Published online by Cambridge University Press: 24 December 2009
The Imāmī Shī'ī theory of the imāmate evolved gradually during the first Islamic century and was given a definitive shape in the middle of the second/eighth century by Hishām b. al-Ḥakam. For the next 100 years or so, until the death in 260/874 of the eleventh Imām, al-Ḥasan al-'Askarī, no significant changes seem to have been introduced. Only in the mid-fourth/tenth century does a major addition appear in the form of a doctrine: it is the belief that there are 12 Imāms, the last of whom remains in a state of concealment (ghayba) until his ultimate return as Mahdī, or Qā'im. This ghayba is divided into two periods: a shorter, ‘lesser’ ghayba (al-ghayba al-ṣughrā), lasting from 260/874 to 329/941, during which the Imām was represented on earth by four successive safīrs; and a longer, ‘greater’ ghayba (al-ghayba al-kubrā), whose duration is known only to God. It is this doctrine which distinguishes Twelver Shī'ism from the earlier Imāmiyya, and it io worth examining in some detail ite origina and the-main-stages of its development.
1 See the article ‘Hishām b. al-Ḥakam’, by W. Madelung, in EI, second ed.
2 Watt, W. Montgomery (‘The Rāfiḍites: a preliminary study’, Oriens, XVI, 1963, 119 f.)Google Scholar has pointed out that the term ‘Imāmiyya’ occurs in a Zaydī source used by Abū 'l-Ḥasan al-Ash'arī (d. 324/935–6) (Maqālāt al-islāmiyyīn, ed. Ritter, H., Istanbul 1929–1933, 64)Google Scholar, and has suggested that it was first employed before 850. This suggestion appears to be corroborated by an additional source, the Kitāb naqḍ al-'uthmāniyya by the Baghdādī Mu'tazīlī Abū Ja'far al-Iskāfī (d. 240/854). At one point al-Iskāfī dissociates himself from the Imāmiyya whose obduracy, he says, leads them to ‘deny well-known things’ (The text is reprinted from Ibn Abī 'l-Ḥadīd's Sharḥ nahj al-balāgha at the end of al-Jāḥiẓ's Kitāb al-'uthmāniyya, ed. Hārūn, ‘Abd al-Salām Muḥammad, Cairo, 1374/1955, 318).Google Scholar The terms qaṭ'iyya and ahl al-nasaq (the latter used almost exclusively by al-Nāshi’ al-Akbar (d. 293/906); see van Ess, J., Frühe mu'tazilitische Häresiographie, Beirut, 1971, 28 f.)Google Scholar are older and broader than ‘Imāmiyya’. The term ‘Ithnā-'ashariyya’ was probably first used around 1000. It does not appear in the Fihrist of the Imāmī al-Nadīm (d' 380/990) (cf. Seilheim, R., Israel Oriental Studies, II, 1972, 428–32)Google Scholar, but is employed by the rabidly anti-Shī'ī ‘Abd al-Qāhir al-Baghdādī (d. 429/1037) to refer to a subsection of the Imāmiyya (al-Farq bayna 'l-firaq, ed. Muḥammad Muḥyi 'l-Dīn ‘Abd al-Ḥamīd, Cairo, 1384/1964, 23, 64). With the increasing predominance of the Twelvers, the terms ‘Imāmiyya’ and ‘Ithnā-'ashariyya’ gradually became synonymous (see Friedlaender, I., ‘The heterodoxies of the Shiites in the presentation of Ibn Ḥazm’, JAOS, XXIX, 1908, 151).Google Scholar
3 For a detailed analysis of the relationship of the two sources, see Madelung, W., ‘Bemerkungen zur imamitischen Firaq-Literatur’, Der Islam XLIII, 1–2, 1967, 37 ff.Google Scholar
4 Al-Nawbakhtī, , Kitāb firaq al-shī'a, ed. Ritter, H., Istanbul, 1931, 90–3Google Scholar; Sa'd b. ‘Abdallāh, Kitāb a1-maqālāt wo, 'l-firaq, ed. Mashkūr, M. J., Tehran, 1383/1963, 102–6.Google Scholar
5 See ‘Abdallāh, Sa'd B., op. cit., 103.Google Scholar
6 cf. Goldziher, I., Abhandlungen zur arabischen Philologie, II. Das Kitâb al-Mu'ammarîn des Abû Ḥâtim al-Siģistânî, Leiden, 1899, pp. lxii–lxix.Google Scholar
7 The sixth in al-Nawbakhtī, 's list (op. cit., 84 f.)Google Scholar, the eleventh in the Kitāb al-maqālāt wa 'l-firaq (114).Google Scholar
8 According to the Kitāb firaq al-shī'a, Muḥammad was two years old when his father died; the information in the Kitāb al-maqālāt wa 'l-firaq is that he was grown up (bāligh) at the time (loc. cit.).
