See: Nawbakhtī, Firaq, pp. 55–56; Saʿd b. ʿAbdallāh, Maqālāt, pp. 81–82.