9 Al-Ash'arī, , op. cit., 17Google Scholar f., 30. Al-Ash'arī (ibid., 14) also mentions a sect of ghulāt who believe in the same 12 persons but who claim that God resides in each of them. It should be noted that while all Twelver Shī'ī doctors agree that Muḥammad b. al-Ḥasan is the hidden Imām, there are traditions according to which it is forbidden to mention his name (see e.g. al-Kulīnī, , Uṣūl al-kāfī, ed. ‘al-Ghaffārī, Alī Akbar, Tehran, 1375/1955–1956–1377/1957–1958, 1, 332 f.).Google Scholar This principle, however, was not universally observed (cf. e.g. Bābawayhi, Ibn, A Shī'ite creed, trans. Fyzee, Asaf A. A., London, 1942, 98).Google Scholar An attempt at harmonization is made by explaining that the Qā'im has two names: one, Aḥmad, is made known, and the other, Muḥammad, remains a secret. See al-Kāshānī, Muḥsin, al-Nawādir fī jam' al-aḥādīth, Tehran, 1380/1960, 148.Google Scholar
10 Al-Barqī, , Kitāb al-maḥāsin, ed. al-Muḥaddith, Jalāl ai-Dīn al-Ḥusayuī, Tehran, 1370/1950–1951, 3–15.Google Scholar
11 Bābawayhi, Ibn, Kitāb al-khiṣāl, Najaf, 1391/1971, 436–51.Google Scholar
12 Al-Barqī, , op. cit., 332 f.Google Scholar
13 ‘al-Qummī, Alī b. Ibrāhīm, Tafsīr, ed. al-Jazā'irī, Ṭayyib al-Mūsawī, Najaf, 1386/1966–1967–1387/1967–1968, II, 44 f.Google Scholar
14 See especially al-Kulīnī, , op. cit., I, 328 ff.Google Scholar, 525 ff. The Khiḍr-tradition appears on p. 525 f., and traditions on two concealments on p. 339 f.
15 See Bābawayhi, Ibu, Ikmāl al-dīn, Tehran 1301/1883–1884, 204Google Scholar, cited by al-Majlisī, , Biḥār al-anwār [= Biḥār], [Persia], 1305/1887–1888–1315/1897–1898, XIII, 236Google Scholar; al-Ṭūsī, Abū Ja'far, Kitāb al-ghayba, ed. al-Ṭihrānī, Āghā Buzurg, Najaf, 1385/1965–1966, 285Google Scholar, cited in Biḥār, XIII, 237Google Scholar; Muḥsin al-Kāshānī, , op. cit., 199 f.Google Scholar And cf. Kitāb Muḥammad b. al-Muthannā, in al-Uṣūl al-arba'u mi'a, MS Tehran University, no. 962, fol. 53b (where the Qā'im is said to be followed by 11 Mahdīs). Al-Majlisī (loc. cit.) suggests two possible interpretations of these traditions: the 12 Mahdīs might be the Prophet and the 11 Imāms, whose rule would follow that of the Qā'im; or else these Mahdīs might be the legatees (awṣiyā') of the Qā'im, who would provide guidance to the community with the other Imāms who will have come back to earth (raja'ū).
16 See al-Nu'mānī, , Kitāb al-ghayba, Tehran, 1318/1900–1901, 2.Google Scholar
17 ibid., 4 f.
18 al-Rāzī, Al-Khazzāz, Kifāyat al-athar, [Persia], 1306/1888, 289.Google Scholar
19 Bābawayhi, Ibn, op. cit., 3 ff.Google Scholar
20 Al-Nu'mānī, , op. cit., 41.Google Scholar See also al-Ṭūsī, , op. cit., 96.Google Scholar The Shī'ī scholar Muḥammad Ḥusayn al-Ṭabāṭabā'ī expresses reservations about the soundness of this exegesis. See his al-Mīzān fī tafsīr al-Qur'ān, DC, Tehran, 1379/1959–1960, 286.Google Scholar
21 Referring either to the 12 Israelites mentioned in Qur'ān V, 12, or to the 12 Companions chosen by Muḥammad. See Bābawayhi, Ibn, Kitāb al-khiṣāl, 463 f.Google Scholar
22 Al-Nu'māuī, , op. cit., 40.Google Scholar
23 ibid., 29–31; Bābawayhi, Ibn, Ikmāl al-dīn, 179 f.Google Scholar; the references in my article ‘An unusual Shī'ī ianād’, Israel Oriental Studies, V, 1975, p. 144, n. 10.Google Scholar
24 Al-Nu'mānī, , op. cit., 24.Google Scholar
25 ibid., 42.
26 Biḥār, IX, 120 ff.Google Scholar See also Bābawayhi, Ibn, Kitāb al-khiṣāl, 445 ff.Google Scholar
27 See the article ‘Ghadīr Khumm’, by L. Veccia Vaglieri, in EI, second ed. The most exhaustive treatment of the subject from a Shī'ī point of view is that of ‘Abd al-Ḥusayn Aḥmad al-Amīnī in his al-Ghadīr fī 'l-kitāb wa 'l-sunna wa 'l-adab, Tehran, 1372/1952–1953.Google Scholar
28 Al-Nu'mānī, , op. cit., 33.Google Scholar
29 ibid., 48 f.; Bābawayhi, Ibn, Ikmāl al-dīn, 149–67Google Scholar; idem, Kitāb al-khiṣāl, 436–45Google Scholar; in general BiḥĀr, IX, 128 ff.Google Scholar For contemporary Shī'ī works consult, e.g., ‘al-Ḥā'irī, Alī Yazdī, Ilzām al-nāṣib fī ithbāt ḥujjat al-ghā'ib, Tehran, 1351/1932–1933, 75 ff.Google Scholar; al-Ghiṭā’, Muḥammad al-Ḥusayn Āl Kāshif, Aṣl al-shī'a wa-uṣūluhā, Najaf, 1369/1950, 99.Google Scholar
30 Al-Khazzāz, , op. cit., 293 f.Google Scholar, cited in Biḥār, IX, 141–4.Google Scholar
31 Al-Khazzāz, , op. cit., 294–7Google Scholar, whence Biḥār, IX, 145.Google Scholar
32 Al-Khazzāz, , op. cit., 305.Google Scholar
33 ibid., 298 ff. Al-Majlisī criticizes al-Khazzāz, for ‘mixing Imāmi traditions with those of the opponents’Google Scholar, and declares that in the Biḥār only reliable traditions are quoted (Biḥār, I, 12).Google Scholar
34 Al-Nu'mānī, , op. cit., 49.Google Scholar In Shī'ī traditions the Imams are often referred to as khulafā', or khulafā' allāh fī arḍihi. See, e.g., al-Kulīnī, , op. cit., I, 193 f.Google Scholar
35 See Rosenthal, F., ‘The influence of the Biblical tradition on Muslim historiography’, in Lewis, B. and Holt, P. M. (ed.), Historians of the Middle East, London, 1962, 35–45Google Scholar; Kister, M. J., ‘Ḥaddithū 'an banī isrā'īla wa-lā ḥaraja’, Israel Oriental Studies, II, 1972, 215–39.Google Scholar
36 For some examples see Kister, , art. cit., 222 f., 232, 233.Google Scholar
37 In various traditions (usually on the authority of Ka'b al-Aḥbār), the Prophet's name in the ancient Scriptures (or in the Torah) is said to have been ‘Mādh Mādh’, meaning ‘good, good’ (al-Qāḍī, ‘Iyāḍ, al-Shifā’ bi-ta'rīf ḥuqūq al-muṣṭafā, Cairo, 1369/1950, I, 148Google Scholar; al-Nuwayrī, , Nihāyat al-arab, XVI, Cairo, 1374/1955, 79)Google Scholar, or ‘Mūdh Mūdh’ (al-Bājūrī, Ibrāhīm, al-Mawāhib al-laduniyya 'alā 'l-shamā'il al-muḥammadiyya, Cairo, 1301/1883–1884, 213)Google Scholar, or ‘al-Ḥādd’ (Biḥār, VIGoogle Scholar (unpaginated)), or ‘Mād Mād’ (al-Ḥā'irī, , op. cit., 38, 45).Google Scholar Most of these forms derive from the Hebrew me'ōd me'ōd (Gen. xvii, 2, 6, 20).Google Scholar It is claimed that the letters constituting the name ‘Mād Mād’ have a combined numerical value of 92 (this would be true if the alif were doubled), and that this is also the combined numerical value of the word ‘Muḥmmad’ (al-Ḥā'irī, , op. cit., 38).Google Scholar
38 See Gen. XXV, 13–16Google Scholar; of. 1 Chron. i, 29–31.Google Scholar
39 Al-Nu'mānī, , op. cit., 49 f.Google Scholar The names of Ishmael's sons as they appear in this story attest to a considerable corruption of the original text. Thus Nebaioth is rendered ‘Baqūnīth’ (or ‘Bāqūbīth’), Qēdār becomes ‘Qadū’ (‘Qaydawū’ ?), Adb'el is ‘Ra'īn’ (or ‘Dabīrā’), etc. The corruption is somewhat less marked in a different tradition, on the authority of Ka'b al-Aḥbär, copied in the Biḥār (IX, 127)Google Scholar from the Mugtatḍab al-athar of Ibn ‘Ayyāsh. Muslim authors in general seem to have been uncertain as to the correct form of the names of Ishmael's sons. Thus al-Ṭabarī (d. 310/923) points to the discrepancy between Ibn Isḥāq's version and that of other sources. See his Tārīkh, ed. de Goeje, M. J. and others, Leiden, 1879–1901, Prima Series, I, 351 f.Google Scholar The tradition about the 12 sons of Ishmael is quoted already in the Tafsīr of Ismā'īl b. ‘Abd al-Raḥmān al-Suddi (d. 128/745). See al-Muẓaffar, , Dalā'il al-ṣidq, II, Najaf, 1372/1953, 314Google Scholar; al-Ṭabarsī, al-Nūrī, Kashf dl-astār ‘an wajh al-ghā'ib ‘an al-abṣār, sine loco, 1318/1900–1901, 106.Google Scholar
40 Al-Nu'mānī, , op. cit., 50.Google Scholar
41 ibid. The Arabic is preceded by a badly corrupted transliteration of the Hebrew original.
42 ibid.
43 Bi;ḥār, IX, 127Google Scholar, quoting from Ibn ‘Ayyāsh's Muqtadab al-athar.
44 Al-Nu'mānī, , op. cit., 51 f.Google Scholar
45 ibid., 54. Cf. al-Kulīnī, , op. cit., I, 629 f., 531 f.Google Scholar
46 Al-Nu'mānī, , op. cit., 90Google Scholar, cited in Bī;ḥār, XIII, 142.Google Scholar
47 ibid. Cf. the somewhat vague formulation in a tradition of the Prophet: ‘He [i.e. the Qā'im] will undergo two ghaybas, one of which will be longer than the other’ (al-Khazzāz, , op. cit., 307).Google Scholar
48 This is said by al-Majlisī to refer to his appearance before his closest associates (khawā;ṣ;ṣ mawālīhi wa-sufarā'ihi), or to the fact that news about him will reach the people (Bi;ḥār, XIII, 143).Google Scholar
49 Al-Nu'mānī, , op. cit., 91.Google Scholar
50 See e.g. al-Faḍl b. al-Ḥasan al-Ṭabarsī, I'lām al-warā fī a'lām al-hudā, ed. al-Khursān, M. Mahdī al-Sayyid Ḥasan, Najaf, 1390/1970, 445Google Scholar; Biḥār, XIII, 142.Google Scholar
51 al-Murtaḍā, Al-Sharīf, Tanzīh al-anbiyā’, Najaf, 1380/1961, 228.Google Scholar
52 ibid., 233f.
53 See Bābawayhi, Ibn, Ikmāl al-dīn, 82 f.Google Scholar
54 Often identified as the Shi'b Abī Yūsuf. See Yāqūt, , Mu'jam al-buldān, III, Beirut, 1376/1957, 347.Google Scholar
55 See ‘al-Suhaylī, Abd al-Raḥmān, al-Rawḍ al-unuf fī sharḥ al-sīra al-nabawiyya, ed. ‘al-Wakīl, Abd al-Raḥmān, III, Cairo, 1389/1969, 354.Google Scholar
56 See al-Ṭūsī, , Kitāb al-ghayba, 61–3.Google Scholar See also Sa'īd b. Hibat Allāh al-Rāwandī, al-Kharā'ij wa 'l-jarā'iḥ, Bombay, 1301/1883–1884, 162.Google Scholar
57 cf. above, p. 525, n. 21.
58 See e.g. al-Ya'qūbī, , Tārīkh, Najaf, 1358/1939–1940, III, 40 f.Google Scholar, whence al-Shaybī, , al-Fikr al-shī'ī wa 'l-naza'āt al-ṣūfiyya, Baghdād, 1386/1966, 25.Google Scholar
59 Al-Nawbakhtī, , op. cit., 34.Google Scholar
60 Cited by Ṭāwūs, Ibn, al-Malāḥim wa 'l-fitan, Najaf, 1383/1963, 26, 147.Google Scholar
61 See Wensinck, A. J. and Mensing, J. P. (ed.), Concordance et indices de la tradition musulmane, Leiden, 1936–1964, I, 306, s.v. thny.Google Scholar
62 Al-Qasṭallānī, , Irshād al-sārī li-sharḥ ṣaḥīḥ al-Bukhārī, x, Būlāq, 1327/1909, 273.Google Scholar In practical terms, the only difference between (i) and (iii) lies in the replacement of Yazīd and Hishām, two of ‘Abd al-Malik's sons, by al-Ḥasan b. ‘Alī and ‘Abdallāh b. al-Zubayr.
63 On the margin of al-Qasṭallānī, 's Irshād, VIII, Bulaq, 1326/1908, 5–7.Google Scholar
64 Perhaps including al-Ḥasan b. ‘Alī.
65 Al-Muẓaffar, , op. cit., II, 314 f.Google Scholar Al-Muẓaffar (ibid., 315–18) rejects this and other interpretations given by al-Faḍl b. Rūzbihān. See also al-Ṭabarsī, al-Nūrī, op. cit., 94 ff.Google Scholar
66 A good example of such an anti-Umayyad, pro-Quraysh attitude is provided by al-Maqrīzī, 's al-Nizā' wa 'l-takhāṣum fīmā bayna banī Umayya wa-banī HāShimGoogle Scholar (passim). In a Shī'ī tradition of a somewhat different character, the Prophet declared that after his death the community would come under the rule of 12 ‘erring Imāms (imām ḍalāla)’, two of them from Quraysh (referring probably to Abū Bakr and ‘Umar) and 10 from the Banū Umayya. See , Aḥmad B. ‘Alī al-Ṭabarsī, Kitāb al-īḥtijāj, II, Najaf, 1386/1966–1967, 4.Google Scholar
67 cf. Wensinck, and Mensing, (ed.), op. cit., VII, 83, s.v. hrj.Google Scholar
68 See the recent discussion by van Ess, J., op. cit., 28 ff.Google Scholar; idem, Das Kitāb an-Nakṯ des Naẓẓām und seine Rezeption im Kitāb al-Futyā des Ğāḥiẓ, Göttingen, 1972, 52 ff.Google Scholar; idem, ‘Das Kitāb al-irğā’ des Ḥasan b. Muḥammad b. al-Ḥanafiyya’, Arabica, XXI, 1, 1974, 32 ff.Google Scholar
69 See al-Ṭihrānī, Āghā Buzurg, al-Dharī'a ilā taṣānīf al-shī'a, Najaf, 1355–7/1936–1938, Tehran, 1360/1941 ff., XVI, p. 74, no. 371.Google Scholar
70 ibid., XVI, p. 76, no. 382.
71 ibid., XVI, p. 78, no. 395.
72 ibid., XVI, p. 76, no. 382. According to al-Kishshī's Rijāl, however (ed. Aḥmad al-Ḥusaynī, Najaf, c. 1964, 344–6), the wāqifī was al-Ḥasan's father, ‘Alī al-Baṭā'inī, who believed that al-Riḍā was the last Imām.
73 See al-Nawbakhtī, , op. cit., 79 f.Google Scholar; , Sa'd B. ‘Abdallāh, op. cit., 106 f.Google Scholar
74 ibid.
75 cf. al-Ṭihrānī, Āghā Buzurg, op. cit., XXI, p. 69Google Scholar, no. 3995.
76 See al-Ṭabarsī, , l'iām al-warā, 443 f.Google Scholar, cited in Biḥār, XIII, 99 f.Google Scholar
77 Al-Nawbakhtī, , op. cit., 68Google Scholar; , Sa'd B. ‘Abdallāh, op. cit., 90.Google Scholar
78 In. an. Imāmī tradition, Ja'far al-Ṣādiq says explicitly that ai-Kāẓim will disappear twice (inna li-Abi 'l-Ḥasan ghaybatayn), so that some men will claim that he has died. In fact, says al-Ṣādiq, Mūsā al-Kāẓim. will not die until he appoints a legatee (al-Ṭūsī, , op. cit., 38).Google Scholar
79 Note the complete title of Ibn. Bābawayhi's work: Kitāb ikmāl al-dīn wa-itmām al-ni'ma fī ithbāt al-ghayba wa-kashf al-ḥayra.
80 ibid., 13. See also al-Ṭabarsī, , op. cit., 443Google Scholar; and cf. the introduction to al-Barqī, 's Kitāb al-maḥāsinGoogle Scholar by Jalāl al-Dīn al-Muḥaddith, p. kāf-alif.
81 Al-Nu'mānī, , op. cit., 47.Google Scholar
82 Kitāb Abī Sa'īd ‘Abbād al-'Uṣfurī, in al-Uṣūl al-arba'u mi'a, fol. 10a.
83 Kitāb Muḥammad b. al-Miithannā al-Ḥaḍramī, in al-Uṣūl al-arba'u mi'a, fol. 53b. As it stands, this sounds like an Ismā'īlī tradition. The Ithnā-'asharī version of this and similar sayings is that Ja'far al-Ṣādiq is the first of the seven last Imāms. Cf. al-Ṭūsī, , op. cit., 36.Google Scholar
84 See al-Ḥillī, Ibn al-Muṭahhar, Rijāl, ed. al-'Ulūm, Muḥammad Ṣādiq Baḥr, Najaf, 1381/1961, 83Google Scholar; Goldziher, , Muhammedanische Studien, Halle, 1889–1890, II, 10 f.Google Scholar
86 al-Ḥimyarī, Al-Sayyid, Dīwān, ed. Shakar, Shākir Hādī, Beirut, 1966, 355–69.Google Scholar
86 See Sourdel, D., ‘La politique religieuse des successeurs d'al-Mutawakkil’, SI, XIII, 1960, 12 ff.Google Scholar
87 See the discussion in Watt, W. Montgomery, art. cit., 119–21Google Scholar; Cahen, C., ‘Buwayhids’, in EIGoogle Scholar, second ed., II (in particular pp. 1350–2); idem, ‘La changeante portée sociale de quelques doctrines religieuses’, L'élaboration de l'Islam. Colloque de Strasbourg, 12–14 juin 1959, Paris, 1961, 16.Google Scholar
88 cf. Laoust, H., ‘La pensée et l'action politiques d'al-Māwardī (364–450/974–1058)’, REI, XXXVI, 1, 1968, 43 ff.Google Scholar; idem, ‘Les agitations religieuses à Baghdād aux IVe et ve siècles de l'Hégire’, in Richards, D. S. (ed.), Islamic civilisation 950–1150, Oxford, Cassirer, 1973, 169–85.Google Scholar
89 cf. Massignon, L., ‘Recherches sur les Shī'ites extrémistes à Bagdad à la fin du troisième siècle de l'Hégire’, ZDMG, XCII, 1938, 378–82Google Scholar; Cahen, loc. cit.